基于T细胞调节的免疫遗传优化算法对逆变器PWM控制序列的优化

针对遗传算法(GAs)收敛速度慢、易于陷入局部最优等不足,基于单相全桥逆变器输出电流与参考电流的误差模型,提出一种改进的免疫遗传优化算法(IGOAs)用于逆变器PWM最优控制序列优化。算法采用0、1编码,自适应突变概率及T细胞调节算子增强算法的快速收敛性和种群的多样性。数值实验中考虑逆变器负载端电阻为定值和受随机扰动两种情形,将GAs和IGOAs用于此两种情形的PWM控制序列优化,仿真结果统计表明：IGOAs具有较好的收敛性和稳定性,负载电阻受随机扰动时,IGOAs较GAs能快速跟踪参考电流,获得较小的THD电流波。

Immune genetic optimization algorithm based on T cell-mediated for sequence optimization of PWM inverter

For overcoming the disadvantage of slow convergence and easy of falling into local optimum of genetic algorithms (GAs), an improved immune genetic optimization algorithm (IGOAs), based on the error model of the output current and reference current of single-phase full-bridge inverter, is proposed for finding optimal PWM control sequence of inverter. In the algorithm, antibodies are coded by using 0 and 1. An adaptive mutation probability and T cell-mediated operators are designed for increasing the diversity of population and improving convergence of IGOAs. Fixed resistance and resistance with random perturbations of the load side of inverter are considered in numerical experiments. IGOAs and GAs are compared to optimizing PWM control sequences with the above two model. Simulation results show that IGOAs has good convergence when resistance is static. In addition, IGOAs quickly track standard current when load resistance is exposed to random value. The THD value obtained by IGOAs is smaller than that of GAs.