加入滞后时间削弱器的大滞后系统的神经网络PID控制  

NN-PID Control for Large Time Delay System with Decreasing Time Delay Controller

作　　者：李钟慎[1] 

机构地区：[1]华侨大学机电及自动化学院测控工程系,泉州362021

出　　处：《电子测量与仪器学报》2008年第3期51-54,共4页Journal of Electronic Measurement and Instrumentation

基　　金：福建省自然科学基金(编号:E0710018);华侨大学高层次人才科研启动费资助项目(编号:07BS510)

摘　　要：Smith预估控制是大滞后系统主要的控制方法之一,通过Smith预估器将大滞后的对象转化为无滞后的对象,从理论上解决了大滞后系统的控制问题,但在实际应用中还存在缺陷。本文首先提出滞后时间削弱器,将大滞后的对象演变成小滞后的对象,然后给大滞后系统加入滞后时间削弱器,采用PID做为控制器,用BP神经网络对PID控制器的参数进行整定。仿真结果表明,加入滞后时间削弱器的大滞后系统取得更好的控制效果,它的鲁棒性高,自适应能力强,具有抗外来扰动的能力。Smith predictive control is one of the main control methods for large time delay system. Using Smith predictive controller, large delay time plant is translated into the plant without delay time, which resolves the control problem of large time delay systems theoretically, but there are still some flaws in practical applications. In this paper, the decreasing time delay controller is firstly presented, which remodels large delay time plant into small delay time plant. And then decreasing time delay controller is added to large time delay system, the small time delay plant is controlled by PID, the parameters of PID controller are tuned by BP neural networks. Simulation results show that the control effect of the large time delay system using decreasing time delay controller is improved, the system has high robustness, auto-adaptive ability and anti-jamming ability.

关 键 词：大滞后系统 滞后时间削弱器 神经网络PID控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]