Fault Detection For Delta Operator Network Systems Based On T-S Fuzzy Model

With the increasing complexity of systems,the traditional point-to-point system no longer meets people's needs for production and life.In view of this,Networked Control Systems(NCSs)came into being.NCS cuts a lot of unnecessary wiring,which makes the connection between the equipment of the system more concise.In addition,it reduces costs and the complexity of system maintenance,facilitating information sharing,so it is widely used in industrial production.Due to limited bandwidth,problems such as delay,packet losses,communication limitation will occur.These factors can affect the stability of the system,which should not be ignored when analyzing the system.Once the system fails,it will cause huge economic losses.Therefore,it is very meaningful to quickly detect system failures.The current literature of networked control systems is mostly about linear systems,but nonlinear systems are common in the actual industrial production.The T-S fuzzy model is a very effective tool for studying nonlinear systems.It can use(nonlinear)membership functions to interpolate the local linear model,and expresses the complex nonlinear system with some simple linear systems,which greatly simplify the difficulty of system analysis.In the networked control system,the analog signals received by the sensor will be discretized,and then transmitted through the digital network.At present,most of the literatures use the shift operator discretization method.But in the case of high-speed sampling,the discretized system by the shift operator will deviate from the original system,and the system may be unstable.However,the delta operator discretization method can avoid the above problem.When the sampling period gradually decreases,the discretized system can still remain stable,and it has better digital characteristics.Based on the above advantages,this thesis mainly studies the fault detection for nonlinear delta operator networked control systems under multiple constraints using TS fuzzy model.The main research contents are as follows:(1)Aiming at the nonlinear delta operator system with delay and packet losses,a fault detection filter that satisfies performance is designed.Two mutually independent Bernoulli random variables are used to represent the bilateral packet losses phenomenon.The packet losses and time delay are described in the same model,which is constructed by T-S fuzzy model.The Lyapunov function and linear matrix inequalities(LMIs)are applied to prove the asymptotic stability of the system,and filter parameters are obtained.(2)The existence of parameter uncertainty will also affect the stability of the system.Therefore,on the basis of the previous research content,a nonlinear delta operator fuzzy model is established considering the uncertainty and limited communication.Also,a fuzzy fault detection filter which satisfies performance is designed.By constructing the Lyapunov function,the sufficient conditions for system satisfying asymptotic stability are obtained.Then,the filter parameters can be acquired by solving LMIs.The comparison with the simulation results of the shift operator proves the superiority of the proposed algorithm.(3)The fault detection for the switched fuzzy system with bilateral packet losses and uncertainty is studied.The T-S fuzzy model is used to establish a global fuzzy model for the nonlinear uncertain networked time-delay switching system,and the packet losses process satisfying Bernoulli distribution.Then,a fuzzy filter is designed.By using the average dwell time method(ADT)and multiple Lyapunov functions,the system is proved to be exponentially stable.Finally,the minimum value of performance and filter parameters are obtained by solving LMIs.Simulation results verify the effectiveness of the method.