Emotional Learning-Based Firing Angle Optimization for Switched Reluctance Generator in Wind Energy Systems

Abstract

Switched Reluctance Generators are promising for wind energy conversion systems due to their robustness and simple construction. However, their nonlinear magnetic characteristics make the control design more challenging. This paper proposes an emotional learning-based controller to optimize the firing angles of the generators under variable wind conditions. Inspired by the limbic system and amygdala model, the controller employs a fuzzy-Bayesian inference mechanism to adaptively regulate the turn-on and turn-off angles. To achieve maximum power extraction and reduce losses, these angles are optimized based on the proposed emotional learning approach. The method is implemented and validated through MATLAB/Simulink simulations, demonstrating improved efficiency and dynamic performance for renewable energy applications.