TechRxiv

Multi-objective hyperparameter optimization of artificial neural networks for optimal feedforward torque control of synchronous machines

Download (7.24 MB)
preprint
posted on 2023-11-27, 13:46 authored by Niklas MonzenNiklas Monzen, Florian StroeblFlorian Stroebl, Herbert Palm, Christoph HacklChristoph Hackl

Multi-objective hyperparameter optimization (MO-HPO) is applied to find optimal artificial neural network (ANN) architectures used for optimal feedforward torque control (OFTC) of synchronous machines. The proposed framework allows to systematically identify Pareto optimal ANNs with respect to multiple (partly) contradictory objectives such as approximation accuracy and computational burden of the considered ANNs. The obtained Pareto optimal ANNs are trained and implemented on a realtime system and tested experimentally for a nonlinear reluctance synchronous machine (RSM) against non-Pareto optimal ANN designs and a state-of-the-art OFTC approach. Finally, based on the most recent results from ANN approximation theory, guidelines for Pareto optimal ANN-based OFTC design and implementation are provided. 

Funding

German Research Foundation (DFG) Open Access Publication Costs (NE 1911/2-1)

History

Email Address of Submitting Author

niklas.monzen@hm.edu

ORCID of Submitting Author

0000-0003-4818-7815

Submitting Author's Institution

University of Applied Sciences Munich

Submitting Author's Country

  • Germany