Research on Fault Diagnosis and Prediction Method about Driving Motor for Seeding Metering of Plot Seeder Based on SVMs Classification and Regression. Issue 3 (April 2020)

• Record Type: Journal Article
• Authors: Hou, Yuntao; Wu, Zequan; Cai, Xiaohua; Dong, Zhongge
• Is Part Of: IOP conference series. Volume 474:Issue 3(2020)
• Publication Date: 2020-04
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/474/3/032012
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: To solve the problem of fault diagnosis and prediction about driving motor (brushless DC motor) for seeding metering of plot seeder, the method based on SVMs classification(SVC) and regression(SVR) is proposed in this paper: firstly, the three-phase current signals of brushless DC motor(BLDCM) are analyzed to extract characteristic values of various samples; then the SVC model based on clustering combine SVMs dynamic pruned binary tree is constructed, and the SVR models about all and each class of training samples are all constructed, too. The first input sample is firstly predicted according to the SVR model about all training samples, then the predicted value is identified that belongs to a certain class of faults by the SVC model, and the next operation will be carried out according to the classification result. This method is verified in the fault diagnosis and prediction experiments of BLDCM, the diagnosis accuracy is 96.37% and the Root-Mean-Square Error is 0.217. It might be seen that the method presented in the paper is of validity.