Research On The Analysis And Control For Networked Fuzzy Systems Based On Piecewise Lyapunov Function

Networked control system(NCS)refers to the closed loop system formed by real-time network as a channel for information transmission.The object of this thesis is to deal with the two typical issues of NCS,data quantization and data dropouts.Considering the objective facts that external disturbance definitely exists and the full knowledge of the state information is rarely avaiable in the actual industrial control system,the problems of stabilization analysis and controller design for nonlinear NCS represented by Takagi-Sugeno(T-S)fuzzy model are studied in depth based on Lyapunov stability theory and piecewise Lyapunov method.The main contents of this thesis are as follows:(1)The state feedback control problem for nonlinear networked stochastic time-delay system with data quantization and single-side stochastic data dropout is studied.Packet dropout stochasticly occurs at single-side communication channel between sensor and controller,and satisfies Bernoulli random process.Data quantization is modelled by the logarithmic static and time-invariant quantizer which exists at single side communication channel between controller and actuator.The quantization effect is represented by sector-bound uncertainty.The plant in the NCS under consideration is nonlinear time-delay system with white noise represented by T-S fuzzy model.Under the assumption that the state is measurable,in order to reduce conservatism of results and deal with time-delay in the plant efficiently,a delay-dependent piecewise Lyapunov function and Non-Parallel Distributed Compensation(Non-PDC)control scheme are used,and the sufficient condition guaranteeing exponential stability of the closed loop system in the mean square and synthesis method of sate feedback controller are proposed by Linear Matrix Inequalities(LMIs).At last,effectiveness of the proposed method is illustrated through an example simulation.(2)H_?state feedback control problem for nonlinear networked system with data quantization,stochastic data packet loss in both links and external disturbance is studied.The packet dropout phenomenon is modelled as mutually independent Bernoulli process,which occurs in both links,from the sensor to the controller link and from the controller to the actuator link.The control input is quantized by static logarithmic quantizer.The plant of the NCS is nonlinear system considering external disturbance described by a T-S fuzzy model.Under the assumption that state is measurable directly,piecewise Lyapunov function and Non-PDC control law are exploited to analyse the performance of the system and design controller.The synthesis method of H?state feedback controller that makes the control system asymptotically stable in the large and meet disturbance attenuation performance is proposed by LMIs.Finally,an example is presented to demonstrate the effectiveness of the proposed method.(3)H_?dynamic output feedback control problem for nonlinear networked system with data quantization,stochastic data packet loss in both links and external disturbance is studied.The network condition and processing methods are similar to(2).Assuming that state information is not directly measurable and external disturbance is considered,the sufficient conditions to make the closed loop system stable with the given H_?disturbance attenuation performance are proposed by LMIs,based on piecewise Lyapunov function and Non-PDC law.At last,the effectiveness of the proposed results are illustrated by an example.