自适应混沌粒子群算法及在PID整定中的应用

Adaptive Chaotic Particle Swarm Optimization and Its Application in PID Parameter Setting

针对微藻自动培养装置的温度控制,常规PID控制器的参数整定需耗费大量人力进行调节,提出一种采用自适应混沌粒子群的PID参数整定算法,充分利用多涡卷广义Jerk混沌序列的随机性及遍历性,对粒子群进行混沌初始化;采用非线性调整机制对惯性权重进行自适应调整;引入基于适应度方差的局部收敛判别机制,以混沌扰动的方式帮助种群跳出局部最优。仿真结果表明,相较于标准粒子群算法,改进算法能始终保持粒子群的多样性,系统响应超调量小,调节时间短,具有更好的全局搜索能力,控制精度较高,适应性和鲁棒性好。

In the parameter setting of the PID temperature controller for the microalgae culture device, lots of practices are needed. A novel adaptive chaotic particle swarm optimization （ACPSO） algorithm was proposed. At first, the algorithm made use of the advantage of randomness and ergodicity of multi - scroll Jerk chaotic system to increase the diversity of the original swarm. Then an inertia weight was adjusted adaptively with nonlinear strategy. At last, the swarm fitness variance was used to distinguish whether the chaotic disturbance was needed to help the swarm jump out the local optimal solutions. Finally, the proposed algorithm was applied for the setting of PID parameters. The simulation results show that the response of the proposed algorithm has a small overshoot and short adjusting time, having better performance than the standard particle swarm optimization algorithm.