question
stringlengths 19
6.88k
| answer
stringlengths 38
33.3k
|
|---|---|
Problem Statement: What is the property GAMPC? | Solution: The property GAMPC refers to the liquid activity coefficient pressure correction and is often mistaken for the Poynting factor, which is the property PHIPC.
The built-in model in Aspen Plus for GAMPC returns 1.0 and there are no other models available. This is supposed to be a pressure correction term to activity coefficient. However, in practice it has not been used. GAMPC existed since the beginning of the Aspen project (mid 1970s), and is kept for version compatibility.
Keywords: GAMPC, pressure correction, fugacity
References: None |
Problem Statement: How are the allowable inlet and tail pipe pressure relief losses calculated? | Solution: In most cases, the allowable loss will be set to the percent of differential pressure specified in the design rules. This will typically be 3% and 10% of differential setpoint for inlet and tail piping, respectively. However, if the vessel pressure for dynamic runs or the inlet pressure for steady state runs does not reach 10% overpressure, the allowable losses will be lower and calculated according to the following equations:
Allowable Inlet Pipe Pressure Loss
Allowable Loss= DSP*Factor*(VP - Patm)/(1.1*(DSP+DischP-Patm))
VP= Vessel or inlet pressure closest to 10% overpressure Factor= Allowable percentage specified in the design rules DSP= Specified differential set pressure
Patm= Atmospheric pressure (14.696 psi)
DischP= Specified discharge pressure
Allowable Built-Up Back Pressure
Allowable= DSP*Factor*{(VP - Patm)/(1.1*(DSP+DischP-Patm))}**2.
VP= Vessel pressure closest to 10% overpressure
Factor= Allowable percentage specified in the design rules DSP= Differential set pressure
Patm= Atmospheric pressure (14.696 psi)
DischP= Specified discharge pressure
Keywords: Pressure Relief, allowable inlet pipe pressure drop, allowable built-up back pressure
References: None |
Problem Statement: What can the Swing Cut utility be used for? | Solution: The Swing Cut utility, which is used in conjunction with the Petroleum Distillation column, allows you to generate and export assay tables with user-specified swing cuts in PIMS format. Aspen PIMS (Process Industry Modeling System) is a production planning and optimization tool that is widely used in the refinery industry. It allows you to determine the best operating conditions at minimum cost using Linear Programming (LP) and simplified assumptions. The Swing Cut utility provides tighter integration between Aspen RefSYS and Aspen PIMS to achieve a wider refinery modelingSolution.
LP crude assays consist of yields and properties of heart cuts and swing cuts. Heart cuts refer to material that must always be allocated to a given refinery stream. For example, the kerosene heart cut is material that will always be taken from the crude column kerosene draw. Swing cuts represent material that can be allocated to two adjacent crude column draws. For example, the naphtha/kerosene swing cut is material that can be drawn out of the crude column with the naphtha stream, or kerosene stream. The allocation of the swing cut is determined by the cut point on the actual crude column; this is set by specifying the TBPs.
Keywords: swing, cut, utility, allocation, PIMS
References: None |
Problem Statement: How smoke point is calculated in Aspen RefSYS? | Solution: Smoke point for a stream is a calculated based on the stream blending rule provided below:
SPidx = Sum [ Vol(i) / SP(i) ]
SP = 1 / SPidx
where:
SP(i) =Smoke Point of Component i
Vol(i) =Liquid Volume Fraction of Component i
SPidx = Smoke Point Index of Stream
SP = Smoke Point of Stream
For more information refer to the Aspen RefSYS 2006.5 Options guide. This document can be accessed online via the DOCUMENTATION link, and is also available on the documentation DVD provided with the software.
Keywords: smoke, point, petroleum, property
References: None |
Problem Statement: In doing a RBatch simulation with a vent stream, the simulation seems to be taking a very long time. The user is not sure wheather it has 'frozen' or is it simply integrating very slowly. | Solution: Increasing the diagnostics will print a one line integration diagnostic in the Control Panel.
In Aspen Plus 10, increase the On Screen diagnostics level to 5 from its default value of 4 on the RBatch / Block Options / Diagnostics sheet. In Aspen Plus 9, increase the Term-level value from 4 to 5 on the RBATCH.Blockops form.
E.g:
-->Calculation begin.....
Block:B1
Model: RBATCH
* 1.000D-01 SEC
2.932D+02 K 3.1234D+03 N/M2
1.4OOOD+4
KG
If the vent is switching on and off use the new vent algorithm. To use the improved Vent algorithm in RBatch, specify the reactor volume and the vent opening pressure to calculate the reactor pressure rather than specifying reactor pressure.
Keywords:
References: None |
Problem Statement: How is the accuracy and reliability of the pressure relief calculations in Aspen Plus? Do the Aspen Plus calculations meet all of the benchmark calculations for valves and flow through pipe sections? | Solution: The pressure relief system calculations meet all the benchmarks. Initially in version 9.2, there were some cases where the benchmarks were not in agreement. These cases contained a vertical outlet pipe because the pressure loss due to elevation was included in calculating the total pressure loss. This is not in conformance with DIERS rules, which ignores that part of the pressure drop since it should be recoverable. This was changed in the code and now Aspen Plus is congruent with the DIERS benchmarks.
There are some limitations to the Pressure Relief calculations in Aspen Plus; therefore, Aspen Dynamics, which is also in complience with the DIERS benchmarks, is recommended.
Keywords: Pressure Relief
Valve
Pipe
References: None |
Problem Statement: What is the need of APED component for the use of Aspen HYSYS Petroleum Refining? | Solution: All the reactor model in Aspen HYSYS Petroleum Refining needs Equation oriented solver (EO) to solve. At present EO models can be solved only through Aspen Properties, in Aspen HYSYS. This requires Aspen Properties Enterprise Database APED (default properties database used by Aspen Properties) to be installed.
Keywords: Aspen HYSYS Petroleum Refining, Equation oriented(EO), Aspen Properties Enterprise Database APED
References: None |
Problem Statement: How do I model a supercritical component like N2 at ambient conditions in a mixture that is not supercritical (such as water or methanol)? | Solution: You may take one of two approaches:
Use an equation of state to model the component. If binary interaction parameters are not present in the data base, then a data regression (DRS) run with PTXY data will be needed to get a good representation of the data. To do this regress the binary interaction parameter (for example RKSKIJ, PRKIJ parameters). Examples of property methoods that use equation of states are RK-SOAVE, and PENG-ROB. The problem is that most of the standard equations of state do not model polar mixtures well. The equations of state with mixing rules such as PSRK can handle polar mixtures more accurately.
Use an activity coefficient model. Light gases (such as N2) are supercritical at ambient conditions and will have to be modeled as a Henry component. To model a light gas a Henry's component, specify the component or components in a Henry Comps group on the Components\Henry Comps form, and then specify that group on the Properties\Specifications form. Henry's constants (HENRY, a temperature dependent binary parameter) will be needed. In Aspen Plus 9 and higher there are binary interaction parameters for many light gases with water and non-water solvents.
You can regress solubility data for the Henry's constants if they are not available. SeeSolution 106147 for an example.
Keywords: henry
supercritical
light gas
activity coefficient gamma model
References: None |
Problem Statement: What can the Petroleum Assay utility be used for? | Solution: The Petroleum Assay utility, which is used in conjunction with characterized assays from the Petroleum Assay, allows you to obtain the results of a laboratory style analysis for your simulation streams. The object for the analysis can be a material stream or a stream phase in any stage of a tray section, separator, condenser or reboiler. Simulated distillation data including TBP, ASTM D86, D2887, D1160(Vac), and D1160(Atm), as well as petroleum properties for each cut point are calculated. The data in the utility can be viewed in a tabular format or graphically.
Note that the Petroleum Assay utility is only available for use when you have added a petroleum assay in the simulation environment. Additionally, when there is a petroleum assay in the simulation environment, the Boiling Point Curves utility will be unavailable.
Keywords: petroleum, assay, BP, curves, utility, allocation, PIMS
References: None |
Problem Statement: It is possible to vary the Feed Rate in a RadFrac Design Spec and Vary. The Feed Rate variables is hardwired to be mole-flow. Is it possible to enter the flow in units of mass-flow. | Solution: Unfortuntely, this can not be done in the current system since the block does not have the information about the basis of the stream flow. Mole-flow units must be used.
Keywords: radfrac
FEED-FLOW
References: None |
Problem Statement: Problem running HYSYS Pipesegment in Aspen HYSYS Dynamics | Solution: The feed and product nozzle objects information does get lost (i.e. empty) and it causes the crash or P/F solver fails, when initializing the segment. This may happen if you have multiple segment added, and you tried to delete them. Or you have feed stream / product stream and it got deleted and you added a new stream.
Wordaround:
1. Go to Dynamics Tab and change the radio button for Pipe holdup type to other option and return back.
The empty feed/product nozzles can be recovered by selecting Pipe Holdup Type (Dynamics tab) to one dp calc/pipe and just selecting back to one dp calc/segment. Then you can run the pipe segment model without any problem.
Keywords: Pipe, segment, dynamics
References: None |
Problem Statement: In the Naphtha Reformer environment / Operation tab / Advanced, there are three factors listed under Activity Profile Constants - what do these three factors do ? When might we have to adjust them? | Solution: The Activity Profile Constants enable you to adjust the activity distribution through a reactor. The only real valid reason for adjusting the activity distribution is to match measured internal reactor temperatures. Great care must be taken in this effort to insure that the exact locations of internal thermocouples are known. It is not recommended to tune to internal temperatures unless duplicate thermocouples are located in the beds at the same levels but in different quadrants of the reactor.
Within the reactor the activity profile is calculated as follows:
Activity Profile = Activity Intercept + Activity Coefficient x exp (Activity Distribution Exponent x ZLOC)
where:
Activity Intercept = Minimum Activity + COC Response Factor * Minimum Activity / Exp (AvgCOC)
Activity Coefficient = COC Response Factor*(1-Minimum Activity)*Distribution Factor/(1-Exp(-Distribution Factor))
Activity Distribution Exponent = Distribution Factor
ZLOC is the dimensionless length (from zero to 1)
Minimum Activity: is the minimum activity of the catalyst
COC Response Factor: is the coke on catalyst response factor
AvgCOC: is the average coke on catalyst for the reactor bed.
The default values of 1 for Minimum Activity, 0 for COC response Factor, and 1 for Distribution Factor yields a flat activity profile so that the whole reactor bed has the same activity.
Keywords: naphtha reformer, activity profile
References: None |
Problem Statement: Why is the Boiling Point Curves utility not listed?
Boiling Point Curves utility is missing from list of utilities. How do I bring it back? | Solution: Petroleum Assay Utility in RefSYS (Aspen HYSYS Petroleum Refining) is equivalent to Boiling Point Curves utility in HYSYS. When you use RefSYS, Boiling Point Curves utility is removed from the list of utilities. It is not a defect. It is removed by design.
If you rather want to use Boiling Point Curve utility, you can use the boiling point curve extension unit operation fromSolution document 110062.
If you want to get rid of RefSys and want Boiling Point Curve utility back, you need to do the following.
1. Delete all RefSYS assays.
2. Delete all Petroleum Assay utilities (if used).
3. Delete all RefSYS unit operations.
4. Save the file.
5. Close all instances of HYSYS to make sure that RefSYS is not even running in a parallel instance.
6. Open the file save in step 4. Boiling Point Curves utility should be available now to add.
Keywords: Boiling Point Curve utility, missing, not listed.
References: None |
Problem Statement: What are some publications that discuss simulation and troubleshooting? | Solution: Listed below are some publications that are available from the literature.
Constant Volume problems
Horwitz, B. A., The Case of the Collapsing Can, Chem. Engr. Prog. June 1997, pp 61-65.
Distillation
Kister, H.Z. Troubleshooting Distillation Simulations, Chem. Eng. Prog. 91, 63 June 1995.
Taylor, R., R. Krishna, and H. Kooijman, Real-World Modeling of Distillation Chem. Eng. Prog. July 2003.
Dynamic Simulation
Cassata, J.R. S. Dasgupta and S.L. Gandhi, Modeling of Tower Relief Dynamics Part I Hydrocarbon Processing, October 1993.
Cassata, J.R. S. Dasgupta and S.L. Gandhi, Modeling of Tower Relief Dynamics Part II Hydrocarbon Processing, November 1993.
Feliu, J.A., I. Grau, M.A. Alos, and J.J. Macias-Hernandez, Match Your Process Constraints Using Dynamic Simulation Chem. Eng. Prog. December 2003.
General
Schneider, D.F. Build a Better Process Model Chem. Eng. Prog. April 1998.
Sowell, R. Why a Simulation System Doesn't Match the Plant Hydrocarbon Processing March 1998.
Glasscock, D. A. and J.C. Hale, Process Simulation: The Art and Science of Modeling, Chemical Engineering, November 1994. pp 82-89.
Gallier, P.W. and T.P. Kisala, Process Optimization by Simulation, Chem. Eng. Prog., Aug 1987.
Heat exchangers and HX networks
Sigal, R.,Challenges in simulating heat exchanger networks, Hydrocarbon Processing, October 1996, pp 125-132.
Physical Properties
Carlson, E.C., Don't Gamble With Physical Properties For Simulations, Chem. Eng. Prog., 92 (10), pp 34-46.
Sadeq, J. H.A. Duarte, R.W. Serth, Anomalous Results From Process Simulators Chemical Engineering Education, Winter 1997.
Liu, Y. and S. Watanasiri, Successfully Simulate Electrolyte Systems Chem. Eng. Prog. October 1999.
Agarwal, R., Y.K. Li, O. Santollani, M.A. Satyro, and A. Vieler, Uncovering the Realities of Simulation Chem. Eng. Prog. May and June (2001).
Chen, C.C. and P.M. Mathias, Applied Thermodynamics for Process Modeling AIChE J. February 2002.
Additional references are listed for specific thermodynamic and transport property models in the Physical Properties Methods and Models reference manual, and in Reid, Prausnitz and Poling. The Properties of Gases and Liquids, 4th ed. McGraw-Hill, New York, NY (1987).
Heat and Material Balances
Agreda, V.H., R.C. Schad, Heat and Material balances: Making It All Add Up, Today's Chemist At Work, 2 pp 24-29 (1993).
Schad, R.C., Don't Let Recycle Stream Stymie Your Simulation, Chem. Engr. Prog., 90 (12) pp.68-76 (1994).
Reactive Distillation
Nijhuis, S.A. and F.P.J.M. Kerkhof, Multiple Steady-States during Reactive Distillation of Methyl tert-Butyl Ether, Ind. Chem. Res. Vol. 32, pp. 2767-2774, 1993.
