Study On Control Issues For Takagi-Sugeno Fuzzy Systems

With the rapid development of technology,control systems have become increasingly complicated.Today's control systems have evolved from simple,easily describe,low computation cost linear systems into complex,high cost nonlinear systems,which bring more difficulties for system analysis.Takagi-Sugeno(T-S)fuzzy model can address the nonlinear systems effectively.In the framework of ''IF-THEN'' rules,complex nonlinear plants can be regarded as the summation of a serious of linear subsystems with corresponding membership functions,and then,can be analyzed and controlled via the linear control theory.However,the type-1 T-S fuzzy model uses certain membership functions while ignoring the uncertainties which exist in membership functions in considered systems,therefore,it is unable to tackle the problem of uncertainties.To this end,the interval type-2(IT2)fuzzy model was born to deal with such an intractable issue.By using the upper and lower membership functions with corresponding weighting coefficients,uncertainties,which exist in nonlinear uncertain systems,can be described effectively.On the other hand,complex transmission environment will result in the occurrences of numerous factors,such as sensor failure,actuator failure,packet loss,event-triggered and etc.Inspired by above discussions,the control approaches of T-S fuzzy systems are investigated from the following aspects:The first chapter introduces the background and advantages of the T-S fuzzy model,the event-triggered strategy and the research significance of uncertain quarter active vehicle suspension systems.Thus the necessaries and significances of the research of this paper are established.Meanwhile,the corresponding techniques and methods utilized in this paper are presented.The second chapter considers a class of uncertain nonlinear discrete-time systems.Based on the IT2 T-S fuzzy model,the IT2 T-S fuzzy discrete-time systems are established,and a fuzzy filter design method,which is more general and does not share the same membership functions of considered systems,is proposed,thus the flexibility of designing fuzzy filter is improved.In addition,sensor failure model is introduced to describe this phenomenon.Based on the above technique,sufficient conditions to design dissipative filter are proposed,and the problem of how to design reliable IT2 T-S fuzzy filter in the case of sensor failure is addressed.Finally,simulation results illustrate the effectiveness of the control approach.The third chapter considers a category of nonlinear networked systems.Based on T-S fuzzy model,the fuzzy nonlinear networked systems model is established.In view of that the partial states are unavailable,the dynamic output feedback control scheme is adopted to control considered systems.Considering the value and finiteness of communication resource,a novel adaptive event-triggered strategy is introduced to decrease transmission burden and save limited communication resource.In view of the complexity and instability of the network environment,the Bernoulli random distribution is used to describe the phenomenon of data loss and the actuator failure model is established to describe the failure of the corresponding component.On this basis,the conditions of designing adaptive event-triggered fuzzy dynamic output feedback controller are proposed.The problem of how to design a fuzzy controller to control the networked system in the case of unavailable state,data loss and actuator failure is tackled while saving the transmission resources.The fourth chapter researches quarter uncertain active vehicle suspension systems in the framework of T-S fuzzy model.Considering the aging of sensor components and the long term utilization,a sensor failure model is introduced to describe the failure that occurs in sensor.Since the most sensitive area of the human body is 4Hz-8Hz,the controller is directly designed in the finite frequency domain and a fuzzy controller model is constructed.On the basis of generalized Kalman-Yakubovich-Popov lemma,projective theorem and reciprocal projection theorem,the conditions of designing the fuzzy controller in the case of sensor failure in the finite frequency domain are given.On the basis,the problems of improving drive comfort level and suspension constrained requirements are tackled.Finally,simulation results verify the effectiveness of the proposed approach.