question
stringlengths 19
6.88k
| answer
stringlengths 38
33.3k
|
|---|---|
Problem Statement: no licenses available when trying to start a controller.
Asplmd.log reports that all of the available licenses have been assigned. | Solution: Stop the License Manager, go to the License Manager Control program,
Start / Programs / License Manager 2.2 / License Server Control Program, stop the service, go to settings and reset the Start time to 0.
Start the License Manager and then re-start the controllers. The re-start of the controllers will free up currently assigned license keys and make them available for re-assignment to the re-started controllers.
Keywords:
References: None |
Problem Statement: How do I configure actuator fail open, fail shut and fail hold options in a control valve? | Solution: The valve actual fail options are available in the Actuator page under the Dynamics tab as shown below.
In the event that the signal from the controller is cut off from the valve, the valve becomes wide open. In Aspen HYSYS, when the Fail Open radio button is selected, the valve modifies the signal received as
ActDesired % = 100% - ActDesired %(from controller)
If ActDesired% (from controller) becomes zero in the event of a signal failure, the actuator becomes fully open.
If the Fail Open mode is selected for the valve, Reverse Acting controllers need to be set as Direct Acting and Direct Acting controllers need to be set as Reverse Acting.
For Fail Shut option, HYSYS does not modify the signal from the controller as with the Fail Open mode. If the signal from the controller becomes zero, so can the ActDesired% value. Since the valve does not modify the signal from the controller, the controller direction does not have to be changed.
For Fail Hold mode, in the event that the signal from the controller is cut off from the valve, the valve fails in the position it is currently in.
Keywords: Fail Open, Fail Shut, Fail Hold
References: None |
Problem Statement: Converting a DMC application to a DMCplus application. | Solution: This is a utility to help in converting a DMC application to a DMCplus application.
Basically... get a PRT file from the DMC application and with the mdl file, use Build to create a DMCplus CCF file. In Build set the correct number of Feedforward variables. Run the PRTUPDATE and enter the name of the PRT file and the name of the CCF file. PRTUPDATE updates the CCF file with tuning and limits from the PRT file.
Go back to build and finish configuring the controller and verify the tuning.
You then have a head start in converting the DMC application to a DMCplus application.
Keywords: DMC, DMCplus, migrating, convert
References: None |
Problem Statement: A corrupt bin file that has a bad creation time in the header will not extract. It would generate an access violation when it tries to generate the begin time for the extraction. This fix bypasses the offending code and prompts for the begin time.....The new extract will be delivered with AMS 70 . | Solution: Download the Extract70.zip file attached to thisSolution.
Move to the extract.exe to the aconline/bin directory where the current extract.exe file resides. You may want to rename the current extract.exe to something like extractold.exe to that it can be used in the event the new extract.exe has any problems.
Keywords:
References: None |
Problem Statement: Not all processes appear to be starting as noted in installation documentation. | Solution: The Autostartlist.txt gets updated as Advanced Control software is installed. If this does not occur properly, the result can be that some required processes are not started. If modifications are required to this file, you will have to perform an acobase_shutdown and then and acobase_startup to invoke the changes.
The autostartlist.txt file resides in program files\aspentech\ac online\sys\etc and all possible entries follow. You may not have some of the products installed on your computer. Cut and paste the appropriate sections from thisSolution into your autostartlist.txt file to start the corresponding processes. The section starting the DMCPLUSCONTEXTSERVICE may not be needed as with newer version of DMCplus this service is started automatically, from Services in Control Panel.
PROCNAME=DMCPLUSMANAGE
SEQNUM=1
STDIN=null
STDOUT=null
STDERR=null
APPPATH=$ACODHOME\bin\manage.exe
ARGLIST=autostart
APPTYPE=3
PROCNAME=SMARTSTEPMANAGE
SEQNUM=2
STDIN=null
STDOUT=null
STDERR=null
APPPATH=$ACODHOME\bin\smartstep_manage.exe
ARGLIST=autostart
APPTYPE=3
PROCNAME=DMCPLUSCONTEXTSERVICE
SEQNUM=10
STDIN=null
STDOUT=null
STDERR=null
APPPATH=NULL
APPTYPE=2
PROCNAME=DMCP_VIEWSRV
SEQNUM=20
APPPATH=$ACODHOME\bin\dmcp_viewsrv.exe
APPTYPE=1
PROCNAME=APOLLO_VIEWSRV
SEQNUM=21
APPPATH=$ACODHOME\bin\apollo_viewsrv.exe
APPTYPE=1
PROCNAME=SMARTSTEP_VIEWSRV
SEQNUM=22
APPPATH=$ACODHOME\bin\smartstep_viewsrv.exe
APPTYPE=1
PROCNAME=BGCSNET
SEQNUM=79
APPPATH=$ACODHOME\bin\bgcsnet.exe
ARGLIST=gcsdmcp C:\Program Files\AspenTech\AC Online\bin\dmcpview.exe APPTYPE=0
PROCNAME=IQLINKSRV
SEQNUM=80
APPPATH=$ACODHOME\bin\iqlinksrv.exe
APPTYPE=1
PROCNAME=ACOMSGSRV
SEQNUM=90
APPPATH=$ACODHOME\bin\acomsgsrv
APPTYPE=1
PROCNAME=DMCPLUSCOMPOSITE
SEQNUM=250
APPPATH=$ACODHOME\bin\composite_manage.exe
ARGLIST=autostart
APPTYPE=3
Keywords: autostart, bgcsnet, processes
References: None |
Problem Statement: Unable to load controller because of communication failure, but all tests, e.g. cimio_t_api, ping, nslookup, examining logical_devices.def file, services file, etc. seems to indicate that the link is healthy. | Solution: Many times, a CCF or collection will reference a non-default device to collect tag information, which is not accessible by the default server.
When DMCplus validates tags in the controller, it tries to validate these tags from the non-default server. If there is a problem with that server, it generates a communication failure, in spite of the fact that the default server is functioning correctly.
Examine the CCF or collect file for any references to an alternative device, e.g. a device explicitly defined in the tag string, this is a references to a non-default io device. Check the status of the device and correct problems if they exist.
Keywords:
References: None |
Problem Statement: [PROBLEM]
The CM50S CIM-IO Server v3.2.3 doesn't behave correctly under CIM-IO Kernel 4.8.
[IMPACT]
This problem is most evident with IP.21, IP-X, SetCIM, and RT-OPT. The problem is not likely to occur with the various DMCplus
clients (View, Collect, Control.)
[SYMPTOM]
When an IoGetDef record is inactivated, as STOPGET message is sent to the CM50S Server. When the IoGetDef record is
reactivated, the next CIM-IO GET request fails with a timeout error. (At that point, the TSK_M_CM50S must be restarted to reconnect to the server.)
The RT-OPT GUI tool will simply fail to access the server.
[ | Solution: ]
Upgrade to Cimio 4.8 engineering release.
[WORKAROUNDS]
The workaround is to install and run the CM50S Server v3.2.3 using CIM-IO Kernel 4.7.1.
[NOTES]
The client can use either Kernel -- 4.7.1 or 4.8. The critical point is that the CM50S server will only function properly using 4.7.1, and will work with that version if the client's CIM-IO Kernel is 4.6B, 4.7.1, or 4.8.
Solution
In order to remove this issue as the source of the problem,
you can use Kernel 4.7.1.
It is recommend, installing in a new directory tree. The steps would be:
1. shutdown the server
2. rename the CIMIOROOT directory to something else (like [.CIMIO_OLD48]
3. deas/group CIMIODEF deas/group CIMIOETC deas/group CIMIOINC deas/group CIMIOCOM deas/group CIMIOLOG deas/group CIMIOCODE deas/group CIMIOROOT
4. Install kernel 4.7.1 in an empty directory (e.g., [.CIMIO471]
5. Install CM50S CIM-IO Server
6. compare the [.CIMIO_OLD48.ETC]CIMIO_CM50S.CFG with the newly installed CIMIOETC:CIMIO_CM50S.CFG and make the appropriate customizations.
7. Change the ownership of the files and directories in the CIMIOROOT tree to be owned by the account used to start and stop the CM50S Server.
8. Start the server and make sure all the processes are running.
Note: The CIM-IO Kernel developers are working on aSolution to the problem. There is an engineering release for 4.8 being developed, but I don't believe it is available yet.
Keywords: CM50s, interface, vax, tdc3000
References: None |
Problem Statement: When DMCplus is turned off at the PCWS, why do the ET's Stay on and RTO stay running? | Solution: Client has two switches on the Production Control Web Server, PCWS, one for ET's and one for DMCplus.
It was observed that when DMCplus is turned off, the ET's stay on , as does the RTO application and the client would like to know why RTO would be on if DMCplus is off?
Development confirmed that DMCplus and RTO are independent. Turning off the DMCplus will not effect the RTO in terms of it running or not.
The ET's are left on, when DMCplus is turned off, but they are not being recognized. They are left on, so that they are available when DMCplus is restarted. Also, the PCWS or DMCplus does not change the ET's. The ET's are only changed by the operator. If an ET causes a problem, an error message is displayed, but the ET itself is never changed automatically.
Keywords: RTO, Watch, DMCplus, External Targets, PCWS
References: None |
Problem Statement: CIM-IO IBM ACS, LOOPST, AWSCOD | Solution: In order to read LOOPST, the source type CBST must be used. CBST is a mnemonic meaning Control Block Status and corresponds to Loop Status (LOOPST).
DBST should not be used for LOOPST. DBST is a mnemonic meaning Data Block Status or Database Status and it corresponds to the data quality of the tag.
The tagname which can be used with CBST and with CBAW (anti-windup) is critical. It must represent a FVV (Full Value Variable) and must be either the setpoint (40) item or output (80) item of that FVV.
For example, the tags DXA001.40 and HIC002.80 are valid for use with CBST and CBAW, but the tags DXA001.10 and HIC002.10 are not.
Using CBST with the tag DXA001.10 is an error and will result in the error message The device data type does not match the ACS (we are using IBM-ACS system) itemtype that you received.
To fix the problem specify item 40 (if controlling the setpoint) or item 80 (if controlling the output.)
Keywords: DMCplus, CIM-IO, CBST, LOOPST
References: None |
Problem Statement: How do I create a strip chart for a variable for all the streams in the simulation? | Solution: In the newer Aspen HYSYS versions (from V8.2) the user can add a variable to a strip chart directly from the stream or for all the streams in the simulation from the Workbook.
For a stream, the user can select the variable, apply right mouse click and then send the variable to strip chart. If the strip chart does not exist then this also offers to create a new chart.
If the user wishes to create a strip chart for a variable for all the streams in the simulation then this can be achieved from the workbook. Please follow the steps below:
1. Open the workbook from the Navigation Pane on the left.
2. You can uncheck the Horizontal Matrix option to display stream in the column view.
3. Highlight a column for the variable to be added in the strip chart.
4. Apply right mouse click to any value in the column.
5. Then select Send to Strip Chart. You can create a new chart or send to an existing chart.
The above steps are highlighted in the screenshot given below.
Keywords: Strip Chart, Stream Variable
References: None |
Problem Statement: How to solve Failed to connect to database error while updating tank inventory | Solution: While trying to update the Tank Inventory, the following error was encountered.
This error occurs mainly due to the huge size of the database. To solve this error, please follow the below mentioned steps.
1) Open your database file.
2) Go to Table “ORION_MGR_CRDINV_IMPORT”
3) Delete all old data, which might not be of use now.
4) Re import again.
Keywords: ORION_MGR_CRDINV_IMPORT
References: None |
Problem Statement: Range Error - CLE Error | Solution: This error is created because the SP, PV and CV limits on the intermediate point were much narrower than on the control point. This is because they are set based on the CCF and the engineer may have made them much tighter than on the original point. Detail both points and then change the range on the intermediate point to mach those on the control point (They cannot set the SPHILM and SPLOLM since they are set by the CL)....Once this is done, the range error should clear...This same thing can happen if the engineer changes the Engineering limits in the CCF and doesn't adjust the range limits on the intermediate point.
Keywords:
References: None |
Problem Statement: What does the Warning Message: Integration time: For the Breakdown option of the EMotor, a final maximum speed value for the speed vs torque curve must be entered which is equal to the full load speed percent of synchronous speed mean? | Solution: According to this message, when the breakdown option for electric motor is selected, you will have to make sure the largest (last) curve speed value entered is less than the full load speed value.
From the Speed vs Torque curve, you can see the maximum x value under: Compressor | Rating | Electric Motor | Speed x Torque Curve.
Keywords: Breakdown, Electric Motor, Compressor
References: None |
Problem Statement: This article describes steady-state trajectory control strategies. | Solution: Although not recommended, Aspen DMCplus external targets--normally a single, steady state target--can be formulated as a desired trajectory through custom calculations or an external program. However, there are several things to consider before implementing such a strategy.
In review, at each execution of the controller, it:
1. Looks at where the process is.
2. Decides where it should be (external targets come into play during this step).
3. Formulates a move plan to ensure safely getting there.
4. Implements the first move of the plan.
5. Goes to sleep
If the external target changes every execution, the move plan at any one execution won't know that the target will change in the future, so it may not be the optimum move plan overall. Also, the controller tuning is largely based on deltas, so you may need to change the tuning to have Aspen DMCplus control a trajectory rather than to a steady state target.
As an alternative, you should consider using Aspen Nonlinear Controller. It has the built-in ability to safely control to a trajectory, and the process doesn't have to be nonlinear in order to use the software.
Keywords: ET, External Target
IRV, Ideal Resting Value
LP, Linear Program
QP, Quadratic Program
SS, Steady State
References: None |
Problem Statement: What is a material pool and how is it used? | Solution: Use Materials Pools to allow for tank pooling.
Once defined, material pools can be added to the Trend List Chart to view and track aggregate information.
For example , if the user Prefer to create a Material Pool, for Diesel group of Tanks, one can add all those Tanks under Material Pool and the procedure is as follows:
To create a new material:
1. From the Event interface, click Model | Materials to display the Material Pools dialog box.
2. Click New. The New Material Pool dialog box appears.
3. In the ID field, enter the tag you want to assign to the material pool.
4. In the Description field, enter a description of the pool. You can also enter or modify a description later from the Material Pools dialog box.
5. If you wish to base this new pool on an existing pool, do the following otherwise skip to step 6:
a. Select Copy from. The drop-down list with current defined pools becomes enabled.
b. Select the pool you wish to copy information from. The details associated with the existing pool will be copied to the new pool
6. Click OK to return to the Material Pools dialog box.
7. Select a Type:
If Type is PROD_SERVICE or FEED_SERVICE, select a Basis ID that indicates the material that the pool is an aggregate of. If Type is CRUDE_UNIT, select the unit name that indicates the unit feed that will be aggregated by this pool.
-Or-
If Type is PROD_TANK_N_STRM, FEED_TANK or CRUDE_TYPE, a grid appears where you can define the finite list of tanks (process/crude) or crudes associated with Type. Repeat this step until all components are added.
8. Optionally enter Minimum and Maximum limits that are applied to the pool. If you do not enter these values, limits are automatically calculated from the component limits.
9. Optionally enter Minimum Working and Maximum Working limits that are applied to the pool. If you do not enter these values, limits are automatically calculated from the component limits.
10. Click OK to save your changes in memory and exit the dialog box. Your changes will not appear in the database until you save your model by clicking on the toolbar.
Keywords: None
References: None |
Problem Statement: Installation of Advanced Control Application could not be completed because of the following message found in the asplm.mif file, Could not create special folder 25. This is a microsoft issue. | Solution: Use the patch attached to thisSolution, unzip the file and run the registry file and the missing folder will be created.
Keywords: asplm, installation
References: None |
Problem Statement: How do you avoid dynamic compressor singularity issue in pressure flow matrix? | Solution: In order to solve a compressor in dynamics, generally two specifications should be selected in the Dynamics Specifications group in the Specs page of the Dynamics tab. Attention should be given while selecting specifications which causes complications or singularity in the pressure flow matrix.
