The Research Of Multivariable Imc-Pid Control Method Based On Non-Square Effective Open-Loop Transfer Function

IMC-PID control simultaneously contains the both advantages of the internal model control and the PID control. So IMC-PID has good robustness and control performance, but also can be widely applied in practice in the form of PID. These features amount to a complex system control problems added new vitality. Multivariable control systems control research in the field has always been a hot and difficult, the key point is how to resolve the issue of interactive between multi-loops. In this paper, IMC-PID controller design method is studied in order to control in a multi-variable system to be more effective, practical and simple control method.First, this paper analyzes the impact of IMC parameters on the system control performance and robustness. Based on this study, the IMC-PID control method is researched. Existing IMC method is rarely mentioned how the tuning filter parameters, mentioned by some empirical formula is given, it is difficult with a wide range of applications, and the argument is exactly the key of control performance and robustness.In order to deal with this problem, in this paper, particle swarm optimization (PSO) applied to the single-variable control system for the tuning of filter parameters. By PSO’s global search ability, but also to ensure that the system in combination with a monotonic step response, stability, and no overshoot of the performance index, the method of IMC-PID controller design is proposed. Simulation studies show that the method for the optimal filter parameter setting can bring good control performance and robustness.Secondly, the paper analyzes multi-variable system. Currently, the system for square multivariable internal model control method has been more, but for the widespread method of non-square control system less and not widely applied, especially for non-square IMC-PID control.Therefore, this paper based on control methods in the square of the system for these problems, proposed an new method of non-square system control. The method is based on the effective open-loop transfer function (NEOTF). Meanwhile, the single-variable system for the PSO-based internal model control method is extended to multi-variable non-square system, to obtain the ideal control effect. For the calculation of higher order items, we use Taylor-PSO model reduction method, making the reduced order model is more accurate. Simulation demonstrate that the proposed method compared with other methods have good robustness and control performance.