针对多电平逆变器SHEPWM(selected harmonic elimination pulse width modulation,SHEPWM)消谐模型的求解问题,提出一种遗传算法和模拟退火算法相融合的新型算法。该算法根据个体适应度值进行自适应交叉和变异操作,采用模拟退火算法进行个体更新,以增加种群的多样性,增强全局寻优能力,避免陷入局部最优,并以实数编码保存个体来提高计算精度。以二极管箝位三电平逆变器SHEPWM为例,给出了全调制度下的开关角度轨迹及较高调制度下的另外两组解,绘制了谐波失真含量(total harmonic distortion,THD)随调制度变化的曲线,并给出了详细的仿真结果。最后通过建立的二极管箝位三电平逆变器实验平台进行了实验验证,仿真和实验结果证明了该算法的正确性和可行性。
In view of solving the SHEPWM harmonic elimination model, a hybrid algorithm was proposed, which was composed of genetic algorithm and simulated annealing algorithm. According to the value of individual fitness, the adaptive crossover and mutation were performed, and the simulated annealing algo- rithm was adopted to improve the individual update, thus the diversity of group was increased. The ability of global optimization was enhanced and the local optimum was avoided, using real coding to save the individual to improve calculation accuracy. Taking the SHEPWM in neutral pointed clamped three level inverter as an example,the trajectory of switching angles under the whole modulations and another two sets of feasible solution under high modulations were given, and a curve for the THD varies modulation was drawn. Simulation and experiment results certificate the angles worked out by the proposed algorithm is correct and valid.