Research On Coupling Compensator And Adaptive Controller For MIMO Control System

Since there always is coupling between the control channels in the MIMO(multiple input multiple output) control system, we will meet some serious difficulties in the course of system design. At present, decoupling method undoubtedly is main design method, which uses coupling compensator to transform the MIMO system into system with independent control channels, and then design the controller one by one to. But designing the coupling compensator needs a premise that the coupling functions in the system should be known. So the coupling functions we must identified. However, the identification of the coupling function is much more difficult than the identification of single object, so far there is no mature method to identify complex coupling function. This thesis put forward an innovation control method for MIMO system, it is named as method without decoupling. It is not necessary to know the coupling functions during whole design course, therefore identification about coupling functions becomes unnecessary.This thesis expounds the principle of the control method without decoupling. Firstly, it deduces the method of parallel compensation method of coupling, which does not know the coupling functions, but the compensators need a lot of energy. Then, it gets a practical control algorithm without decoupling through the block diagram transformation.The system can work normally under condition that compensator only need a small amount of energy. Nerual network can be as algorithm of coupling compensators, and various availble controllers can be as the main controllers of every channelin yhe system. Therefore, this kind of system is feasible and can be realized. The method in the thesis avoids the identification of coupling function, so it provides a new way for the design of the MIMO control system with coupling. My method solves an important problem that has plagued the academic community of automation for long time.The thesis studies the control algorithm of nonlinear MIMO system. In the control method without decoupling, an adaptive control algorithm is employed in the main control channel, and a CMAC network algorithm is employed in the coupling compensator.This thesis has done research on linear system with linear coupling function and nonlinear system with slow time-varying coupling function. Conventional controller and adaptive controller are respectively employed as the main controller.This paper has done research on linear system and nonlinear system, which have linear coupling function and slow time-varying coupling function. The main controller try to use conventional controller and adaptive controller respectively.The thesis has made improvements on incremental GPC(Generalized Gredictive Control) algorithm for MIMO systems, to make it more suitable for MIMO system with time delay. The method enhance the robustness of the system and provide an effective method for selecting the optimal incremental coefficient.In this project, some examples are simulated by MATLAB tool to validate the control algorithms. The results of simulation show that these control algorithms are feasible, and they can effectively improve the performance of MIMO control systems.In the thesis, achievements can be applied to a variety of process control systems and other control systems with coupling. If being spreaded, the achievements can generate considerable benefits. My thesis has the innovation in the control method of MIMO system, and it has a contribution to the development of control theory.