Jiao Shang-Bin, Liu Jing-wei, Zhang You-min, Zhang Xiao-liang, Xie Guo
{"title":"Nonlinear Modeling of Double-barrel Eddy Current Coupler","authors":"Jiao Shang-Bin, Liu Jing-wei, Zhang You-min, Zhang Xiao-liang, Xie Guo","doi":"10.1109/ICIEA.2019.8833842","DOIUrl":null,"url":null,"abstract":"Firstly, the parameter analysis of the double-cylinder permanent-magnet eddy current coupler is carried out by using Maxwell finite element analysis software. Considering the theoretical and practical situation, the relationship between the main structural parameters and the transfer torque is determined. Then, in order to reduce the calculation cost of the subsequent optimization, BP neural network and support vector machine are used to establish the nonlinear model between the main structural parameters and the transfer torque. Comparing the two methods, it is found that the SVM method is superior to the BP method in predicting performance and running time. Finally, a test system is built to verify the accuracy of the SVM nonlinear model.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"194 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICIEA.2019.8833842","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Firstly, the parameter analysis of the double-cylinder permanent-magnet eddy current coupler is carried out by using Maxwell finite element analysis software. Considering the theoretical and practical situation, the relationship between the main structural parameters and the transfer torque is determined. Then, in order to reduce the calculation cost of the subsequent optimization, BP neural network and support vector machine are used to establish the nonlinear model between the main structural parameters and the transfer torque. Comparing the two methods, it is found that the SVM method is superior to the BP method in predicting performance and running time. Finally, a test system is built to verify the accuracy of the SVM nonlinear model.