Create a model for your process unit

The Process Model tab of the Process Units page shows a list of the process models currently associated with the process unit selected in the Process Unit drop-down menu.

The list of globally available process models can be reviewed in the Model Explorer.

Build a process model from scratch

Follow these steps to build a process model from scratch. For background information on how process models are associated to process units, see Process Units and Process Models.

1. Add a process unit

   In the left pane, click Flowsheet in the Asset section. The Flowsheet page opens.

   Click Process Unit in the Model Toolbox window and drag to the flowsheet canvas. A new process unit is created. If you cannot see the Model Toolbox window, make sure that the Toolbox button in the Flowsheet group of the Home ribbon tab is active.

   Add inputs and outputs and rename pipes as needed.

   

2. Open the process unit page

   Double click on the unit in the flowsheet to open the unit page. Alternatively, click Process Units in the Operational section of the left pane, and choose the correct unit from the left-hand side list in the Overview tab. Click the list of units and press Ctrl + F to display a search box and quickly locate a unit by name.

   

   By default, the units in the left-hand side list are grouped by unit type. Right-click on the list and choose Group By Type from the context menu to disable this option and list the units in alphabetical order.

   The flow units of measure in the Input Sockets and Output Sockets panes are those configured in the Flowsheet page, in the Weight Flow and Volume Flow controls of the Flowsheet ribbon tab.

3. Add a process model

   Open the Process Model tab. Select Add > Create New from the Process Model group of the Home ribbon tab. The Create New Process Model dialog window opens.

   Enter a Name and optionally a Description for the model. Choose a Type from the drop-down menu. Click OK when finished. The new model appears in the Process Models list.

   

4. Set stream associations

   Open the Model Structure tab and make sure that the Stream Associations page is selected. For each model connection in the Inputs and Outputs lists, choose the pipe to associate it with. Double-click an item in the Flowsheet Pipe column to show a drop-down menu listing the available pipes.

   

   To add or delete a connection, do either of the following:

   • Click Add Input or Add Output (Delete Input or Delete Output) in the Model Connections group of the Home ribbon tab.

   • Right-click on either the Inputs or Outputs pane, and choose Add Input or Add Output (Delete Input or Delete Output) from the context menu, respectively.

     Once the connection has been created, you can type a name for it and select the associated flowsheet pipe.

     

5. Add drivers

   The properties of your feed affect the properties of your output products. Feed properties that affect the product properties are known as drivers. As the values of drivers change, so do the model predictions. The detailed way in which model predictions are governed by drivers is configured in the last step of this workflow.

   To add drivers to your model:

   1. Stay on the Model Structure tab and open the Drivers page.

   2. Click the Add button in the Drivers group of the Home ribbon tab. A list of driver types appears.

   3. Click on the type of driver you want to add to open the corresponding editor.

      

6. Add predictions

   While drivers are the input variables of a model, predictions are its output values. Adding predictions is done in the same way as adding drivers.

   4. Stay on the Model Structure tab and open the Predictions page.

   5. Click the Add button in the Predictions group of the Home ribbon tab. A list of prediction types appears.

   6. Click on the type of prediction you want to add to open the corresponding editor.

   See Predictions for more information on the types of prediction and the corresponding editors.

   

7. Edit model structure

   The Model Structure tab shows a list of the models available for your process unit.

   In the Model Structure tab, open the Default page under Base Delta Structure. Note that, in the image below, Default is the name of the base defined in the Bases page. You can rename this base and add new bases from the Bases page. On this page you can edit the details of how prediction values are calculated by the model.

   You can disable predictions by clearing the check box in the Active column for the corresponding row. Similarly, you can disable drivers by clearing the check box in the top row for the corresponding column.

   If you have multiple bases, enabling or disabling a driver or prediction for one base also affects the other bases.

   When you disable a driver or prediction in this table, this is reflected in the Drivers or Predictions page, and vice versa.

   

   For distillation units, the Model page replaces the Base Delta Structure page. In the ECP/SI Model subpage you can define the distillation parameters of the unit.

   

   Some Base + Delta models use multiple bases, that is, different models at different points in their operational ranges. This is used to overcome non-linearity in models, by breaking them into smaller models which can be modeled linearly. The exact model used during simulation and optimization by Plan is controlled by an input variable, and this is used to choose the appropriate model.

   If a Base + Delta model has multiple bases, the value of the interpolation variable for each base is shown in the base list.

