Intermittent Control, Observation And Application Of Several Types Of Hybrid Stochastic Systems

Compared with continuous control,intermittent control can effectively reduce the control cost/energy,compared with impulsive control,intermittent control can capture more internal information of the system.As a result,intermittent control has attracted more and more researchers's attention and has become a hot spot in the field of modern control.Based on the deterministic system of intermittent control theory has been perfected,which contains random noise,random uncertainty,Markov jump of the mixed stochastic model is one of current hot research topic in the field of con-trol,mixed stochastic model is not only to supplement of deterministic model,and can accurately response dynamic characteristics of industrial system and nature,and mixed with the presence of random noise is not only change the dynamic characteris-tics of original system,reduce the control performance,and even damage the system stability.In this thesis,the hybrid stochastic system as the research object,explore the structural characteristics of internal system,developed the effective intermittent control method,in order to further reduce the conservative,use the switch system the average dwell time method,and convex combination skill,puts forward a set of suit-able Lyapunov analysis framework.In this thesis,the specific research work includes the following aspects:1.The extended dissipative performance of distributed parameter systems(DPS s)with stochastic disturbances and multiple time-varying delays is studied by us-ing a new fuzzy aperiodic intermittent sampled-data control strategy.Different from the previous fuzzy sampled-data control results,the state sampling of the proposed sampled-data controller occurs only in space and is intermittent rather than continu-ous in the time domain.By introducing a novel multi-time-delay-dependent switched Lyapunov functional to explore the dynamic characteristics of the controlled system,and by means of the famous Jensen's inequality,reciprocally convex approach,etc.,the criterion of the system's mean square stabilization is established based on LMI technique,which quantitatively reveals the relationship between the control period,the control length,and the upper bound of the control sampling interval.Especially,the optimal control gain is given by an optimized algorithm designed in the thesis,which greatly reduces the cost.Finally,a numerical example is presented to demon-strate the effectiveness and superiority of the proposed approach.2.The non-fragile intermittent extended dissipative stabilization and synchro-nization problem for a class of uncertain stochastic reaction-diffusion neural networks with discrete and distributed time-varying delays is investigated.By combining the switching time-dependent Lyapunov functional method with the Wirtinger inequal-ity technique,novel stability and synchronization criteria and extended dissipation analysis can be obtained.Then,by solving a set of delay-dependent linear matrix inequalities,we obtain the required intermittent non-fragile controller,which can be used to satisfy the extended dissipative behavior.Finally,two numerical examples are given to verify the validity of the stable and synchronous results.3.The intermittent quasi-synchronization problem of chaotic delayed neural net-works under parameter mismatching and random perturbation mismatching is stud-ied.First,a novel switching time dependent Lyapunov function with Razumikhin technique is introduced to establish the new quasi-synchronization criteria.The new results remove the limit of control width versus time delay,which greatly improves the results that have been achieved so far.Then,based on the quasi-synchronization criteria,the sufficient conditions on the existence of periodically and aperiodically intermittent controllers are presented.Finally,two numerical examples are used to demonstrate the effectiveness and feasibility of the proposed results.4.The intermittent non-fragile control problem for a class of stochastic neutral time-varying delay systems with random uncertainties is studied.Firstly,the variable substitution method is used to transform the problem into the intermittent stabilization problem of stochastic singular systems.Secondly,according to the dynamic char-acteristics of the intermittent control singular system,the switching time dependent Lyapunov functional is introduced,and the convex combination technique is applied to provide sufficient conditions for the system stability.Then,it is proved that these conditions can be transformed into a set of feasible solutions of linear matrix inequal-ities.Finally,a numerical example is given to verify the proposed control scheme.5.The L1 stabilization problem for switching positive systems with random in-terval delay is discussed by using intermittent static output feedback control strategy.Firstly,a kind of intermittent static output controller is designed to solve the problem that the system state information is only partially measurable when the sensor fault is affected by external disturbance.Then,the piecewise Lyapunov functional is in-troduced,and the modal-dependent mean dwell time technique is used to obtain the corresponding stability criteria.On this basis,the L1 performance of the switched positive system is analyzed and the design method of controller gain matrix is pro-posed.Unlike the previous switched positive system which only considered the de-terministic time delay,the time delay considered in this thesis occurs randomly in the interval,which has more research significance and value.6.The extended dissipative anti-interference control problem for a class of switched singular semi-Markov jumping systems with multiple perturbations and time delays is studied via disturbance observer.Different from the existing results,a nov-el and comprehensive system model is introduced in this thesis,that is,the switched semi-Markov jump model,which is more general than the Markov jump model in de-scribing the random mutations of the system structure and parameters.In addition,the uncertain transition rates considered for switched semi-Markov processes include two cases where the uncertainty is bounded and completely unknown,rather than the com-mon case where a single part is unknown.Especially,the composite controller that enables the closed-loop system to reach random tolerance is easier to calculate than the associated working method.Moreover,this thesis also proposes an optimization algorithm of constrained control gain,which can effectively reduce the actual control cost to a certain extent.Finally,an example is given to verify the effectiveness of the proposed method.7.In view of the multi-noise,random actuator failure and sensor packet loss,the hybrid passive and H? control method based on the intermittent estimator is studied.First,for only partial information caused by sensor failure,An intermittent estimator is designed to track the undetectable state of high-speed trains.Then,two different Lyapunov function/functional strategies are adopted to establish two different stability criteria.Simultaneously,in order to reduce control costs and speed up convergence,two different algorithms are designed.It is worth emphasizing that,unlike the exist-ing HST results that are affected by actuator failures,this thesis uses a more flexible fault representation,that is,the semi-Markov switching mode,which is more realistic background and higher use value.Particularly,the Lyapunov function designed in this thesis can drive the system state monotonically decreasing in the working interval and the rest interval,thereby avoiding the state pulse jump phenomenon.Finally,using the test parameters of the Japanese Shinkansen HST,carry out simulation verification and comparative analysis with related work to verify the advantages of the control technology proposed in this work.