变时滞系统的辨识自适应控制算法的改进与仿真研究

The Modification and Research of Identification Adaptive Predictive Control for Variable Time Delay System

城市供水出水浊度过程控制是大时滞难控对象,Smith预估控制方法的研究有效地改善了时滞带来的控制困难的问题,但是当系统模型不精确时,很难获得好的控制品质。采用时滞辨识自适应方法,有好的控制效果,但当被控系统的时滞参数是时变时,在一定范围内可能存在的时滞辨识的效果不理想,所以提出采用变搜寻区域的辨识方法,可以实时地搜索出真实过程的时滞。该算法用于变时滞的系统,具有良好的控制品质和较强的自适应能力。

Because the turbidity control system of municipal water supply plant had the characteristic of large time delay,it was always difficult to be controlled.Smith predictive control was studied in order to solve the problem caused by delay.When the model of system was not accurate,it was difficult to acquire good control quality.Identification adaptive control algorithm was proposed to identify the delay time and had good control effect.But when the time delay was time-variant,the effect of delay time identification might not satisfactory within a certain range,so the algorithm with changing search range was applied.It could search real delay time in time.The simulation test results show the proposed approach is suitable to the variable time delay system and has excellent control performance and enhanced self-adaptability.