XRG-Library SMArtInt+ for smart integration of artificial intelligence in Modelica released
Our latest development - the SMArtInt+ Library - enables you to integrate your AI models into Modelica from different sources and of different types in a simple and user-friendly way. Neural networks from TensorFlow, TensorFlow Lite and ONNX are supported. Due to the standardized and open interface of the SMArtInt library, which has been available for some time, the following types of neural networks can be integrated:
- Quasistatic Feed Forward Neural Networks (FFNN)
- Dynamic Recurrent Neural Networks (RNN), stateful or non-stateful
- Neural Ordinary Differential Equations (NODE)
- Generative models, working with Encoder/Decoder
- Complex applications like BNODE
The integration is done in compiled form and can be changed at any time independently of the Modelica model. The library is supplemented by the SMArtInt+ modules, which support you in the creation, integration and use. Important features include:
- Simple generation of Modelica blocks with the correct I/O interface
- Dynamic extrapolation warning based on the training data used
- Dedimensionalization of the neural network to enable use for scalable objects
Would you like to find out more? Please feel free to contact us and send us an email. We look forward to your inquiries!
News
1
XRG-Library SMArtInt+ for smart AI-integration in Modelica released
Wir are proud to announce the release of our newest exciting product with thrilling festures for integration of artificial intelligence in Modelica. MORENews
1
AIT uses HumanComfort CFD Toolchain
XRG Simulation is excited to announce another purchase order of HumanComfort licenses in the automotive field. MORENews
0
Historischer Erfolg für XRG Staffeln beim Hamburg Triathlon
XRG gelingt ein historischer Triumph beim diesjährigen Hamburg Triathlon mit zwei Staffeln auf Podiumsplätzen. MORENews
0
New direct dial numbers for our headquarters in Hamburg
Good news: The dial for our headquarters and fax remains the same. MORENews
0
CLARA+ ADDON FOR ROBUST TWO-PHASE FLOW MODELLING AND PHASE SEPARATION
For some applications it is important to take buoyancy effects into account. MORENews
0