自适应遗传算法在数控伺服系统控制参数优化中的应用

PID参数整定是一个多参数组合优化的问题，针对目前常用的工程整定法无法在全局范围内对PID参数进行组合优化，只能从系统的单项性能指标出发进行整定，而标准的遗传算法又容易出现过早收敛等问题，为此，提出了基于改进的自适应遗传算法的PID参数整定方法。这种方法能够随适应度值自动改变交叉概率和变异概率，这种方法既能够确保算法的收敛，也能够很好的保证种群的多样性。将该方法应用于数控伺服系统，控制效果良好，最后将Ziegler-Nichols算法与自适应遗传算法整定的PID控制系统的动态响应性能作了对比分析，仿真试验结果证明了基于自适应遗传算法的PID参数整定方法的优越性。

Application of Control Parameters Optimization of CNC Servo System based on Self-Adaptive Genetic Algorithm

The parameters adjustment of PID controller is a problem of optimization for multiple parameters. In order to solve the shortcomings of current engineering methods for parameters adjustment that can only deal with it according to the single requirement of system and can＇t optimize it in the whole range, as well as the standard genetic algorithm is prone to premature convergence and other problems, therefore, an improved adaptive genetic algorithm based on the PID parameter adjustment method was presented in this paper. This approach enables crossover probability and mutation probability automatically changes along with the fitness value, not only can it maintains the population diversity, but also can ensure the convergence of the algorithm. Using this method in CNC servo system and achieved good control effect, and compare the dynamic response of the PID control system based on Ziegler-Nichols algorithm and genetic algorithm. Simulation results show that the PID parameters adjustment method based on self-adaptive genetic algorithm has much superiority.