Other
Gruber, G. and J. L. Rak, Model of a Wiped Film Still Chem. Engr. Prog. Dec 1989.
Rahbar, M.T., Imporve the Design and Operation of Desalination Plants by Computer Modelling and Simulation, Desalination 92, pp. 253-269, 1992.
Keywords: steady-state
References: None |
Problem Statement: There are multiple options through which you can zoom in or out of the Planning Board. This | Solution: discusses some of these options.
Solution
The following methods can be used to zoom in/out of the Planning Board:
1. By using the Planning Board contextual ribbon | Format tab | Time Range panel. This panel has 3 options:
a. Predefined
b. From-To Date
c. Around Selected
a. Predefined: Provides a list of time-based options (like First day, First month, Entire Horizon, etc.) clicking on which the Planning Board shows only the relevant time periods.
b. From-To Date: Provides a calendar widget for the From Period and another for the To Period. Selecting these options would show the relevant time periods on the Planning Board.
c. Around Selected: Selecting an Activity and clicking on the “Around Selected” button would select the time periods around a selected activity. The amount of days and hours before and after can be configured by clicking on the dog-ear (little arrow at the bottom of the panel):
2. By using the slider at the bottom to zoom in/out. Clicking on the number provides more options to custom zoom.
3. By right clicking and dragging the mouse and drawing a rectangle would also zoom into the Planning Board
4. By using the keyboard shortcuts:
a. Zoom in: Ctrl + Z
b. Zoom out: Ctrl + Shift + Z
Keywords: PB
Zoom
Keyboard Shortcut
Time Range
References: None |
Problem Statement: Can solids components be included in the Pressure Relief block calculations? | Solution: The input stream to the vessel can contain solids. However, solids must be included in the MIXED substream, and not in one of the solid substreams. Solids are not included in the phase equilibrium calculations or the pressure drop
calculations for the vent stream.
The amount of solids in the feed stream will be normalized and and re-calculated based on the the solid molar density.
Fixed in Version:
V7.1 - solids can be included in the solid substreams
Keywords: None
References: : CQ00325091,CQ00314346, CQ00171139 |
Problem Statement: Why do some rupture disk simulations show toggling between open and close status? | Solution: In a rupture disk simulation, the vessel pressure drops very steeply when the rupture disk bursts. When the pressure falls to the same value as the vent discharge pressure, the integrator runs into trouble and tends to fail.
In order to get around this problem in the calculations, the rupture disk is set to close when the pressure gets very close to the discharge pressure and set to open again when the vessel pressure increases. In reality, of course, the rupture disk cannot close. This algorithm helps to get around numerical problems.
Keywords: pres relief, rupture, disk
References: None |
Problem Statement: Using the same TBPCUT points with the same assay as Pro/II, pseudocomponents with somewhat similar properties are generated, but the volume distribution of the pseudos (to represent the entire assay) are grossly different on the light ends | Solution: The Pro/II default temperature for D1160 is 1 atmosphere, instead of the industry standard 10 mm Hg. Aspen assumes the D1160 data is at 10 mm Hg, unless the user modifies it. If the user forgets to modify it, the light end distribution of the psedo components is totally different.
Keywords: petroleum
ada
assay
psedo
pseudocomponent
D1160
PRO II
volume
References: None |
Problem Statement: What is Data Regression (DRS) testing when it says TEST LIQUID LIQUID EQUILIBRIUM RESULTS AGAINST THREE PHASE FLASH ALGORITHM? | Solution: All data regression systems based on minimization of Gibbs free energies perform only local minimum analyses. Thus, they only find local minima instead of the global minimum. This leads to potential cases where the local minimum conditions may be satisfied for more than one set of parameters. Therefore, as with other numerical techniques (e.g., Newton), the use of different initial guesses may lead to different local minima. In such cases it is advisable to try different initial guesses and explore the full range of parameter values. The acid test is the evaluation of the parameter descriptions using flash calculations.
Testing the LLE equilibrium results against a 3-PHASE FLASH ALGORITHM is related to the discussion above. Liquid-liquid equilibrium systems are more prone to show multiple local minima. Therefore, the 3-PHASE test is an attempt to resolve the question of whether or not the converged parameters satisfy the condition for a global minimum of Gibbs free energy. Assuming consistent thermodynamic data, this test (comparing flash calculations vs. experimental data) is successful if the DRS actually converged in the vicinity of a true stable minimum. The predicted compositions are generally not satisfied in the case of a metastable or unstable local minimum.
Recall that DRS does not minimize the Gibbs free energy and just matches the fugacities. This is a necessary but not sufficient condition for stable LLE. Therefore, the flash calculations are made to ensure that a Gibbs free energy global minimum has been reached.
Keywords: None
References: None |
Problem Statement: I want to increase the octane number of some of my gasoline blending stocks in order to run quick 'what-if' scenarios. I don't want to run a full simulation of my refinery for each scenario. How can I do that? | Solution: TheSolution here is to use the Assay Manipulator unit operation in Aspen RefSYS. The Assay Manipulator is used to shift or change petroleum properties for a stream in your simulation.
Add an Assay Manipulator through the RefSYS Object Palette (F6) or through the Flowsheet | Add Operation | Refinery Ops | Manipulator item in the main menu.
To use the Manipulator to specifically change the Research Octane Number (RON) for the straight-run Naphtha stream in the attached case, follow the steps below (the completed example is attached):
1. Go to the Options page of the Assay tab.
2. Select the RON (Clear) property using the drop-down list. Select the Shift Prop option (instead of the Change Props option).
3. Move to the Shift Props page. Shift the product RON number to 90 (from the current value of 40-50) as shown. The Prod ? Feed is automatically calculated.
You can shift or change as many petroleum properties as you wish using only the Assay Manipulator.
Keywords: Manipulator, octane shift, petroleum property
References: None |
Problem Statement: How to make a linear change of properties using Assay Manipulator? | Solution: In the assay manipulator you can specify the targets either entering the value for the target or a value for the difference between the product and feed. For example, if the sulphur content in the feed is 2% and if you want this to be changed to 2.5%, you can either enter this value in the target field for the product or a delta, 0.5 in the target field for Prod - Feed.
If there is a requirement to modify the property using a multiplicative factor or other linear calculations this can be implemented in a HYSYS spreadsheet and the value can be exported to any of the target fields in the manipulator. The attached example illustrates this functionality.
Keywords: Property Shift, Crude, Assay Manipulator
References: None |
Problem Statement: Why does the input Dynamic Viscosity not match the Aspen Assay Manager (AAM) Results? | Solution: In general, AAM in V8.8 is a better tool to do assay characterization because it provides more properties and more flexibility. For instance, it works well to match yield (given only three distillation data point up to 20% only), and bulk properties such as flash point, pour point, sulfur content, etc.
For viscosity, the AAM only takes Kinematic Viscosity input to do data match and only calculates the Dynamic Viscosity as output.
In order to use the Dynamic Viscosity input data, it’s necessary to convert Dynamic to Kinematic viscosity using the following formula:
KinVisc = DynVisc * 1000.0 / bulk density (in kg/m3).
If there is no bulk density input, the bulk density from AAM can be used as estimate.
Keywords: Dynamic viscosity, Kinematic viscosity, Assay Manager, AAM
References: None |
Problem Statement: Is there an example model for Dock Scheduling? | Solution: Yes! When V8.5 is installed, a new sample model is also installed that demonstrates Dock Scheduling. The sample model is DemoDSS.mdb and is located with the other sample models under
C:\Users\Public\Documents\AspenTech\Aspen Petroleum Scheduler\Demo\Access
Keywords: Dock Scheduling Example
Sample model
References: None |
Problem Statement: A phase may have flipped between the vapor and liquid slots in operation Vessel | Solution: A phase may have flipped between the vapor and liquid slots in operation Vessel ?.
This can happen with supercritical mixtures. If a spike or discontinuity has also occurred in your dynamic simulation, please check the model choices in regions that may be affected. For example, make sure valves use the simple resistance equation and not one that treats vapor and liquid phases differently.
If the fluid is at supercritical condition, it is neither gas nor liquid. The simulation will still assign a phase to it. This in itself can cause problems, as the phases can flip from vapor to liquid and vice versa in the simulation. Now the calculation in the valve models uses the vapor fraction to calculate the pressure drop / flow. However this is meaningless if you have a supercritical fluid. The flow through the valve will jump from one value to another, based on the present vapor fraction HYSYS calculates.
To circumvent this, you can change the valve equation that is used.
You can also make your simulation give more accurate results by changing the following integrator options:
1. In the Flowsheet menu, Integrator, go to the Options sheet, and select the following options:
2. Uncheck the option Truncate large volume integration errors
3. Check the option Close Component material and energy balances
4. On the Execution sheet, change Control and Logical Ops to 1
5. Change Energy calculations to 1
6. Change Composition and flash calculations to 1
By forcing these calculations to take place more frequently we decrease the numerical integration error.
Keywords: A phase may have flipped
References: None |
Problem Statement: We are trying to bring Crude Run events via the staging tables in APS. Nevertheless, we don't know which are the values that needs to be populated for this kind of events. | Solution: For a crude run event these are the values that must be populated in ORION_MGR_ tables:
1. ORION_MGR_EVENT_IMPORT
a. START_DATE
b. STOP_DATE
c. QUANTITY
d. MOVEMENT- this number is important because will be used to relate the other tables
e. EVENT_TYPE- for crude run events the number is 13
f. BLENDER – name of crude unit
2. ORION_MGR_EVSOURCE_IMPORT
a. MOVEMENT- this number must be the same as the movement defined in ORION_MGR_EVENT_IMPORT
b. SOURCE-Here we list the tanks that will feed the crude run unit
c. VALUE.- fraction
3. If the crude unit has parameters you can add them in ORION_MGR_UNITPARAM_IMPORT
The resulting event with that information is next:
Keywords: crude run events, staging tables, ORION_MGR_EVNT_IMPORT
References: None |
Problem Statement: Upgrade to PCLS0113 & PCDM0112 to provide the full CL functionality compatable with the TDC v1.0 schematics. The parameter IPNSUB was added to PCDM0111 and the file was renamed. Logic for the mode switching was improved in PCLS0113.
The TDC v1.1 schematics have other CL versions. | Solution: To replace the program PCLS0112.CL with PCLS0113.CL on existing LCN controllers you need to ;
Print system entities of the sub point to an EB file. It is important to print the current sub entity because most of the parameters are built NOT BUILD_VISIBLE and configuration data is lost when reconstituted to the screen or an IDF file.
2.
Edit the EB
file replacing the CDS
package names as follows:
Replace
Old Package
New Package
PCDS0110
PCDS0111
PCLS0112
PCLS0113
Inactivate the existing subpoint, then Exception Build the EB file and LOAD with OVERWRITE.
DETAIL the main point and reenter the subcontroller name.
Confirm the MV configuration for the controller using the Build screen. For example point type is AM, PM, HW, FF, or NA.
Link the PCLS program to the sub entity. Activate the subpoint, process the point a couple of times, inactivate the subpoint, unlink the CL, then relink the CL and activate the point.
Keywords: PCDS0110, PCDS0111, PCLS0112, PCLS0113, EB, Build, CL, subpoint, activate, OVERWRITE, LOAD, IPNSUB, PCDM0111, PCDM0112
References: None |
Problem Statement: Messages are not appearing on the production control webserver | Solution: Run the AcoBase shutdown and then the AcoBase Startup.
An alternative method would be to stop and then restart each controller individually in the hopes of stopping the controller that has the lock on the message queue.
This condition is caused by a process in the control system having locked the message queue, preventing it's use by the other control processes. It is not certain what causes the lock, however having more than one instance of DMCplus Manage open at any given time is one condition. The stopping and restarting of the controllers, releases the lock and the messages resume.
Keywords: msgs, web server, control, DMCplus, messages
References: None |
Problem Statement: How can GMULTs be deleted from DMCplus Build | Solution: This is relatively simple to accomplish.
Step One: Capture the CCF and Model.
Step Two: Capture the non-zero gmults and update the model curves with a model curve operation.
At this point you have some choices.
Option 1: Leave the gmults in the controller with a value of 1 and with each gmult set to 1.0. {This is one of the alternatives you suggested.} This gives you the opportunity at some future date to use a GMULT to adjust a controller model gain in a pretty simple fashion.
Option 2: Remove the GMULTs from the CCF. This can be done by a Tag replacement operation. I tried this on one of my sample CCF's and here are the steps I followed:
1. Using Build, open the CCF.
2. Bring up the tag replacement tool. Under 'Which sections' select Independents and select (ALL), Under 'Which Entry(s)' chose Entry Types and USER. In the 'Replace what:' area, select the keyword tab. in 'Find what' select READ and in 'Replace with', select (None). Be certain to Preview to make sure that you are only getting the GMULTS.
3. In the left hand navigation pane, select Configure andin the right hand pane search for GMULT. Change its value to 0.
4. Save the CCF.
5. Recommission the controller with the new model and CCF.
Keywords:
References: None |
Problem Statement: How do I manipulate the user group rights for the model accuracy screen? | Solution: The table GROUPS defines which group of users has permissions to add screens. If a user belongs to this group he can add or delete Model Accuracy screens. Once a screen has been added, any user can modify the Event Facilities and Stream Facilities by right clicking and choosing the relevant option.
Keywords: User rights for screen
Permission to add screens
PSMA screens
User rights
References: None |
Problem Statement: Calculate NEWPV to indicate availability of new analysis. | Solution: The
following CCF calculation will enable this:
CVtag
R4 DEP
DBVL
CVtag
I4 NEWPV
LOCAL Init = 0
CVtag
R4 TOLANZ LOCAL
(user-defined entry)
CVtag
R4 DEPOLD LOCAL
(user-defined entry) Init = -9999
Input Calc
DEP:CVtag|NEWPV =
@IF( @GT( @ABS(DEP:CVtag|DEPOLD - DEP:CVtag|DEP), DEP:CVtag|TOLANZ), 1, 0)
BAD value handling in the above calc will give a zero when either DEP or DEPOLD are BAD. Initialize DEPOLD = -9999
to avoid setting NEWPV=1 on start-up.
Output Calc
DEP:CVtag|DEPOLD = DEP:CVtag|DEP
where,
TOLANZ = change in analyzer required to trigger the setting of NEWPV DEPOLD = saved value of DEP from last execution
Note: Don't forget to set MXUSPR based on cycle time and how long you want
to run without update.
