Title :
Modeling inductance for bearingless switched reluctance motor based on PSO-LSSVM
Author :
Xiang, Qianwen ; Sun, Yukun ; Ji, Xiaofu
Author_Institution :
Coll. of Electron. & Inf. Eng., Jiangsu Univ., Zhenjiang, China
Abstract :
The least square support vector machine (LS-SVM) inductance model optimized by the particle swarm optimization (PSO) algorithm is presented for bearingless switched reluctance motor (BSRM). The training sample is first obtained using the 3D finite element model (FEM) of the prototype, and then LS-SVM model is built, whose hyper-parameters are optimized using PSO algorithm. The absolute error and relative error are computed, which demonstrate the high accuracy of the proposed model.
Keywords :
finite element analysis; inductance; least squares approximations; particle swarm optimisation; reluctance motors; support vector machines; 3D finite element model; LSSVM; PSO; PSO algorithm; absolute error; bearingless switched reluctance motor; hyper parameters; inductance model; least square support vector machine; particle swarm optimization; relative error; Computational modeling; Data models; Inductance; Magnetic levitation; Reluctance motors; Support vector machines; Switches; Bearingless switched reluctance motor; Inductance characteristic; Least square support vector machine; PSO-LSSVM; Particle swarm optimization;
Conference_Titel :
Control and Decision Conference (CCDC), 2011 Chinese
Conference_Location :
Mianyang
Print_ISBN :
978-1-4244-8737-0
DOI :
10.1109/CCDC.2011.5968291
