改进遗传算法PID参数整定及在漂白温度控制中的应用

遗传算法（GA）是一种基于群智能的全局随机优化算法。针对简单遗传算法（SGA）收敛速度慢、易于早熟等缺点，采用改进的自适应交叉算子和自适应变异算子。结合兼顾性能指标和响应过程平衡的适配函数，以多种改进方式相结合的遗传算法对PID参数进行寻优整定。并将该控制器应用于纸浆漂白温度控制中，仿真结果表明：改进遗传算法能够明显改善收敛速度和寻优效果，当被控对象存在较大纯滞后、时间常数特性较大时，采用本方法优化PID控制器参数可获得比较满意的控制效果。

Self-tuning of PID Parameter Based on Modified Genetic Algorithm and its Application to Bleaching Temperature Control

Genetic algorithm（GA） is a global random optimal method based on swarm intelligence. To overcome the shortcoming of the slow convergence speed and easy premature of single GA, a modified genetic algorithm（MGA） based on .self--adjusting cross operator and mution operator and real coding has been proposed. This method has been used to adjust the PID parameter. Considering the stability index and dynamic response, a new fitness function has been brought out. Lastly, the PID controller based on MGA tuning has been applied to temperature control in paper bleaching, the sumulation resuits show that the MGA can impove the convergence speed and optimal effect, the large delay and big inertia system adopting the PID controller based on the MGA can gain a satisfied control effect.