Please follow the following guideline to avoid singularity issues:
1. The Pressure rise checkbox shouldn’t be checked if the inlet and exit stream pressures are specified under “Woksheet”.
2. The Speed checkbox shouldn’t be checked if the Use Characteristic Curves checkbox is cleared.
Key Words
Compressor, Singularity, Speed, Pressure
Keywords: None
References: None |
Problem Statement: How can I Create a Memory Dump for 32bit AspenTech Applications? | Solution: For Orion, MBO and Olefins scheduler, sometimes you may see a dialog showing the reason why the application is crashing.
In order to get more information for the crash, a dump file may be required. Since, APS, MBO and Olefins scheduler are 32-bit applications, a 32-bit dump file may be useful for analyzing the issue. Following steps can be followed in order to generate a .dmp file.
1. User will need to start Task Manager at C:\Windows\SysWOW64\taskmgr.exe. This is a 32bit program better for dumping our 32Bit applications. Do not use the option to open taskmgr from Taskbar or Ctrl+Alt+Del.
2. Right-mouse-click your application and then Create dump file.
3. Send the dump file to Aspen Tech to analyze the crash.
Keywords: .dmp file
32bit application
References: None |
Problem Statement: An application is crashing and not leaving enough troubleshooting information. | Solution: Debugging applications using the Microsoft Debugger
To debug failures where there is insufficient information in the output files, or returned messages, or the program simply aborts without leaving any information, but the failure is not reproducible by AspenTech developers or testers, clients can use the ?Microsoft Debugger? to produce a ?dump? file that might have information necessary for developers to diagnose the problem..
The debugger must be downloaded from http://www.microsoft.com/whdc/devtools/debugging/installx86.mspx
and installed on the client's system where the failure is occurring.
In order to utilize this mechanism, a special version of the executable for the failing program must be created by development or support personnel. The specific failing executable must be, or have been, created with ?symbols?. There is a list at the bottom of the page of the executables and version numbers that have been created with symbols, and can be provided by support personnel after acquiring them from development. Other applications will need to be addressed as the need arises.
The client can be instructed to do the following. On the machine where the error is occurring;
1. Download the latest version of the Microsoft debugger and install this in c:\debuggers (or D:\debuggers). Overriding the original path will simplify later steps.
2. Stop all instances of the executable to be debugged and rename the existing executable file to something else so we can restore it later. Unzip the new executable that is built with symbols into the same directory as the original.
3. Start the new copy of the executable
4. open a command prompt and cd to c:\debuggers
5. Run one of the following commands, depending on circumstances:
Each command will open another cmd window that must remain open, or the debug process will exit.
[NOTE: that in each command, <executable name> is the name as it appears in Windows Explorer, e.g. manage.exe]
(The following command cannot be run from Terminal Services)
ADplus -crash -pn <executable name> -o c:\dumps
(If terminal services must be used, then change the command to)
ADplus -crash -NoTsCheck -pn <executable name> -o c:\dumps
(If the executable is used to generate several processes, and only one is troubling, e.g. control.exe for a specific DMCplus controller, then specify the Process ID instead of the executable name. Get the PID from Windows Task Manager. Obviously you must be able to identify the appropriate process. Note -pn is changed to -p
ADplus -crash -p <Process ID> -o c:\dumps
The following example starting a debug process for manage.exe on a machine where the debuggers and dump files are on D:.
D:\Debuggers>adplus -crash -pn manage.exe -o d:\dumps
Attaching the debugger to: MANAGE.EXE
(Process ID: 4544)
This will start up a process that attaches to the manage.exe that will create a dump file on the crash. Remember, the command window can be minimized, but it should not be closed or the attached process and the manage.exe will close.
The dumps directory should be placed on a disk with gigabytes of free space. When the manage crashes, the adplus will create a series of files in the dump folder that will need to zipped up and sent to support. It may be necessary to put the file(s) on the FTP site (ftp.aspentech.com) because of the size.
The behavior of the command ?adplus? is that on a crash, it will immediately produce a crash dump, and release the program, unlike other versions which will produce a dialog and wait for the user to respond, effectively preventing the process from being restarted, either manually, or with a ?restart? capability until the user answers the prompt. For example the Context Server can be configured in the Services applet to ?restart? when it fails. If it is so configured, it will produce the crash dump, and then restart.
These are currently instrumented versions for:
Executable
AMS 6.0.1
2004/2004.1
2004.2
2006
Acoviewdp.exe (the PCWS data provider)
168
134
122
Cimio_dmcp_dlgp.exe (the context server)
139
196
137
Dmcp_viewsrv.exe
137
Control.exe
139
196
104
137
Manage.exe
139
196
137
The numbers in the boxes where a version and executable come together is not what appears in the ?Properties/Version tab? but represents the released build that most is most closely associated with the ?debug? version.
Keywords:
References: None |
Problem Statement: Is there a document available that specifically details to how to configure the App node for use with DMCplus? The focus should be based on the needs of DMCplus and the capabilities of the App node. | Solution: Yes there is a document available. The example used in the document details a configuration of a Honeywell App node. Please down load the attachment to thisSolution titled: Honeywell App Node.doc.
Also seeSolution ID 103189 - DMCplus -Tips and Tricks for the Honeywell App Node
Keywords: appnode, cache, TPN, opc
References: None |
Problem Statement: On Microsoft Operating systems, you can start Collect from the command prompt by specifying the command collect and the correct parameters (path to, and name of collection input file).
As an example let's use the collection input file that comes with the demo controller col5x3. Put the input file in a directory called test on your c: drive. The path will be:
C:\test\col5x3
You could issue the command:
%acodhome%\bin\collect c:\test\col5x3
in a DOS window and start the collect process, but you would not be able to close the DOS window, or logoff that user without causing the collection to stop. | Solution: Solution 102680 gives instructions how to utilize the Schedular in a Windows NT system. The technique only works for NT. A similar set of steps can no doubt be devised to utilize the schedular for other Microsoft platforms.
A more generalized alternative is to start the collection task as a background process using mpf_manage, and should work with any operating system.
It is important to know that there are string buffers utilized in the mpf_manage utility that place some restrictions on the length of the parameter strings you can pass to an executable. This can be accommodated by making a copy of the collect.exe file in the test directory mentioned in the problem statement (e.g. c:\test)
Now you can issue the command:
mpf_manage appstart collect c:\test\collect c:\test\col5x3
where:
mpf_manage - is the name of the utility you are invoking
appstart - is the first argument for mpf_manage which tells that you want to start an application.
collect - is a character string you are going to use to identify the process
c:\test\collect - is the path to the collect executable
c:\test\col5x3 - is the path to the input file.
Now that you have the collection process running, the next task will be to stop it when you have decided the collection is finished. Usually, the Windows Task Manager is not able to stop it. There are some utilities that are able to stop it.
NOTE: Collect has always been stopped abruptly by something similar to a kill command. This hasn't presented any problems we are aware of, though in theory, if the kill command comes precisely when the application is writing the data to disk, the bin file could be corrupted. To guard against this, monitor the Windows Task Manager Performance tab and kill the process in between the CPU usage peaks which indicate the collection is taking place.
1. There is a web site www.sysinternals.com, and on that web site are many useful utilities, and some have the capability of killing any process including the collect process when it is running in background.
a. Process Explorer will show a representation of the executables you have running on your system as well as some useful information about them. It will also allow you to kill any process you have running.
b. PSKill is another utility from Sysintermals, and it simply allows you to kill any process.
2. If you are utilizing the Honeywell App Node, they have a privileged task manager. It is called PrivTaskMgr. The user should determine for himself if there will be a violation of any Honeywell agreements if the priviledged task manager is copied to another computer within the user's network. If it is determined that it is OK with Honeywell, then there are two files that must be copied to the computer where you wish to stop the collect process; privtaskmgr.exe, and privtaskmgr_rc.dll. Simply navigate to the executable and double click on it - AFTER stopping the regular task manager. Once you have killed the process it is a good idea to switch back to the normal task manager.
Keywords: Collect
Background
References: None |
Problem Statement: What is the difference between Derived and Complex properties in UBML? | Solution: Derived
Derived properties are calculated from other blend properties (Non-component properties). For example, road octane is calculated as the average of RON and MON. EPA and CARB emissions are also derived properties.
Complex
Complex properties are calculated from multiple component properties. For example, to calculate distillation properties (IBPD, T10D, T50D, T90D, and FBPD) the distillation properties for each of the blend components must be provided.
Keywords: Complex properties
Derived properties
UBML
References: None |
Problem Statement: Where Can information / documentation be found for the CCF Switcher | Solution: Locating the CCFSwitcher documentation.
In windows explorer, navigate to where the DMCplus Desktop applications are installed.
You should find a Utilities folder.
In the Utilities folder, you will find a DMCplus Utilities Help file.
Double clicking on this file will invoke the online help facility for the DMCplus Utilities and you will find a book on the CCFSwitcher.
Keywords: DMCplus, CCF Switcher, utiilty, doc
References: None |
Problem Statement: AMS6.01 ER4 added the ability to turn off a final output check just prior to writing
MV's to the DCS by declaring a user defined variable, DLSTCHK=1.
For AMS6.01 DLSTCHK must to be defined in the CCF's General section.
For AspenOne 2004.x DLSTCHK must be defined in the CCF's Configuration section. | Solution: To verify whether the DLSTCHK parameter is defined properly (for AMS6.01 only):
1 Create a prt file
2 All user defined variables, including DLSTCHK, should be listed at the very end of this file. If
DLSTCHK is missing, then its probably defined in the Configuration sections and needs to be
moved to the General section using DMCplus Build.
3 Once the CCF is updated, copy/ftp the new file to the online computer
4 Performs a Manage Delete or Unload, and Load the new CCF fresh
5 Finally verify that the DLSTCHK parameter is listed in a newly created prt file.
NOTE: For AspenOne 2004.x since DLSTCHK is defined in the CCF's Configuration section, it
will not be shown in the .prt file
Keywords: DMCplus
DLSTCHK
References: None |
Problem Statement: The messaging and diagnostics text written to composite .log, .err, and
.eng files may fall behind and get out of synchronization with the real-time
execution of the suite. This makes interpretation difficult. | Solution: Limit the maximum number of DMCplus controllers in any single composite tier to
no more then 8. Simply add a new tier group and redefine the controller/tier-group
accordingly.
Keywords: DMCplus Composite
Composite
References: None |
Problem Statement: The strip chart displays valve opening/closing interval of 20 seconds irrespective of user specified actuator linear rate (%/second). | Solution: The strip chart has a default interval sample setting of 20 second. This means that data is recorded on the strip chart at intervals of 20 second. For example a valve has been specified to close in 4 seconds, even though the valve actually closes in 4 seconds the strip chart will record closing time as 20 seconds because of the default interval sample setting.
To display the actual time on the strip chart the sample interval needs to be changed to a much shorter interval. This can be done when adding a strip chart as shown below.
.
Keywords: Strip chart, Sampling Interval, Actuator Linear rate (% / second), Valve opening, Valve closing
References: None |
Problem Statement: How has the handling of small crude fractions by the crude unit changed between V7.3 and V8.7? | Solution: There is a change in V8.7/V8.8 that impacts how small crude fractions are handled by the Crude Unit. In V7.3.1 crude fractions smaller than 0.005 (0.5%) were ignored, in V8.7 the Crude Tolerance value (the crude composition threshold setting from model settings) from the Config Dialog is used.
Â
While comparing results between V7.x and V8.x, change the setting in the Config dialog to 0.5 (%), this will make the results off the crude unit to be similar, excluding for gas cuts, where the calculation methods have been enhanced in V8.x
Keywords: Crude Composition Threshold, Crude Unit, CONFIG, settings
References: None |
Problem Statement: While pooling PROD_SERVICE material pools for product tanks, the trends stop trending while the blends are being made, that is, during the time duration of the event, the trends are flat lines, and they update in the period after the blend event.
In the Dist and FO blending screen, notice that there are two material service events called “TEST� which span the entire model horizon of two days:
Next, create two separate blend events against product tanks TJTA and THSF (with the same durations and the same rates and quantities) from 6/1/2011 03:59 to 6/2/2011 08:47
Now create a prod_service material pool combining service “TEST� (Which is the material service for both the above tanks THSF and TJTA)
Adding the material pool to the trend list and simulating for the schedule, the trend screen for the material pool volume MPCTJTA+THSF shows no activity during the course of the blend event, and updates correctly at the end of the event: | Solution: This is correct behavior for APS. Whenever you have a blend event into the tank, the tank service will automatically be set to the service as that product (which is Jet in this case); it will only be updated by material service event (if the service is different as Jet like in this case test) after the blend event is ended.
One suggestion as a workaround would be to create a material that is the same name of the product (say JET in this case), and then the trend updates dynamically for the duration of the blend event, using a prod_service material pool, with JET as service.
However, this workaround seems limited, such that you can only pool and track same products (JET), coming out at different blend headers, at the same time. So, using Prod_Service, for product tanks, it seems currently not possible to pool and track multiple products.
Keywords: PROD_SERVICE, material pool, blend event, material,trends
References: None |
Problem Statement: Controller has been modified and will not load. | Solution: On many occasions the extent of the modifications to a controller and / or model file
will prevent the controller from loading after the changes.
It is recommended to stop the controller, delete the controller, reload it and then restart it.
This process can be successful in removing references to the previous controller configuration and
enable the recognition of the changes.
Note: it is always recommended to save a copy of the CCF and MDL file of the previous controller configuration in the
event that it is necessary to revert to the previous configuration.
Keywords: CCF, Model, load
References: None |
Problem Statement: What change has been made in the way SMC calculations are handled since v7.3? | Solution: It was found that in some cases, calculated results for SMC models were different in PIMS and APS. The cause was that PIMS performs these calculations using double values and APS used float in v7.3. In later versions, APS was modified to also use double values so PIMS and APS have consistent results. Users might get differences in results from v7.3 and v8.7/v8.8 because of this change. However, typically these differences are small.
Keywords: SMC calculations
References: None |
Problem Statement: Curve Operations do not work on an Imported model in DMCplus | Solution: The Curve Operations option does not work on an Imported model in DMCplus Model, but the workaround for this is very simple.
TheSolution is to assemble a new model and copy the existing Imported model to the new model. To do this, simply right-click the Models folder under Project Data and select Assemble a New Model . Give the new model a unique name and make sure that is has the same Time to SS (min.): and Number of Coefficients: as the model that was originally imported. After entering this data, click the Apply and OK buttons to obtain a new model with an extension of Assembled. Then, double-click both the original Imported model and the new Assembled model so that both model files are visible on the screen. Then, on the Imported model, left-click and drag the mouse over the entire set of model curves. Release the mouse, and then left-click and drag the entire selected set of curves onto the blank Assembled model. A dialog saying Add selected curves to the model <Model-Name> [Assembled]? will appear. At this point, click the Yes button and the software will then tell you The independent tag <tag-label> does not exist in the target model and give you an option to Add or Ignore the tag. Select the Add radio button for each independent tag and then do the same for each dependent tag as asked by the program. At the end of this process, the Assembled model will have the same set of curves as the original Imported model, and the Curve Operations dialog will then be available.
Keywords: DMCplus Model
Curve operations
References: None |
Problem Statement: How do I use the model accuracy interface? | Solution: First of all, there are four different bias available in stream reconciliation:
1. Calc Delta: Using the difference between the average plant and simulated value and add it to the simulated value
2. Calc Avg: Using the average plant value and overwrite the simulated value
3. User Delta: Providing a delta to be added to the simulated value
4. User Avg: Providing a value to override the simulated value
For methods 1&2, the Plant Values are from the database table ORION_MGR_PLANT_VALUES
When using the update bias dialog, we will be using one of these methods to update our simulated values, based on the above described calculations
I've taken a simple example, using the User Delta bias:
1. Go to the model accuracy tab
2. Right click on “model accuracy screens” -> new screen, and name it
3. There are no stream trends added to the screen yet. Right click on the screen name and go to “Stream Facilities”
4. This opens up the “stream facilities manager” to add a new trend, right click inside the box and select “add trend”, then select the stream name and the property to trend. Here I have chosen a Diesel stream from the Cat Cracker
5. Clicking on the “+” handle to the left of the entry opens up a dropdown, where we can chose the biasing method, the unit, the source and destination stream and property
6. I have applied the bias method “user delta” on the CCD stream, which is of the form y=mx+c where m=0, so the “bias_value” column is the constant value by which we increase/decrease the simulation value
7. Clicking on OK, now generates a “Bias” trend on the trend chart of the stream
Seeing above, the blue dots are the plant values (which are not included in the biasing method used here).
