多目标超参数优化人工神经网络，实现同步电机的最佳前馈转矩控制

IF 5.2 Q1 ENGINEERING, ELECTRICAL & ELECTRONIC

IEEE Open Journal of the Industrial Electronics Society Pub Date : 2024-01-22 DOI:10.1109/OJIES.2024.3356721

Niklas Monzen;Florian Stroebl;Herbert Palm;Christoph M. Hackl

{"title":"多目标超参数优化人工神经网络，实现同步电机的最佳前馈转矩控制","authors":"Niklas Monzen;Florian Stroebl;Herbert Palm;Christoph M. Hackl","doi":"10.1109/OJIES.2024.3356721","DOIUrl":null,"url":null,"abstract":"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.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"41-53"},"PeriodicalIF":5.2000,"publicationDate":"2024-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10411011","citationCount":"0","resultStr":"{\"title\":\"Multiobjective Hyperparameter Optimization of Artificial Neural Networks for Optimal Feedforward Torque Control of Synchronous Machines\",\"authors\":\"Niklas Monzen;Florian Stroebl;Herbert Palm;Christoph M. Hackl\",\"doi\":\"10.1109/OJIES.2024.3356721\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"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.\",\"PeriodicalId\":52675,\"journal\":{\"name\":\"IEEE Open Journal of the Industrial Electronics Society\",\"volume\":\"5 \",\"pages\":\"41-53\"},\"PeriodicalIF\":5.2000,\"publicationDate\":\"2024-01-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10411011\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Open Journal of the Industrial Electronics Society\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10411011/\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Open Journal of the Industrial Electronics Society","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10411011/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}

引用次数: 0

摘要

多目标超参数优化被用于寻找用于同步电机最佳前馈转矩控制（OFTC）的最佳人工神经网络（ANN）架构。所提出的框架允许系统地确定与多个（部分）相互矛盾的目标有关的帕累托最优人工神经网络，例如所考虑的人工神经网络的近似精度和计算负担。获得的帕累托最优方差网络在实时系统上进行了训练和实施，并针对非帕累托最优方差网络设计和最先进的 OFTC 方法对非线性磁阻同步电机进行了实验测试。最后，本文基于仿真网络逼近理论的最新成果，为基于帕累托最优仿真网络的 OFTC 设计和实施提供了指导。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

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 (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.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

IEEE Open Journal of the Industrial Electronics Society ENGINEERING, ELECTRICAL & ELECTRONIC-

CiteScore

10.80

自引率

2.40%

发文量

审稿时长

12 weeks

期刊介绍： The IEEE Open Journal of the Industrial Electronics Society is dedicated to advancing information-intensive, knowledge-based automation, and digitalization, aiming to enhance various industrial and infrastructural ecosystems including energy, mobility, health, and home/building infrastructure. Encompassing a range of techniques leveraging data and information acquisition, analysis, manipulation, and distribution, the journal strives to achieve greater flexibility, efficiency, effectiveness, reliability, and security within digitalized and networked environments. Our scope provides a platform for discourse and dissemination of the latest developments in numerous research and innovation areas. These include electrical components and systems, smart grids, industrial cyber-physical systems, motion control, robotics and mechatronics, sensors and actuators, factory and building communication and automation, industrial digitalization, flexible and reconfigurable manufacturing, assistant systems, industrial applications of artificial intelligence and data science, as well as the implementation of machine learning, artificial neural networks, and fuzzy logic. Additionally, we explore human factors in digitalized and networked ecosystems. Join us in exploring and shaping the future of industrial electronics and digitalization.