A Novel Performance Enhancement Process for Single-pulse Controlled Switched Reluctance Generators

This paper presents a performance enhancement process for switched reluctance generators (SRGs) to facilitate the power regulation and efficiency optimization by controlling the excitation angles. First, an adaptive gradient algorithm (AGA) is employed to adjust the turn-on angle of SRGs online to ensure the accuracy of power regulation. Then, the relationship between efficiency and turn-off angle and the freewheeling period is modeled based on Latin hypercube sampling (LHS) and Gaussian process regression (GPR). The maximum efficiency and the combination of control angles are calculated offline using the established model based on a Sine Cosine Algorithm (SCA). The simulation results verify the effectiveness of the proposed method. The proposed optimization process does not require complex formula derivation and avoids the use of nonlinear flux-linkage characteristics to control the SRGs, which is simple and easy to be implemented for the industry application.