To address the issues that the Moth-Flame Optimization (MFO) algorithm tends to fall into local optima and suffers from insufficient convergence accuracy, a multi-strategy improved Moth-Flame Optimization algorithm (MMFO) is proposed. First, the Sine-Tent-Cosine chaotic map is employed to initialize the population, enhancing population diversity. Second, the oscillation mechanism of the Sine Cosine Algorithm (SCA) is integrated to balance global exploration and local exploitation. Finally, the somersault foraging strategy is introduced to strengthen the algorithm's ability to escape local optima. The proposed MMFO is applied to optimize the parameters of two types of Power System Stabilizers (PSS), namely Lead-Lag and PID controllers. Simulations are conducted on a Single-Machine Infinite-Bus (SMIB) system under four error criteria: IAE, ISE, ITAE, and ITSE. Comparative analyses with MFO, Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), Flower Pollination Algorithm (FPA), and Jaya algorithm demonstrate that the PSS optimized by MMFO achieves the best comprehensive performance in suppressing low-frequency oscillations, reducing overshoot, and shortening settling time, thereby verifying its effectiveness and superiority.
Ye Lyuyang (Thu,) studied this question.