The Improved Design And Simulation Of Model-Free Adaptive Control Method

Classical control theory and modern control theory is based on the system model, such as PID control, predictive control, fuzzy control, neural network control and so on. However, these control methods have some defects, for example, there are so many parameters to determine, and these methods do not have a satisfactory effect when dealing with the large time-delay system. So, the control methods based on model are facing the high requirement of practical problems in engineering and engineering application. The new control methods should meet the requirements of controlled objects, such as imprecise model, adaptability, overcoming the large time-delay system and so on. Model free adaptive control (MFAC) which does not base on the mathematical model of the controlled object is another important aspect of automation. MFAC is control object oriented, it is a functional combination control method which does not consider the mathematical model of the controlled system first.Model-free adaptive (MFA) control, as its name suggests, is an adaptive control method that does not require process models. There are not any information of controlled process in the controller, only the I/O data of the controlled system should be available when designing the controller. It is the products of the integration of modeling and controling, a nonlinear controller designed by the approach of functions combination. Unlike the traditional control methods which consider the mathematical model of the controlled object first, MFAC combinates the modeling and feedback control, real-time modeling and real-time feedback control correction online. It has been widely used recently.This paper makes research on the basic model-free adaptive control method and proposes the improved method which combinates the single output tracking differentiator with the basic MFAC. The new controller makes full use of the filter characteristics of the tracking differentiator and improves the anti-interference performance of the MFAC. Specific studies are as follows:1,The basic theory of model-free control method. Introducing several major branches of model-free control theory, such as model-free fuzzy control, expert control, neural network model-free control, learning control, auto-disturbance rejection control and model-free adaptive control. 2,The basic theory of model-free adaptive control method. Analyzing the related theory of model-free adaptive control method. First, introducing the concept of universal model and its expression. Second, deriving model-free adaptive control law and the estimated algorithm of the pseudo-partial derivative which is the unknown factor of the control law. Finally, getting the model-free adaptive control program.3,Designing the model-free adaptive controller. Mainly making research on model-free adaptive controller. First, analyzing the convergence and rationality of MFAC with the mathematical methods. Second, Explaining why the model-free adaptive controller can be applied to a wide class of nonlinear systems without modeling, finally, presenting the model-free adaptive control program and making simulation comparison with PID control method. The simulation results show the model-free adaptive control method is better than PID control method in the aspacts of tracking, adaptability and overcoming the large time-delay obviously. Reaching the following conclusions: First of all, model-free adaptive controller is like a black box, second of all, model-free adaptive control can handle non-linear discrete-time systems.4,The improved design of model-free adaptive control method. Making improvement on the basic model model-free adaptive control method to further enhance the control quality of model-free adaptive controller. MFAC with single output tracking differentiator (TD-MFAC) is proposed which combinating the basic model-free adaptive controller and the single output tracking differentiator. The improved controller which combinates the filter characteristics of the single output tracking differentiator enhances the anti-jamming ability of model-free adaptive control method. Finally, making the compared simulation with the basic model-free adaptive controller. The simulation results show that TD-MFAC has the unique advantage which is suitable for dealing with the controlled system with interference.