Fuzzy Modeling And Control For A Class Of Inverted Pendulum Systems

Nowadays, with the rapid development of the modern science and technology and productive forces, the controlled objects become more and more complex in the practical process control system, which caused the people would have more and more high request to the control system. Fuzzy modeling based on T-S fuzzy model has attracted the experts and scholars’ attention, and how to set up less fuzzy rules and the model with high precision to be an important task on fuzzy modeling.This thesis considers the model for a class of a single inverted pendulum system. The thesis realized several fuzzy modeling for a class of single inverted pendulum system by combining the past fuzzy modeling method. This thesis studies modeling problems for the fuzzy system, in order to obtain relatively simple T-S fuzzy model, the thesis combined with the traditional approximate linearization of fuzzy modeling method. The nonlinear term of unable to approximate will be established three kinds of fuzzy modeling with the fan range, uncertainties and to take the world’s methods. Each kind of fuzzy modeling method with less fuzzy rules model and higher modeling precision advantage through balance the number of fuzzy rules and the dimensions of the matrix inequalities on the control calculation burden.Aiming at the uncertainties selected, the T-S fuzzy system will be obtained based on the robust control technology. Although the presented modeling method will obtain larger dimension matrix inequality, it can be decreased the number of fuzzy rules. By combining the improved fan range method and the linear approximation method, then we can obtain the T-S fuzzy model and use the PDC technology is utilized to design the fuzzy controller. Numerical simulation results, obtained by Matlab, demonstrate the well control effectiveness based on the proposed method for the model of T-S fuzzy system and fuzzy controller, this king of single-stage inverted pendulum system can be stable control. For convenience, the thesis only choose one of the fuzzy modeling method to design the controller and the computer simulation.