Keywords: NEWPV, CCF, Calc, analysis
References: None |
Problem Statement: I've installed a stand alone key on the correct machine, but cannot get a valid license key for application. | Solution: Go to Start / Programs / AspenTech / Common Utilities / SLM Configuration Wizard
Use the Configure button to reveal the SLM configuration information for that machine.
If the entry for the license source directory does not read c:\Program Files\Common Files\Hyprotech\Shared, then go to the registry and delete the source directory entry.
This is accomplished by going to Start / Run and type regedit
When the registry opens, go to HKey_Local_Machine / Software / HyproTech / SLM / Configuration
Delete the LicenseSourceDirectory key, without this key in the registry, it will default to the Hyprotech \ Shared directory.
Close the registry and re-open to confirm the change has been made.
Close registry again and try application.
Keywords: None
References: None |
Problem Statement: For some unknown reason, the TDC3000 parameters RESETCMD and STRTSTOP values are set BAD in the controller common buffer. Subsequent runs of the controller process will fail with messages in acodhome:[etc]valid.err like those shown below:
Sun Apr 19 07:57:09 1998
ctlr_name Error writing tag: ORD tag_name.RESETCMD : Sun Apr 19 07:57:09 1998
ctlr_name Error writing tag: ORD tag_name.STRTSTOP : Sun Apr 19 07:57:09 1998
ctlr_name Error writing list 7: IDB: IDB_Write failure: PDB: Write failed for 1 or more values
One of the controllers got turned off and could not be turned ON by the operator. Operator message indicated Error 54 PCSPUT error.
Using MANAGE, save the controller context to the CCF. You will observe that the RESETCMD and the STRTSTOP entries have BAD values in the CCF.
.STRTSTOP~~~AWRITE~~~I4~~~-9999~~~::tag_name.STRTSTOP:ORD:
.RESETCMD~~~AWRITE~~~I4~~~-9999~~~::tag_name.RESETCMD:ORD: | Solution: Change the values for RESETCMD and STRTSTOP to 0 and load and start the controller.
.STRTSTOP~~~AWRITE~~~I4~~~0~~~::tag_name.STRTSTOP:ORD:
.RESETCMD~~~AWRITE~~~I4~~~0~~~::tag_name.RESETCMD:ORD:
The controller should run aok now. The work-around to ensure that this problem does not recur is to place output calcs in the CCF to set these values to zero at each execution.
.CALC0002~~~CALPUT~~~CH(12)~~~RESETCMD = 0~~~
.CALC0003~~~CALPUT~~~CH(12)~~~STRTSTOP = 0~~~
Keywords:
References: None |
Problem Statement: Item Count: 3
Item: 1
Summary: Model loses double-precision model gains when a model assembly (.dpa) file is used to create the model.
Topic: Model v1.10.04, Double-Precision Model Gains
When a model is exported, the steady-state gain is written to the text .mdl file as a double-precision number for use by the DMCplus online controller v1.10 and later. When the model is exported as a model assembly file, any curve operations that set the steady-state gain are exported in single-precision. The work-around is to either export the model as a .mdl file or to set the gain in the built model after using the model assembly file.
End of Item: 1
Item: 2
Summary: Model does not correctly verify ramp curve operations
when a model assembly (.dpa) file is used to create the model.
Topic:ÂÂ Model v1.10.04, Model Assembly Files
When the model is exported as a model assembly file, any curve operations that exist are written into the .dpa file. When the .dpa file is imported, a second operation (after replace) on a ramp variable will cause the import to fail.
End of Item: 2
Item: 3
Summary: Model does not provide a message box on errors
when a model assembly (.dpa) file is used to create the model.
Topic:ÂÂ Model v1.10.04, Model Assembly Files
When the model assembly file is imported, errors / failures should trigger a message box indicating the cause of the error. Due to a bug, this does not occur.
End of Item: 3 | Solution:
Keywords:
References: None |
Problem Statement: When I’m defining the number of treating and total beds in a Hydrocracker Configuration Wizard, how does the configuration look like? | Solution: For the Hydrocracker Reactor Model in Aspen HYSYS Petroleum Refining, you select the TOTAL number of beds (Number of beds cell at the right of the Configuration Wizard). Then, you specify how many of those beds are treating beds.
By default the reactor will put the treating beds before the cracking beds.
Example:
Keywords: Hydrocracker, number of treating beds, cracking beds
References: None |
Problem Statement: What does the Extend Stream Functionality button do? | Solution: The Extend Stream Functionality button (found on the Worksheet | Composition pagetab of the stream view) is designed for use with Aspen RefSYS, and allows you to associate an assay (created with the Aspen RefSYS Petroleum Assay Manager) with a given stream in the flowsheet.
If Aspen RefSYS has not installed, this feature cannot be used (although the button will still appear active).
Keyword
extend, stream, functionality, button, assay, define
Keywords: None
References: None |
Problem Statement: Aspen RefSYS can represent all oils in a single case with a single component slate. This is very different from the Aspen HYSYS method of representing each oil with its own set of hypothetical components. How does this affect the properties associated with each component and each stream? | Solution: This is one of the major distinguishing features of modeling with Aspen RefSYS versus Aspen HYSYS. By using a single component slate in Aspen RefSYS, the simulation time can be greatly reduced. Each component in the Aspen RefSYS component slate has a vector of properties associated with it, and these properties change as the component moves through the flowsheet (i.e. through distillation towers , reactors, etc.). A full description of the lumped and blended properties follows:
Lumped and Blended Properties
Crude Properties are summed in two ways that provide distinct advantages over earlier Oil Characterization methods in Aspen HYSYS. Crude properties may be lumped or blended. For example, properties summed 'vertically' and component wise for a single stream are considered 'lumped':
Prop = Crude Property
Fi = Mole/Volume/Mass Fraction
nc = Number of components
Lumped properties are summed Vertically
Crude properties summed component wise and 'horizontally' for two or more streams are 'blended':
Prop = Crude Property
Fi = Mole/Volume/Mass Fraction
S = Number of Streams
The advantages of this approach are seen in that only a single oil component slate is needed to characterize any number of oils in Aspen RefSYS. Oil properties are summed within a pre-defined temperature range determined from the oil component TBP. This procedure only applies to oil components; library component properties (e.g. C1, C2, C3, ?) are summed directly. This approach for oil characterization eliminates the need for multiple oils (as in Aspen HYSYS) which can significantly slow down the speed of the simulation.
Keywords: property, component, oil, characterization, stream property, petroleum property
References: None |
Problem Statement: What is the recommended transition type in a stream cutter when moving from assay basis to the Hydrocracker Product Transition (HCRSRK) basis? | Solution: When going from an assay basis to an HCRSRK basis, the defualt transition type is the Fluid Package Transition, but it might not maintain all the assay properties (e.g. sulfur and nitrogen content).
The preferred transition when going from the assay basis to the HCRSRK basis is the Refining Reactor Transition. Once that is selected, you would select Hydrocracker as the type and you would select a feed fingerprint (e.g. default).
For the product transition, you can also use a fluid package transition. As with the feed transition, though, this would be very difficult to set up to not lose the properties you are interested in.
The recommended transition when going from the HCRSRK basis back to the assay basis is the Standard Hydroprocessor Product Transition that makes the boiling curve much smoother due to the lumpy nature of HCRSRK.
However, there are a small number of pure components required in the assay component list:
Hydrogen, Methane, Ethane, H2S, Propane, i-Butane, n-Butane, i-Pentane, 22MPropane, Ammonia and H2O.
If any of these are not present, there will be a warning message in the trace window that will let you know what components are needed.
Keywords: stream cutter, HCRSRK, transition type
References: None |
Problem Statement: How to work within a simulation with various sets of property values for pure components using calculator blocks. | Solution: Most parameters used in property models can have more than one set of values for a given component or pair of components, this allows users to assign appropriate values for different range of temperatures or pressures for instance. Conversely, in the case of properties of pure compounds that possibility is not available. Although there is not a direct way to do this, thisSolution suggests a way so that users can still work with different set of values for properties of pure components within the same simulation.
Calculator blocks can be configured to change properties in a part of a simulation, and then undo the changes for the rest of the flowsheet. Below there is an example of a flowsheet showing how the calculator blocks modify properties in the simulation:
The following lines explain in detail how to set up your simulation. In this case we will change for example the entalphy of formation of phenol in one part of the flowsheet (highlighted in red), and then we will undo the changes downstream.
1) Insert the first calculator block and create a new export variable for that block. As shown in the picture below, select “Property Parameters”, “Unary-Param” for type and “DHFORM” as the variable referred to the component you want to change.
2) Go to “Calculate” tab and enter the new value for the variable. Fortran or Excel can be used to define the value. Please note that in the calculator block the variable units have to be expressed in SI, so in the case of enthalpy the units would be J/kmol.
3) You can select when the calculator block is placed in the execution sequence (referred to below). Select according to your needs. In this case we will select after the heater “B1”.
4) Add the second calculator block and repeat the process but setting the value of the variable to its original value. Choose as well where to place the calculator in the sequence. In this case after block “B4”
5) If we run the simulation we will see how the calculator blocks affect the properties of pure components following the sequence selected.
Keywords: Calculator block, data sets, properties for pure components, Aspen Properties.
References: None |
Problem Statement: The .ocr file syntax for designing user forms is unclear. What should be on each line? | Solution: The syntax is as shown below. These details augment the description in the Getting Started Customizing Unit Operation Models manual.
Line 1: Blocks\blockname, blockname.form_name, tabname, index, 1
(repeat Line 1 for n = number of input tabs)
Line n+1: Blocks\blockname, blockname.form_name, tabname, Default, 1
Line n+2: Blocks\blockname, blockname.form_name, tabname, index, 1, ResultsOnly
(repeat Line n+2 for m = number of result tabs)
Line (m+n)+2: Blocks\blockname, blockname.form_name, tabname, ResDefault, 1
Notes:
Index starts with zero.
Line n+1 declares default input tab to be displayed when form is accessed; only required once.
Line (m+n)+2 declares default results tab to be displayed when form is accessed; only required once.
Result tab starting index should be greater than last input tab index
ResultsOnly causes tab to appear only when results are present
Additional advanced functionality exists which is not defined here
Keywords: ocr
user forms
custom forms
References: None |
Problem Statement: Why the amount of NH3 is present in the Product Yields results for the Hydrocracker Unit when there is no NH3 in the outlet stream? | Solution: Typically, the NH3 is removed with wash water before the HPS unit (high-pressure separator) and this is modeled as a component splitter internally in the Hydrocracker. This is the reason why there is no NH3 in the outlet stream but NH3 yield is shown in Product Results.
Keywords: Hydrocracker, Results, Product Yields, NH3
References: None |
Problem Statement: How can I re-characterize an assay using the Petroleum Shift Reactor? | Solution: First you need to have your original assay with Petroleum Assays and not with Oil Manager.
If you have your assay ready, the following step by step shows the way to re-characterize an assay using the Petroleum Shift Reactor.
1) Having a simulation with an assay from Petroleum Assays, create a stream and make sure to attach the assay to the stream.
2) Go to the Model Palette and then to the section of Refining. Select the Petroleum Shift reactor.
3) Define your stream with the Assay as the Main Feed and create a new stream for the Product Cuts.
4) Go to the Model Data section and since we only have one stream we need to set the yield in 1. You can also define the Flow Rate as the same of your inlet.
5) Now we are going to specify the new distillation curve. To do this go to the Product Specs tab and then to Product Cuts. Enter the Product temperature of your outlet stream, IBP and FBP of the new curve. At this point a calculation should be ready.
6) Now go the Distillation Curves section and insert all the necessary rows to fill in with your distillation curve. This will allow a better specification of your results.
7) Copy all the point in between from your new distillation column to fill in the rows that you inserted.
8) With this your reactor should have full results according to your new TBP curve. If you observe that the calculations did not refresh, disconnect and reconnect the inlet stream.
9) Finally, if you want to observe the new TBP curve, you need to go to the Product stream and in the Attachments tab go to Analysis and Add a Petroleum Assays analysis.
This will show you the TBP calculated by Petroleum Assays with base on the results of the shift reactor.
Keywords: RefSys, Petroleum Shift Reactor, Assay re- characterization, TBP
References: None |
Problem Statement: What can the Delta Base utility be used for? | Solution: The Delta Base utility generates the required derivatives for use in delta base linear programming models, which in turn provides a simplified model that can be imported into programs such as PIMS (PIMS is an Aspen product that allows you to determine the best operating conditions at minimum cost using numerical optimization methods; PIMS calculation consists of linear models and simplified assumptions).
Sets of independent and dependent variables are selected from within the flowsheet. Base values are specified for the independent variables and HYSYS will calculate the change in the dependent variables with respect to changes in the independent variables. The utility works by perturbing each of the independent variables from the assigned base value, solving the flowsheet at the perturbed value, and determining the delta change in each of the specified dependent variables.
Keywords: delta, base, utility, model, export, import, PIMS
References: None |
Problem Statement: Sometimes, users have multiple versions of Aspen Plus installed on their PC. Even if the user visits the TOOLS | | Solution: VBA/ActiveX will open the latest installed version of Aspen Plus - this is a registry parameter. To easily see which version of Aspen Plus is being opened by default, after the open simulation code executes, add a message box to display the version number:
msgbox The following Aspen Plus version is running & vbcrlf & & go_simulation.name
where go_simulation is the name given to the Aspen Plus application inside VBA
The way to change the default version, is version dependent.
To change to Aspen Plus version 10.x or 11.x, do the following:
1. Open a session of Windows Explorer
2. Navigate to the directory where the Aspen Plus GUI for the desired version is located
a. for version 11.1: c:\Program Files\AspenTech\Aspen Plus 11.1\GUI\xeq
b. for version 10.1 & 10.2 the path is similar to part a, except for the version number.
To change the default version to Aspen Plus 12.x or 2004.x:
1. Close down all open Aspen Plus sessions.
2. Click on the Windows START button, click on Programs, click on Aspen Tech, and then click on Aspen Plus xxxx.x (where xxxx.x is the version number)
3. In the list of Aspen Plus programs, click on Aspen Plus Registry Fix Utility. Note: this option may not display unless you expand the list by clicking on the double headed down arrow (last item on the list).
4. When the utility starts, there will be a list of all installed Aspen Plus versions. Click on the version that you would like to make the default version.
