Research On H_∞ Filtering Of Interconnected Systems Under Communication Constraints

Many practical engineering examples can be modeled as networked interconnected large-scale systems.In recent years,control and filter theory of interconnected large-scale systems have attracted intensive interest of many scholars,which has led to considerable theoretical achievements and has been successfully applied in many fields,such as large-scale smart grid systems,computer communication network,chemical engineering and so on.The networked interconnected large-scale system is a complex system composed of several interrelated subsystems.The data of all interconnected subsystems are transmitted through the communication network,it inevitably leads to many problems,such as packet collision,network attacks and limited bandwidth.As a result,the problem of communication restriction is particularly prominent in interconnected systems.At present,the controller design for the networked interconnected large-scale systems under communication restriction has been relatively perfect,but the filter design is rarely involved.Therefore,it is valuable and significant to investigate the stability analysis and filter design of interconnected systems under communication restriction.Based on networked interconnected large-scale systems,this paper study the H_∞filter design problem of the interconnected Markovian jump systems and T-S fuzzy interconnected systems under communication restriction.The main research contents include the following two aspects:(1)The filter design problem for interconnected Markovian jump systems with the event triggering mechanism and output quantization is studied.Firstly,in order to save network resources,a static event-triggered mechanism with delay is introduced.In addition,a logarithmic quantizer is set in front of the filter to quantize the measurement output,which will further reduce the occupation of network resources.Considering the influnence of event-triggered mechanism and output quantization,a filtering error system model is established for interconnected Markovian jump systems.Then,by using Lyapunov functional theory and inequality technique,some sufficient conditions are presented to guarantee the system stochastically stable with an H_∞norm bound,which will be transformed into equivalent linear form.The desired filter and event-based parameters can be obtained via solving a group of LMIs.Finally,a numerical example is given to verify the validity and rationality of the proposed method.(2)The finite-time H_∞decentralized filter design problem of nonlinear interconnected system based on T-S fuzzy model under hybrid network attacks is studied.Firstly,considering that network attacks can occur in communication channels at any time,we use two independent Bernoulli distribution variables to describe Do S attacks and FDI attacks,and a delay-dependent decentralized H_∞filtering error system model is established for T-S fuzzy interconnection system.Then,by constructing the fuzzy line-integral Lyapunov function independent of the time derivative of the membership function and combining with the finite-time stability theory,the sufficient conditions are presented such that the error system is finite-time stochastic stability.The filter parameters can be obtained by the help of LMIs.Finally,the effectiveness and advantages of the proposed method are verified by two simulation examples.