基于人工免疫多目标寻优算法的PID自整定

PID SELF-TUNING BASED ON ARTIFICIAL IMMUNE MULTI-OBJECTIVE OPTIMIZATION ALGORITHM

提出了一种基于人工免疫多目标寻优算法(AIMOA)的PID参数自适应整定的设计方法。利用生物免疫系统的免疫机理设计系统响应的目标函数,再通过AIMOA算法搜索PID控制器的优化参数组,最后将基于AIMOA算法同基于遗传算法(GA)和齐格勒—尼柯尔斯(Zi-Ni)方法的PID自整定进行了仿真比较。结果表明:AIMOA算法具有快速收敛性,能够较快地搜索到PID参数自适应整定的最优或者次最优解,体现了算法的优越性、实用性和有效性。

A novel artificial immune multi-objective optimization algorithm （AIMOA） is presented, which is applied successfully to selftuning of PID controller. Firstly, natural immunological mechanisms are used to design the optimum objective function of control system. Then, the global optimum parameters combination of PID self-tuning is located by AIMOA algorithm. Finally, the advantages of AIMOA algorithm for PID self-tuning are further highlighted through the comparison between the classical Ziegler-Nichols and GA method. AIMOA algorithm shows fast convergence, and it can be utilized to quickly locate the optimal or sub-optimal parameters of PID controller. Experiment results demonstrate that the proposed algorithm is valid, superior and effective.