5. Run the registry utility, and allow it to complete (should take a few minutes or less).
One last final step, regardless of which version. You should test the default version association by starting a Windows Explorer session and then double clicking on any Aspen Plus backup file (*.bkp). As Aspen Plus loads, observe the version number on the splash screen -OR- after Aspen Plus finishes loading, click on the HELP pull down menu and then click on About Aspen Plus
Keywords: VBA, default version, application association, changing default version, activex, COM
References: s form in the VBA Editor and changes the Aspen Plus GUI Type libray to another version, the same version is spawned each time the user use VBA/ActiveX to open a simultion. |
Problem Statement: Is it possible to predict Petroleum properties such as Research Octane Number (RON) or Motor Octane Number (MON) for a Chromatography analysis data? | Solution: If you have a Chromatography analysis data and directly enter the composition in a stream, you will not be able to read the petroleum properties. These will be shown as empty.
This type of analysis is not included in Petroleum Assays which is the tool that allow the users to get petroleum properties such as RON or MON. However, HYSYS has pure components properties for most of the hydrocarbons.
You can check these properties if you double click on a specific component in the component list as shown below (i.e. double click on Pentane and select the Edit Properties button to show the available properties, for this one RON is 92.3).
In order to calculate the RON or any other petroleum property for a stream, you need to import first an assay from the HYSYS assay library to make sure there would be a petroleum refining component slate.
Follow the steps below to import the assay library.
1. Go to Properties Environment and open Petroleum Assays, and then add a new assay using the option Import from Library. A list with the assays library will be shown. Select one of these.
2. At this point a component list and fluid package are added. Go to the component list to confirm that you have the desired components (the components of your Chromatography) and if needed add the missing components.
3. Go to the Simulation Environment and add a feed stream. Then specify the composition to match the desired composition.
4. Now you will see that the stream can predict the RON and MON.
By default, HYSYS uses the volume blending method, but the Healy method is recommended for more accurate results. Follow the steps below to change the method.
1. Use the correlation manager to change the method. To access the Correlation Manager, from the Home ribbon tab, in the Simulation group, click Correlation Manager.
2. Find RON under the petroleum properties in the correlation manager and set the method to what you like.
Keywords: Petroleum Properties, RON, MON Chromatography, Blend
References: None |
Problem Statement: In a 3-phase (vapor-liquid-liquid) distillation, if there is only one liquid phase present how does Aspen Plus determine whether it is in the 1st liquid (L1) or 2nd liquid (L2) phase? | Solution: When Aspen Plus is writing the report, it must identify the liquid phase as L1 or L2. This is not a problem when two liquids exist. However, it becomes an issue when only one liquid is present. To decide, Aspen Plus checks the concentration of the key components to identify the 2nd liquid phase as specified on the RadFrac\Setup\3-Phase sheet. If the mole fraction of the key components (L2-COMPS) is larger than 0.5 (default), then Aspen Plus will display and report the liquid flowrate as the second liquid phase (L2). The default value of 0.5 can be changed by visiting the RadFrac\Convergence\Algorithm sheet and changing the value of the Mole fraction threshold for 2nd liquid key component (L2-Cutoff).
Keywords: column tower
References: None |
Problem Statement: What internal specifications are used/assumed by Aspen HYSYS when switching a petroleum distillation column from zone by zone to stage by stage calculations? | Solution: When switching from the zone-by-zone to stage-by-stage solver in the petroleum distillation column, the following active specs are provided:
Flow rates for top vapor, bottoms and all side products.
Local reflux ratios for all bottom stages at borders between zones. In the short cut model we assume that there is no liquid flow between the upper and lower zones (there is however significant vapor flow from lower to upper zones). For practical reasons we set these local reflux ratios to a small number, e.g. 0.01.
Temperature change for all pump arounds.
Duty specifications for all reboilers.
The pumparound duties are also calculated but are only provided as estimates.
Keywords: petroleum distillation, zone, stage, specification
References: None |
Problem Statement: In Aspen Plus 10, one sometimes encounters the following problem when running a file containing Fortran code:
*** SEVERE ERROR
ERROR DURING DYNAMIC LINK OF USER ROUTINE(S) OR IN-LINE FORTRAN PLEASE CHECK FILE _2734lfo.LD FOR LINKER MESSAGES.
*** SEVERE ERROR
COULD NOT RESOLVE USER OR IN-LINE FORTRAN SUBROUTINE(S):
SUBROUTINE ZZFORT IS MISSING
*** SEVERE ERROR
PROGRAM TERMINATED DUE TO UNRESOLVED ROUTINES
! Errors while processing input specifications | Solution: When Aspen Plus encounters a Fortran statement that can not be interpreted, Aspen Plus creates a temporary Fortran file that contains a subroutine with the name ZZFORT. The subroutine contains all the fortran block statements and is compiled and linked dynamically (during the run) by the external Fortran compiler installed on user's machine.
There are several possible causes for the ZZFORT is missing error message:
The external fortran compiler is not installed or the incorrect compiler is installed. In this case, user either has to make sure the correct compiler version is installed or try to rewrite the fortran blocks to make sure that all the statements are interpretable. SeeSolution 104149 for a list of Fortran statements that can be interpreted.
User is using a Aspen Plus 9 file that contains old Fortran code. User must go through Ap9to10 to convert the input file.
User might run out the environmental space. In this case, Aspen Plus GUI will not issue any warnings. If you try to compile any fortran manually using aspcomp in the Engine environment, you will get run out environmental space message. In this case, you need to increase the allocation by editing config.sys file and then reboot. In the config.sys file, edit this line
shell=c:\win95\command.com/p/e:1280
You may increase the allocation. Typically this is done in multiples of two to 4096, 8192, 16384, etc.
User left some object files (obj) in the working directory that may contain errors or are from a prior version. When Aspen Plus tries to link the fortran obj files, it links ALL obj files residing in the same directory, regardless they are going to be used or not. So when an incorrect obj sits in the working directory, Aspen Plus won't be able to link them thus producing the error message. In this case, user must remove the offending obj file(s). The names of these files can be fournd in the *.ld file mentioned in the Error message.
There are user database files or fortran subroutine object files which are missing or are not in the expected directories. Copy these missing files to the correct directory.
Keywords: ZZFORT
object file
obj
compiler
fortran
Fortran block
References: None |
Problem Statement: What type of distillation method is used for variable CutYieldByVol(%), CutYieldByMol(%) and CutYieldByWt(%)? How should I enter the distillation data? | Solution: The CutYieldByVol(%), CutYieldByMol(%) and CutYieldByWt(%) in both Input Summary form and Results Summary form it's always showed as TBP distillation type.
If you go to Distillation Data tab you can enter different type of curves, such as: TBP, D86, D1160, D2887.
Keywords: CutYieldByVol, Input Summary, Distillation Data, TBP.
References: None |
Problem Statement: Are there any tips for importing HYSYS D86 curves into RefSYS? | Solution: If you define an assay using a D86 curve in the HYSYS Oil Environment and want to use this assay in RefSYS you must use the API 1994 D86-TBP Conversion Method. This must be done to each individual assay.
To change the conversion method follow the steps below:
Enter the Oil Environment from the Basis Manager.
Open the assay you want to modify.
In the assay's property view navigate to the Calculation Defaults tab.
Select API 1994 from the D86-TBP Interconversion drop-down list.
Keywords: D86, Conversion, Assay
References: None |
Problem Statement: How do I use calculated UA for air cooler in dynamic simulation? | Solution: The UA value in the air cooler is taken as the steady state value. In dynamic mode, the UA value can be derived from the steady state value as follows:
UA-Dynamic = F x UA-Steadystate
Where, UA-steadystate value is the value entered on the Parameters page of the Design tab and F is the flow scale factor. This scale factor is calculated as follows:
F = (2 x f1 x f2)/(f1 + f2)
where,
f1 = (mass flowrate / reference flowrate)^0.8 for air
f2 = (mass flowrate / reference flowrate)^0.8 for fluid
Note: The calculated UA value used in the dynamic mode is not presented in the Air Cooler interface. This can be presented in a spreadsheet as shown in the attached simulation.
Keywords: Air Cooler, Dynamics, Dynamic UA
References: None |
Problem Statement: How is it possible to find true critical point for a mixture? | Solution: The Aspen Plus Prop-set properties PCMX for critical pressure, TCMX for critical temperature, (as well as VCMX for critical volume and ZCMX for critical compressibility factor) give simple mole fraction averages (pseudocritical) for the critical properties of a mixture which should not be considered very accurate. Finding the true critical properties is harder.
Often the best method is to enter the desired mixture as a stream on a flowsheet and to generate a PT-envelope for the stream from the Tools/Analysis/Stream analysis menu. For many mixtures, stream analysis will give a good picture of the critical conditions. It is not possible to get a fully continuous curve up to the critical point, since at the critical point the equation-of-state (EOS), being cubic, fails.
For some mixtures, the PT-envelope will stop at some distance from the critical point. If there is a real need to calculate critical conditions, the user can try one or more of the following methods.
1. Run the analysis with additional vapor fractions instead of just using the dew and bubble curves. Vapor fractions of 0.2 and 0.8 may calculate out closer to the critical, and even if they do not, they will help show where the critical point must lie. (The curves for every vapor fraction will intersect at the critical point.)
2. Instead of using analysis, set up a Flash2 block and vary pressure with a sensitivity study. Set the flash tolerance tight (1E-7 is good) and be sure that each flash starts from the previous results. Near the critical point, make the Pressure intervals small.
3. Make a table of the bubble and dew point pressures properties, PBUB and PDEW, as a function of temperature. They are calculated by a different algorithm, and often will go up to and even beyond the critical point. (Note that the critical point is the point where the two pressures are equal. Any values calculated above the true critical point have no physical significance.)
Other things to think about:
1. These sorts of calculations only make sense with a property method based on an equation of state.
2. Ordinary cubic equations of state such as the popular RK-SOAVE and PENG-ROB are not particularly accurate near critical.
3. If attempting calculations near critical, be sure that the EOS is using actual critical properties for the components in the simulation. Property methods such as Peng-Rob have fitted parameters (TCPR and PCPR) that are fitted to better represent behavior at 'normal' conditions. If the curves are to go through the actual pure component critical points, TCPR should be set equal to the actual critical temperature.
Keywords: None
References: None |
Problem Statement: How can I see which stream is related to a Petroleum Assay? | Solution: This situation may happen when in a big simulation user wants to locate all streams with Petroleum Assay analysis quickly. The following steps will help to locate such streams.
In the main Flowsheet go to the Flowsheet/Modify Tab and in the right hand side you will find the Conditional Formatting part.
Display the list related to the Default Color Scheme and click on <Add New>
A window will appear, select for the variable “Petroleum Assay Type”
Keep the default colors
Go back to the PFD and user will be able to locate the stream(s) with a Petroleum Assay analysis attached.
Keywords: Petroleum Assay, Colour Scheme
References: None |
Problem Statement: Can I specify viscosity data for my assays using Aspen RefSYS? | Solution: YES. Starting with version 2006, users can input viscosity data for their assays in Aspen RefSYS. Attached with thisSolution is a recorded demo with detailed instructions.
This video will demonstrate the difference the new assay parameter adds to the viscosity calculation of the petroleum assay stream.
1. Start a new simulation case.
2. Import the sample assay visc100test provided.
3. Make a copy of the sample assay and add the viscosity data at 100?C from the spreadsheet Data-Viscosity100, so that you end up having two assays and can compare the calculated results generated by the additional viscosity data.
4. Change the viscosity calculation method from HYSYS Viscosity to Indexed Viscosity.
Provide the following values for the Viscosity Index Parameters:
- A = 1.0
- B = 0.0
- C = 1.2
5. Enter the Simulation Environment.
6. Add two material streams with the following properties:
- Temperature = 20?C
- Pressure = 200 kPa
- Molar Flow = 100 kgmole/h
7. For stream 1, apply the composition of the assay sample containing the Viscosity @ 100 data.
8. For stream 1 With Vis100, apply the composition of the assay sample without the Viscosity @ 100 data.
9. Generate a case study to view how the viscosity of the stream varies with different temperatures.
- Two independent variables = temperatures of stream 1 and stream 2
- Two dependent variables = kinematic viscosity of stream 1 and stream 2
- Enter a temperature range of 40?C to 100?C with step size of 10 degrees.
10. The calculated viscosity results are also affected by the specified Viscosity Index Parameters.
Please note that this video is over 37MB in size. If you are interested in download and play it, you are advised to check with your system administrator first to see if such a large file is allowed for your system.
Keywords: assay, viscosity data, viscosity
References: None |
Problem Statement: Assistance with setting up Perfmon | Solution: Perfmon is a tool for monitoring various statistics about system operation. It can look at the system as a whole, or individual processes that are running. The following procedure configures PerfMon to collect some key memory statistics:
1. Go to Windows Start -> Run
2. Type in perfmon and click Ok. The Performance monitor program will start.
3. In the tree view, open the Performance Logs and Alerts folder
4. Right-click on Counter Logs and select New Log Settings...
5. Enter a name like High Memory or something similar and click Ok.
6. On the dialog that comes up, click Add Objects.
7. Select the Memory object and click Add. Then click Close.
8. Now click the Add Counters button.
9. Select Process in the Performance Object combo box.
10. Ctrl-click to select these counters from the left-hand list box: Handle Count, Private Bytes and Virtual Bytes. Make sure all three are selected.
11. Click the All Instances radio button above the right-hand list box.
12. Click the Add button, then click the Close button.
13. Verify that the counters you selected are shown in the list box.
14. Change the sampling interval to 300 seconds.
15. Switch to the Log Files tab. Make sure the file type is Binary File.
16. Notice the directory where the log files will be saved. You can change it if you want by clicking the Configure button.
17. Switch to the Schedule tab. Click the After radio button and set the log to stop after 6 hours.
18. Check the Start a new log file check box.
19. Click Ok.
This procedure will initiate logging of system memory usage. It will create a new log file every 6 hours. You can zip these up and e-mail them to AspenTech Support. (They should compress about 90%.) The uncompressed files are about 1-3 MB each, so naturally the log shouldn't be allowed to run indefinitely! When we are through with our testing, you should:
1. Run perfmon again
2. Open the Performance Logs and Alerts folder
3. Click on Counter Logs. Running logs will have a green icon.
4. Right click on the log you created and click Stop.
5. Delete the old log files on disk.
If you want to look at the data yourself, follow these steps:
1. Run perfmon.
2. Select System Monitor in the treeview, if it isn't already selected.
3. Click the disk icon in the toolbar (the 4th icon). The System Monitor Properties dialog will appear.
4. Click the Log files radio button.
5. Click the Add button. Browse to the first log file and add it. (You remembered to make a note of the directory, right?)