At a random time-point on the screen, the simulated value, the biased value (which as we added, will be the (simulated value +2) at all points) are shown. The OldBias is as of now at NO_TREND_DATA as we have not biased/updated sim values yet. The OldBias keeps track of the value from which the new bias is calculated and it will start taking values from after the first bias update.
8. Now that we have a biasing method defined and the curves generated based on that, we can next “Update Bias” of trends by clicking on the icon as shown below:
9. This opens up the update bias/schedule biasing dialog. See below that it has one entry as of now, where you can choose to edit the factor, constant and duration of the bias
So here we have the stream and its volume which is going to be updated as -
CCDnew= (0) CCDold + 2, which is the bias method we have chosen
10. Clicking on OK does an automatic simulate, to update the CCD stream values
11. After simulation, notice that the simulated values have gone up by the specified value, and now merge with the biased curve
Now looking at the same timestamp as shown before, look at the simulated value (which has been updated), the biased value (which the sim values have merged with). The OldBias here takes the value from which we biased from.
12. If the update bias dialog is opened again, there will now be two entries in place of one
About this screen, the un-editable entry with the “done” entry checked says that this bias has been applied already. The other entry can be edited, to apply a new bias to the stream.
13. It is better to do another bias to understand better. Going back to the stream facilities manager window, and using a bias_value of 3
This means that the bias of 3 is applied from the original simulated value (19.08) and not the biased value (21.08)
14. Adding a bias_value of 3, and clicking OK generates a new bias curve on the trend, which exceeds the current values by 1
Now looking at the same timestamp as shown before, the simulated value and OldBias are the same as shown before, Biased value is the new curve created by adding the new bias (+3 increase)
15. Now opening the update bias dialog box has two entries: the old bias which was applied (cannot be edited) and the new bias which we are going to update now. All the updates are from the initial simulated value (19.08)
16. Clicking on OK re-simulates APS and then the new bias is applied to the simulated values.
17. The updated values are shown in the trend
Now looking at the same timestamp as shown before, the simulated value is updated at 22.08, merging with the biased value at 22.08 and the Oldbias is the value from which we applied the bias from, which is 19.08
18. Looking at the update bias dialog again, we see the updated bias factors and constants
Again, there is one fixed entry (showing the applied bias) and one edit-able entry (for applying a new bias)
Keywords: Model Accuracy, User Delta Bias, Stream Reconciliation, Update Bias
References: None |
Problem Statement: How can I coordinate multiple DMCplus controllers? | Solution: The DMCplus Composite Suite allows a user to coordinate multiple DMCplus controllers within an operating unit. DMCplus Composite has been sold as a layered product of DMCplus since DMCplus version 2.0.
The intent of DMCplus Composite is to allow very large unit-wide problems (such as ethylene plants) that are too big to fit into one single DMCplus controller to be solved in a coordinated way. Such large problems cannot be solved in a single DMCplus controller due to computer limitations and limits on the ability of operations personnel to understand extremely large controllers.
A standard DMCplus controller has an open-loop prediction, a steady-state optimization, and a dynamic move plan calculation. When such a DMCplus controller operates as part of a DMCplus Composite suite, that controller's open-loop prediction and steady state optimization are replaced by the combined or composite open loop prediction and steady state optimization computed by DMCplus Composite. The individual controllers then compute their own dynamic move plan calculations and pass the results back to Composite after they have output their moves to the process control system.
DMCplus Composite makes use of a tiering concept that allows controllers in the front end of a plant to honor constraints in the back end of a plant by including front end manipulated variables, such as furnace feed rates, as feedforward variables in the back end controllers. The Composite suite ties these relationships together such that the composite steady stateSolution honors the constraints of all participating controllers.
All controllers in a DMCplus Composite suite must execute on the same host computer and at the same frequency. There are specific Composite configuration parameters that must be entered in the CCF for every controller that participates in the Composite suite, and there is a Composite initialization file that must be configured that cross-references all controllers that are configured for that suite, as well as specifying the tier in which each controller is executed by Composite.
For more information on configuring DMCplus Composite, consult the DMCplus online help files.
Keywords: DMCplus
DMCplus Composite
References: None |
Problem Statement: Will apc applications see a new license server if old server is no longer available? | Solution: Yes, if the server is configured and available to the online server looking for the valid key.
In general, when an APC online application starts up, it looks first to itself for a local, stand along license key in the HyperTech / Shared directory. If it does not find one, it looks to the license servers configured in the SLM Configuration Wizard. Once it gets a valid key, it remembers where the key came from and continues to check for connectivity to the source of the key. Every four hours, the application confirms the acquisition of a valid license key. If it cannot get one from the remembered source, it will initiate a 10 day countdown. It continues to search the license servers for a valid key, and when it does, it will report that a valid license key has been found and the countdown will terminate and be re-initialized to 10 days. This could take a few cycles of confirming connectivity and realizing the source is no longer available and a new one is needed.
A few things to consider. If a license server is replaced by a new one, it is recommended to stop and restart the online applications in order for them to connect to the new license source.
It is also recommended to identify only the license servers in the SLM Configuration Wizard, that you know have the license keys your applications need. It is also recommended to Configure Advanced Settings in the SLM Configuration wizard, and to UNCHECK the Enable Broadcast option, so that only the servers configured will be used to search for valid license keys.
This has been known to optimize the time it takes to acquire a valid license key.
Keywords:
References: None |
Problem Statement: As blending calculations are done on volume basis in a weight-based model, should the blend event recipes be entered in % weight or % volume? | Solution: From UI, user can define blend event in WGT unit, and with WGT% in recipe. APS will automatically use 0.8 as SPG to convert to VOL when doing blending calculations.
Keywords: Blend Calculations
Blend event
References: None |
Problem Statement: User could not load controller, DMCplus Manage reported too many user entries defined in CCF Section: CALC. | Solution: Replace the existing dmcplusdataentry.xml file with one of the xml files attached to thisSolution.
The only difference in the attached xml files is the amount of extra entries, 750 will provide more than 150.
You should find the dmcplusdataentry.xml file in the ACOnline\cfg directory on the DMcplus machine. You will want to strip the 150 or 750 from the end. I would also recommend renaming the current dmcplusentrydata.xml in the event it will need to be put back in.
Also, before doing the replacement, first stop or close your instance of DMCplus Manage, replace the file and then Start up a new instance of DMcplus Manage. This will read the new file you have put in the cfg directory.
Keywords:
References: None |
Problem Statement: Can the valid.err file be deleted when DMCplus is running? This file is helpful but is hard to use when it gets very large. | Solution: Yes, the valid.err file can be renamed or even deleted and a new one will be opened on the next error message.
Keywords: \etc\valid.err
error file
log file
References: None |
Problem Statement: DMCplus Operating Displays (Operator CV display, Operator MV Display, Subcontroller Display, Message History display) always show the first controller of Controller Directory. DMCplus is controlling several units and some consoles should show another controllers information when it is invoked. | Solution: Add following Initial Tag move to DMCPBASE display.
From Tag : C M 3 ( If third controller has to be a default controller) To Tag : L M 11
Keywords: Controller
DMCplus Operating Displays
Controller Directory
Local Tag
Initial Tag Move
References: None |
Problem Statement: Aspen DMCplus Model 2006
Error when doing Save As with Aspen DMCplus Model 2006 | Solution: Make a copy of your dpp and try Ctrl-Shift-v on the open Model file. Answer 'Yes' and see if this repairs your file.
Keywords: save, model, dmc, error
References: None |
Problem Statement: The situation may arise that a controller can be loaded but will not start. Two techniques can be used to run the controller independently of Manage to gain more information for debugging the problem. | Solution: The controller can be run interactively with error messages written to the screen by using the following command: At a dos prompt, move to the ACONLINE/BIN directory:
$ control
ctlr_name
Y (this is the answer to 'run on-line?')
Stop the controller execution with CTRL-C.
Starting with v1.10, the controller can be run detached in a fashion that mimics the Manage Load, Manage Start sequence with the following command:
$ dmcpstart ctlr_name Y
Stop the controller execution with $ stop ctlr_name.
Keywords: None
References: None |
Problem Statement: Aspen DMCplus CLP Issue - Generation of doqp.dat files | Solution: There is a flag (EPSMVPMX) that the user sets to 3 in the CLP.ini to generate these files. Here's the excerpt from Aspen DMCplus helpfile for EPSMVPMX.
Maximum Move Penalty. This entry is used to select and/or tune the steady-state QP algorithm.
To enable the original DMCplus 2.x QP engine, set this value less than or equal to 1.
To enable the interior point QP engine (recommended) set this value to 2. This is now the default value (as of Aspen DMCplus 6.0).
A value of 3 will use the interior point QP algorithm and also generates a QP diagnostic file every cycle. This value is not recommended as it can quickly fill up the disk.
For the original QP engine (EPSMVPMX <= 1) this parameter is used as the maximum move penalty to improve numerical performance. It must be set to a value greater than the minimum move penalty (EPSMVPMN). The maximum value allowed for the original QP engine is 1.0
Keywords: None
References: None |
Problem Statement: How to set up External Targets | Solution: Use the Build application.
Open a CCF file.
Click the OPTIONS box on the toolbar, The button with the hand on it, or go to Tools and Options. Go to the External Targets box on the General page and click Full Use (RTO). Select the ETENB box to Enable External Targets in the Online Controller. Click OK.
Go to the Independent folder and select an MV you want to make an External Target. Click on the EXTERNAL TARGET Box to set.
Put the parameters in alphabetical order by clicking on the NAME bar above the parameter names. Set the desired ET - External Target variables, particularly ETMV and ETMNEW. The ETMRNG can only be used on MV's that have not been transformed.
Set the Equal Concerns and LP parameters.
Set ETMCRT and ETMCST.
These arrays must be completed for each MV included as External Targets. Continue this process for all MV's to be used as External Targets.
Go to the Dependent folder and select a CV you want to make an External Target. Click on the EXTERNAL TARGET Box to set.
Again put the parameters in alphabetical order.
Again set the ET variables.
Go to the Calculations Folder.
Select ET.
Set the parameters that appear.
Establish your MV and CV ranks.
Save the CCF.
Reminder: Information for the variables is available by double clicking on the variable name.
Keywords: External, Targets, Ranks, RTOPT
References: None |
Problem Statement: How do I import integrator parameters such as current time in the Aspen HYSYS spreadsheet? | Solution: The integrator parameters are not accessible in the HYSYS spreadsheet via right mouse-click and Import Variable method. The integrator parameters such as Current Time can be populated in a HYSYS spreadsheet via the Drag and Drop method as described below.
1. Open the HYSYS spreadsheet and then designate the cells in the spreadsheet where the parameters are to be imported.
2. Open the Integrator from the Menu bar (Simulation -> Integrator).
3. Select an integrator parameter such as Current Time.
4. Apply the right mouse click and hold the mouse button and drag and drop the current time value into the spreadsheet cell.
The screenshot below shows the Current Time imported to Cell B2 using the Drag and Drop method.
Keywords: Integrator, Spreadsheet
References: None |
Problem Statement: Do I have to install any Aspen DMCplus Server patch for the new USA DST? | Solution: There are no patches for the Advanced Process Control applications themselves because they use GMT instead of the system clock. However, there is still a need to apply the Microsoft DST patch applicable to your operating system. This allows the time change to happen at the correct time.
NOTE: After installing the Microsoft patch, you should restart the machine to insure that all processes that use the affected time functions load them from the new libraries. Timestamps for certain applications will be incorrect until this is done.
See Knowledge Base article 120732 for patches to the Production Control Web Server versions 6.0.1, 2004.1, 2004.2 and 2006.
There is going to be a problem with IP.21 as it uses the historian, which means AspenWatch has concerns, but they are due to the IP.21 issues.
The Information Management team has provided the following information:
Knowledge Base article 118715 has links to the Microsoft patches for the different OS systems.
Because of the way Microsoft did it, Aspen had to come out with patches of their own so as to be able to correctly see data from year 2006 and earlier during certain specific weeks.
So the same article
- already tells about the Client patches, Process Explorer, Web.21 etc.
- will 'shortly' be updated with information about the new IP.21 server patches (including AspenWatch)
We are recommending customers to keep checking back with that article (118715) as it gets updated pretty frequently these days as more and more things surface - both with changes by Aspen and Microsoft.
AspenTech patches are for AMS 6.0.1, 2004.1 and 2004.2. The patch for aspenONE v. 2006 will be in aspenONE v. 2006 Service Pack 1.
Keywords: DAYLIGHT SAVING TIME
USA
DST
PATCH
FIX
References: None |
Problem Statement: This technology tip provides some CCF Switcher usage tips and suggestions. | Solution: The CCFSwitcher executable can be run either as a Windows Task Service, or right from an open DOS Window.
AspenTech suggests that the CCFSwitcher configuration file be tested using the
CCFSwitcher.exe -debug command from an open DOS Window. Each cycle the CCFSwitcher runs, a wealth of information is written to the DOS window. This information is provided in addition to the information written to the CCFSwitchConfig.log file. The information written to the .log file can be reduced depending on the Verbosity setting (ie 0 - reports Errors only, 7 - writes maximum diagnostics information).
Upon each execution, the current value for the Controller name, Feed Selection Tag(s) and value(s), are written to CCFSwitchConfig.dat file. This .dat file is refreshed each cycle and its last modification time stamp is updated. Should this file be deleted, then the CCFSwitcher routine will create a new .dat file using the current values for the Feed Selection Tag(s). This implies that any changes in the Feed Selection lineup (from last cycle) are lost, and no CCF swapping occurs.
Similarly, if the CCF_SWITCHER_ON_OFF_TAG is set to 0 (or OFF), then the .dat file is not refreshed (as can be determined by file's last modification time stamp). Thus any changes in the Feed Selector values are lost. Indeed, if control Engineers have manually loaded the appropriate CCF's, then it may be better to delete the .dat file before toggling the CCF_SWITCHER_ON_OFF_TAG to 1 (or ON). Any values other then 0 or 1 for the CCF_SWITCHER_ON_OFF_TAG are ignored and interpreted as 1.
Once the CCFSwitcher's functionality has been fully tested and commissioned, then its recommended to schedule its execution as a Windows Task Service. Use the following command from a DOS Window to define the executable as a Windows Task Service: CCFSwitcher.exe -install The CCFSwitcher routine should now be listed as a Windows Task Service and be listed on the Services Dialog as, DMCPlus CCFSwitcher Service. Ensure to Start its execution and also ensure to toggle the Startup Type from Manual to Automatic. A Startup Type of Automatic causes the
CCFSwitcher Task to be scheduled for execution automatically upon computer reboot.
Using the Windows Task Service to execute the CCFSwitcher.exe causes the configuration file, CCFSwitch.ini, to be read each cycle. Any typographical errors, or accidental Feed Stock references to non-existing CCF's will cause the Switcher algorithm to stop; albeit the Windows Task Service status will remain Started. Error messages will be written to the .log file; and the .dat is not updated (ie as can be seen by its the 'frozen' time stamp)
A CCF Switch is triggered when a change in any Feed Selector value is detected. Should no matching CCF name be given in the .ini file (for the Feed Selector change), no CCF swap occurs and the appropriate messages are written to the .log file, and the Feed Selector's Status (if defined).
Should a Feed Selector change be detected, but the appropriate DMCplus controller is either not loaded, or its image is not executing, then the appropriate CCF will be loaded. However; the user must manually start the DMCplus controller image. The name of the (last) loaded CCF file can be viewed with the Manage menu in the Production
Control Web Viewer. Once the correct CCF has been verified the controller's execution can also be started using the
Production Control Web Viewer.
