INTRODUCTION TO SMARTINT+ LIBRARY

 

There is an increasing need for hybrid models consisting of physical and surrogate as well as data-driven models. XRG has developed an interface, the SMArtInt+ Library, that enables the easy integration of neural networks from different sources and types. SMArtInt stands for Smart Integration of Artificial Intelligence in Modelica. The SMArtInt+ library enables you to integrate machine learning models from different sources in a standardized, user-friendly and efficient way via an external source that can be changed independently of the Modelica code. Currently, SMArtInt+ supports TensorFlow and ONNX models, which can be exported to Dymola and OpenModelica or imported from Modelica blocks.

Course instructor

The course is held in English by experienced by experienced SMArtInt+ users or the library developers.

Prerequisites for attending

For a successful participation the participant should have a fundamental knowledge of using the simulation software Dymola as well as common functional principles of the description language Modelica (see also our standard Modelica training offer) and TensorFlow. Furthermore, a personal computer with installed Dymola software, TensorFlow and a valid license of the SMArtInt+ Library is needed.

Training course objectives

The objective is to introduce the participants to the Modelica SMArtInt+ Library and the supplementary Python modules.

Content

  • Purpose of the library
  • Installation and package structure, Python module content
  • Explanation of different types of neural networks supported 
  • Generation of neural networks from measurements or simulation data
  • Integration into Modelica and FMUs
  • Use of features like scaling, extrapolation detection, I/O filtering and dedimensionalisation for a more general use 

Course duration and place of event

The course will take 1 day (9 hours each including lunch break). It is either provided online or in presence. Special offers are available for onsite trainings.

Bookings

For your requests give us a call (+49 40-76629 2630) or send an email (training@xrg-simulation.de). You will receive a quotation via email and make your booking. The booking will be confirmed by email, again. Until 14 days before the course you can cancel without any charges. After that period we bill a cancellation fee about 50 pecent of the costs.