6. Add the other log files the same way, if any.
7. Optionally at this point you can click the Time Range button and use the funny slider control to set the time range. (Drag the ends of the slider to increase or decrease the range.)
8. Click Ok.
9. Click the + sign in the toolbar to add new trends. It works the same way as adding Counters when configuring a log.
10. You may want to add a trend for the Private Bytes of each instance of RTE Application.
11. You will probably need to set the plot scale. Right click on any line in the table below the plot and select Properties.
12. Choose an appropriate scale and click Apply or Ok.
Keywords: Performance, Monitor
References: None |
Problem Statement: What is the significance of bundle top height and bundle bottom height entered on the Dynamics | Heat Exchanger page of the Separator property view? | Solution: HYSYS is using the Bundle Top Height and the Bundle Bottom Height values to calculate the liquid Cover percentage and from that it calculates the Liquid Area and the overall Vapor Area . Afterwards these areas are used to estimate the overall Shell heat transfer coefficient UA_shell and the overall Tube heat transfer coefficient UA_tube. The logic is as follows:
bundleHeightPercent = BundleHeight / vesselHeight * 100
bundleBottomHeightPercent = BundleBottomHeight / vesselHeight * 100
if (vesselPercentLevel > bundleHeightPercent)
LiquidCover = 1
else if (vesselPercentLevel < bundleBottomHeightPercent)
LiquidCover = 0
else
h = bundleHeightPercent - bundleBottomHeightPercent
LiquidCover = (vesselPercentLevel - bundleBottomHeightPercent)/h
ShellU = (VesselLiquidU * LiquidCover + VesselVaporU*(1-LiquidCover))
TubeU = (TubeLiquidU * Tubelevel + TubeVaporU *(100-Tubelevel)) /100
Overall heat transfer coefficient U = 1/(1/ShellU+1/TubeU)
Keywords: Separator, Dynamics, Bundle, Height, Heat, Transfer, Exchanger
References: None |
Problem Statement: How do I read LOOPST and AWSCOD from EXAOPC to DMCplus through CIMIO for OPC ? | Solution: Refer to attached document.
Keywords: DMCplus, connect, EXAOPC, cim-io, opc, smart data types
References: None |
Problem Statement: How is the LP solved? | Solution: The LP, due to ranking, is solved in a sequence of steps. These steps can be categorized into two main areas: 1. A feasibility calculation, and 2. an optimality calculation. The feasibility calculation solves a sequence of linear programs. The decision variables in these linear programs are the give-up on the CV constraints. The linear programs are constructed by looking at successive rank sets. All the linear programs contain manipulated variable (MV) constraints.
The first linear program that is solved contains the information on only the most important rank set (the one with the smallest rank values). The MVs are free variables in the linear program at this point. The only question that is being asked is whether or not a feasibleSolution exists. If one does, then the answer to the first LP will be zero. This means that there will be no give-up on this rank. If there is some nonzero give-up, (i.e. a nonzeroSolution) then the value of theSolution is the amount that must be either added or subtracted to the constraint to make it feasible. The CV constraint is adjusted and then the next rank set is considered.
The linear program for this set is similar to the previous one. In this case, however, the decision variable are the give-ups on the second (or next) rank set. The LP still has the MV constraints, and it has the CV constraints (which are now considered hard because the give-up was already calculated) from the previous LP. TheSolution then tells us how to modify the constraints of this, the second, rank set. This procedure is repeated for all the rank sets.
Once this part is done, we have a set of constraints that have been optimally adjusted in accordance with the priority ranking that guarantee a feasible input exists. The next stage is to solve the actual economic optimization. In this stage, the decision variables are the MV values (actually the MV moves). The LP has the same manipulated variable constraints and all the (now hard) CV constraints from the first stage of calculation.
This LP is solved to find the actual steady-state targets.
A couple of notes and clarification:
Ramps
Regarding the effects of ramps in ranks, please note that the LP is solving for the ramp to balance, not be within limits. Balance may occur at strange ramp values in the LP. You should probably have ramps in their own rank and that should be the most important rank. The LP must solve a ramp or the controller aborts, with exceptions based on allowed imbalances.
The steady-state target shown for a ramp variable has nothing to do with the LPSolution. Since the LP is solving for the steady-state rate of change, the true LP Target for a ramp variable is always zero. This is not useful for the operator, so what the controller shows as the SS target is the end of the closed-loop prediction vector. This value may or may not be within the limits, largely depending on the dynamic ECEs for the ramp variable, and the move suppressions. For a ramp CV, a SS target outside the limits DOES NOT mean that the LP has failed to balance the ramp.
ETs
External Targets (ETs) are actually calculated after the CV rank feasibility stage and before the economic optimization. The ETs are handled in the exact same manner as the CVs. The ETs are currently implemented by adding upper and lower constraints that pinch the CV or MV at specific value. The give-up calculation to find a new feasible set of constraints is the same as above.
Additional Notes:
Yes. If the constraint is feasible, the give-up is zero, and the constraint is a hard constraint for all following calculations. Nothing else gets fixed during.
If a constraint is infeasible, the consequence is a positive give-up on a constraint (see above). The feasibility stage treats the MVs as free variables tries to find the minimum amount of give-up. If things are feasible, the give-up is zero. If infeasible, the nonzero give-up is added to the constraint (the constraint is moved to guarantee feasibility) and the next constraint is then considered. There is a chance that some ETs may be ignored, but if the controller was well thought out, these instances can be avoided. The main case when this happens is when there are only a couple MVs that affect a given ET. If some other, higher ranked CV requires the MV move to keep the CV within limits, the ET will be given up on.
Keywords: LP, External Targets, CV, Ranks, Ramps
References: None |
Problem Statement: We defined descriptions in all MTGIND and MTGDEP entries. Before we
did so, we had the correct message in the message overview in DMCview.
But now, we only have the first 12 characters of the description, and
then nothing anymore. In the ENG file the whole message is still
available. What did we do wrong? How can we get the messages back in
DMCview? | Solution: For v1.04.xx, change all MTGIND and MTGDEP entries to be 12 characters. This problem is fixed for v1.10 to be out in early November.
Keywords:
References: None |
Problem Statement: How do I configure an anti-surge controller in Aspen HYSYS? | Solution: To model an anti-surge controller Aspen HYSYS is using a modified PID Controller logic, as the controller need to act very rapidly.
Refer to the attached PDF document for full instructions and to the attached hysys file and spreadsheet for and example.
Instructions
AspenTech product documentation can be copied to your server or client computer, or accessed directly from the documentation DVD. If you do not have access to the documentation DVD, you can download all documentation from the online Technical Support Center.
To access the documentation attached to thisSolution, follow the Instructions below.
.PDF Files
Printable documentation is published in Adobe Portable Document Format (.pdf). You must use the Adobe Acrobat Reader to view these read-only product-specific documents.
To view a .pdf document in your browser, select a document link. To print a .pdf document, select the print icon in the Adobe toolbar after the document loads in your browser. To download a document to your local drive:
From Firefox/Chrome, right-click on the document link and select Save Link As.
From Internet Explorer, right-click on the document link and select Save Target As.
Adobe Acrobat Reader allows you to navigate and print Portable Document Format (PDF) files. Visit the Adobe web site at http://www.adobe.com and download the latest version of the free reader.
If you have an older version of Acrobat Reader, you must first uninstall it with the Adobe Acrobat uninstaller available from the Start menu.
Keywords: Compressors, control, speed, flow limits, surging
References: None |
Problem Statement: What are the guidelines for converging and troubleshooting the Delayed Coker model and Visbreaker model in Aspen HYSYS Petroleum Refining? | Solution: The delayed coker model or The visbreaker model may fail to converge or calibrate for a number of reasons
· Incompatible property slate in feed assay
· Errors in calibration data (mass balance, property balance, etc)
· Large step change in key feed properties, operating conditions
· Solver settings
Feed Assay Data
The petroleum assay will automatically generate a full property slate and predict unknown properties. The visbreaker model is sensitive to the following properties, so these must be defined:
· Density
· Conradson Carbon Content
· Sulfur Content
· Nitrogen Content
· Distillation
The feed to the visbreaker is typically residue stream, so the model doesn’t expect light ends in the feed.The model will have trouble solving if there is a large fraction of non-resid components in the feed.
Calibration Data
The feed attached to the model is the feed for calibration calculations so please verify the data entered in calibration correct or not without any errors.
· Mass Balance: Make sure that the mass of products you are reporting is balances with the amount of feed
· Property Balance: Make sure that the amount of sulfur and nitrogen reported in the feed is balances with the amount reported in the calibration products. The H2S stream is calculated by difference, so the sulfur balance will not be an exact balance, but the feed should have atleast as much as is in the products.
Large Input Changes
Once the model has been calbrated and you have pushed the calibration factors into the simulation model, the model will re-solve automatically. After this, you can use the model to predict behavior at varying operating conditions. Some times, when making significant changes to process variables, the model will have trouble solving. The model is very sensitive to “Feed Assay� data properties & operating conditions like furnace temperature.
If you need to make a significant change in one or more of these variables it is recommended to take smaller steps to help find theSolution
Solver Settings
Sometimes, the visbreaker model will have trouble converging only because the solver settings are not allowing the model time to solve. If the model is not converging, try some of the following:
· Increase creep step iterations: This will increase the number of creep iterations used by EO solver. Make sure also to increase the maximum number of iterations. Try increasing the creep iterations to 50, and the maximum iterations to 100.
· Adjust the creep step size: Usually decreasing the step size will be the mode effective. This forces the solver to take smaller steps to find theSolution, which is helpful especially for non-linear problems. Try decreasing the step size to 0.05
Keywords: calibration failed, delayed coker model not converged, visbreaker model not converged
References: None |
Problem Statement: You are trying to load a controller and although you have checked and double-checked your Cimio_logical_devices.def file and your .cncdev parameter in the Aspen DMCplus controller configuration file, and they are correct, continue to get logical device not found error messages. This is most likely to happen after a controller has been off for a while (weeks or months) or have been migrated to a new platform. | Solution: The most likely cause for this problem is that there are non-default logical devices specified for a few tags in the controller. They could be anywhere in the .ccf.
Clients will no doubt recognize the tag name syntax for many DCSs which looks something like the following:
::FC2001.PV:DBVL:
If we expand it with explanatory text, it might look like this:
LogicalDevice:unit:Tagname.param:datatype:
The values for LogicalDevice and :unit: will default to the value for .cncdev and :.cncunit: unless they are overridden by explicit values. So, in your .ccf, you might have a value for .cncdev and .cncunit as follows:
.CNCDEV~~~CONSTANT~~~CH(31)~~~IOHCI~~~
...
.CNCUNIT~~~CONSTANT~~~I4~~~1~~~
and most of your tags will look like the above example:
::FC2001.PV:DBVL:
but back before your current recollection, the controller may have been designed to retrieve one or two tags from a different device, so that you may have a tag or two that look like this:
IOHCI2::HDLEVEL.PV:DBVL:
Where IOHCI2 refers to a different device which either is no longer defined in your cimio_logical_devices.def file, for some reason, or in the case of a migration to a new platform, has not been put in the cimio_logical_devices.def file.
Unfortunately, there is no simple way to search the .ccf using a search function in an editor to find the explicitly defined devices. You must engage in a tedious visual search of the file to see which logical devices have been explicitly defined, and then either define the new logical device, or determine if the value is actually needed, and if not remove it, or determine if it can be retrieved from the default logical device.
Keywords: None
References: None |
Problem Statement: How do I configure the set point ramping in the PID controller? | Solution: The PID controller in Aspen HYSYS Dynamics can be configured to include the set point (SP) ramping functionality. This facility is available in the Advanced page of the PID controller page. See the screenshot below.
The Advanced page contains the following four groups:
Group
Description
Set point Ramping
You can specify the ramp target, duration and mode.
Set point Options
This contains the options for setpoint tracking. You can specify the tracking options when the controller is manual mode, local mode or remote mode.
Sp and Op Limits
You can set the upper and lower limits for set point and output targets.
Algorithm Selection
This contains the PID controller algorithms for output calculation.
The set point ramping group contains the following two fields:
Target SP. Contains the set point you want the controller to have at the end of the ramping interval. When the ramping is disabled, the Target SP field displays the same value as the SP field on the Configuration page.
Ramp Duration. Contains the time interval you want to complete set point change in.
There are also two buttons available in this group:
Enable. Activates the ramping process
Disable. Stops the ramping process.
When the ramp mode is set Enable the set point changes over the specified period of time in a linear manner. This is only available when the controller is in Auto mode.
While the controller is in ramping mode, you can change the set point as follows:
Enter a new set point in the Target SP field, on this page.
Enter a new set point in the SP field, on the Configuration page.
During the set point ramping the Target SP field, shows the final value of the set point whereas the SP field, on the Configuration page, shows the current set point seen internally by the control algorithm.
During ramping, if a second set point change has been activated, then Ramping Duration time would be restarted for the new set point. The set point ramping can be used for ramp-up as well as ramp-down (see the screenshot below).
Keywords: Ramp, Set Point Tracking, PID Controller
References: None |
Problem Statement: What is zero speed flow resistance [k] sizing parameter for reciprocating compressor? | Solution: In HYSYS Dynamics, when the speed of the compressor is exactly zero, then the flow through the unit is governed by a typical pressure flow relationship:
Flow = K zero speed* Sqrt (Density * Frictional Pressure drop)
Where K zero speed = zero speed flow resistance
The zero speed flow resistance [k] parameter is only used in Aspen HYSYS Dynamics, it is not provided by compressor manufacturer. User can calculate this by pressing Size k... button available on Design II Setting page of reciprocating compressor
Keywords: zero speed flow resistance [k], Compressor
References: None |
Problem Statement: Why are derivative events - blend component (type 38) not published in _EVENTS table even when the option publish derivative event is selected? | Solution: Event Type ID: 38 was purposely, eliminated from Publishing with recent versions in order to reduce the amount of data that is published in the _Events table, and increase the publishing speed. The required information is also published in EV_XXXX tables. Therefore we suggest to use EV_XXXX tables to extract information related to type 38.
Keywords: type 38
publish
blend components
derivative events
_Events
References: None |
Problem Statement: How do I test an automation code for Aspen Petroleum Scheduler or Aspen Multi-Blend Optimizer? | Solution: Below are steps that can be used to test an automation code.
From Excel, press ALT-F11 to open VBA editor.
Click ‘ThisWorkook’ as shown below. This would open an empty panel on the right where VBA automation code can be written. You can also copy/paste code from the APS HELP.