Keywords: DMCplus CCFSwticher
CCFSwitcher
CCF Switcher
References: None |
Problem Statement: We want to simulate a process/model mismatch in DMCplus Simulate in order to test the performance of a DMCplus controller under the type of real-world conditions that will be seen in an operating plant. | Solution: Three types of advanced simulation capabilities were introduced in DMCplus Simulate 5.0 and remain available in subsequent versions, as follows:
1. Advanced simulation capabilities using gain multipliers
2. Advanced simulation capabilities using different model files
3. Advanced simulation capabilities using measurement noise
ThisSolution looks at the features of each of these types of simulation
Advanced Simulation Capabilities Using Gain Multipliers
DMCplus Simulate has always honored gain multipliers (GMULTs) in the controller if they were enabled. In addition, the prediction calculated by the plant in Simulate did not honor the GMULTs. This design allowed users to set up simulations with a basic type of model mismatch. For users wishing to take advantage of the model mismatch, a new user interface for editing the GMULTs was added in DMCplus Simulate version 5.0.
The first step in setting up the use of GMULTs in this interface is to select the Controller | Model Options menu and click the Enable gain multipliers in the controller model checkbox under the Enable Controller Gain Multipliers section of the menu screen. When this checkbox is enabled, all of the potential options in the Enable Plant Gain Multipliers section of the menu screen can now be used as well (they are grayed out if controller GMULTs are not enabled). The three radio buttons available under the Enable Plant Gain Multipliers section are:
Disable gain multipliers in the plant model - This option automatically sets all of the plant GMULTs to 1.0, meaning that the gains from the plant model will be used.
Use the controller gain multipliers in the plant model - This option uses the controller GMULTs in the plant model as well.
Use user-defined gain multipliers in the plant model - This is the most powerful of the three options in that it allows the user to specify different GMULTs in the controller and plant models. For example, specifying a larger GMULT in the plant model for a particular model curve will cause the plant to move more than the controller expects, which could lead to oscillatory control action if very aggressive DMCplus tuning is used. This can be very useful in coming up with optimal DMCplus tuning parameters in the face of a pure gain model mismatch.
By default, all GMULTs are limited to values between 0.5 and 2.0. These are the default values of the GMULTL and GMULTU parameters that are found in the Configure section of the CCF. These values must be set in DMCplus Build and cannot be viewed via the Controller | Internal Variables... menu in DMCplus Simulate.
Finally, it is very important to point out how the GMULTs transfer to the online controller. If non-unity GMULTs are specified for the controller model in DMCplus Simulate, these will automatically transfer to the target CCF as user-defined variables upon a Controller | Update Configuration... operation. The GMULTs will transfer to the appropriate independent under the Independents folder in DMCplus Build as LOCAL variables with a tag name of GMULTxxx, where xxx represents the order in which the appropriate CV appears in the Dependents folder. Non-unity plant GMULTs do not transfer at all since they are used only in DMCplus Simulate. The parameter GMULTE, which resides in the Configure section of the CCF and governs the use of GMULTs in the online controller, also does NOT transfer from DMCplus Simulate. This means that the Enable gain multipliers in the controller model checkbox in DMCplus Simulate ONLY impacts offline simulation. If the user desires to use GMULTs in the online controller, the value of GMULTE must be set equal to 1 in the controller CCF. This is done to clearly separate the use of GMULTs between online and offline use.
Advanced Simulation Capabilities using Different Model Files
Although testing model gain mismatch does help in evaluating controller performance, it is also desirable to test dynamic model mismatch as well. Users can specify a completely different model (.MDL) file to represent the plant model. Given that capability, it is quite simple to use the Curve Operations feature in DMCplus model to modify curves by changing dead times, adding pure first and second-order Laplace transfer functions, and others. The user can then export the modified model to a new .MDL file. Once the modified model has been exported, it can be imported into DMCplus Simulate by selecting the Controller | Model Options menu, checking the Use imported model radio button under the Specify Plant Model section of the menu screen, and then using the Browse button to browse to the desired model file.
Advanced Simulation Capabilities using Measurement Noise
The first two methods described are very useful in testing the controller under model mismatch situations. However, in both of these methods, the controller will eventually line out at a steady state value if there are no MV/CV limit changes made or there is no attempt to perturb the feedforwards. That is not the way that a real-world application, that is constantly subject to measurement noise, will perform.
It is possible to force measurement noise effects into the controller in DMCplus Simulate via the parameter SIMCVSTDDEV. This can be done in DMCplus Build by setting the SIMCVSTDDEV parameter for any CV in the Dependents section of the CCF, and in Simulate under the Noise Std Deviation column under the Controlled Variables tab. Once a non-zero value of SIMCVSTDDEV is in place, you can activate the noise in Simulate via either the Add Noise button on the Simulate toolbar (just to the right of the Run and Pause buttons), or by going to the Controller | Model Options menu and checking the Add noise to controlled variables box under the Enable Noise section of the menu screen.
The SIMCVSTDDEV entry is only used by Simulate and is ignored by the online controller. Mathematically, a Gaussian distribution with a mean of zero and a standard deviation of SIMCVSTDDEV is added to the dependent variable DEP value at the end of each execution in Simulate. This parameter is stored in the CCF so that subsequent visits to Simulate preserve the desired values.
This method is guaranteed to show noise and oscillatory-type behavior at all times, even if there are no perturbations made to feedforwards and no changes made to CV targets. However, care must be taken in specifying the value of SIMCVSTDDEV to avoid the possibility of a CV prediction violating a validity limit and possibly causing the simulation to go open-loop. Specifying a SIMCVSTDDEV of as low as 1% of the normal control range should be enough to observe real-world behavior with both noise and some oscillatory behavior.
Also note that all three of these methods can be used together if the user so desires, to give the most thorough possible tuning checkout.
Keywords: DMCplus Simulate
SIMCVSTDDEV
GMULT
References: None |
Problem Statement: When is it appropriate to specify an MV as reverse acting in a DMCplus controller or SmartStep tester? | Solution: IREVRS is the manipulated variable reverse action indicator for each MV in an Aspen DMCplus controller or Aspen SmartStep tester. Its possible values are 0 (direct acting: increase setpoint, output increases) and 1 (reverse acting: increase setpoint, output decreases). In general, you should specify IREVRS as 0 whenever possible. ThisSolution identifies situations where IREVRS needs to be specified as 1.
When DMCplus looks at windup protection during variable validation, it looks at the values of both IREVRS and the anti-windup status code (AWSCOD) for each MV. AWSCOD can have four possible values: 0 (no windup, setpoint can be moved in either direction), 1 (wound up low, setpoint can only increase), 2 (wound up high, setpoint can only decrease), and 3 (output path blocked, setpoint cannot be moved in either direction). When IREVRS is set to 1, the AWSCOD values of 1 and 2 are flipped when a windup condition is encountered.
Next, let's talk about direct and reverse action from the point of view of both DMCplus and major DCS vendors. DMCplus considers a regulatory controller to be direct acting if you want the manipulated variable to increase when the setpoint increases. A good example of this is a PID flow controller. If you want more flow, then you generally increase the flow control valve output.
DMCplus considers a regulatory controller to be reverse acting when you want the manipulated variable to decrease when the setpoint increases. Consider the example of an overhead vapor temperature controller which cascades to a reflux flow controller. In this case, when you increase the setpoint of the temperature controller, the reflux flow setpoint needs to be decreased, which makes the temperature controller reverse acting. However, the reflux flow controller itself is direct acting.
This all sounds simple enough. However, it gets complicated when you consider the DCS vendors' point of view. DMCplus calculates the control error of a loop as:
ERROR = SP - PV
where SP is the setpoint and PV is the process value. However, most DCS vendors (Honeywell, for example) calculate control error in the opposite direction, namely:
ERROR = PV - SP.
For example, using Honeywell logic, our normal direct acting flow controller would be configured using REVERSE control action in the DCS. The top temperature controller which DMCplus thinks is reverse acting would be configured using DIRECT control action in the DCS.
Fortunately, Honeywell uses the same thinking as DMCplus with respect to anti-windup status. A Honeywell DCS is smart enough to flip the anti-windup status values of 1 and 2 to present the correct anti-windup status code to DMCplus. This enables IREVRS to be set as 0 in DMCplus.
However, not all DCS systems support windup protection. An example of this type of system is an ABB MOD300 DCS using version 10 ABB software. Since it is unwise to implement DMCplus without windup protection, it is necessary for the user to calculate the windup status external to the DMCplus control engine. This can be done by using input calculations in a CCF.
A typical input calculation would calculate windup status by checking the output of the regulatory controller against low and high values, and declaring a windup condition if the output is outside one of these limits. An example of this type of calculation for regulatory loop TC1, where TC1.SP is an MV and TC1.OP is a CV, might be:
{IND:TC1.SP|AWSCOD} = @IF( @LE( {DEP:TC1.OP|DEP}, 0), 1, @IF( @GE( {DEP:TC1.OP|DEP}, 100), 2, 0))
This basically sets the windup code to 1 (wound up low) if the controller output is less than or equal to 0, and sets the windup code to 2 (wound up high) if the output is greater than or equal to 100. For a DCS that does not support windup protection, this type of calculation should be carried out for each MV in the controller.
However, in order for this to work correctly with regulatory controllers that require reverse action (as defined in DMCplus), the user must specify IREVRS equal to 1 for that MV in the CCF. This is because of the way that the DMCplus control engine responds when confronted with an AWSCOD of 1 or 2. Consider the case of TC1.SP above, where the current value (VIND) is 225, the lower operating limit (LLINDM) is 220, the upper operating limit (ULINDM) is 240, and the AWSCOD from the input calculation is returned as a 2 because the TC1.OP is at 101.5. If this controller is truly reverse acting, then the TC1.SP cannot be decreased without aggravating the windup condition. In order for this to happen, IREVRS for TC1.SP must be configured as 1 to allow DMCplus to flip the AWSCOD from 2 to 1. If IREVRS is specified correctly, then the effective value of LLINDM will be clamped at the current VIND of 225, meaning that DMCplus cannot further aggravate the situation by decreasing TC1.SP. If IREVRS is erroneously configured as 0, then the engine clamps the effective value of ULINDM at the current VIND of 225, and the controller can still decrease TC1.SP to the lower limit of 220, even though this aggravates the windup condition.
Keywords: IREVRS, AWSCOD
References: None |
Problem Statement: Wanting a list of License keys internal feature numbers to identify client needs with software received. | Solution: Here is a list of all feature numbers with their feature names.
10, ASPEN PLUS,
11, BATCHFRAC,
12, RATEFRAC,
13, BPS,
15, POLYMERS PLUS,
16, B-JAC INTERFACE,
17, Aspen PEP,
30, ASPEN PLUS End User Model,
40, PROPERTIES PLUS,
45, BATCHPLUS,
50, Aspen Pinch,
51, Aspen Split,
55, ADSIM,
56, Aspen Chromatography,
60, Aspen Modeler,
61, Aspen Custom Modeler,
62, Aspen Dynamics,
63, Aspen Traflow,
65, SMB PLUS,
70, HETRAN,
71, HETRAN-INTEGRATED,
72, AEROTRAN,
73, AEROTRAN-INTEGRATED, 74, TEAMS-ASME,
75, TEAMS-ADM,
76, TEAMS-CODAP,
80, Otis,
81, Xeng,
82, Builder,
91, TOKEN POOL 1,
92, TOKEN POOL 2,
93, TOKEN POOL 3,
94, TOKEN POOL 4,
95, TOKEN POOL 5,
96, TOKEN POOL 6,
97, TOKEN POOL 7,
98, TOKEN POOL 8,
99, TOKEN POOL 9,
100, Aspen IQ - IQmodel,
101, Aspen IQ - IQconfig,
102, Aspen IQ - IQonline, 103, Aspen IQ - IQmodelRT, 104, Aspen Target Model, 106, Aspen Target Control, 108, Aspen Online,
109, Aspen Teams - Heat Exchanger Mechanical Design, 110, Aspen Properties,
111, HTRI Interface, 112, MIMI Database,
113, MIMI Database Server, 114, Aspen Water,
115, MIMI Expert System, 116, MIMI Communications, 117, MIMI Graphics,
119, MIMI Planning Board, 121, MIMI Discrete Scheduling, 123, MIMI Continuous Scheduling, 125, MIMI Demand Management,
127, MIMI Linear Programming, 129, MIMI Robust Optimization, 131, MIMI Inventory Management Plan, 133, MIMI SQL Bridge,
135, MIMI Client Seats, 137, MIMI Distributed Computing Demo, 139, MIMI Request Broker,
141, MIMI Gasoline Blending, 143, MIMI ActiveX Components, 145, MIMI Web Server, 147, MIMI Object Bridge, 149, MIMI Logistics Interfaces, 151, MIMI SAP Integration, 153, MIMI ASA Template, 155, MIMI SAP POI Template, 157, MIMI Semiconductor Global Planner Template, 159, MIMI Refinery Planner/Scheduler Template, 161, MIMI Polymers Scheduler Template, 163, MIMI Specialty Chemicals Template, 165, MIMI Chemicals Planner Template, 167, MIMI Logistics Manager Template, 169, MIMI Trim Optimizer Template, 171, IBM OSL Optimizer,
173, Ilog CPLEX Optimizer, 175, Xpress Optimizer, 177, Maximal Software Math Programming Language, 178, PIMS,
179, PPIMS, 180, RefSked,
181, MPIMS,
182, PASSMAN,
183, XPIMS,
184, PIMSXCHG,
185, PIMS-SX,
186, PIMSBCI,
187, PIMS-MIP,
188, SDPIMS,
189, PIMS-SI,
191, LPIMS,
193, XPRESS,
195, CPLEX,
197, TPIMS,
198, Aspen OLI,
200, Plantelligence,
250, Aspen Zyqad,
300, InfoPlus.21,
301, Point Count,
302, Definition Editor,
303, Event.21,
304, CIM-IO,
305, InfoPlus.21 Administrator, 306, OPCDA for InfoPlus.21, 307, SQLplus Server,
308, ODBC OAM,
309, Desktop ODBC Server, 310, SQLplus Client,
311, Desktop ODBC Driver,
312, ERP Connect,
313, PPPI Module,
314, Batch.21 Server,
315, Batch.21 Client,
316, Batch Connect,
317, Process Explorer,
318, Multivariate,
319, Process Data Add-In, 320, Process Recipe,
321, Recipe Explorer,
322, Aspen Calc,
323, Aspen Advisor,
324, Aspen Advisor Expert,
325, Aspen Reporter,
326, Aspen DBTools,
327, Aspen SSInterface, 328, Aspen Advisor Connect, 329, Aspen Advisor ERP, 330, Aspen Advisor SnR, 331, Aspen Advisor Plan, 332, Aspen Process Order Server, 333, Aspen Process Order Client, 334, Compliance.21,
335, Accounting.21,
336, Grade-IT,
337, Human Interface,
338, LIMS Interface,
339, GCS Client,
340, Aspen Product Tracking Server, 341, Aspen Genealogy Server,
342, Aspen Genealogy Client, 343, Profile.21 Server, 344, Q Server,
345, Cimview,
346, OPCDA for Cimview, 347, Aspen Product Tracking Client, 350, DMCplus Desktop,
351, DMCplus Build,
352, DMCplus Model,
353, DMCplus Simulate,
354, DMCplus Online,
355, DMCplus Control,
356, DMCplus Composite, 357, Aspen Watch Client, 358, Aspen Watch Inspector, 359, Aspen Watch Server, 360, Aspen Watch Maker, 361, Aspen IQpredict, 362, Aspen IQupdate,
363, Aspen IQ Desktop, 364, Aspen IQ Online, 400, Scan21 for DDE (Windows Component), 401, CIM-IO for Fisher-Rosemount ChiP, 402, CIM-IO for Moore APACS via NIM (RNI), 403, CIM-IO for \@aGlance,
404, CIM-IO for ABB 1180/1190 via DIU, 405, CIM-IO for Bailey semAPI,
406, CIM-IO for DDE, 407, CIM-IO for Eurotherm Gauge via DCP, 408, CIM-IO for Fisher-Rosemount RNI, 409, CIM-IO for Foxboro FoxAPI, 410, CIM-IO Client for G2, 411, CIM-IO for Setcim/InfoPlus-X/InfoPlus.21, 412, CIM-IO for Measurex MCN,
413, CIM-IO for OPC, 414, CIM-IO for Westinghouse, 415, CIM-IO for Wonderware Intouch, 416, CIM-IO for Honeywell TDC 3000 via HTL Access , 417, CIM-IO for Intellution FIX, 418, CIM-IO for Hitachi EX Series, 419, CIM-IO for RSLinx,
420, CIM-IO for Yokogawa EW3, 421, CIM-IO for PI,
422, CIM-IO for Ulma 3D, 423, CIM-IO for GE Fanuc via HCT, 424, CIM-IO for Toshiba TOSDIC, 425, Enterprise Objects, 426, CIM-IO for Yokogawa ACG10s, 427, AFW2,
429, AFW3, 431, AFW4,
433, AFW5,
434, IQmodel Powertools, 435, PIMS Report Writer, 436, Aspen Retail Automatic Stock Replenishment, 437, PIMS Multi-period Blend Optimizer, 438, Aspen Bulk,
439, Web.21, 441, Plan vs Actual Refining, 443, LP Accuracy Management for Refining, 445, Aspen Price for Refining,
447, Plan vs Actual for Petrochemicals, 449, LP Accuracy Management for Petrochemicals, 451, Aspen Price for Petrochemicals, 453, Production scheduling for Polymers, 455, Aspen Transition Manager Client, 457, Aspen Transition Manager Server, 459, Aspen Plus Online,
461, Aspen Plus EO Optimizer, 463, Aspen Retail Replenishment Scheduling Optimization, 465, Aspen Retail Common Interface Viewer, 467, Aspen Retail Table Structure Auditer, 469, Aspen Retail Credit Utility, 471, Aspen Retail Remote Access Terminal, 473, Aspen Retail Supply Chain Analyzer, 475, Aspen Retail NT Services,
477, Aspen Polyquest, 479, Utilities Library, 481, MIMI Transportation Planning, 483, Aspen FCC,
485, Aspen CatRef, 487, Aspen Hydrocracker, 489, Aspen Hydrotreater, 491, Advisor Mathematical Reconciliation Engine, 493, Aspen_eBRS,
495, Aspen_eBRS_Designer, 497, MapInfo MapX,
499, CIM-IO for Measurex ODX, 501, Aspen Alarms and Events, 503, Aspen SEM,
505, webgraphics,
507, Aspen SmartStep Simulate, 509, Aspen SmartStep Online, 511, Aspen SmartStep Desktop, 513, Production Control Web Server, 515, UPID Base,
517, UPID APC, 519, UPID SIM,
521, Otiss Development, 523, FDA 21 CFR Part 11, 525, ORION,
527, PIMS Sub-model Calc, 529, Aspen Operations Log Server, 531, Aspen Operations Log Client, 533, Tank Management Server,
535, Tank Management Client, 537, Pumper Log Server, 539, Pumper Log Client, 541, Aspen Enterprise Model Server, 543, Aspen Enterprise Model Client, 545, Aspen ORION Plan,
547, Aspen ORION XT, 549, Aspen ORION MBO, 551, Aspen Blend Component Library, 553, PIMS Extended Tags,
555, RTOWatchServer, 557, RTOWatchCimio,
559, RTOWatchMaker,
561, AspenBlend,
563, Web.21 Graphics Converter, 565, PIDWatch,
567, Apollo Model,
569, Apollo Control,
571, Apollo Powertools,
Keywords: ALM, License, Keys, internal, names, ACS, AMS
References: None |
Problem Statement: Gain Scaled Cost Tuning Worksheet | Solution: This Excel worksheet has the ability to read a DMCplus model file (*.mdl) and provide insight into the behavior of the steady-state solver. The user must enter the MV costs (CST) and MV cost criteria (LPCRIT). The functionality in this spreadsheet is made available to meet the obligation to the AC&O User's group development contract for AMS 6.0. The users voted that this functionality be added to the AMS 6.0 version of DMCplus. Due to difficulties in determining the best form and place for the strategy features (DMCplus Simulate, Aspen Watch, Production Control Web Server, etc) it was decided that this spreadsheet be made available for evaluation as a stand alone tool. Note that the RGA functionality was added to DMCplus Model in AMS 6.0. The spreadsheet can aid in the following analyses:
Identify MV/CV pairs that have a RGA number that exceeds a user specified value. If RGA numbers are excessive then the gains can be adjusted to eliminate any near collinear submatricies. .
