Development of a Framework Based on PyFMI for Optimization and Fault Detection

In this article we describe a methodology and a tool chain that allows reusing simulation models for optimization and fault detection during building operation. We illustrate the integration of simulation models, set up in Modelica, in a Python environment using PyFMI. Then, considering as a demonstrator a complex energy supply system for a school campus in Germany, we show first results for the optimization of the ice storage management of that system. Finally, we discuss further applications of the described setup for offline and online fault detection, optimization and predictive control.