Advanced Data Management

Process Model Setup explains how you set up the model structure, and how you enter the data in order to create a full mathematical model of a process. This section introduces you to the data management features that are of particular interest for off-design modelling and for case studies. It describes how you create more than one mathematical model for a single process scheme. You will learn how you can transfer data between the mathematical models, and how you can use these features to maintain consistency in the models, for different load cases.

Model Refinement and Off-Design Modeling

During the course of a project you will frequently have a varying demand on the models that you are using. Initially, you have only preliminary data available. Many details are yet to be determined. At this stage you will most likely prefer to use relatively simple component models. These models should only require the input of data that is actually available and relevant, in order to avoid unnecessary overhead. They also should minimize the overhead involved in guaranteeing fast calculations and rapid convergence. The following description will refer to this case as the design model of the process.

When you start investigating a thermodynamic process, for example, you will not know the dimensions of the pipes. They will depend on many other pieces of equipment that are not yet defined in detail. As a consequence, you will not be able to calculate the pressure drops. However, from experience, you might be able to provide a good estimate for the pressure loss. In this case, the most convenient pipe model is one that allows you to enter the pressure drop explicitly. The pressure drop will be calculated with the trivial equation

\[\varDelta p = [value]\]

In consequence, you might be interested in the off-design characteristics of the plant. If detailed information is missing, you can base your off-design models on the results obtained with the design model of the process.

In the example of the pipe, for calculating the pressure drop the model will now take into account the design pressure drop and the design mass flow as well as the actual mass flow. The equation for calculating the pressure drop will now look like the following:

\[\varDelta p = \varDelta p_{0} \big( \frac{m}{m_{0}} \big) ^{2}\]

Where \(\varDelta p_{0}\) and \(m_{0}\) are values obtained from the design model of the process. \(m\) might actually be the result of the calculation.

At a later point, you may need even more detailed models. In the case of the pipe example, you may need to take into account the diameter, the length, the roughness of the surface, etc.

Datasets

In principle, you can carry out the sequence of calculations cited before by creating separate projects with the respective models. However, you would have to transfer data manually between the projects, which is a time consuming and error prone procedure.

IPSE supports cases where the flowsheet for the process remains the same, no matter which model you are using. It allows you to create more than one numerical model for a single flowsheet. Such a numerical model is called dataset. The datasets can be viewed as different design layers of your project.

A dataset is a mathematical model of a process, at a certain load point or configuration. It contains the component models to be used and the data (variables, parameters, etc.), required to solve the process model.

Note that for different data sets, the process scheme is the same, whereas the individual unit models and settings can be changed. Thereby it is possible to store different operating points of the same process in one project.

Relationship among Datasets

The datasets of a project are usually closely related to each other. In the example of the pipe given before, the off-design dataset uses values from the design dataset (e.g., and are values from the design dataset). In a case study, to obtain different design models, you need to vary only one or a few process parameters by using datasets. You will most likely create a reference dataset. All other datasets will use the data from the reference dataset and only modify some of the data items.

IPSE manages the datasets in a tree structure, the dataset tree. Each dataset is a node in the dataset tree. Except for the dataset located in the root of the tree, each dataset has a parent dataset. IPSE implements a mechanism for updating datasets from their parent dataset. Updating a dataset means to copy data from the parent dataset. The parent dataset is referred to as update source, while the dataset that is updated is called the update target.

The path from the root of the dataset tree to a dataset is called the update path. Each dataset in the update path depends on its precedent. In order to achieve valid results, the source dataset must be consistent before it can be used for an update. To be more precise, it must satisfy the following requirements:

It must have been updated (unless it is the root dataset).
The process model must have been solved since the last update or the last manual modification, relative to any of the data items.

Whenever you update a dataset, IPSE checks the consistency of all datasets in the update path. If it finds an inconsistent dataset, it issues a warning.

Figure 1 shows a typical dataset tree for a project that includes design and off-design calculations. The design dataset is the root of the dataset tree. The “100 % load” dataset depends on the design dataset. These two datasets use different models, but the results of the design dataset will be used as design data in the models of the “100% load” dataset.

The “75 %, 50 % and 25 % load” datasets all depend on the “100 % load” dataset. They will use the same models, and nearly identical input data. Only those values that actually define the load case will be different (for example, the prescribed power output).

Figure 1. The hierarchical relation ship of datasets in IPSE projects.

Mapping Data Items

The dataset tree defines a basic structure for transferring data between the datasets. However, it is important to know how a single data item is copied from a model that is in the source. Figure 2 shows how the items must be copied in the example of a pipe. The values are not only copied between items of the same name. Frequently, the values must be copied between items with different names and even from items of other objects.

Figure 2. Mapping of data items

The detailed mapping of data items is defined in the model library that you are using. The items are updated according to the following rules:

If the model used in the source dataset is the default model, the items in the target dataset are updated from the items that are defined in the model library.
If the model used in the source dataset is different from the default model, the items in the target dataset are updated from items with the same name in the source dataset.
It is always possible to disable the update for a data item. This is necessary, for example, to avoid that the values describing a load point are overwritten in an update.