Provide insight into the controller's strategy for relieving violated constraints. The 'strategy' tab will show which MV will be preferred for each CV low and CV high constraint and all MVs with models to that CV will be ranked. It should be noted that this analysis does not take into account the multivariable nature of the controller. For example, the #1 handle for a given constraint may be at a limit or may be needed to prevent another CV violation. This spreadsheet does not weight these factors, however, it is useful to see the order that the controller will try to react to a given violation. It provides a common sense way to validate the LP costs that are in use.
Keywords:
References: None |
Problem Statement: Some clients utilize a DCS for which there is no Aspen DMCplus Build template provided. As a result, they may sometimes have difficulty configuring IO data points. This template is for Emerson Delta-V | Solution: Important: The attached template file is to be understood as an example. Every effort has been made to make it accurate and usable, however, this is an offering procured from Services personnel, and not a product and therefore has not gone through the development and testing process applied to products. Support for this will be limited to a best faith effort to make it work on a client's site via telephone support.
If you discover and correct any problems with this file/Solution, please notify support so that the modifications can be considered for incorporation into this file.
Keywords: Delta-V
Template file
TCC
References: None |
Problem Statement: Use the attached Excel Add-in to aid in DMCplus Model analysis.
Unsupported tool, use at your own risk. | Solution: Excel Add-in v7.5
Keywords: DMCplus, Model, Excel
References: None |
Problem Statement: OpenVMS systems
When it is possible to load a small controller but trying to load a larger controller gives the following message:
Loading controller xyz from CCF xyz1
Scratch OTF allocation failed.
Failure creating context header
CTL: Error Loading Controller
Control load failed for controller: xyz, CCF xyz1
Go to sys$system and run authorize, show user_name. User pgflquo is often in the area of ~50,000 , increase it to 250,000. Logout and in and user should be able to get the load to proceed to give validation errors, which is what is what is expected for a first load.
Honeywell AxM
from a re-load on a running controller:
scratch otf allocation failed
failure creating context header
ctl: error loading controller
from a first time start via Manage:
can't become daemon (exited Manage and re-started aok)
The client found that his second swap file had been deleted.
Restore second swap file, for total swap file size = 2 * physical memory | Solution:
Keywords:
References: None |
Problem Statement: Cim-IO 4.7 is built with DECC not VAXC so they no longer include the VAXCRTL.EXE in the TCPIP.OPT file that we link with. This is not a problem for linking control or manage ... but it is a problem for dmcpview_* because the GCS libraries are built using VAXC. | Solution: Need to add the following as last line to CIMIOROOT:[IO]TCPIP.OPT
SYS$SHARE:VAXCRTL.EXE/SHAREABLE
Keywords: DMCplus, CIM-IO, VAX, TCPIP, DECC
References: None |
Problem Statement: GCS error 21 specific to DMCplus View v1.0x.xx. | Solution: GCS error 21 is specific to DMCplus View v1.0x.xx. It occurs on validation of the displays. A possible fix is to send displays that have been validated against the a machine of the target type, i.e., VAX or alpha.
Keywords: GCS, Error 21, DMCplus View, View
References: None |
Problem Statement: What could cause problems opening Excel files within Aspen Petroleum Scheduler (APS) / ORION in the Citrix server? This problem is also characterized by the error messages Windows cannot find FileName.xls. Make sure you typed the name correctly and try again followed by Shell command FileName.xls failed as shown in Figures. | Solution: 1. Copy the file locally to the machine and see if the problem persists
2. Check if this happens only if some other user has that file opened
3. Try mapping the network drive rather than directly accessing as \\ and see if that works
Key words
ORION APS unable to open Excel file, Citrix server Excel file, ORION APS shell command failed
Keywords: None
References: None |
Problem Statement: What is the significance of 'Minimum duration of continuous operation events' option in APS? | Solution: This feature is available from V7.3 onwards; this can be found in Model Settings | Events Default dialog
Continuous operation events can be applied only for unit operations. A unit operation can be assigned to operate in continuous mode by right checking the Continuous Operation option box in the Unit Definition dialog. The Unit Definition Dialog can be invoked by right clicking over the Unit (for example alkylation unit as shown in Figure) and the choosing Properties menu.
This specific option Minimum duration of continuous operation events ask for several questions from a scheduler?s perspective. The following section analyzes some of the key questions.
Suppose my alky unit is defined for continuous operation, will APS prevent me from adding an event that is less than the minimum duration?
In this example Minimum duration of continuous operation is set as 60 min, this prevents the creation of events spanning less than 60 min. This applies only on the events screen; therefore this restriction avoids creating some minuscule and overlapping events un intentionally from the screen. On the other hand, if needed events spanning less than 60 min can be created form the dialog.
How to represent unit down time if it does not allow discontinuity?
It is better not to consider units as continuous if it has discrete occurrences.
Why unit operation events cannot have overlaps?
Unit operation cannot have overlaps. For example, reformer is a unit operation and it cannot be operated at two different severities at the same time.
Keywords: Continuous operation
Minimum duration
Unit operation events
References: None |
Problem Statement: We don't see how to easily make PCLS0114 generic to handle the AO point on the PM. | Solution: One alternative is to insert a ratio or PID block at the PM on top of the AO block. This will have shed timer and shed mode.
Other sites have handled this with an associated point that provides the information needed to set LOOPST and AWSCOD via the Cim-IO for CM50S smart data types. Extra CL must be used with these associated points. An example for you to customize is attached.
PACKAGE
-- PROGRAM:
CCC_CL.CL
-- AUTHOR :
John Meline
-- PURPOSE:
This package is attached as CCC_CL.CL
--
to a dummy CUSTOM AM point in place of a REGULATORY
--
AM point in a DMCplus controller. This code enabled me to
--
LINK PCLS0114.AO (the watchdog code) to a sub-controller
--
point containing a point with no MODE.
--
This package allows MODE shedding (on the TDC side)
--
of several points and allows the DMCplus to read the MODE from
--
the dummy CUSTOM AM point because the HIGHWAY ANALOG OUT
--
and ANALOG IN (AO and AI) points do not have a
--
parameter called MODE.
--
This code is named after a CCC controller (an external
--
compressor computer controller that is external
--
to TDC PID control).
--
This code sets the MODE of the CCC_PT to CAS when:
--
point.DMCSRV
--
sub-pt.DMCON
--
main-pt.DMCON
--
are all in the ON state; otherwise the code sets the MODE
--
to MAN.
-- SUGGESTIONS: (1) Modify the CUSTOM section below, adding extra parameters
--
as necessary.
--
(2) Build a CUSTOM AM point with one CL block
--
and one custom data segment (PACKAGE).
--
(3) Link the CCC_CL.AO file to the CUSTOM AM point.
--
(4) Enter the appropriate tag names into the custom parameters
--
on the CUSTOM AM point.
--
(5) Activate the custom AM point.
PARAM_LIST CCC1_PL
PARAMETER
END CCC1_PL
DMCSRV --service switch
PARAM_LIST CCC2_PL
PARAMETER
END CCC2_PL
MODE : MODE -- mode of control point
PARAM_LIST CCC3_PL
PARAMETER DMCON : LOGICAL -- ON/OFF switch on sub, main ctrlr points
END CCC3_PL
CUSTOM (class GENERAL; access engineer; bld_visible)
PARAMETER CCC_PT
:
CCC2_PL
AM Regulatory Pt
PARAMETER CCC_ID
:
CCC1_PL
DMC Interface Pt
PARAMETER CCC_SUB
:
CCC3_PL
Sub Controller Pt
PARAMETER CCC_MN
:
CCC3_PL
Main Controller Pt
END CUSTOM
--
--
PROGRAM BLOCK
Keywords: None
References: None |
Problem Statement: How can I start the ACOMSGSRV process manually without running ACOBASE shutdown and then ACOBASE startup. | Solution: Download the attached Acomsgsrv.Zip file, move the 2.bat files provided into the Program Files \ Common Files \ AspenTech Shared directory.
The AcoMsgSrv_startup.bat and AcoMsgSrv_shutdown.bat can be used to start or stop the ACOMSGSRV process. This process is required for the Dais trader to see the AcoMsgSrv offer for display with the Aco Message Viewer application.
Keywords:
References: None |
Problem Statement: re-linking View gives 3 undefined symbols:
gethostbyaddr, getpeername, sockets | Solution: For DMCplus 1.04 when Cim-IO 4.7A is installed, the DMCPLUS_LINK.COM file must be modified in 3 places as follows:
Under $VIEW: label under line $! Cim-IO make change as follows:
$ else
$! Cim-IO
$ REALLIB = ,ACODHOME:[lib]'MACHTYPE'_GCSREALS/lib
$ EXTTCPOPT =
$ endif
becomes:
$ else
$! Cim-IO
$ REALLIB = ,ACODHOME:[lib]'MACHTYPE'_GCSREALS/lib
$ EXTTCPOPT = ,acodhome:[lib]'MACHTYPE'_TCP_'TCPLIB'/OPT $ endif
Beginning with line $ echo for DECNet, change as follows:
$ echo for DECNet
$ LINK/nodebug/exe=acodhome:[bin]dmcpview_decnet.exe -
ACODHOME:[LIB]dmcpview_decnet.obj, -
ACODHOME:[LIB]view/lib, -
ACODHOME:[LIB]'MACHTYPE'_GCSLINK_DECNET/lib 'REALLIB' , - ACODHOME:[LIB]view/lib, -
'DMCPLUS_LIBS', -
ACODHOME:[lib]CONNECT/lib,-
acodhome:[lib]acodutil/lib -
'CONNECT_LIBS'
becomes:
$ echo for DECNet
$ LINK/nodebug/exe=acodhome:[bin]dmcpview_decnet.exe -
ACODHOME:[LIB]dmcpview_decnet.obj, -
ACODHOME:[LIB]view/lib, -
ACODHOME:[LIB]'MACHTYPE'_GCSLINK_DECNET/lib 'REALLIB' , - ACODHOME:[LIB]view/lib, -
'DMCPLUS_LIBS', -
ACODHOME:[lib]CONNECT/lib,-
acodhome:[lib]acodutil/lib -
'CONNECT_LIBS', -
acodhome:[lib]vax_crtl/opt
Beginning with line $ echo for SERIAL, change as follows:
$ echo for SERIAL
$ LINK/nodebug/exe=acodhome:[bin]dmcpview_serial.exe -
ACODHOME:[LIB]dmcpview_serial.obj, -
ACODHOME:[LIB]view/lib, -
ACODHOME:[LIB]'MACHTYPE'_GCSLINK_SERIAL/lib 'REALLIB' , - ACODHOME:[LIB]view/lib, -
'DMCPLUS_LIBS', -
ACODHOME:[lib]CONNECT/lib,-
acodhome:[lib]acodutil/lib -
'CONNECT_LIBS'
becomes:
$ echo for SERIAL
$ LINK/nodebug/exe=acodhome:[bin]dmcpview_serial.exe -
ACODHOME:[LIB]dmcpview_serial.obj, -
ACODHOME:[LIB]view/lib, -
ACODHOME:[LIB]'MACHTYPE'_GCSLINK_SERIAL/lib 'REALLIB' , - ACODHOME:[LIB]view/lib, -
'DMCPLUS_LIBS', -
ACODHOME:[lib]CONNECT/lib,-
acodhome:[lib]acodutil/lib -
'CONNECT_LIBS', -
acodhome:[lib]vax_crtl/opt
Keywords: DMCplus, CIM-IO, 1.04, 4.7A
References: None |
Problem Statement: What version of UCX is required for CIM-IO and DMCplus? | Solution: The requirements for DMCplus 1.04.xx or 1.10.xx and Cim-IO on an OpenVMS platfom using UCX are:
OpenVMS 6.1, 6.2, 7.1 CIM-IO Kernel 4.6b, 4.7.1(preferred) UCX 4.1 with ECO 4 , 6 or 8 or UCX 4.2
Keywords: CIMIO, OpenVMS, UCX, DMCplus
References: None |
Problem Statement: User Blend Model Library (UBML) is a custom built model library. It providesthe user the ability to incorporate user-defined blending correlations into an APS model. The question is how to run APS with a custom built UBML.DLL? | Solution: The steps to follow are,
1. Copy the custom built UBML.DLL to the APS program directory. For example, on Windows XP platform, the program directory is,
C:\Program Files\AspenTech\Aspen Petroleum Scheduler
2. Run APS from this directory by double clicking the executable file 'ORION.EXE', or from Start | Programs | Aspen Tech | ...
