An ELM identifier and inverse controller based algorithm for dynamic decoupling control of bearingless switched reluctance motor

This paper proposes a new dynamic decoupling control algorithm for a single winding bearingless switched reluctance motor (BSRM), using extreme learning machine (ELM) identifier and inverse controller. Firstly, the working principle and mathematic model of BSRM are described and finite element analysis (FEA) is employed to get characteristic profiles. Then, the invisibility of mathematic model is analyzed and an ELM identifier is applied to identify the inverse model. To illustrate the superiority of ELM inverse model, comparisons are conducted with neural network (NN) and support vector machine (SVM). Finally, simulation results validate that the proposed algorithm can actualize dynamic decoupling control.