From Debug Menu, select ‘Compile UnitsCode’. This option will be greyed out if the code is already compiled. Once you pass Compiling, you can click a breaking point in the code, then click run, the run will stop at the selected breaking point.
Now you can highlight the variable of your interest to check if there is any return values. Alternatively, F8 key can be used to debug the automation code line by line.
Keywords: None
References: None |
Problem Statement: Sum of MV moves <> steady-state | Solution: Solution
The message Sum of MV moves <> steady-stateSolution is the result of either the maximum move being set too small, or the move suppression is set too large. It may also require a visit to how the equal concerns are configured. In general this is a tuning issue. It may be advantageous to run the controller in Aspen DMCplus Simulate and adjust the setting until acceptable performance is reached.
It would also be recommended to check move suppressions.
Keywords: dmcplus, error, tuning, simulate, ece
References: None |
Problem Statement: Serious Bug in Copying a Model with DMCplus Model 1.10.04
This is a serious issue. This is based on a previous Advisory and is renamed as an Alert due to concern that the recipients did not take it seriously enough and that the writer did not express the serious nature of the problem concisely. It is shortened and re-written to impart importance. Please read carefully and follow the guidelines for copying a model in DMCplus Model 1.10.04.
Deleting a variable(s), or possibly even re-ordering variables, in a copied model before initially exiting the model properties box may result in model corruption with no evidence other than visual curve placement. Use of this model will be dangerous in an online controller and unexpected results will occur. | Solution: The correct procedure for modifying a copied model in DMCplus Model 1.10.04 follows:
copy the model
change the name in the properties dialog
press OK to exit the properties dialog
select the model
edit the properties to remove or re-order variables without problem.
The above procedure will not corrupt the model.
Carefully evaluate any models you have copied; compare them with the source model to verify that the curves are in the correct place. If corruption is found or suspected, correct the model by:
select the model
right mouse-click
select the option to re-assemble the model.
Keywords: DMCplus Model
References: None |
Problem Statement: Some of the CIM-IO servers support the acquisition of engineering units and
descriptions into the CLE file. Some of the servers have functioning smart data
types for AWSCOD and LOOPST. Some have been field-tested. Which ones? | Solution: Desc
Smart Field
Eunits
Listsz
Data
Tested
Biles AIM
Centum Exapapi
Centum Exapapi
no
no
yes
no
no
yes
yes ? v1.0
no -1 v 2.4
no -1 v 2.5 Cim-IO kernel 4.7.1
CM50S
CM50S
CM50S
--- -
yes
yes
yes
-
yes
yes
yes
-
yes 299 v 2.0 (Irevers=1)
yes 600 v 3.0
yes -1 v 3.2 CIM-IO kernel 4.7.1,
UCX 4.1 eco 6
Foxboro Bridge
Fisher Chip
Fisher Chip
??
no
yes
??
no
yes
?? ? v 2.0.2
yes -1 v ?? (in use, Sep 97)
yes -1 v 1.2
IBM ACS RAF/RXF yes
Mod 300
yes yes
??
1 v 1.0
??
yes 250 v ??
OpenDDA
OpenDDA
OpenDDA
OSI PI
no
yes yes
yes
no
yes
yes
yes
no
yes 99 v 1.2 (Irevers=1)
300 v 1.3 (Collect listsz=99)
yes 300 v 1.4 (Collect listsz=99)
yes ? v ??
Keywords: CIM-IO, CIMIO, Smart Data Types, Collect
References: None |
Problem Statement: When is it proper to use MV or CV tracking in DMCplus? | Solution: The intent of thisSolution is to document AspenTech best practice recommendations for the use of MV and/or CV tracking in DMCplus controllers.
MV Tracking - How It Works
Every MV in a DMCplus controller has a tracking flag called TRKMAN. The value of TRKMAN determines how an MV will be treated by the controller when an MV is requested to be turned on, and the current MV value (VIND) is outside the MV operating limits (ULINDM and LLINDM). An MV is being requested to turn on when the old independent variable status (INDSTA) for that MV is not good (that is, the previous value of INDSTA was something other than 0), and the current value of the MV loop status (LOOPST) equals 1 (the regulatory loop can now accept a supervisory setpoint).
TRKMAN has two legal values: 0 and 1. When TRKMAN = 0, MV tracking is not enabled, and an MV will NOT be turned on by the controller when the value of VIND is outside the operating limits. When TRKMAN = 1, MV tracking is enabled, and the violated operating limit will be internally overwritten with the current value of VIND if the MV is requested to turn on and VIND is outside the operating limits. If this is a controller off to controller on situation when MV tracking is performed, the current value of VIND will also be written back to the process control system (PCS) as the new operating limit.
Any new operating limit implemented by MV tracking will be clamped by the appropriate engineering limit if necessary.
MV Tracking - Recommendations for Use
AspenTech does NOT recommend MV tracking as standard practice. MV limits tend to be very important for operators since those limits determine the range over which the controller is allowed to move an MV. The operators should be trained that the MV operating limits (ULINDM and LLINDM) belong to them, and that those limits will not be overwritten under any circumstances.
However, if operators are trained that the operating limits belong to them and will never be overwritten by the controller, they must also be trained to make sure that the value of VIND is within the operating limits before turning on an MV since an MV outside operating limits will not be turned on if MV tracking is not enabled (TRKMAN = 0).
CV Tracking - How It Works
Every CV in a DMCplus controller has a tracking flag called TRKDEP. The value of TRKDEP determines how a CV will be treated by the controller when the controller goes from open-loop (ONSTS = 0) to closed loop (ONSTS = 1) and the current CV is outside the CV operating limits.
TRKDEP has three legal values: 0, 1, and 2. When TRKDEP = 0, CV tracking is not enabled, and the controller will then attempt to bring the CV back to the violated operating limit. When TRKDEP = 1, CV tracking is enabled, and the current value of DEP will be written back to the PCS as the new value of the violated operating limit. When TRKDEP = 2, a special form of CV tracking is enabled that causes the current value of DEP to be written back as the new value of both upper and lower operating limits (UDEPTG and LDEPTG).
Any new operating limit implemented by CV tracking will be clamped by the appropriate engineering limit if necessary.
CV Tracking - Recommendations for Use
AspenTech does NOT recommend CV tracking as standard practice. If the operators have been trained that they are to set the CV operating limits, the controller should not overwrite these limits. Some possible exceptions to this general recommendation are shown in the following paragraphs.
For a CV that is a true process constraint (such as a tube skin temperature, a metallurgical limit, a flooding limit, a valve output limit, or another type of process safety limit), the controller must take action to drive the CV back to the constraint limit. For such a case, always set TRKDEP = 0 to disable CV tracking.
For product quality targets that are controlled to a setpoint (that is, UDEPTG = LDEPTG), it could be acceptable to set TRKDEP = 2 to force the quality target in DMCplus to track the actual process value when the controller is turned on. However, DMCplus is probably not going to truly bump the process when the controller is turned on and DEP is different from the operating limits. In this case, it is acceptable to set TRKDEP = 2 IF the operators have been trained that this CV is a special case where we want the limits to track the actual value when the controller is turned on. However, it is certainly not a requirement to use CV tracking in this case.
One case where a setting of TRKDEP = 1 could be acceptable would be a feed rate CV, where feed rate is also an MV for the controller and both the VIND for the MV and the DEP for the CV read the same PCS setpoint as their current values. This would enable the operator to set wide operating limits for the MV but more narrow limits for the CV. This could be a way to implement a quick feed rate reduction without turning off the feed rate MV (to avoid a situation where the operator changes an MV limit to more than 1 SSSTEP away from the current VIND value, thus causing the controller to turn off due to an infeasible MV problem). In this case, it would be acceptable to implement TRKDEP = 1 to prevent the controller from cutting feed rate if the controller was turned on and the feed rate was above UDEPTG.
As an example, consider a case where the controller is off, the feed rate MV limits are 40 and 120, the CV limits are 95 and 105, and the current value of the feed rate setpoint at the PCS is 108. If the controller is turned ON and TRKDEP is not equal to 1, the controller will cut feed rate from 108 to 105. If TRKMAN = 1 in this case, the UDEPTG for the feed rate CV is written back to the PCS as 108 and feed rate is not cut.
Keywords: TRKMAN
TRKDEP
MV tracking
CV tracking
References: None |
Problem Statement: There are three mechanisms that can enable a write from a DMCplus system to the DCS directly or via an intermediate database product. The obvious mechanism is that the DMCplus controller has been turned on by the plant operators and begins writing to the MVs.
The other two mechanisms are less well understood. These are from DMCplus View and from DMCplus Manage Download. DMCplus users should be aware of these and exercise caution where appropriate.
DMCplus View reads and displays information directly from the DMCplus controller context. Some fields (not VIND, VINDSP, DEP) are shown in cyan color and allow entry of new values on the View screen. The value is changed in the controller context. If these variables have database / DCS connections, either read or write, then View will update that database / DCS connection with the changed value. Because of the limitations on what values will allow entry, data changes from View are not normally cause for concern.
The DMCplus Manage Download option allows the user send the current controller context values to the database on the host DCS directly or via an intermediate database. This is useful in the following situations where updating many values is required:
upon initial configuration of a controller
when an operating scenario changes needing new limits
for new parameters after re-tuning with DMCplus Simulate.
Only values for the parameters listed in the file $ACODHOME/cfg/download.dat that have database / DCS connections will be downloaded by the Manage Download option. Do NOT add the parameters VIND, VINDSP or DEP to the download.dat file. There is no built-in logic to prevent you from writing to any parameter listed in this file.
WARNING: It is advised to turn the controller off before using the DMCplus Manage Download option. Abrupt changes in operation may occur if the controller is on during the download. | Solution:
Keywords:
References: None |
Problem Statement: If you issue a delete command from the command line, manage allows you to delete a running controller without stopping the process.
There is no problem if you use the menu. However, if you use a command line like:
$ manage delete ctlr_name
while the controller is running, the process will continue but will not be evident from Manage. You have the control process running and you can no longer stop it or see it with a Manage stop or a Manage list command.
If you then load the controller via the Manage menu, you can stop it via the Manage menu. | Solution: The successful command line method to delete the controller is to first stop the controller via:
$ manage stop ctlr_name
then, delete the controller via:
$ manage delete cltr_name
Keywords:
References: None |
Problem Statement: Selecting File from the DMCplus View screen menu, then selecting Print from the pull down menu does not do anything -- the screen is not printed. | Solution: Copy the Config.key file from the GCS32\Displays directory into the directory where the DMCplus View screens are stored. This will enable the print option on the File pull-down menu and will also make tool bar buttons available that provide abilities to zoom in and out on a screen, to page forward and backward through screens displayed, to submit and cancel password validation, and to print a screen.
Note: A printer setup/selection window is not displayed when a print job is requested. The printer must be selected and configured to print in landscape mode from the Print Setup option on the File pull-down menu prior to the Print option being selected.
Keywords: DMCplus, View, GCS, config.key
References: None |
Problem Statement: This document explains how equal concerns errors (ECEs) in the DYNAMIC CALCULATION work. | Solution: Equal concern errors in the dynamic calculation are unlike ECEs in the steady-state or LP calculation. In the steady-state calculation, the ECEs are weights on constraint violation. They provide a way to make trade-offs in constraints of the same rank set.
In the dynamic calculation ECEs are used as weights on the controlled variables. Once a steady-state target is calculated, the controller goes through a sequence of steps to determine an 'average' equal concern error. The DMCplus controller offers the ability to input lower-, middle-, and upper-dynamic equal concern errors as well as transition zones. The actual Model Predictive Control (MPC) calculation then makes use of only a single ECE value for each controlled variable (CV) in the controller.
It does this as follows:
Create a first order trajectory from the current manipulated variable (MV) position to its steady-state target.
Using this trajectory, determine the resutling CV trajectory.
Use weighted values (based off of the CV trajectory and the input blocking) of the equal concerns (lower, middle, and upper) to determine an 'average' equal concern error for use in the move calculation (a.k.a. dynamic calculation).
Repeat this for all the CVs.
Once an 'average' ECE is found for each CV, the controller uses this value in the dynamic move calculation. It uses a constant value across the horizon. It does not vary the ECE value over the horizon.
The transition zones provide ways to modify the ECE based on the the value of the CV. The ECE is a piece-wise linear function of the operating space when transition zones are used. The value of the ECE is generally a constant in each region of the operating space. This results in a jump discontinuity of the ECE outside the limits and inside the limits. It jumps from the value of the lower-dynamic ECE to the middle-dynamic ECE to the upper-dynamic ECE. This is true when the transition zones are zero.
When the transition zones are nonzero, the ECE ramps in a linear fasion from one value to the next. So the value of the ECE changes in a linear way from the lower dynamic ECE to the middle ECE. There are two cases if the transitions zones overlap. If the transitions zones overlap and the ECE lines extending from the lower and upper constraints meet, then the ECE takes on values in a continious way from one line to the next, with a slope change at the intersection. If the lines do not meet, then a straight line is drawn from the lower ECE to the upper ECE. The middle value is ignored.
Keywords:
References: None |
Problem Statement: DMC v5.0 to (DMC v5.3 or DMCplus) Migration Issues | Solution: Each controller will be in a separate controller-named directory.
Only one custom input transform (CTRANI) and only one custom output transform (CTRANO) is implmented. This is unique to each controller and must reside in the controller directory. Use of CCF calculations is encouraged.
Equal concern and transition zone entry names have changed to ECELPU, ECELPL, ECECMU, ECECMM, ECECML, TRANZU, and TRANZL.
CVRANKU and CVRANKL were added to enable ranking of CVs for the LPSolution.
Composite Linear Program (CLP), External Targets (ET) and sub-controllers were added with DMC v5.3 and many new CCF entry names now exist.
CRIIND and CRIDEP have changed from 2-state to 3-state.
SREIND and SREDEP have been added as engineering-level service switches.
TRKMAN (2-state) and TRKDEP (3-state) were added to allow adjustment of limits at the master controller transition from off to on.
DMC 5.02 CCF
GENERAL SECTION
5.02 parameter DCS tagname
.WRITE I4 THISHH 0 DMC99THISHH !Time of last run (hours)
.WRITE I4 THISMM 0 DMC99THISMM !Time of last run (minutes)
.WRITE I4 THISSS 0 DMC99THISSS !Time of last run (seconds)
The time variables have changed, but you can look at THISTM, LSTTIM, LSTDAY, LSTSEC in the entry dictionary.