APS will automatically link the UBML.DLL.
Keywords: UBML
ABML
UBML.DLL
custom
References: None |
Problem Statement: How does publish daily results work? | Solution: The daily results publisher calculates the daily results and writes them to the database so that they are available for reports. To run daily results publisher, use Simulator | Publish | Publish Daily Results. All results are contained in tables that start with _ (underscore), for example: _STRMS, _TANKS, _RECEIPTS, _RUNS, _PL, _CRDRUNS.
The _TANKS table contains daily results for tanks. Only tanks for which the PUBLISH field is non-zero are reported. For crude tanks, the total volume of the tank at the beginning of the day and the fraction of each crude are published. For process tanks, the volume of the tank at the beginning of the day and the tank properties are published.
The _STRMS table contains daily results for streams. Only streams for which the PUBLISH field is non-zero are reported.
The _RECIEPTS table contains daily crude receipts
The _PL table contains the content of each pipeline at the beginning of each day
The _CRDRUNS contains daily crude runs for each crude unit.
Keywords: Publish daily results
Tables with underscore
_TANKS
References: None |
Problem Statement: When going to the Pipeline Fill screen I can see that all my batches for my crude pipelines are of the same color. How can I correct this so I see differing colors? | Solution: For Crude Pipelines the color of the batch will be the color of the crude that has the most % in the batch.
So in the next screen I see that all my batches are colored blue:
This is because all of the batches have as majority crude, 4TS. In order to change the color of this crude we will need to go to Model/Material attributes and specify the color that we want to use:
So if I change the color from blue to green now I will see the next on my screen:
Keywords: crude, color, batches, pipeline fill screen, material attributes
References: None |
Problem Statement: Why is it that the values in table EV_QUALITIES differ from the SBO dialog? | Solution: The EV_QUALTIES and the AB_BLN_QUALITIES record blend data under different conditions and thus give different values. The help file says that the EV_QUALITIES table stores ‘normalized' header and tank specification data for each quality. The AB_BLN_QUALTIES data come from the SBO dialog box. Now let's consider this situation:
There are three component tanks A,B and C and a blend destination tank D. Let us say the blend event duration is from t1 to t2.
1. AB_BLN_QUALITIES DATA: The data is from single blend simulation. Records blend data at the start of the blend event (t1). At t1, the A,B,C component tank data are taken into consideration and the final blend data is calculated based on the data collected at t1. Any other events that may involve these tanks A,B,C,D from t1 to t2 are not considered here.
2. EV_QUALITIES: This table holds the overall simulation values, considering events of all units and tanks from the start to the end of the scheduling horizon. So if A,B,C or D were to be involved in any other activity between t1 and t2, those would also be taken into consideration in this table. Thus there could be a slight difference between these two tables.
Keywords: EV_QUALITIES, AB_BLN_QUALITIES, Aspen Petroleum Scheduler, SBO dialog,
References: None |
Problem Statement: The DMCplus controller turns off but the TDC controllers don't shed. | Solution: If the alarms status of the timer and on/off flag points are disabled, then the CL code will not shed the local PID controllers to the specified shed mode. They will time-out on their own at the time specified for each PID. This time is normally 2.5 times the DMCplus controller execution cycle. If it is less than 2.5 times, then the PID may shed on a single missed cycle.
Keywords:
References: None |
Problem Statement: Under certain conditions, the DMCplus controller could accept a -9999 (Cimio BAD flag) and be used as a valid value for an intermittent variable and base its online calculations on this. This could occur when the analyzer update logic erroneously sets the NEWPV=1 trigger. | Solution: In previous DMCplus versions AspenTechnology stressed the use of vetted logic calculations to trigger an analyzer update (i.e. setting NEWPV=1). Under certain conditions, it was difficult for some users to always guarantee that the correct logic is used. Thus, and additional validation check has been added to the online controller to reject BAD values as valid numbers.
Install Engineering Release 2 for AMS6.01 (seeSolution 114336).
Keywords: DMCplus
References: None |
Problem Statement: A user is unable to load a product receipt and product shipment event through EIU to an oracle database. The exact same test using Oracle export to Access, and the product event (shipment, receipt) loaded from EIU without a problem. | Solution: If we enter a valid TRAN_MODE in SHIP Via field, then the Product event can be audit successfully. See below the red circle area.
For Oracle, sqlldr is used to import data to database, and we set TRAN_MODE as the last column for importing ORION_MGR_EVENT_IMPORT. If the final field in a given row of your data file is unspecified (NULL), sqlldr will reject that records.
Keywords: EIU
Oracle
SQL
Receipt
Shipment
TRAN_MODE
References: None |
Problem Statement: Converting the CCF from CM50S to AppNode / OPC | Solution: Changes to be made to CCF for APP node
Change following line, BLDTEMP in Config section
.BLDTEMP~~~BUILD~~~CH(11)~~~DEFAULT.TCC~~~
to
.BLDTEMP~~~BUILD~~~CH(15)~~~TDC3KAPPWIZ.TCC~~~
Change CNC Device name
.CNCDEV~~~CONSTANT~~~CH(31)~~~IOCM50S~~~
should be changed to
.CNCDEV~~~CONSTANT~~~CH(31)~~~IOHCI~~~
Change CNCUNIT
.CNCUNIT~~~CONSTANT~~~I4~~~3~~~
should be changed to
.CNCUNIT~~~CONSTANT~~~I4~~~1~~~
Change ONSTS Flag point as follows:
.ONSTS~~~WRITE~~~I4~~~0~~~::VT701FLG.PV:ORD:
should be changed to
.ONSTS~~~WRITE~~~I4~~~0~~~::VT701FLG.PV.internal:ORD:
Change Timer related points as follows:
.STRTSTOP~~~AWRITE~~~I4~~~0~~~::UNT5TIMR.STRTSTOP:ORD:
should be changed to
.STRTSTOP~~~AWRITE~~~I4~~~0~~~::UNT5TIMR.STRTSTOP.internal:ORD:
.RESETCMD~~~AWRITE~~~I4~~~0~~~::UNT5TIMR.RESETCMD:ORD:
should be changed to
.RESETCMD~~~AWRITE~~~I4~~~0~~~::UNT5TIMR.RESETCMD.internal:ORD:
Additional Changes may be necessary depending on old ccf
CIMIO to APP node does not support smart data types CBST for LOOPSTATUS and CBAW for Anti-reset windup code (AWSCOD)
.AWSCOD~~~READ~~~I4~~~0~~~::FFY3307D.SP:CBAW: should be changed to
.AWSCOD~~~READ~~~I4~~~0~~~::WFY3307D.AWSCOD:DBVL:
And
.LOOPST~~~READ~~~I4~~~1~~~::FFY3307D.SP:CBST: should be changed to
.LOOPST~~~READ~~~I4~~~1~~~::WFY3307D.LOOPST:DBVL:
where WFY3307D is an (Honeywell) interface point that is generated by CCF2EB program.
If you are reading controller Mode in Calc section or otherwise, it must be changed as follows:
.FPC101_MODE~~~READ~~~R4~~~2.~~~::FPC101.MODE:DBVL: should be changed to
.FPC101_MODE~~~READ~~~R4~~~2.~~~::FPC101.MODE.internal:DBVL:
For IQ-Applications
.UPDSRCUSED~~~WRITE~~~I4~~~2~~~::KT2SDBNZ.SRCUSED:ORD:
should be changed to
.UPDSRCUSED~~~WRITE~~~I4~~~2~~~::KT2SDBNZ.SRCUSED.internal:ORD:
.NEWLABSTSNUM~~~WRITE~~~I4~~~2~~~::KT2SDBNZ.LABSTS:ORD:
should be changed to
.NEWLABSTSNUM~~~WRITE~~~I4~~~2~~~::KT2SDBNZ.LABSTS.internal:ORD:
Keywords: CCF, AppNode, DMCplus, OPC, Config
References: None |
Problem Statement: User can add an executable command with an argument in the 'User Tools' under Tools tab. But what is the syntax in APS or MBO? | Solution: The following screenshot shows the syntax you need to enter. In the 'Executable File' field, enter the executable file name with double quotes around it, then a space, following by the argument.
Keywords: Command
Command line
Execute
argument
User tools
References: None |
Problem Statement: Some entries in the CCF need to be changed from LOCAL to WRITE. The CCF is based on the template which can be modified. How do those changes get applied to the CCF? | Solution: You need to modify the template, update the CCF with the template and run the tag wizard again to substitute the correct tagnames. Here are the steps:
Save and / or copy your CCF before starting to modify.
Using Build, modify the template for the changes that are required. For future moves, you would change the LOCAL to WRITE and add the tagname entity syntax similar to other entries shown in the template. You will want to Save As for the modified template under a different name.
Using the Entry / Replacement editor in Build, use the Keyword options to change all entries in the necessary sections from LOCAL to from Template. You may have to run the Entry / Replacement editor again and have it change the Source as well. You will also need to run it one more time to change the tagname so that you get the ~tag from the template; you do this by changing the first three entries to not used and the fourth to template tag.
Now use the Tag Wizard to substitute tagnames for the ~, ^, and * characters.
The Entry / Replacement editor in Build is very flexible. You will find many options available for modifying the CCF with different kinds of changes to the template.
Keywords:
References: None |
Problem Statement: We have determined that installing DMCplus (offlines)
and Aspen IQ Config on the same machine will cause
DMCplus Build to crash when using the calc facility.
The symptom is that clicking on the Add button or any
part of the spreadsheet control in the Calculations -> Equations form of Build causes the following message
to appear:
Run-time error '380'
Clicking the OK button crashes Build.
The problem is that IQ Config installs a newer,
incompatible version of the file:
C:\WINNT\System32\SS32X25.OCX (on NT)
C:\Windows\System\SS32X25.OCX (on 95) | Solution: To identify which version of the file you have on your system, right-click on the file and select Properties. The following are the version numbers for the SS32X25.OCX installed by each product:
DMCplus Build: 2.5.020
IQ Config: 2.5.025
Our testing shows that IQ Config works with either version of the file. Only DMCplus Build relies on the older version.
To correct this problem, you must revert to the DMCplus version of the SS32X25.OCX file. You can do this in one of two ways:
Unzip the DMCplus version (version 2.5.020) of the OCX and copy it over the one that IQ Config installed (no registering is needed). You must not be running either application when you do this.
Contact ACO support for a copy via email or AT employees may check the AT network at:
Unity\\acodsw\DMCplus_110\patch files.
Delete the existing copy of the file and re-install DMCplus. Both applications must be stopped before doing this.
Keywords: DMCplus Build
References: None |
Problem Statement: How to handle DMCi - Smart DAta Types - MVTOL - DMC2CCF | Solution: DMCi controllers provide a Smart Data Type that has not been implemented yet in DMCplus. This is related to MVTOL and can result in lost moves if the MVTOL value in not manually entered in the new DMCplus CCF. The DMC2CCF utility does not move this value into the new CCF. This is used primarily as a tolerance about the limit to determine if the MV is within limits or should be down-graded to a feedforward.
Keywords: DMCplus, Smart, CCF, MVTOL, Data, Type
References: None |
Problem Statement: How to open an APS units file as a separate instance in Excel 2013 | Solution: There is a concerned setting in Excel for this issue that APS takes care of. In Excel Options -> Advanced Tab -> General Section, there is a setting called “Ignore Other Application that use Dynamic Data Exchange (DDE)”
This setting is toggled by APS through the CONFIG table with the keyword:
EXCEL_IGNORE_REMOTE_REQ = 'Y' or EXCEL_IGNORE_REMOTE_REQ = 'N' (Y-> Box is checked, N-> box is unchecked)
EXCEL_IGNORE_REMOTE_REQ is a configuration to decide whether we check DDE for the Excel instance APS is using on startup. This will decide whether other Excel workbooks are allowed to be attached to the APS Units Excel instance. When EXCEL_IGNORE_REMOTE_REQ = Y, other workbooks will not be attached to APS Units Excel instance. When EXCEL_IGNORE_REMOTE_REQ = N, other workbooks can be attached to APS Units Excel instance.
EXCEL_IGNORE_REMOTE_REQ = 'Y' in CONFIG table && DDE is off for new Excel instance
Please follow these steps to make the units file in Excel 2013 open as a separate instance
Do not open APS and make sure there's no Excel instance running. Open a blank Excel file and make sure DDE is off. If not, uncheck it and close.
1. Make sure EXCEL_IGNORE_REMOTE_REQ = 'Y' in CONFIG table.
2. Open APS and validate the DDE settings for Units Excel file. It has DDE checked. This is what we expected (owing to the entry in the CONFIG table)
3. Double click another Excel file to open. It can be successfully opened in a new instance. This is what we expected.
4. Close APS and the other Excel file you opened (You should make sure the DDE for the other Excel file is off before closing it in order to validate if APS is restoring the DDE when it is closed. Or you can close the other Excel file first and then close APS.) Check that DDE is off. So APS is restoring the original DDE status when it is closed as we expected.
Keywords: Excel 2013, Instance, Ignore Other Application that use Dynamic Data Exchange (DDE), EXCEL_IGNORE_REMOTE_REQ, table CONFIG
References: None |
Problem Statement: How has the behavior of splitters changed with respect to handling negative stream inputs? | Solution: The way that the splitter unit functions has changed in two ways since 8.4. Previously, the algorithm that the program followed while handling a splitter with a negative feed was different
Until 8.4, the output would be both negative and positive flow. The splitter, as of V8.4 will push all of the negative flow to the last stream out of the splitter.
To give an example, consider the splitter below, taking in feed F1 and yielding Y1, Y2 and Y3.
Until version 8.4:
In general a splitter with n-output streams will have n-1 parameters telling APS how much of the input should go to the first n-1 streams with the nth stream being a ‘balance’ stream that takes any leftover flow. Consider that the parameters of the splitter are -1, 0 (meaning we want Y1 to take all the feed)
The corresponding flow is seen below:
Until v8.4, as seen above, the older algorithm will send out the same negative feed to each yield stream until the stream the disposition asks it to go, and then the last stream will account for the material balance.
In the above case, the first stream is the one that it is designed to go, but see below what happens when the parameters of the splitter are 0,-1 (meaning we want Y2 to take all the feed)
Looking at the above flow, since the disposition was on the second stream, the negative feed is assigned to each stream till it reaches the stream that was designed to get the feed. So we have Y1=-30 and Y2=-30. The stream Y3 has a +30 to account for the material balance of -30.
After version 8.4:
As mentioned before, the splitter will now push all of the incoming negative flow to the balance stream of the splitter. Revisiting the two cases from above,
The parameters of the splitter are -1, 0 (meaning we want Y1 to take all the feed)
The corresponding flow is seen below:
The above would be the same even when the parameters are 0, -1. This is because the new algorithm just looks at the incoming feed and if it’s a negative value, it pushes it directly out to the last stream as seen above. This change can have a big effect while comparing streams between concerned versions, and this change has an amplified effect when there is a cascade of these yields through further splitter units downstream.
