Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines
Multiobjective hyperparameter optimization 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 (partl...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of the Industrial Electronics Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10411011/ |
| _version_ | 1797321059438952448 |
|---|---|
| author | Niklas Monzen Florian Stroebl Herbert Palm Christoph M. Hackl |
| author_facet | Niklas Monzen Florian Stroebl Herbert Palm Christoph M. Hackl |
| author_sort | Niklas Monzen |
| collection | DOAJ |
| description | Multiobjective hyperparameter optimization 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 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. |
| first_indexed | 2024-03-08T04:52:58Z |
| format | Article |
| id | doaj.art-d7a1046175e44ea795ffe262337b3597 |
| institution | Directory Open Access Journal |
| issn | 2644-1284 |
| language | English |
| last_indexed | 2024-03-08T04:52:58Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of the Industrial Electronics Society |
| spelling | doaj.art-d7a1046175e44ea795ffe262337b35972024-02-08T00:02:23ZengIEEEIEEE Open Journal of the Industrial Electronics Society2644-12842024-01-015415310.1109/OJIES.2024.335672110411011Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous MachinesNiklas Monzen0https://orcid.org/0000-0003-4818-7815Florian Stroebl1https://orcid.org/0000-0002-9468-1956Herbert Palm2https://orcid.org/0000-0003-4721-1414Christoph M. Hackl3https://orcid.org/0000-0001-5829-6818Institute for Sustainable Energy Systems (ISES), Hochschule München University of Applied Sciences, Munich, GermanyInstitute for Sustainable Energy Systems (ISES), Hochschule München University of Applied Sciences, Munich, GermanyInstitute for Sustainable Energy Systems (ISES), Hochschule München University of Applied Sciences, Munich, GermanyInstitute for Sustainable Energy Systems (ISES), Hochschule München University of Applied Sciences, Munich, GermanyMultiobjective hyperparameter optimization 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 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.https://ieeexplore.ieee.org/document/10411011/Artificial neural network (ANN)hyperspace exploration (HSE)multiobjective hyperparameter optimization (MO-HPO)optimal feedforward torque control (OFTC)reluctance synchronous machine (RSM) |
| spellingShingle | Niklas Monzen Florian Stroebl Herbert Palm Christoph M. Hackl Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines IEEE Open Journal of the Industrial Electronics Society Artificial neural network (ANN) hyperspace exploration (HSE) multiobjective hyperparameter optimization (MO-HPO) optimal feedforward torque control (OFTC) reluctance synchronous machine (RSM) |
| title | Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines |
| title_full | Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines |
| title_fullStr | Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines |
| title_full_unstemmed | Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines |
| title_short | Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines |
| title_sort | multiobjective hyperparameter optimization of artificial neural networks for optimal feedforward torque control of synchronous machines |
| topic | Artificial neural network (ANN) hyperspace exploration (HSE) multiobjective hyperparameter optimization (MO-HPO) optimal feedforward torque control (OFTC) reluctance synchronous machine (RSM) |
| url | https://ieeexplore.ieee.org/document/10411011/ |
| work_keys_str_mv | AT niklasmonzen multiobjectivehyperparameteroptimizationofartificialneuralnetworksforoptimalfeedforwardtorquecontrolofsynchronousmachines AT florianstroebl multiobjectivehyperparameteroptimizationofartificialneuralnetworksforoptimalfeedforwardtorquecontrolofsynchronousmachines AT herbertpalm multiobjectivehyperparameteroptimizationofartificialneuralnetworksforoptimalfeedforwardtorquecontrolofsynchronousmachines AT christophmhackl multiobjectivehyperparameteroptimizationofartificialneuralnetworksforoptimalfeedforwardtorquecontrolofsynchronousmachines |