不稳定和积分时滞过程的Smith预估器设计

Smith Predictor Design for Unstable and Integrating Processes with Dead Time

针对不稳定和积分时滞过程,提出了一种设计修正的Smith预估器的新方法．首先利用内环控制器镇定不稳定或积分时滞过程,然后通过等效变换将修正的Smith预估器结构简化为内模控制(IMC)结构,并利用内模控制原理和积分平方误差(ISE)指标设计给定点跟踪控制器．最后,根据负载扰动下保持闭环系统稳定的要求,利用最佳相位裕度标准或Nyquist稳定性判据设计负载扰动控制器．多个仿真示例验证了该设计方法比现有的方法具有更好的动态特性与鲁棒性．

A design method for modified Smith predictor is presented to control unstable and integrating processes with dead time. First, an inner-loop controller is used to stabilize the unstable or integrating processes with dead time. Then the modified Smith predictor is simplified to internal model control （IMC） structure by equivalent transformation, and the controller for setpoint tracking is derived by IMC principle and the integral square error （ISE） criterion. Finally, to keep the closed-loop system stable under load disturbance, the optimum phase margin criterion or the Nyquist stability criterion is adopted to design the load-disturbance controller. The proposed method is verified to be better than the existing ones in dynamic characteristics and robustness by simulation of several examples.