Artificial Neural Network

To add the NeuralNet to the Class Library, click Manage Class Library > Libraries > Tools > Neural Networks on the Home ribbon tab.

You can use artificial neural network models to deduce a function from observations. This is particularly useful in applications, where the complexity of the data or task makes the design of such a function by hand impractical. Most neural networks are trainable systems that are able to learn to solve complex problems from a set of exemplars.

In Plant Simulation we use these functions of the artificial neural network:

• Function approximation, or regression analysis, including time series prediction and modeling.

• Classification, novelty detection and sequential decision making.

• Data processing, including filtering.

  Note:

  We modeled the NeuralNet as an application object in a Frame. For this reason opening help with F1 and What’s This Help in the dialogs of the objects do not work. To open help for the object, select Help > Help on Neuralnet in its dialog.

We modeled the dialogs of the objects with the object Dialog. The windows of the objects do not use the standard Siemens PLM theme, but your Windows theme.

To assign the object another name or a label, click it with the right mouse button and select Rename .

Compare the sample models: Click the Window ribbon tab, click Start Page > Getting Started > Example Models and click Small Examples. Then, select the respective Category, the Topic, and the Example in the dialog Examples Collection and click Open Model.

These textbooks provide an easy-to-read introduction to neural networks:

E. Rich, K. Knight: Artificial Intelligence McGraw-Hill, New York et.al. 1991

H. Ritter, Th. Martinetz, K. Schulten: Neuronale Netze Addison, 1992

Related Topics