积分时滞系统的Fuzzy-Smith预估器的研究

Analysis of Fuzzy-Smith Predictor for Integrating Plus Dead-time Plants

针时积分时滞系统,基于PID的Smith预估器需要被控对象的精确数学模型,对于实际工业过程中难以建模和参数变化的被控对象,其鲁棒性差,控制性能也往往达不到期望的效果。模糊PID控制器因其鲁棒性强的优点,在工业界得到了广泛应用,但目前仍然缺乏充分的解析理论分析。针对这些现状,将可解析的模糊PID控制器引入Smith预估器以提高其鲁棒性,并从理论上证明了这种Fuzzy-Smith预估器的鲁棒性优于传统的基于PID控制器的Smith预估器。基于二阶积分时滞系统,结合模糊PID的滑模特性,采用李亚普洛夫稳定性理论,在理论上分析和证明了由于模糊PID控制器中非线性项的存在从而能补偿更多的不确定性,因此其鲁棒性更好。最后的仿真结果进一步证明了这点。

For integrating plus dead-time plants,the PID-based Smith predictor needs an accurate model.But in industrial processes, it is very difficult to obtain the accurate model and the parameters often change,so it results in poor robustness and control performance for the PID-based smith predictor.Although fuzzy PID controller has been widely used in industrial processes to its good robustness,the sufficient analytical theories are still lack.Thus,an analytical fuzzy PID controller is introduced to the Smith predictor in order to improve the robustness,and it can be proofed that the robustness of the Fuzzy-Smith predictor is better than the PID-based Smith predictor. After obtain the second order integrating plus dead-time model of the plant,based on Lyapunov theory and sliding mode control,it can be proofed that the nonlinear term can compensate more uncertainty and the robustness of the Fuzzy-Smith predictor is better.The simulation results further demonstrate it.