Using the Dataset Manager

The dataset manager allows you to perform changes in the dataset tree. You use the dataset manager to select the dataset that is displayed in a particular project window. The dataset manager is also used to create and delete datasets, as well as to change the relation between them, and to modify the text that describes them.

Figure 3. Dataset Manager

The dataset manager is located in the manager bar. You activate the dataset manager by selecting the Dataset Manager tab in the manager bar, see Figure 3.

Selecting a Dataset

Each project window displays one dataset. The active dataset is the dataset that is displayed in the active window. When you open a project, IPSE selects the dataset with the lowest number as the active dataset.

If you want that IPSE selects a particular dataset when the project is opened, make sure that this dataset has the lowest of all dataset numbers.

All operations described below are carried out on the active dataset:

When you invoke the solver, it calculates the solution for the model described by the active dataset.
When you edit the data of an object, the changes are applied to the data in the active dataset.
You select a dataset in the following way:
Activate the window for which you want to change the displayed dataset, (for example by clicking on the window).

In the dataset tree of the dataset manager select the dataset that you want to use in the active window by clicking on the respective dataset in the dataset tree.

Creating a New Dataset

To create a new dataset you do the following:

Make sure that the active window displays the dataset that you want to use as parent for the new dataset.
Select Project New Dataset. PSE opens the window shown in Figure 4.
You can now enter a description for the dataset. This description must be unique, since it is used to identify the dataset within the project. The dataset manager also allocates a unique identification number in the range from 0 to 9999. The identification number is used for easier access when data is exported. You can change the identification number, if required.
Click on OK to close the window. PSE now displays the new dataset in the active window. The new dataset is created by duplicating the source dataset. If required, you must now edit the object data.

Alternatively, you can do the following:

Right-click in the dataset manager on the dataset that you want to use as the parent for the new dataset and select the command New from the popup menu. It is important to notice, that this can be different from the dataset that is displayed in the active window.
IPSE opens the window shown in Figure 4.
Proceed as explained above.

Figure 4. Creating a new dataset

Modifying a Dataset

To change the description of a dataset or its update source:

Select the dataset that you want to modify.
Select Project Edit Dataset. IPSE opens the window shown in Figure 4.
Similarly as for a new dataset, you can now change the description and the update source.
Click on OK to close the window.

Alternatively, you can do the following:

Right-click in the dataset manager on the dataset that you want to modify and select the Edit command from the popup menu. It is important to notice, that this can be different from the dataset that is displayed in the active window.
IPSE opens the window shown in Figure 4. Proceed as explained above.

Updating a Dataset

With the update mechanism data is copied from the source dataset to the active destination dataset.

To update a dataset:

Project Update Dataset. If any of the datasets in the update path is inconsistent, IPSE issues a warning. You should remove the inconsistency by updating and recalculating the inconsistent datasets in the update path. Once the update source of the dataset is consistent, you can update the active dataset. IPSE displays a window where you choose how to update the estimates.
Make the appropriate selections and click on OK to update the dataset.

When you update a dataset, you can choose how to update the various data items. Unless you have special requirements, the default updates are typically appropriate. Once you have chosen an update pattern for a dataset, IPSE will store it for future use.

Figure 5. Updating a dataset

Disabling the update for selected items

In certain cases, it is necessary to prevent IPSE from updating specific data items. For instance, when a variable defines an off-design point, retaining the manually selected value is often desired. If updating for a particular variable is not disabled, the manually set value will be overwritten by the value from the source dataset. To handle such scenarios, IPSE provides the option to disable updates for selected data items.

To disable updating of a selected data item:

Open the data dialog for the respective object.
Remove the Update check for the item that you do not want to be updated.

The model update option

This option controls if a model (not only the model data) is changed when you update a dataset. Enable this option if you want that the same model is used in the base dataset and in the depending dataset after updating.

To enable/disable the model update option:

Open the data dialog for the respective unit.
Set/remove the check in the box right from the model selection combobox.

Checking the consistency of the update path

To check for inconsistency in the update path:

Select Project Check Update path.

Disabling Objects in Individual Datasets

All types of objects (units, connections or global objects) can be disabled in an individual dataset. IPSE treats disabled objects as if they do not exist in the respective dataset with the following effects:

Disabled objects are greyed out in those datasets where they are disabled.
If the solver is called for a dataset, disabled objects are not included in the system submitted to the solver.
If the user executes the command Projects Update Dataset disabled objects are ignored and not updated.
If a disabled object would be used as update source the update target is not updated.
If the user executes the command Project Load Default Values, disabled objects are ignored and no default values are loaded for them.

By default, all objects are enabled in all datasets.

To disable an object or a group of objects in a dataset:

Select the dataset for which you want to disable objects.
Select the object(s) that you want to disable or enable.
Select the command Objects Enable

or

Press Alt+E to enable/disable the selected objects.

Individual objects can also be disabled via the object data window, see Figure 6.

Figure 6. Disable object in the object data window

Global objects can not be selected in the graphics editor. To disable a global object open its data input window.

The mechanism of disabling objects can be used in various ways. It is particularly useful if complete sections of a plant are switched off in a dataset, as shown in the example in Figure 7 and Figure 8.

Figure 7. Combined cycle plant in regular operation

Figure 8. Combined cycle plant in bypass operation