【作者】 孟祥华；

【导师】 孙灵芳；

【作者基本信息】 东北电力大学 ， 控制理论与控制工程， 2011， 硕士

【摘要】 在工业控制领域中,被控对象往往存在很多难以控制的因素,如非线性、强耦合、大时滞、非最小相位特性等,导致系统超调量变大,调节时间加长,有些甚至出现振荡、发散,系统的动态品质明显下降。大多数已有的控制方法也很难应用于这样的过程,所以多变量大时滞过程的解耦控制问题一直是控制领域中的一大难题。本文针对多变量系统强耦合和大时滞的特点,将当前迅速发展的内模控制方法应用于多变量大时滞系统的解耦中。并对非方形系统的解耦控制问题进行了初步的研究。本文的主要工作如下:(1)根据内模控制原理和解耦控制理论推导内模解耦控制器矩阵,分析时滞和非最小相位零点对控制器的影响。(2)采用遗传算法对模型进行降阶处理,简化设计过程;并且在此基础上对模型进行最小相位和非最小相位分离,然后根据解耦后闭环回路和控制器对角元素应该满足的时滞和RHP零点的条件设计内模解耦控制器,并且在反馈回路中添加低通滤波器,滤波器参数的选择兼顾了系统的稳定性和鲁棒性。(3)基于Matlab/Simulink平台,对内模解耦控制器的控制效果进行了研究。分别对系统的解耦性能、鲁棒稳定性以及在干扰和参数摄动下对工艺指标的控制效果进行了仿真实验研究。(4)创新地把多元统计中建立最优回归方程的理论应用到耦合度分析中,根据系统的输入输出数据建立回归模型,通过建立最优回归方程来确定系统输入和输出之间的最佳配对关系,达到了弱化系统耦合的效果。仿真实验证明了方法的可行性。更多还原

【Abstract】 In the industrial control field, many uncontrollable factors exist in the controlled object, such as nonlinear, strong coupling, large time-delay, non-minimum phase characteristics etc, the factors lead to large quantitative overshoot, settling time longer, and even, some even appear to flap and dissipation,so the system dynamic quality significantly decreased.It is difficult to use most of the control methods in the control system, so decoupling control in the multi-variable big time-delay process has been a major problem in the control area. According to the characteristics of strong coupling and large time-delay of the multivariable system, the internal model control method, developed rapid at present, is applied to the decoupling of the multi-variable big time-delay system. And make a preliminary study for the decoupling control issues of non-square system.The main contents of this paper are as follows:(1) According to the internal model control principle and decoupling control theory, the decoupling controller matrix is derived and the influence of the time-delay and non-minimum phase zero for the controller is analyzed.(2) The reduction of time-delay model is processed with genetic algorithm, and the design process is simplified, the minimum phase and the non-minimum phase are separated based on the model reduction, then design the internal model decoupling controller based on the diagonal elements of the decoupled closed-loop transfer function and the controller should meet the requirements of the time-delay and RHP zero. And then low-pass filter is added in the feedback loop, Both the system stability and robustness are considered to choose the filter parameters.(3) Based on platform of Matlab / Simulink, the control effects of internal model decoupling controller was studied.Decoupling performance of the system, robust stability ,disturbance and parameter perturbation performance are tested in the Matlab / Simulink platform.(4) The theory, which the optimal regression equation is established in the multivariate statistics, is innovatively applied to coupling analysis. According to input and output data of the system, regression model is established, the optimal matching relation is determined between input and output of the system by establishing the optimal regression equation, reach to the effect of weakening the system coupling, the simulation results show that the method is feasible.更多还原