Keywords: splitters, negative streams, balance streams, mixer, APS
References: None |
Problem Statement: In the planning targets interface, where do the start and end dates come from? (These need not be confined to the model horizon) | Solution: These values are not part of any database tables and they come from the user registry:
You can find them as registry keywords “Start Date� and “End Date� in the following path in the registry:
HKEY_CURRENT_USER/SOFTWARE/ASPENTECH/ORION SCHEDULING/KPI SCREEN OPTIONS/<Model Name>/ ADMIN/Screen 1
( the last three fields are model specific)
An example is as below from the sample model
Keywords: Planning targets, User Registry
References: None |
Problem Statement: Which models can be chosen to model a heat exchanger in a Aspen HYSYS Dynamics simulation? | Solution: There are three models available in the Heat Exchanger unit operation (‘Dynamics’ tab, ‘Model’ page):
1-Basic: This model uses an Overall UA factor, which can reflect dependency on the flowrate, if reference values for this variable on the shell and tube side are introduced.
2-Intermediate: This model allows the introduction of information pertaining heat transfer in each of the spatial domains (area and transfer coefficient in shell and tube, as well as tube mass and specific heat).
3-Detailed: This model also requires the introduction of these individual phase coefficients, but incorporates more details regarding the exchanger configuration (namely, number of shell and tube passes, and zones per shell pass).
Keywords: Heat exchanger, Dynamics, Model, Detailed, Intermediate, Basic, UA
References: None |
Problem Statement: If I am undergoing the pipeline conversion wizard, and if there are no PLINV entries for a particular model start date, how does v8.7/v8.8 populate beginning linefills for a crude segment? | Solution: In v8.7/v8.8, we can see the beginning linefill of a crude pipeline from the Events-> Beginning Linefill -> Crudes.
The above dialog box has three major parts:
1. Batch number: There can be multiple batches (material movements) in the pipeline at a time
2. Crude Composition: What makes up this batch
3. Destination Tank: Which tank the batch is headed for
During DBupdate and using the pipeline conversion wizard, the PLINV entries are now used in a new table called PIPELINE_INV (PLINV is no longer used in the new pipeline system). However, if for the model start date, there are no entries in PLINV, the application will populate a beginning linefill (the assumption made in the new system is that pipeline are never empty). This is the behavior of the application for populating a crude pipeline beginning linefill, with no corresponding PLINV entry during the update.
1. BATCH: It creates one batch automatically, with
2. CRUDE: It picks the first crude alphabetically, that is active from t.CRUDES
3. DESTINATION TANK: It picks the LAST tank from t.TANKS, which has an unassigned location in the LOCATION_XSEQ field
Keywords: Beginning Linefill, Crude Pipeline, PLINV, PIPELINE_INV, CRUDES, TANKS, LOCATION_XSEQ
References: None |
Problem Statement: This message appears when trying to install Excel Integration Utility add-in:
File Interop.orion.dll has different computed hash than specified in manifest.
How can I resolve it? | Solution: Go to the registry, to the key HKEY_CURRENT_USER\Software\Microsoft\Office\Excel\Addins\AspenTech.APSMBO.ExcelIntegrationUtilities
Make sure that the Manifest string has the following value:
C:\Program Files (x86)\Common Files\AspenTech Shared\ExcelUtilities\AspenTech.APSMBO.ExcelIntegrationUtilities.vsto|vstolocal (for Windows 7, 64bit)
Also, the LoadBehavior string should have a value of 3.
Keywords: None
References: None |
Problem Statement: How are the tables PL_LOCATION and LOCATIONS used? | Solution: PL_LOCATION:
In v8.7 and beyond, when you create a tank from the Model->tanks window, you see an extra field named “Location”
These are pipeline locations (sources/destinations), which are defined in the “locations” dialog box, which can be found from model->pipeline->locations
Based on the tank's location (as defined in the model-> tanks dialog) and the definition of the pipeline route, we can schedule crude/process pipeline events with these tanks.
This event would read of a pipeline crude shipment between T012 which is in Houston and T257 in Mobile Alabama, using the pipeline route defined.
In the Tanks table, the LOCATION_XSEQ stores the location sequence number of the tank. These sequence numbers come from the table PL_LOCATIONS.
All locations related to the tank definition, will refer to the X_SEQ in the PL_LOCATIONS table
LOCATIONS:
This table is used in dock scheduling. In a dock screen, go to model-> locations. The locations dialog appears
These locations show up in the table LOCATIONS.
For instance, these locations are used to specify customer information (you can go to a dock screen, model->customers)
You see the locations defined in the table LOCATIONS. When you look at the database for these entries, they are found in two tables CUSTOMERS and CUSTOMERLOCATIONS, where LOCATION_XSEQ points to the table LOCATIONS and not PL_LOCATIONS
So LOCATION_XSEQ in dock scheduling related tables should point to T.LOCATIONS
Keywords: LOCATIONS, PL_LOCATION, LOCATION_XSEQ, Pipeline Scheduling, Dock Scheduling, CUSTOMERS, CUSTOMERLOCATIONS
References: None |
Problem Statement: The HELP facility in the Demand Manager Configured Application (DM CAP) environment can not be viewed in your Citrix Farm. The problem is that Citrix runs as a client application on the various servers in the FARM. As a result, a SYS command cannot be submitted to the server that contains an address of a module as a UNC (Universal Naming Convention), e.g. \\mimiserver5\mimischd\mimi\help\.
Issuing a command using SYS with a UNC gives the error of ILLEGAL OPERATION. Commands using SYS must be of the form SYS C:\....., using a drive letter assignment. | Solution: To resolve the problem:
On each server to be accessed in the Farm, create a directory C:\MIMI\ and placed the DMHELP\USEnglish folder in it.
In each (DM CAP), set the HELP entries in FMTFDEF to be C:\DMHELP\ AND C:\DMHELP\%1 respectively.
In the <MLOAD user MACRO for execution at case load time, enter the command CD C:\MIMI\
The DM CAPS will then access the HELP facility.
Keywords: Demand Manager
Help
References: None |
Problem Statement: Subcontrollers -
Difference between using one controller with several sub-controllers versus several smaller controllers. Subcontrollers in DMCplus, SONREQ, NGDSUB, Ramp Imbalance | Solution: Subcontrollers --
When asked about the differences between using one controller with several sub-controllers versus several smaller controllers. The best answer is that all constraints are considered at one time in the larger controller.
The sub-controllers are an easy way to logically group variables together for taking them out of the controller as a group. But the entire problem is solved together rather than in several, small independent controllers.
When the scope of the control problem gets even larger, the use of a CLP (composite linear program)Solution can be explored. This coordinates the LPSolutions of several large controllers.
Note 1: Sub-controller design should be reviewed with AspenTech control engineers, at least the first application, so that a thorough discussion of the pro's and con's can be had.
Note 2a: Also note that any ramp in any ON sub-controller will still trip the entire controller. Pseudo-ramps will trip the controller as well. An advantage of pseudo-ramps is that they trip only on number of imbalances, not severity of imbalance. A programmed imbalance ramp will trip on both number of imbalances and severity of imbalance.
Placing the ramp and it's associated MVs in a sub-controller allows that ramp to be excluded from the controller, but does not prevent that ramp from tripping the controller if the ramp is in service. In addition, it is important to ensure that either DMCplus or the regulatory loop or the operator are taking care of the ramp.
When more than one subcontroller includes ramps among its variables, it becomes a tricky isue to get the second and later subcontrollers with ramps turned on. The issue is that the controller must have enough of the variables to solve that ramp the first time it runs after the subcontroller is turned on or the entire controller will turn off. The DCS logic to switch the MVs into the correct mode must execute after the SONREQ is set and before the controller next runs on the host.
Note 2b: Subcontrollers can be very powerful for critical variable management but there are limitations with regard to ramp handling, prediction modeSolutions, etc. Threre are sufficient methods of alleviating most ramp imbalances such that those that do occur warrant operator intervention. It is believed that the advantages of large applications vastly outweigh the limitations
Note 3: The use of subcontrollers serves two purposes:
can exclude a ramp or other variable and still run remainder of controller
can logically group the variables that should be taken out of the controller at the same time, i.e., take the level out when the primary handles are taken out of the controller.
Beginning of document:
DMCplus allows a controller to be organized into sub-controllers. The purpose of a subcontroller is to allow a large controller to be partitioned into several small sections, which can improve service factors for a large controller.
Subcontrollers provide tools to manage controllers that include parallel trains where a single train may be out of service while the remainder of the process is operating.
Think of a subcontroller as a mechanism to take a pre-defined group of manipulated and controlled variables off-control simultaneously. It would be possible to take the variables in the subcontroller off-control one by one, but the order would be inconsistent, and it might take several controller cycles to accomplish the task.
The inconsistent order and multiple controller cycles can create unpredictable and undesirable results. Thus, subcontrollers were developed.
Subcontrollers allow for manipulated and controlled variables to be divided into engineering groupings, but they do not affect the control model or the method used to solve the control problem.
A controller with subcontrollers is still one big controller, it is not a collection of several small independent controllers. This is important to remember: it is still one controller, so all constraints are considered simultaneously.
There is an ON/OFF switch associated with each subcontroller, but this sub-ON/OFF switch only affects the status of the variables in that subcontroller when the controller is ON.
Regardless of subcontrollers, there is still one and only one master on/off switch for the controller. When the master switch is OFF, no setpoints will be sent to the field. When the master switch is ON, MVs in the subcontrollers that are ON will be active and get new setpoints/valve positions.
The operator can turn subcontrollers on or off independently of the main ON/OFF switch. The controller can turn a subcontroller OFF, but will not turn a subcontroller ON.
Turning on all the subcontrollers will not auto-ON the controller, the operator must still turn on the master switch. If the subcontroller is defined as critical, and the subcontroller is turned off by the operator or by the controller, the entire controller will be turned off.
When a subcontroller is turned off, the CVs in the subcontroller are shifted into a predict- only state. The CV prediction is maintained, and transformations and NEWPV flags are handled, but they are not part of the control problem.
The CVs are valid for the prediction module, but not included in the controller LP and dynamicSolution. When a subcontroller is turned off, the MVs in the subcontroller are treated as feedforward variables.
All MVs must appear in one subcontroller only. All CVs must appear in at least one subcontroller; CVs can appear in more than one subcontroller.
Note that a critical variable is critical only for the subcontroller. When a critical variable is BAD, all subcontrollers associated with the variable will be turned OFF. If the critical variable is critical for the entire controller and appears in only one subcontroller, then that subcontroller must be critical.
If a critical CV is critical for the entire controller and appears in more than one subcontroller, then either one of those subcontrollers must be critical. Or, the minimum number of good subcontrollers or process handles outside of DMCplus must be used to ensure that when the critical CV is invalid, the controller will be turned off.
Example: Subcontrollers
There are two main sections in the process coupled through an intermediate inventory tank. There are constraints in both sections of the process. The process objective is to maximize production.
The first section of the process is configured as one subcontroller, while the second section of the process is configured as another subcontroller. The intermediate inventory tank level appears in both subcontrollers.
The tank level is a critical CV in each subcontroller. Neither subcontroller is critical, and the minimum number of good subcontrollers required is one.
It is possible to run the controller in three configurations:
Both subcontrollers ON
Front end subcontroller ON, Back end subcontroller OFF. The controller will maximize feed to the front end, subject to the tank level constraint that is set by the flow to the back end.
Back end subcontroller ON, Front end subcontroller OFF. The controller will maximize production (feed to the back end), subject to the tank level constraint that is set by the feed to the front end.
Obviously, the first scenario is the most profitable. But if there are equipment problems that prevent this mode of operation, it is possible to capture the benefits associated with the portion of the controller that can be operated with the other two modes of operation.
In the first scenario, under normal conditions it should be possible for the controller to balance the tank level.
In the second and third scenarios, it may not be possible for the controller to balance the tank level. The controller turns itself off because there would not be enough good subcontrollers to keep the controller running.
Additions to documentation: (Oct 31, 1997)
Are there SONSTS and SONREQ parameters like the main controller? If so, are they documented?
* * The only on/off switch for each subcontroller is SONREQ. The master ON/OFF result indicator (ONSTS) is not affected by turning SONREQ to OFF unless it is a critical sub-controller (SUBCRT=1).
When a first subcontroller is turned on, is the main controller automatically turned on to start the main controller or is it necessary for the interface to turn on ONREQ?
* * The main controller does not come on conditionally. The main controller on request flag ONREQ must be set through the interface.
Can a subcontroller shut itself off without the whole controller shutting off?
* * Yes, if the subcontroller does not turn off due to a critical variable going bad, a critical variable turned off or a ramp imbalance situation.
If the last subcontroller is turned off or shuts itself off, will the main controller shut itself off? Will it shut off ONSTS?
* * Yes, if a violation of the minimum number of good MVs NGDMAN, dependents NGDDEP, or subcontrollers NGDSUB occurs.
Keywords: Subcontrollers, DMCplus, SONREQ, NGDSUB, Ramp, Imbalance
References: None |
Problem Statement: Applicable Versions: DMCplus 1.10.01 (fixed in 1.11)
It is reported that simulation won't run with the following in the CCF:
.XFORM~~~XFORM~~~CH(25)~~~SHIFTRATEPOWER 0 10000. 1~~~
It has been confirmed that code for Simulate and Control takes the log of the power term ( 1 above) and aborts.
The users' options are to scale the value at the DCS, use CCF calcs on all vars requiring transformation or use the Piece-Wise Linear transform. | Solution:
Keywords:
References: None |
Problem Statement: Client is running DMCplus controllers on multiple nodes and has another AspenTech database product on one of these nodes. Engineer wants to see all controllers and database information. | Solution: For two controllers on two nodes, it is required that you run two Opcon sessions that have independent desktop short-cuts that point to separate display directories. The displays in each directory must all use the same node. The node configuration from each directory will use the appropriate node.
For one controller and one database product, you can use the same Opcon session. Put the DMCplus displays in the same directory with those for that database. Configure two nodes: one for the database and a second for DMCplus, even if both are on the same host. You will need to select the DMCplus display files and change the node to the correct one. Add a button somewhere in the existing database display system to access dmcpbase.
Keywords: DMCplus, view, GCS, OPCON
References: None |
Problem Statement: What can cause APS with an Oracle database to crash when changing model settings? | Solution: This problem is usually caused by inconsistent naming conventions between APS and Oracle user name / password. For example if the user name is created in capital letters, then the Oracle login should be also in capital letters, otherwise APS may crash on the model settings dialog box.
If APS user name is ADMIN:
then Oracle login and password should be upper case as well (ORION).
Keywords: APS
Oracle
crash
References: None |
Problem Statement: In the published tables _STRMS and _ZSTRMS the crude unit feed stream volume varies by a small amount from the scheduled rate. The _ZSTRMS table has the correct volume for all the periods in the day but the _STRMS table shows a volume that is lower. It is always lower and this anomaly only happens on the first day that the model is published. So if you publish from dates 6/20 - 6/30 the crude feed volume for 6/20 will not match in the publish tables but for 6/21 - 6/30 the volume will match. What is the reason for this? | Solution: In Aspen Petroleum Scheduler (APS), the first 10 seconds are not part of the daily values. This accounts for the difference between the _ZSTREAMS and _STRMS values.
To verify this we can do a percentage difference between the two tables _STRMS and _ZSTRMS for the first day of the simulation period.
For example, in one of the sample models, the following were the volumes for a crude feed stream on the first day of the model horizon
_ZSTRMS value: 218.448
_STRMS value: 218.422
% difference between the tables ~0.0116
% of 10 seconds in a day is (10/86400) *100 ~ 0.0116
This would account for the small difference in the volume entries between the two publish tables.
Keywords: _ZSTRMS, _STRMS, simulator, publish tables
References: None |
Problem Statement: OpenVMS systems
When it is possible to load a small controller but trying to load a larger controller gives the following message:
Loading controller xyz from CCF xyz1
Scratch OTF allocation failed.
Failure creating context header
CTL: Error Loading Controller
Control load failed for controller: xyz, CCF xyz1
Go to sys$system and run authorize, show user_name. User pgflquo is often in the area of ~50,000 , increase it to 250,000. Logout and in and user should be able to get the load to proceed to give validation errors, which is what is what is expected for a first load.
