Rotor Position Estimation for Switched Reluctance Motor Using Support Vector Machine

Switched reluctance motor (SRM), which has simple construction, high reliability, high efficiency and low cost, has shown its strong competition in many fields. However, mechanical position sensors add to the cost, complexity and potential unreliability at high speed. This paper presents an approach of rotor position estimation for switched reluctance motor based on support vector machine (SVM). For the nonlinear property of SRM, this approach takes advantage of SVM with better solution for small-sample learning problem and well generalization property. Through the off-line training, a better support vector machine structure in which phase current and phase flux linkage are inputs and the corresponding position is the output, is built with to form an efficient nonlinear mapping, and then it facilitates the rotor position estimation. The simulation and experimental results show that this method can achieve correct rotor position estimation, and thus the sensorless control of SRM is realized.