A Novel Particle-Swarm-Optimization-Based Large Signal Analysis Method for Power Electronic Systems

Large signal stability evaluation is important for designing and optimizing power electronic systems. Less conservative large signal analysis method helps decrease weight and costs of power electronic systems especially in transportation and aerospace fields. With the development of power electronic systems, the traditional Takagi-Sugeno (TS) multimodeling method are hard to be used in large power electronic systems as accurate ranges of the nonlinear variables are required. Thus., in this paper., a large signal stability analysis method based on particle swarm optimization (PSO) algorithm is proposed to fill this gap. Firstly, generate a particle swarm, and initialize randomly position of each particle. Position of particles should satisfy restrains of the linear matrix inequality (LMI). Secondly, calculate attraction domain index (ADI) of each particle. Thirdly, update individual and global optimal solutions of the particle swarm by searching the maximum ADI. Make sure the updated position of each particle satisfies the LMI restrains. Finally, find out the best matrix and calculate the candidate energy function of the studied system. Simulations and experiments verify the effectiveness of the proposed method. Compared with traditional TS multimodeling method, the proposed method is less conservative. The accurate range of the nonlinear variables is not required, which is promising to be used in transient stability analysis of higher-order power electronic systems.