Honeywell AxM
from a re-load on a running controller:
scratch otf allocation failed
failure creating context header
ctl: error loading controller
from a first time start via Manage:
can't become daemon (exited Manage and re-started aok)
The client found that his second swap file had been deleted.
Restore second swap file, for total swap file size = 2 * physical memory | Solution:
Keywords:
References: None |
Problem Statement: Is Windows 2003 R2 supported for Advanced Process Control applications | Solution: The following is an explanation of the versions of the Windows 2003 server, confirming that the Windows 2003 R2 is appropriate for versions of Aspen Process Control software prior to version 2006.5. Windows 2003, SP2 is supported on version 2006.5 as well as the previous versions of the 2003 server.
Here is a brief explanation of the Windows 2003 release numbers:
Windows Server 2003 was the original release, the successor to Windows 2000 Server.
Windows Server 2003 SP1 is a hotfix rollup that can be applied to Windows Server 2003.
Windows Server 2003 R2 is an extended version of Windows Server 2003 with SP1 already applied. It added a few new features (e.g., branch office management, identity access and management, and storage management) and merged some that were previously available as add-ons (e.g., Windows Services for UNIX) into a new distribution. I really wish they had named Windows Server 2003 R2 something else. It is really a package that includes Windows Server 2003 SP1 and some additional features. You don't have to apply Windows Server 2003 SP1 to Windows Server 2003 R2.
Windows Server 2003 SP2 implements security and administration changes that were promised as part of the Microsoft Trustworthy Computing initiative. It is not a mandatory update or a hotfix rollup the way SP1 was. It does change the way certain basic OS features function. Hence, applications that ran on Windows Server 2003 SP1 or Windows Server 2003 R2 may not run properly after Windows Server 2003 SP2 is applied.
Windows Server 2003 SP2 can be applied to all Windows Server 2003 systems and Windows Server 2003 R2 systems, as well as Windows XP Professional x64 Edition systems. Windows Server 2003 SP1 does not have to be installed first before installing Windows Server 2003 SP2. The updates are cumulative.
Keywords: None
References: None |
Problem Statement: and | Solution: Standard DMCplus Implementation
In general, implementing DMCplus controllers via a Cim-IO connection to the DCS allows the controller to take advantage of the smart data types in Cim-IO and the controller logic for outputting to PID controllers only when appropriate. Mode-change logic and operator schematics are custom in this case. The insertion of a database product between the DMCplus controller and the DCS will introduce the need for additional custom work and the project will need to account for this in cost and schedule. There are three topics covered in some detail:
A. Standard DMCplus Implementation Platforms (TAN 4424)
B. DMCplus Implementation with Database (TAN 4141)
C. Customization for typical DMCplus Implementation (TAN 4411)
All of the issues discussed in part C are covered under the standard implementations discussed in part A. For Cim-IO servers where the smart data types for PID loop status and windup status are not available, then additional customization will be required, as discussed in part C below. Careful consideration of the implementation platform is required at the proposal stage to ensure that the project is not started with unrealistic budget / time expectations.
A. Standard DMCplus Implementation Platforms
=====================================================================================
DMCplus is designed for a Cim-IO connection to a DCS system. It contains numerous checks for input validity and manipulated variable status. The controller determines which of the manipulated variables should be written at each cycle. Some of the Cim-IO servers provide smart data types for the PID mode and windup status while the controller itself determines when to write setpoints out to the DCS. Operator graphics and PID mode changes for the set of manipulated variables are generally custom. For two common DCS platforms, there are standard implementation environments available that take care of these details as well. These are:
Honeywell TDC3000 (provided by AspenTech)
Foxboro I/A (from Foxboro via the DMCplus Bridge)
For clients with either of these DCS systems, the most cost-effective DMCplus implementation is via these direct Cim-IO methods. For other systems, this is also generally true, but some customization for operator schematics, mode change logic, setpoint outputting logic and other miscellaneous details will be required. The project (AspenTech or client) will bear the cost of this customization. Services of the DCS Implementation Group in APC Development may be required for these customizations.
There may be compelling technical reasons for the introduction of IP.21 or IP-X as an intermediate between the DMCplus controllers and the DCS. When this occurs, the standard implementations for the systems listed above are not available and significant custom work will be required. Additional time/money should be allocated in the project budget for these issues (current estimate is 2 - 5 man-weeks first controller, 1 man-week per controller thereafter). Services of the DCS Implementation Group in APC Development may be required for these customizations, as well. Some examples of compelling technical reasons:
hardware limitations on data transfer (alternative is to buy hardware to circumvent limit)
need for complex calculations and / or scheduling of DMCplus with external calculations
DCS difficult to modify or has no scripting language available
Keywords:
References: None |
Problem Statement: What tables are considered as Ancillary Tables in an Aspen Petroleum Scheduler model? | Solution: While archiving an Aspen Petroleum Scheduler model the users has the option to either include or exclude ancillary files as shown in Figure
The Archive Model feature uses the OrionDBGen.mdb to identify whether it is an Ancillary table or not. When the Include ancillary tables option is not checked during the archiving process, we get all the Orion tables which are listed in TableNames of OrionDBGen.
The tables that are not listed in TableNames of OrionDBGen will be considered as ancillary tables.
Keywords: Ancillary table
Archive model
References: None |
Problem Statement: Cl link error
Error message from linking the PCLS0115 to sub point. ?INVALID TO INDIRECT THROUGH OFF NODE POINT? | Solution: The SUB point cannot be on a different unit than MAIN, FLAG, and TIMER. Make sure the SUB point is on the same unit as MAIN, FLAG, and TIMER. SUCCESS.
The 1st thought is an MV problem. But we checked the SUB, MAIN, FLAG, and TIMER. We also double checked the UCN in FULL control, PM box in FULL control, MVs configured properly (RCASOPT).
Removed all but 1 or 2 MVs then relinked, same problem,
Noticed that the SUB point was in different unit than MAIN, FLAG, and TIMER. So we moved SUB point to same unit as MAIN, FLAG, and TIMER. SUCCESS.
We tested again to be sure that the unit was the problem and not just rebuilding the SUB point. Created dummy SUB point with same info as real, and placed into a different unit. When linking we got the same error as before. Fortunately we usually do not place these in different units because the CCF2EB puts all DMCplus interface points in the same unit.
Keywords: DMCplus, TDC3000, CL
References: None |
Problem Statement: How to install the v1.3 CIM-IO for OpenDDA, CimDDA | Solution: Here are the instructions and the zip file.
First unzip the file on the PC and transfer the kit_cimdda_1.3.a to the AxM
using a binary file transfer. Place it in a
temporary directory like /tmpdmc then:
Login to the dmc account. Stop all dmcplus controllers and shutdown cimdda
and cimio.
$ $ACODHOME/acod_shutdown
$ $CIMIOCOM/iodda_stop.sh
$ $CIMIOCOM/cimio_k_stop.sh
Switch to super user
$ su
Start a korn shell, extract and run the installation script
$ ksh
$ cd /tmpdmc
$ ar -xv kit_cimdda_1.3.a
$ . ./inst_cimdda.sh
Edit the configuration file to disable CIO files.
$ cd $CIMIOROOT/io/dda/etc
$ vi io_dda.cfg
set the NUM_CIO_TASKS=0;
If you are starting cimio from the dmc account, reset the cimio file ownership to dmc, otherwise leave the ownership the same.
$ cd $CIMIOROOT
$ chown -R dmc:axm *
exit from the root and restart the cimio kernel and iodda. If you normally start the cimio software from the root don't exit until after you've started cimio and iodda.
$ exit
$ $CIMIOCOM/cimio_k_start.sh
$ $CIMIOCOM/iodda_start.sh
Restart the controllers from manage.
Keywords:
References: None |
Problem Statement: Standard Named Transform, Piece-Wise Linear, PWLN | Solution: Must be valid transformations across a range that includes eng units plus SSSTEP
The Standard Named Transform, Piece-Wise Linear, must provide valid transformations across a range that includes the engineering limits plus SSSTEP so that the LP can always get valid numbers.
Model Bug (v1.10.04)
In onl\xfm_pwln.c, the following lines appear in the
transform function and in the anti-transform function...
if ((bxAscending && fValue <= pfParam[ix + 2]) ||
(!bxAscending && fValue >= pfParam[ix]))
It should be:
if ((bxAscending && fValue <= pfParam[ix + 2]) ||
(!bxAscending && fValue >= pfParam[ix + 2])) /* fixed */
The effect of the bug is to choose the wrong interpolation segment if the X's in the list of (X,Y) pairs are decreasing in value.
There is no pblm if the X's are increasing in value (which is the more common case.)
This problem doesn't occur in the onlines. The onlines still use the FORTRAN routines. (We can meld them in 1.2 :-)
Keywords: Transform, Piece-Wise, Linear, PWLN, SSSTEP, LP
References: None |
Problem Statement: In steady state simulations, the Balance operations are a quick, easy way of transferring/duplicating stream conditions from one stream to another whilst maintaining the information link.
However, the Dynamics environment does not allow the Balance object. | Solution: In order to maintain simulation functionality when switching to the Dynamics environment, it is necessary to use an alternative to the Balance object. The Spreadsheet object can accomplish this. Relevant process conditions from the source stream(s) can be imported into spreadsheet and then exported to the suitable stream(s).
The table below illustrates the equivalent spreadsheet import/export for a particular Balance parameter:
BALANCE Parameter
SPREADSHEET Import Variable(s)
Component Molar Flow
Master Comp Molar Flow variable for EACH component
Mass Flow
Mass Flow
Heat Flow
Heat Flow
Component Mole and Heat Flow
Master Comp Molar Flow variable for EACH component
AND
Heat Flow
Mass and Heat Flow
Mass Flow
AND
Heat Flow
General
Relevant Molar/Mass/Volume flow(s)
Calculation as per defined ratio
The relevant cells can then be exported to the appropriate stream conditions.
Keywords: Balance, Spreadsheet, Transfer, Duplicate, Steady state to dynamic,
References: None |
Problem Statement: GCS is not working with DMCplus 6.0.1 | Solution: GCS is in the process of being replaced by the Production Control WebServer. The dmcpview.exe is provided in the aconline/bin directory, however it is no longer started automatically. Users of GCS should add the following lines to the autostartlist.txt which can be found in the aconline/sys/etc directory.
Users will see that the GCS displays are no longer provided in the aconline/viewdisp directory, the displays will have to be retrieved from the previous version installation and then revalidated for use with upgraded software.
PROCNAME=BGCSNET
SEQNUM=79
APPPATH=$ACODHOME\bin\bgcsnet.exe
ARGLIST=gcsdmcp C:\Program Files\AspenTech\AC Online\bin\dmcpview.exe APPTYPE=0 APPTYPE=0
Keywords:
References: None |
Problem Statement: The DMCplus Bridge may use a slot in the OM's Import Table for each MV in each controller and the DMCplus Bridge sizing guidelines do not discuss how to address this issue.
This can lead to a full Import Table and, therefore, increased broadcast traffic.
Discussion:
The DMCplus Bridge sizing guidelines can be found in B0193VX in Section
The electronic version of the DMCplus Bridge manual Version 2.5 or later (November 98) does contain a discussion of this sizing issue. Since this manual is the one shipped with the product, future installations should have the required information.
Actions:
See if the sight has a problem by using the show_params command. Look at the OM_NUM_IMPORT_VARS line. A sample printout follows:
1AW51C# show_params
Current Station Configuration FOX CONFIG 001.00
Number
Free
In-use
CMX_NUM_CONNECTIONS
96
94
2
# of connections
URFS_NUM_CONNECTIONS
96
96
0
# of connections used by RFS
OM_NUM_OBJECTS
20000
19508
492
# of OM object variables
OM_NUM_CONNECTIONS
90
90
0
# of connections used by OM
OM_NUM_IMPORT_VARS
100
98
2
# of OM import variables
OM_NUM_LOCAL_OPEN_LISTS
200
199
1
# of OM local open lists
OM_NUM_REMOTE_OPEN_LISTS
50
50
0
# of OM remote open lists
IPC_NUM_CONN_PROCS
96
87
9
# of IPC connection users
IPC_NUM_CONNLESS_PROCS
150
74
76
# of IPC connectionless users
Apps
Objs
Attrs
APPLICATION OBJECTS
1AW51C#
0
0
0
# of Application Objects
The line of interest is underscored. If the Free column is near or at zero. The table should be resized.
The I/A Series Manual B0193MJ describes how to increase the size of the Import Table. | Solution:
Keywords:
References: None |
Problem Statement: Add Custom Params to LCN | Solution: The general process to add custom parameters to the LCN points is to:
Create new tuning CDS packages for the independent and dependent points containing all the custom parameters. Do not add to the existing DMCplus packages so that these remain standard for the next upgrade.
Attach these packages to the correct associated data points. A TDC point can hold up to 10 CDS packages.
Create displays to update the custom parameters on the LCN.
Update the DMCplus BUILD template file to point to these custom parameters on the LCN rather than keeping them as local variables.
Create a new CCF with the modified template.
Keywords: LCN, CDS, BUILD
References: None |
Problem Statement: Under certain conditions the DMCplus controller would implement controller moves
exceeding MAXMOV by small amounts.
This issue was reported in AMS5.0 and traced to be a combination of selecting internal move accumulation (INTSUM=1), but failing to provide the necessary MV tolerance (MVTOL=0). A logic check was added to Service Pack 1 for AMS5.0 to resolve this.
A more subtle issue was reported in AMS6.01 (that is AMS6.0 with SP1). In this particular case, the controller might try to implement a move exceeding MAXMOV due to a discrepance in VIND and VINDSP, having selected INTSUM=1, and trying to go closed-loop. The logic check and its location in the code was modified to address this issue. | Solution: For AMS6.01, install the Engineering Release2 (Solution 114336).
Keywords: DMCplus
References: None |
Problem Statement: Testing the Cim-IO installation. Reads work AOK, writes don't. Error is facility 17, status return 393322 which means that the AxM doesn't have access to write to the AM side. | Solution: Go to TDC and change parameter to enable AxM to have write access to the AM side.
Use PERFMENU, DATACHNG and enter $PRTSTSnn.$XACCESS where nn is the AxM node number. Change the status from READONLY to READWRIT.
Keywords: DMCplus CIM-IO, AXM
References: None |
Problem Statement: Changes made to CCF configuration section show no apparent affect on re-load. | Solution: Configuration section changes require that the controller be stopped, un-loaded and then loaded and the controller restarted.
Keywords: CCF, reload
References: None |
Problem Statement: What is the use of the new WHOLECRUDEPROPS table in v7.3?
How is it different from the old CRDPROPS table?
How does APS use this database table vs the Excel table?
Why do we need the data in both places (actually three places Excel, WHOLECRUDEPROPS, and CRDPROPS)? | Solution: WHOLECRUDEPROPS table is used for the new functionality in V7.3 called Passing ASSAYS into APS. There is a mapping worksheet (Excel Integration Utility template-EIU) with same name that it is required for loading of assay information because it maps crude codes and values from Aspen PIMS to Aspen Petroleum Scheduler.
Once you put information on the WholeCrudeProps worksheet, and select the Audit Inventory option from the corresponding template, select Integration | PIMS to APS | Import Assay in APS, this information will be showing at WholeCrudeProps tab (Import assay dialog box).
After you apply the Import option on the WholeCrudeProps tab, the information will be saved on WHOLECRUDEPROPS and CRDPROPS tables.
The WHOLECRUDEPROPS table is a staging table for the import of assays (as are all of the _XREF tables). The Excel file has these tables for initial import and for updating the mapping or whole crude props when necessary. When importing from Excel (EIU) you are given the option to load the mapping data or not. If you have no updates to the mapping then you would just skip the step.
More details related to this matter can be found in the APS help file, item called ASSAY WORKSHEET
Keywords: -Passing Assays from PIMS to APS
-Excel Integration Utility (EIU)
References: None |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.