.RDWRT I4 NEWOTF 0 DMC99NEWOTF !Load new OTF
.RDWRT I4 RESCAN 0 DMC99RESCAN !Rescan .CFG
NEWOTF and RESCAN have no equivalent due to the changes in the system architecture. Manage Load generates an OTF equivilant and scans the CCF as it loads the controller.
& ... IN THE CV SECTION (CV#1)
.RDWRT I4 USEPRD01 0 DMC99USEPRD01 !Use Prediction Flag
NEWPV is the flag set to indicate a fresh value for updating the model bias term.
Migration from DMC 3
Here are some more comments on migrating from DMC 3:
The model file should work.
Tuning params are some different. Re-commissioning will be required to find new values for ECE, move suppression, etc.
The on/off switch logic is different and, along with other parameter changes, the operator graphics will likely need revising.
A computer running DMC 3 is not likely to be powerful enough to run DMCplus. Version 3 supported off-line tuning. DMCplus does this on-line at each cycle and puts a lot more load on the CPU. The memory and CPU both are likely to be limiting on the existing machine, if bought for DMC 3.
Keywords:
References: None |
Problem Statement: Note: Negative values of GMULT are not allowed in DMCplus 1.10.01 and before.
This issue is resolved in Patch001.
Some situations can require the use of variable gains. With DMCplus, you can manually adjust the gain using a multiplier applied to a response curve for a dependent/independent variable pair.
You may apply a gain multiplier to any or all of the response curves in the model. The muliplier is applied to the entire curve, not just the steady state gain.
However, this needs to be implemented with greater care than a static-gain model because the validation routines in the controller do not protect against bad or nonsensical gain changes. The gain changes need to be done smoothly or the controller may not react properly.
Note:
The CLP in DMCplus v1.0 does not handle GMULT.
Do not apply GMULT with this version. | Solution: Add GMULTE to the Configuration section as a user-defined, integer constant with a value of 1 to enable the use of Gain Multipliers.
Add GMULT as a user-defined real entry to the CCF in Build as a READ, or LOCAL if a CCF calc. It can be placed in the General Section or the specific Independent variable section.
In the General Section, it will look like this:
GMULTxxxyyy
where xxx is the DEP number and yyy is the IND number
OR,
In each Independent Variable Section, it will look like this:
GMULTxxx
where xxx is the DEP number (Build will append the IND number)
By default the controller will only accept gain multipliers with values between 0.5 and 2.0. To change these limits you must add two more parameters to the .ccf, GMULTL and GMULTU. GMULTL sets the lower bound for GMULT and GMULTU sets the upper bound. Use Build to add these two parameters to the Configure section of the .ccf. Make their data types Real. You must use a keyword of LOCAL for these parameters and set the default values to the desired number.
The absolute value of GMULT is limited between GMULTL and GMULTU. GMULTL & GMULTU should be expressed as positive real values. If either GMULTL or GMULTU is exceeded, the controller turns off and exits so that the predictions are not destroyed.
NOTE:
If you add, remove, or reorder the variables in your model Build will not automatically renumber the GMULTxxxyyy parameters. You must do that manually.
Regarding error Existing point cannot be AWRITE. tag_name
It turns out that GMULT is not a user-defined variable in the common buffer, only in Build. Since it already exists in the common buffer, you can't make it an AWRITE. If it is not written out as a WRITE, then you will have to create another user-defined variable, transfer the value via a CCF calc, and AWRITE the new user-defined variable.
Keywords: WRITE, AWRITE, CCF, GMULT
References: None |
Problem Statement: After running the DBupdate utility, do I need to manually change any of the table fields? | Solution: After running the DBUpdate tool, the user can click the Help button at the bottom right corner. There is a detailed procedure in the HELP explaining the procedure.
The last step (step 5) indicated the DBupdate utility also generates a script (highlighted on the HELP screen shot). Please find the generated upgrade script. The database administrator can then run the script to update those fields in the database.
Keywords: DBupdate
upgrade
database
utility
script
table
fields
field
References: None |
Problem Statement: s and | Solution: s:
Issues with the use of hiway gateway points:
1. Ensure that the box is in Full Control.
2. HG (highway) points. There are -TWO- timeout parameters on each MV. Make sure that you use the correct time-out parameter. There is T1 and T2. Check to see that you are using the correct one by pulling up a detail on the point and investigating the parameters. Also, drill down to the detail info on the HG from the System Status display, select the HG, select Netwk/Hwy status, and drill down further to see which of T1 or T2 that it is using. Aspen Technology, Inc. recommends using a value of 120 seconds or higher for the timeout (either T1 or T2).
Changing the timeout fixed the problem. This explains why the block was going into BCAS instead of AUTO since the block itself was shedding instead of the controller. Of course once the blocks were out of AUTO we could no longer write to them which is why the PUT errors.
MV is:
BC/MC
EC
INTERVAL*
SP
PIDSPC
PIDNORM T1 or
T2
OP
PIDCM
PIDDDC T1
or T2
PIDCMA
T1 or T2
? Set Interval to the one that is larger than the controller execution frequency.
3. The final control variable is an .OP on the HG. HG points will not recognize operator clamps on the output and ARWOP shows normal at all times when it is in PIDDDC mode, RCASENB=ON and Mode=CAS.
You can get some extra protection by building an intermediate point which will accept operator clamps. Write to the .SP of an AM point, which will ship the value of the .OP to the lower loop, but then the display shows an .SP as the MV, which is confusing to the operator. The preference would be to write to an .OP of the AM to make the display more understandable.
You can not write to the OP of an AM point in CAS which is required to use the Cim-IO smart-data types for anti-windup and loop status. You will need to write to the SP value but can have that pass the value through and display it as the OP value for the operators information.
The summary of the required steps is:
a. If the secondary (HG) will always remain cascaded to the primary (AM), then you have changed the actual MV from the HG point to the AM point and the standard CL can be used to manipulate it.
or,
b. If the secondary (HG) must also be cascaded to another primary or the operator must manipulate it directly, then do the following:
set the sub.INDTYPE(x) = NA to prevent the standard CL from touching the mode unless.
create separate CL to manage the mode changing / shedding of the HG point and the AM point that you will write to, this CL must shed the HG and/or AM point when either -- the controller is off, or -- both (the controller is on and the CDS_point.INDSTA is not good),
set the sub.MVPVID(x) = AM point name
set the sub.PT_TYPE(x) = 2 to indicate that display is for OP value
Keywords: Gateways, TDC3000, CM50s, cimio
References: None |
Problem Statement: What are Pipe Flow Models used for in HYSYS PIPE segment dynamic page? | Solution: In Aspen HYSYS Pipe segment dynamics the user has the choice to use two Pipe Flow Models.They are explained as below:
1) Simple pipe friction model method:
a) Turbulent: The calculation is fast and simple. This method calculates the friction factor once and uses that value irrespective of the Reynolds number (the calculated friction factor value is not correct if the flow is laminar).
b) Full Range Churchill: The Full-Range Churchill method calculates the friction factor as a function of the Reynolds number. This method is slower but calculates a unique friction factor for the turbulent, laminar and transitional regions. If the flow through the Valve is too low, Aspen HYSYS uses a low limit of 10 for the Reynolds number. For details on the equation you might refer toSolution 121729 or page 6-170 of Aspen HYSYS Operations Guide.
2) Pipe Model Correlations: Allows you to select the pipe flow model based on the available pipe flow correlation selection from the Parameters Page in the Design tab. The calculation time for this method is long and rigorous, however, the results are more accurate.
Keywords: Pipe Flow Model, Pipe Friction.
References: None |
Problem Statement: The AxM services file was modified for the installation of DMCplus. On system reboot, the 3 cdsdaemons required for the AxM wouldn't start. | Solution: >On my recent trip to Europe I encountered 2 more of the renegade AxM boxes
>that didn't want to start the 3 cdsdaemons required for the AxM. After
>talking with TAC in Europe Rafi and I discovered that there is some sort of
>problem with the services files in the /etc directory. The TAC folks asked
>us to move a few lines in the file, like they did successfully at another
>location, but that didn't work for us. Instead we copied the origianl file
>back into place and reapplied our edits for the IODDA and CIMIO processes.
>Visually the two files looked identical, but rest assured only the re-edited
>file worked properly.
>
>It looks like we should insure that there is always a good backup of the
>services file before we start our installs and that we should fall back to
>it if we encounter any problems. You should always perform a reboot to
>ensure the file will work once you leave....
Keywords:
References: None |
Problem Statement: Comm Buf error between DMCplus and Composite
CLPTGT 15024 Composite - Comm buffer connect error | Solution: The Comm Buffer error is caused when there is an error connecting to the shared memory that Composite uses to communicate. When a controller communicates with the composite it maps and locks this buffer. When it has completed its communications, it disconnects and frees the lock. The error can occur at the beginning of the cycle, when the controllers report in initially, or during the PUTPLN when the move plan information is being communicated. In this case it appears to happen during the PUTPLN, and always with the FEEDBIAS controller that is in the second tier.
This could happen because the machine is getting a bit more loaded causing the controllers in the previous tier controllers to take a little longer releasing the buffer for the second tier.
This problem was observed at 2 sites where they had 14 furnaces in the first tier. Artificial tiers were introduced to spread the load out over time a bit. The first 5 furnaces were assigned to tier 1, the second 5 furnaces to tier 2 and the last 4 to tier 3. The FEEDBIAS controller was then moved to tier 4. This helped spread out the communication load and the Comm buffer problem went away.
Keywords: CLP, DMCplus, Composite, CommBuf, combuf, 15024
References: None |
Problem Statement: What considerations should be taken into account when selecting the correct feedback controller? | Solution: The following is a flowchart that outlines a method for selecting a feedback controller:
In general, if an offset can be tolerated, a proportional controller should be used. If there is significant noise, or if there is significant dead time and/or a small capacity in the process, the PI controller should be used. If there is no significant noise in the process, and the capacity of the system is large and there is no dead time, a PID controller is appropriate.
Keywords: feedback, controller, PI, PID, P, proportional, integral, derivative
References: None |
Problem Statement: Using Wollongong TCP/IP software. | Solution: Wollong Pathways
(aka Attachmate Pathways)
Must be 2.5 or later. This command shows the version:
$RUN TWG$TCP:[PATHWAY_ADMIN]PATHWAY$LIST_SOFTWARE
LISTSZ must be 600 or less (with 4.6B.) LISTSZ should be -1 with kernel 4.7.1.
Pathways consumes one file quota per socket. Each of the CIM-IO processes uses two file quotas for the log files. A PQL_MFILLM of 2 means that, after the log files, no quota is left for sockets.
Modify the cimiocom:cimio_cm50s_start.com procedure to set /FILE_LIMIT=150 on each RUN/DETACH command in the procedure. This overrides the SYSGEN-defined process quota limit for open files.
Defining services for CIMIO
Amoco has standard numbers for the services
Edit TWG$TCP:[NETDIST.ETC]SERVICES. and add these entries:
CM50DLGP
3070/tcp
# AspenTech CimIO CM50S DLGP
CM50READ1
3071/tcp
# AspenTech CimIO CM50S READ1
CM50WRIT1
3073/tcp
# AspenTech CimIO CM50S WRIT1
GCSDMCP
3074/tcp
# DMCplus View
CM50READ2
3075/tcp
# AspenTech CimIO CM50S READ2
CM50WRIT2
3077/tcp
# AspenTech CimIO CM50S WRIT2
CM50RSLV
3078/tcp
# AspenTech CimIO CM50S RSLV
CIMIOLOGGER
3079/tcp
# AspenTech CimIO Logger (optional w/kernel 4.7+)
CM50UTIL
3080/tcp
# AspenTech CimIO CM50S UTIL
Defining services for VIEW
Edit TWG$TCP:[NETDIST.ETC]SERVERS.DAT and add this entry:
# DMCplus View Server
service-name
GCSDMCP
program
$1$DIA1:[DMCPLUS.BIN]DMCPVIEW_TCPIP.COM
socket-type
SOCK_STREAM
socket-options
SO_ACCEPTCONN | SO_KEEPALIVE | SO_DEBUG
socket-address
AF_NET , 3074
username
DMC_USER
device-type
UCX
INIT
TCP_Init
LISTEN
TCP_Listen
CONNECTED
TCP_Connected
SERVICE
Run_Program
Note: the SO_DEBUG is probably not needed.
Note: you must specify the full path to program -- no logicals
please.
Note: you will have to stop and start INETSRV to get this
entry picked up by Pathways.
Setting socket message size limit
These commands are required to emulate UCX (which has allows up to 65KB message size.)
$ SET COMMAND TWG$TCP:[NETDIST.CLD]NETGEN.CLD
$ NETGEN SET TCP_RECVSPACE 65534 /WRITE=ACTIVE
$ NETGEN SET TCP_RECVSPACE 65534 /WRITE=CURRENT
$ NETGEN SET TCP_SENDSPACE 65534 /WRITE=ACTIVE
$ NETGEN SET TCP_SENDSPACE 65534 /WRITE=CURRENT
It has been discovered that under 2.5.1 of Pathways these parameters are reset to 8192 on TCP/IP restart -- no matter whether you WRITE=ACTIVE or WRITE=CURRENT. A procedure should be setup to set these before starting any CIM-IO processes.
Keywords: DMCplus, CIM-IO, Wollongong, UCX
References: None |
Problem Statement: How to add a level controller to downstream of Reboiler weir | Solution: Unfortunately the reboiler operation does not have the Chamber 2 liquid percent as one of the variables. Thus the user is not able to control the liquid level downstream of weir. However, there are two ways in which you can work around this to use the level in chamber 2 as a control variable.
The first method is to use a separator operation which allows you to configure the liquid percent downstream of weir. Please follow the instructions below to control the liquid level downstream side of the weir.
1. Delete the default kettle reboiler and add a Separator.
2. Add your feed and product streams. Go to Rating tab and enter the dimensions.
3. Enable the Weir and specify the dimensions of the weir.
4. Go to dynamics tab | heat exchanger and configure the tube details.
5. Add a level controller and select the PV as chamber 2 liquid level.
The second method is to use the Spreadsheet block as a bridge to retrieve that value from the Weir definition window. You can just right-click on the variable you are interested in and then drag and drop it into the spreadsheet. You can then use the variable in any cell of the spreadsheet as a controller variable.
Keywords: Level, weir, chamber 2
References: None |
Problem Statement: How do I change default sampling interval in HYSYS Dynamics? | Solution: The default sample interval is 20 seconds and number of data points in the logger is 300. For any individual stripchart this can be changed via Right-mouse click and selecting the Graph Control page.Click the Set-up Logger button to open the page as shown below.
The default setting can also be changed in the Preferences settings as shown below.
