多变量时滞非方系统的分数阶Smith预估控制

针对多变量时滞非方系统,提出一种基于反向解耦的分数阶Smith预估控制方法.首先,将反向解耦方法推广应用于$m\times n$非方系统中,给出非方解耦矩阵的设计方法,同时为了保证解耦矩阵的稳定正则,给出其实现的必要条件以及条件不满足时的补偿方法;然后,针对解耦后的各个单回路系统设计分数阶Smith预估控制器,根据内模控制与Smith预估控制结构上的等价关系简化控制器的设计,克服时滞环节对系统性能的影响,并且基于最大灵敏度推导出一种控制器参数解析整定方法;最后,通过典型的Shell标准控制问题对所提出方法进行验证.仿真结果表明,反向解耦方法设计简单易于实现,能达到系统完全解耦,控制器参数较少,整定方便,并且具有良好的跟踪能力、抗干扰性和鲁棒性.

Fractional order Smith predictor control for non-square systems with time-delay

A fractional order Smith predictive control approach based on the inverted decoupling is proposed for non-square systems with time-delay. Firstly, inverted decoupling is extended into $m\times n$ non-square systems. The design method of the inverted decoupling matrix is proposed. At the same time, to ensure the decoupling matrix stable and regular, the realizability conditions and the compensation method of the controlled object are provided. Then, we design a fractional order Smith predictive controller for decoupled signal-loop systems. The design method of the fractional order controller is simplitied using the equivalence relation between the IMC(internal model control) and the Smith predictive control. Furthermore, we propose a tuning methodology for controller parameters based on the maximum sensitivity. Finally, the typical Shell standard control problem is studied to verify the effectiveness of the proposed method. The simulation results show that the proposed method is not only simple in design and easy to implement, but also convenient in parameter tuning, and has a better tracking performance, disturbance rejection property and robustness.