Modeling And Synthesis Of Networked Fuzzy Systems Under Event-triggered Scheme

Networked systems have high practical values and theoretical meaning because of its advantages of easy to implement,low cost,high reliability and so on.The introduction of event-triggered scheme to networked systems not only reduces the energy loss,but also reduces the waste of network bandwidth,which further improves the significance of networked systems.Additionally,nonlinearities are universally existing in most systems,and T-S fuzzy model can handle nonlinearities with simple progress and arbitrary precision,then this thesis employs T-S fuzzy method to deal with nonlinearities.On account of networked fuzzy systems subject to event-triggered scheme,we focus on the influence of network induced issues such as data quantization,network induced delay,data packet losses and so on,and the effects of the issues to event-triggered scheme and fuzzy design,the detailed works are summarized as follows.1.We investigate the filter design for fuzzy systems under the effects of eventtriggered scheme and network induced issues.Considering the data quantization induced by a logarithmic quantizer and the data losses modeled by Bernoulli process,as well as the event-triggered scheme for reducing the loss of energy and network bandwidth,we get the states space expression of the augmented system via a specific auxiliary random series.Then,based on the Lyapunov stability theory,we obtain the stability conditions of the filter error system and the design method of the filter via T-S fuzzy method and LMI based method.2.We investigate the controller design and improvement for fuzzy systems under the effects of event-triggered scheme and network induced issues.Considering the effects of event-triggered scheme and data losses,we get the states space expression of the augmented system via a specific auxiliary random series.Referring to the influence of networked issues to the fuzzy controller design,we present a differential mean value theorem based method to convert the membership functions of the controller into the sum of membership functions of system and an bounded uncertainty part which improve the design process of the controller.3.We investigate the fault detection for fuzzy systems under the effects of eventtriggered scheme and network induced issues.Considering the effects of event-triggered scheme and data losses,we describe the network induced issues via a specific auxiliary random series.According to the features of the systems and the external influence such as disturbance,we construct the fault detection filter to suppress the disturbance and detect the faults.In consequence,the state space expression of the residual error system is obtained,then the stability analysis and filter design is processed.Thus,choosing appropriate threshold,associated with the residual signal generated by the fault detection filter,the sensor fault can be detected if it happened.4.We investigate the modeling and control synthesis for continuous fuzzy systems under the effects of event-triggered scheme and network induced issues.Considering the effects of event-triggered scheme,transmission delay and data losses,referring to the features of continuous system,and the effects of the involved issues under different situations,the state space expression is obtained.According to the system model,construct a new Lyapunov functional,then the stability analysis and controller design are processed.5.We investigate the modeling and control synthesis for networked systems under a new event-triggered scheme.The quantization based event-triggered scheme can convert the effects of data quantization and event-triggered scheme into only one of the effect of the two issues,thus simplifies the analysis.Consequently,we analyze the modeling methods of the systems under the proposed scheme and networked issues(transmission delay,data losses).Then the controller design methods are obtained.The above works present a series of methods to handle some basic problems in eventtriggered control and synthetic analysis of networked fuzzy systems such as co-modeling of event-triggered scheme and data losses,fuzzy design under networked issues.Moreover,the application and performance of the methods are further studied.Finally,we verify the effectiveness of the presented methods by simulation examples.