   

   Example: A reformer may have several different operating models, each of which represents running the reformer with a different degree of reforming severity. One model is used to simulate running the reformer to produce 96 RON reformate, while a different model is required to simulate running the reformer to produce 102 RON reformate. Different models are required when a single model cannot accurately describe the results of operating the process unit using differing conditions.
   A reformer model may also have several bases, depending on the final output RON. One model might be used if the target RON is 100, and a different model is used if the target RON is 102. This ensures more accurate calculations of the reformate yield and properties. During optimization, a target RON is defined, and this is used to choose the appropriate model. If the target RON is 104, the 104 model is used. If the target RON is 103, a weighted combination of the 102 and 104 models is used (that is, a 50:50 combination of the models).

   An Offset is a value added to the calculated value for use in any other calculations. It may be negative. Offsets are used to adjust the final calculated value of a Base + Delta calculation, causing the value calculated by the model to be shifted slightly. They are commonly used to make small adjustments to the predictions of Base + Delta models, to make them closer to the current plant operating conditions.

   A single process unit may use several models, representing different operating modes of the process unit. To use multiple process models, see Use Multiple Process Models Within a Single Process Unit.

   Once you have set up your model structure:

   7. Save your changes by clicking the Save icon near the top left corner of the application window.

   8. Share the model by committing the workspace. This creates a new global process model.

When creating a new process model, the model is created with only inputs and outputs matching the containing process unit and only output yield predictions. Therefore, configuring the structure may take some time. It is often quicker to copy existing process models and modify these.

To add a copy of an existing process model:

8. In the Process Models pane, select the model you want to copy.

9. Click Add in the Process Model group of the Home ribbon tab. Click Add Copy in the menu that appears. The copy of the selected model is added to the Process Models pane.

10. Click Model Info in the Process Model group of the Home ribbon tab. The Edit Process Model dialog box opens. If needed, change the name, description, share and type of the model.

11. Edit the model drivers, predictions and structure.

12. Save your changes by clicking the Save icon near the top left corner of the application window.

13. Share the model by committing the workspace. This will create a new global process model.

Edit the basic info of your process model

Note: If a process model has been imported from a password-protected file, it cannot be edited.

Edit the name, description and share for the process model using the Edit Process Model dialog.

To change this information:

14. Select the process model you want to edit in the Process Models pane.

15. If the Status of the process model is set to Read-only, enable editing of the process model by clicking Enable Edit in the Process Model group of the Home ribbon tab. (See also: Enabling Shared models for editing). You can only enable editing for models if you have the correct permissions.

16. Click Model Info in the Process Model group of the Home ribbon tab. The Edit Process Model dialog box will open (see previous image). If needed, change the name, description, share and type of the model.

17. If needed, edit the model drivers, predictions and structure.

18. Save your changes by clicking the Save icon near the top left corner of the application window.

19. To create a new independent process model based on the current structure, create a new version of the model in the required share.

20. To update the original process model and create a new version, you must update the Global Reference Data.

Make existing, shared models editable

Note: If a process model has been imported from a password-protected file, it cannot be edited.

Normally the process models used in Plan are approved shared models. However, it may be necessary to change existing shared models. In this situation, a shared model can be made editable, and then all of the model's configuration may be edited.

To enable a shared model for local editing:

21. Select the model in the Process Models pane.

22. Click Enable Edit in the Process Model group of the Home ribbon tab.

Once a process model is converted to an editable version, you can change any of the model settings such as the number of inputs or outputs, or the values for property drivers in Base + Delta models. These changes can be made in a tool such as Process Model Manager. You can also re-publish an edited model to a shared location. See: Model Contents.

Note: When you make edits to a process model, the edits are local to the current supply chain workspace. You are not editing the global Reference Data and the changes are available only within the current workspace. Changes apply to the current case and its descendants, unless these changes are also overridden in the descendant. To make the change permanent in the workspace and the global Reference Data, you must version the change. This can be done by publishing a new object. See: Creating a global versioned copy of a process model.

Tip: You can compare versions of process models between model workspaces and cases using Update Case. This will highlight the differences between versions of the same object.

Make a process model available in a particular mode

Normally a process model is associated with one or more operational modes, and each operational mode is associated with one or more process models. The Model Mode Connectivity grid allows you to define the link between models and modes. To make a process model available in a particular mode, select the check box at the intersection of the model column and mode row.

When this check box is selected, it means that when the unit is operating in that particular mode, material can be routed to the particular process model. The pipe mapping that directs actual material flow into a particular model is defined using the Stream Associations.