Online Magnetization Surface Identification for a Switched Reluctance Motor

The precise control of switched reluctance motor requires information about the magnetization surface as a function of rotor angular position and current. The surface can be obtained performing offline tests; however, it demands specialized mechanical equipment and time for commissioning. Moreover, the magnetization surface may be changed with the varying of the rotor and stator temperatures, which affect their dimensions and, therefore, the airgap variation. That is why the online estimation of the magnetization surface is desirable. This paper proposes the identification method operating in parallel with the model predictive current control system utilizing the magnetization map. Any error between the referenced and obtained currents is used to correct the reference points of the two-dimensional array representing the magnetization surface. The proposed method was examined using a simulation model and operated online in parallel with the primary control system driving the electrical machine.