Lumped Disturbance Estimation-Based Dynamic Surface Control for 8/6-Type Switching Reluctance Machines

This paper concentrates on two problems in controlling the 8/6-type Switching Reluctance Motors (8/6-type SRMs or SRMs for short) in the presence of unknown external disturbances and uncertain factors. The first is studying the stabilization problem for SRMs’ rotational speed, and the other is the disturbance rejection. The stabilization issue for SRMs’ velocity is known as steering the rotational speed to be stable at the desired value, which is chosen arbitrarily, while disturbance rejection’s primary mission is to eliminate external disturbances that harm SRMs’ control performance. The first problem is handled by Dynamic Surface Control (DSC), in which the Low-Pass Filter (LPF) is integrated into the control scheme to perform the derivative operation of virtual control signals. This feature helps to reduce the computational burden and avoid the undesired phenomenon called "explosion of terms". The external disturbances presenting during the SRMs’ operation are compensated via an online training Radial Basis Function (RBF) neural network. The control cooperation regime between the DSC and the proposed neural network for uncertain SRMs lifts the control performance compared to the traditional DSC. The effectiveness of contributions and control schemes is demonstrated through impressive mathematical proofs and numerical simulation platforms.