基于最小二乘辨识模型的PID自整定应用研究

Application research of PID auto-tuning based on least squares identification model

针对工业数据模型辨识精度不高及PID不能在线自整定问题,提出一种闭环在线自整定控制策略。控制策略利用递推最小二乘辨识方法获得模型,基于该模型采用改进的遗传算法进行PID自整定,构成闭环在线自整定系统。通过MATLAB中的App designer平台验证了所提方法的可行性和有效性。仿真结果表明:辨识得到传递函数可以很好地拟合跟踪工业数据,PID自整定参数闭环阶跃响应控制效果良好。

Aiming at the problem that the identification accuracy of industrial data model is not high and the PID can’t be self-tuning online,a closed-loop online self-tuning control strategy is proposed.The control strategy uses an recursive least squares identification method to obtain a model.Based on the model,an improved genetic algorithm is used to perform PID auto-tuning to form a closed-loop online auto-tuning system.The feasibility and effectiveness of the proposed method are verified by the App designer platform in MATLAB.The simulation results show that the identified transfer function can fit and track industrial data well,and the PID self-tuning parameter closed-loop step response control effect is good.