Position sensorless control of switched reluctance motor based on a linear inductance model with variable coefficients

To solve the problem of inaccurate position estimation due to the influence of magnetic saturation in sensorless control of switched reluctance motor (SRM), a new position estimation method is proposed. By analysing the linearity of the inductance curve, the region with better linearity can be found. Then, in this region, the mathematical model between rotor position and inductance can be built to estimate the rotor position by the linear fitting. However, the inductance curve will shift due to the influence of magnetic saturation in the case of heavy load, which makes the estimation error larger. Therefore, this study carries out a linear fitting of this region under different loads and establishes a linear inductance model with variable coefficient. This method can effectively eliminate the influence of saturation inductance on position estimation and have better estimation accuracy. Simulation and experimental results show the effectiveness and correctness of the proposed method.