We describe initial attempts to model the dynamic thermal behavior of electrical machines by evaluating the ability of linear and non-linear (regression) modeling techniques to replicate the performance of simulations carried out using a lumped parameter thermal network (LPTN) and two different test scenarios. Our focus falls on creating highly accurate simple models that are well-suited for the real-time computational demands of an envisioned symbiotic interaction paradigm. Preliminary results are quite encouraging and highlight the very positive impact of integrating synthetic features based on exponential moving averages.
ZAVOIANU, A.-C., KITZBERGER, M., BRAMERDORFER, G. and SAMINGER-PLATZ, S. 2020. On modeling the dynamic thermal behavior of electrical machines using genetic programming and artificial neural networks. In Moreno-Díaz, R., Pichler, F. and Quesada-Arencibia, A. (eds.) Computer aided systems theory: EUROCAST 2019: revised selected papers from the proceedings of the 17th International conference on computer aided systems theory (EUROCAST 2019), 17-22 February 2019, Las Palmas de Gran Canaria, Spain. Lecture notes in computer science, 12013. Cham: Springer [online], part 1, pages 319-326. Available from: https://doi.org/10.1007/978-3-030-45093-9_39