Research And Application Of Nonlinear Systems Fuzzy Modeling

The intelligent control is the focus research in the study of control field, and it is the advanced stage of development of control theory. With the advanced technology and development of industrial production technology, the control precision of the object is much higher. In the actual industrial process, the problems of process object includes: multi-variable, time variability, strong coupling,and the information that people can informed is fewer, which bring problems of the complex nonlinear system modeling, Therefore , it needs to find a effective way to solve these problems.The nonlinear systems fuzzy modeling theory of this article can solve these problems better, with the purpose of increasing the precision of the fuzzy model and the interpretability of fuzzy systems,the article make a few improvements to the fuzzy modeling methods . In this article, the analyses focus on the improvement of the algorithm theory and the application of the actual object.Theoretical research contains fuzzy structure identification of fuzzy model,including the choice of input variables,the number of membership functions and rules extraction, fuzzy partition of input space and so on. Fuzzy ant colony clustering algorithm make up for the problem of local minimum and the central issue of redundancy in traditional method; for the consequent parameters of T-S fuzzy model, using recursive least- squares parameter implement identification of fuzzy model,Finally, using fuzzy system and neural network to form a neural fuzzy inference network, the system has the self-learning ability, and thus to optimize and adjust the T-S fuzzy model parameters, so that the output of fuzzy model can approximate the actual system. In the fuzzy modeling method validation, using the well-known gas furnace data to fuzzy modeling and simulation, obtains the feasibility of fuzzy modeling method. In practical application,choose the second order water tank for the modeling of objects, used research methods in this paper to fuzzy modeling, when the fuzzy model is completed, select a scheme to control the lower water tank level. According to the simulation, verify the effectiveness of the fuzzy modeling method.