Modelling Dynamical Systems using Transition Constraints
|Termin||Fr 16. Juli 2021|
|Kurzfassung||Despite promising performance of data science approaches in various applications, in industrial research and development the results can be often unsatisfactory due to the costly experiments that lead to small datasets to work with. Theory-guided Data Science (TGDS) can solve the problem insufficient data by incorporating existing industrial domain knowledge with data science approaches.
In dynamical systems, like gas turbines, transition phases occur after a change in the input control signal. The domain knowledge about the steepness of these transitions can potentially help with the modeling of such systems using the data science approaches. There already exist TGDS approaches that use the information about the limits of the values. However it is currently not clear how to incorporate the information about the steepness of the transitions with them.
In this thesis, we develop three different TGDS approaches to include these transition constraints in recurrent neural networks (RNNs) to improve the modeling of input-output behavior of dynamical systems. We evaluate the approaches on synthetic and real time series data by varying data availability and different degrees of steepness. We conclude that the TGDS approaches are especially helpful for flat transitions and provide a guideline on how to use the available transition constraints in real world problems. Finally, we discuss the required degree of domain knowledge and intellectual implementation effort of each approach.