For HYSYS V7.3 and older versions:
For HYSYS V8.0 and newer versions this can be changed from Simulation Options page as shown below.
Keywords: Sample Interval, Data Logger
References: None |
Problem Statement: DMCplus Collect?IACO BASE???????i?????e???B | Solution: Collect?ITest API?????l?E????????CIM-IO?N???C?A???g?\?t?g?????B???????A?A?AACO_BASE???K?v?????U?????B
?b????????????.exe???>?I?u?I?e?`?t?@?C??(input file)???????ECIM-IO???Y?u???e?A???e?}?V???E?R?s?[???e???????A?@Collect?I?i?????A.bin?t?@?C?????i?e?????????
Keywords: DMCplus
Collect
ACO BASE
CIM-IO
JP-
References: None |
Problem Statement: Our PIMS model has recently been updated with a new crude. How do I add this crude to the APS model? | Solution: Given that there is no specific workbook logic regarding specific crude types - which is usually not the case - we will demonstrate below one possible and fairly simple way to add a new crude to the APS model:
1) Open your model
2) Find table CRUDES under ASSAYS branch of TABLES
3) Insert data for the new crude, here we used the code NEW:
4) Open EIU and make sure there that you are connected to the right model by clicking on the connect button shown below.
5) Open template – since we are showing example on Demo model it will be default template:
6) Find Assays sheet (or more of them if there are multiple) and insert the PIMS assay data, Here we will copy the TJL data as it is just an example:
7) In PIMS_CRUDE_XREF sheet map the assay data as shown below:
8) Validate spreadsheet by Clicking on the appropriate (Validate Spreadsheet) EIU button while in Assays sheet (repeat this step if more than one assays sheets will be updated, also: accept to validate mapping data as well by clicking YES when propmpted):
Once this is finished you should see this message:
9) After this a similar step needs to be done with “Audit Data” EIU button which is situated just next to “Validate Spreadsheet”. Again you need to be in the Assays spreadsheet and choose also mapping data.
10) Validation step will open up an Assay Import Window. In its last tab you need to choose the assay or assays which you have updated. After this the “Import” button will become available and needs to be pressed. Screenshot below shows a successful import and also the Assay Data tab with appropriate selection - this depends on your actual case and the crude untis to which you want to add the new assay:
After this you only need to save your model and you should be able to use this new crude from now on. You can verify that the crude has been modeled by checking the CRDCUTX table of the ASSAYS branch.
Keywords: APS, Crude, Assay, New
References: None |
Problem Statement: Pooling crude tanks is easy if you just want pool volume/composition by crude. However, what if I want the properties of that mix of crudes? | Solution: For crude pool types, the properties by component crude are reported. The available volume and ullage are based on the fraction of the crude in the pool. However, for viewing the properties of the pool, you can add a new trend line of the pool for each property, after reloading the simulator. (Found in the Trend->Trend List menu)
Keywords: Tank pooling,crude tanks, crude mix properties
References: None |
Problem Statement: Frequently, a controller will need to be changed, and that change could involve removing a variable that has been determined to not be useful in the control strategy. How is that variable removed from the model? | Solution: To remove any variable pair from a model,
Import the model from which you wish to remove the variable
Create a new assembled model
Set the two models where both are visible on the desktop.
Highlight some of the variables you wish to have in the new model, careful to not include the variables you want to delete. Drag and drop the highlighted variables into the assembled model.
Continue to highlight variables you wish to copy to the new model, until all the variables except for the ones you wish to remove are in the new assembled model.
Save the new assembled model.
An alternative to the above:
Make a copy of your model first. Right click on the copy of the model and select Model Properties Go to the Tags tab in the pop up window and remove the desired CV or MV. You can also add comments at this point to say that you've removed a CV or MV from a previous model. Now you are done, no dragging, dropping, or saying yes to add CVs & MVs to the new assembled model.
Updating the controller:
To update the controller, open the ccf file in DMCplus Build
Save the CCF with the updated mdl file, ( rename original mdl file and copy new mdl file to the same location )
Stop the controller using Manage
Unload the controller using Manage
Copy the new ccf to aconline/app/controller folder ( after renaming the ccf file that is already there, if necessary )
Use Manage to load the CCF
Use Manage to Start the updated CCF
Keywords: None
References: None |
Problem Statement: Selecting each object from the variable navigator to add to the Dynamic Initialization Manager one at a time is tedious. Is there a way to select multiple objects all at once? | Solution: To add several objects in one go, follow the steps below.
1. Select in the flowsheet the several objects you want to initialize.
2. With the mouse right-button click in one of them (while all selected), and choose Dynamic Initialization
3. From the menu that appears, select the desired option (Add Objects to Default).
Then, in the Dynamic Initi Manager, you should be able see selected objects and edit your choice (remove units you don’t want, add another, etc.)
Alternatively, this menu can be accessed through the right-click panel (show above), by choosing 'View Dynamic Manager'.
Keywords: Initialization, Dynamic, Multiple objects, Flowsheet
References: None |
Problem Statement: How are pipe fittings modeled in Aspen HYSYS Dynamics? | Solution: In steady state, appending fittings to a pipe affects pressure drop calculations in the the pipe and a fitting loss factor (K-factor) is determined for the fitting (refer toSolution 109436) and subsequently an equivalent length will be assigned to the fitting. Take note that the K-factor for swages are determined differently. When the model is taken to the dynamic environment the K-factor will affect the P-F relationship in Dynamics.
In dynamic mode the fitting type segments do not give rise to holdups, i.e there is no real dynamics effect from them. The only effects the fitting type pipe segments show in dynamics is therefore their contribution to the pressure drop, based on the K-factor which was pre-determined in steady state mode.
Keywords: ,
Pipe fittings, Holdup, Dynamic
References: None |
Problem Statement: I have installed the latest ER2 to my DMCplus Controllers and they are not showing up on the WebServer. | Solution: Unload ALL controllers, even the controllers that are not running, or were loaded originally for testing, such as the COL5X3. Reload and restart the controllers as needed.
Then go to Program Files \ Common Files \ AspenTech Shared and run the dmcpview_shutdown.bat to ensure the dmcp_viewsrv process is stopped. Verify in the task manager. Then run dmcpview_startup.bat to restart the dmcp_viewsrv process. Confirm the process is running in the task manager. Then go to the Production Control WebServer and go to the Preference tab, and use the Scan Now button to re-establish the connection with the dmcp_viewsrv process.
Keywords: DMCplus, PCWS, Production, WebServer
References: None |
Problem Statement: Apollo Viewer is not working, reports No controllers are running to attach to cannot save run file. | Solution: The work around of this issue is using the Production Control Web Server to SAVE the RUN File.
If you go to the Configuration Page of the PCWS, Select the Apollo Column sets, Select General from the Category drop down list, and Controller Detail(engineer) from the Apollo Column set to edit. There are two Save_config options, one is save_config: Save Run File and save_config_on_exit: Save Run File On Exit. Make your selection from the Available list and move it over to the Selected list. Return to the Online page and on the Overview, you should be able to click the Apollo application and see a list of the column sets, where you can then turn on or off the Save Run File option.
If these newly added options are not available, try a page Refresh, or close the web session and open a new one.
Keywords: Apollo, viewer
References: None |
Problem Statement: GCS will not run, BGCSNET not running | Solution: Add the following lines to the autostartlist.txt which can be found in the ACONLINE/SYS/ETC directory.
PROCNAME=BGCSNET
SEQNUM=79
APPPATH=$ACODHOME\bin\bgcsnet.exe
ARGLIST=gcsdmcp E:\AspenTech\AC Online\bin\dmcpview.exe
APPTYPE=0
This will start the required process on a reboot of the machine or when an Acobase_Startup is performed.
The attached zip file contains a .bat file which will enable the user to start the process manually without shutting down and restarting the control system.
You will want to place the file in the PROGRAM FILES / COMMON FILES / ASPENTECH SHARED directory with similar .bat files used to start and stop a variety of processes used by the advanced control system.
Keywords: GCS, BGCSNET, dmcview
References: None |
Problem Statement: DMCplus Controller reporting QP Internal errors | Solution: This is to be considered a first pass at an attempt to trouble shoot any QP errors:
We recommended use of the SVD Tool which can be found in the DMCplus Desktop / Utilities directory for checking the model file. It is also recommended to reset the paramater EPSMVPMX to 2, you will have to stop, unload, reload and restart the controller to see the change. In later versions of the application this has already been done. In older versions you will want to reset and then run the controller. If the problem persists, set this to 3 for additional debug information.
If use of the SVD Tool and the debug information does not provide enough information to correct, send the CCF, MDL, .prt, and .eng file to [email protected]
Keywords: DMCplus, QP, Internal, SVD
References: None |
Problem Statement: DMCplus Model fails to plot and/or create the Model Uncertainty plots for datasets whose sampling interval exceeds once per minute. In addition, regional settings other then English (US) may cause the plotting utility to fail (i.e. data and timestamp format other then English (US).
The attached plot untility executables address the Regional Settings issue, and allow the plotting of uncertainty information for FIR identification results regardless of data collection interval. Unfortunately trending of Subspace ID results are not supported by this attachment. | Solution: Copy the attached zip file and extract both plotting executables into the DMCplus Desktop folder.
Keywords: DMC
DMCplus
DMCplus Model
Model Uncertainty
References: None |
Problem Statement: Is DMCplus 4.1 supported on Windows XP? | Solution: DMCplus 4.1 is currently supported on Windows NT, Windows 98 and Windows 2000 Only. Version 5.0 of the DMCplus Desktop tools, which will be part of the AMS 5.0 deliverable expected by the end of 2002, will be supported in the Windows XP environment as well. Keep in mind this will be the Desktop applications only, DMCplus Build, DMCplus Model and DMCplus Simulate, NOT the online control software.
Keywords: DMCplus, XP, Windows, Desktop, Online
References: None |
Problem Statement: I compiled my User Blend Model Library file (UBML.dll) and it works well with APS on my computer. However when I sent the file to my colleague, it does not work with APS on his computer. | Solution: The problem is most likely caused by compiling UBML file in Debug mode. In order to be able to use UBML file on any machine it needs to be compiled in Release mode:
Recompile the UBML.dll in the Release mode and provide that to your colleague.
Keywords: UBML
compile
References: None |
Problem Statement: How do I see the velocity and temperature profiles in the pipe segment in Aspen HYSYS dynamics? | Solution: There are two possible ways to view property profile in pipe segment when running the HYSYS in dynamic mode.
1. In Dynamic mode the profile is based on holdup in the pipe segment. A pipe can be configured with multiple segments. This will provide holdup information for each segment. More number of pipe segments will give better result accuracy and more flexibility in terms of information the user requires from the pipe. However, increasing the segments will increase the calculation time and therefore slow down the simulation speed. Screen shot from the attached HYSYS Dynamics case.
2. For rigorous pipe simulation the Aspen Hydraulics can be used. This also gives profiles for each increment including phase velocity. There are sample cases for Aspen Hydraulics in the HYSYS installed folder as shown below.
Keywords: Velocity profile, Temperature profile, Pipe segment, Holdup, HYSYS Dynamics, Aspen Hydraulics
References: None |
Problem Statement: We have found that some firewall software applications may prevent the ACO desktop software from obtaining a license although the license server is installed and configured on the very same computer. Typically an error dialog opens showing the user that the particular desktop application is trying to access the internet and thus prevented from accessing the network. | Solution: Typical installation procedure is to specify the node name of the license server on the AspenTech License Manager Selection dialog. Thus in order to resolve the ip address for the selected server name, the licensing software must access the network to obtain the ip address from the host table. It is this network access that is intercepted by some firewall software and prevents the ACO desktop application from obtaining its necessary license.
We can bypass the need to obtain the ip address from the host table by specifying the server name as: localhost. As an alternative, we may specify the following ip address as well: 127.0.0.1 (which is the resolved ip address for localhost).
NOTE: thisSolution is valid only where the license server and desktop software are in stalled on the same computer.
Keywords: DMCplus desktop, SmartStep desktop, AspenIQ desktop, AspenWatch desktop, DMCplus, SmartStep, AspenIQ, AspenWatch
References: None |
Problem Statement: Firewall Configuration recommendations and considersation | Solution: The basic objective of any firewall is to limit the connectivity between the outside and the inside so that you can exercise as much control as possible on the traffic traversing the firewall. Obviously, you can make a totally secure firewall by disconnecting the outside from the inside. But, while totally secure, that is generally not too useful.
So, you try to limit what you can and still maintain functionality. One way to do that is to establish a gatekeeper on the outside of the firewall who is trusted by those inside and let him vet the traffic that wants to traverse the firewall. This is what a proxy server does. It lives in the dangerous outside and evaluates and authenticates the requests that originate from the outside, passing on the ones deemed to be safe to the inside residents. Ideally, you can limit access to one machine (by MAC address or maybe IP address) and one port (often port 80 in the case of web proxy servers). All other traffic is refused and the vulnerability is severely limited. Less ideal is the idea of allowing one machine multiple-port or even total access (by MAC address or otherwise). The more paths you have open through the firewall, the greater your potential vulnerability.
We recommend having all the APC functionality inside the firewall. That way, all can be treated as trusted, there is no reason to limit access and no interference with functionality. Where that is not a reasonable choice, the network design should limit the traffic across the firewall to as few paths as possible so as to be able to monitor and control the activity between the outside and the inside. Normally, this means identifying and minimizing the connections that are required between the outside and the inside. Limiting the traffic across the firewall to only what is necessary will insure the maximum degree of security possible.
You can write firewall rules to filter access by MAC address (or IP address if the AW and PCWS machines have static addresses instead of using DHCP). You would then allow access on all ports between any machine on the PCN with the specified machines on the Office network but not with any others. The only point of contact for machines on the Office network would then be the AW and PCWS servers. All other access to the PCN would be disallowed.
Filtering by specific port address is not possible because the DAIS trader randomly assigns port addresses when access to a service is requested. If you didn't want to leave the AW server outside the firewall, you could also move the PCWS inside the firewall and then use a proxy server to access the PCWS from the outside. This would limit the traffic through the firewall to a single port (which wouldn't even have to be port 80) and a single machine pair (proxy server <==> PCWS), which is about as secure as you will ever get. MAC address filtering is preferable to IP address filtering because it is harder to spoof, but it means that if a NIC is replaced the rules need to be revised to allow the new MAC address access. IP address filtering is not practical if DHCP is used to assign IP addresses.
Also see:Solution 123707 - Commonly used Firewall ports for APC Applications
Keywords: None
References: None |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.