基于自适应遗传算法的非线性PID控制器

A Nonlinear PID Controller Based on an Adaptive Genetic Algorithm

生物发酵利用夹套进行加热和冷却,其温度控制精度非常重要.但这种以温度为被控对象往往存在严重的积分纯滞后现象,采用常规的控制算法难以达到所要求的控制品质.基于PID控制器各增益参数与偏差信号之间的非线性关系,采用出一种非线性PID控制器.同时,针对经典遗传算法收敛速度慢,易陷入早熟,局部最优等缺点,提出一种自适应遗传算法,自适应遗传算法通过建立适应值差值函数来适时调整交叉、变异概率.这使算法具有自适应性,将其应用于非线性PID控制器参数寻优.仿真结果表明,基于此遗传算法寻优的非线性PID控制器可以有效的抑制大滞后系统的超调缺陷,具有较好的控制精度.

The precision of temperature control is a very important factor to fermentation process. However, because the serious integral and time delay exists in controlled object, the conventional control algorithm could hardly obtain satisfactory quality. Thus, a nonlinear PID controller is designed based on the nonlinear relationship between PID control parameters and error signals. Furthermore, an adaptive genetic algorithm is proposed in this paper in order to overcome the disadvantages in classical genetic algorithm, such as slow early convergence, local optimum, etc. The new adaptive algorithm adjusts the probabilities of crossover and mutation by construct function of the changes of fitness value. Simulation results show that the nonlinear PID controller with the genetic algorithm for tuning the parameter is very effective. It has good dynamic, static response and strong robustness.