Cooperative Control And Distributed Consensus For Wireless Networked Control Systems

With the advance of computing,sensing and wireless communication technology,wireless networked control system overcomes the disadvantages in traditional networked control system,such as high cost of installment and maintenance,poor expansibility,which has become the most promising choice of networked engineering applications.Since the structure and scale of controlled plant keep enlarging,the devices in wireless networked control system(such as sensors,controllers,actuators)exchange information via the wireless communication network according to the designed coorperative schedule in order to complete ubiquitous sensing task or a cooperative control task.This cooperative schedule helps the integration of physical process,information processing and communication in networked control system,which leads to the improvment of effectiveness,control performance and reliability.Due to the adoption of wireless network,the flexibility of the control method in wireless networked control system is increased.However,the opening-up affects the accuracy of information in wireless communication network,and the huge amount of wireless communication easily lead to the exhaustion of communication resource.The opening-up and the limitation of communication resource become the main reasons which affect the performance in wireless networked control system.Thus,the important issue in wireless networked control system is how to design the coordination scheme to cooperate with sensors,controllers and other device.In this thesis,the coordination schedule among sensors/controller/actuator in wireless networked control system has been investigated in the views of feature of networked devices,the topology in communication network and other aspects.The main contributions of this thesis are listed as follows,.1.From the perspective of nonlinear controller,the cooperative control problem in wireless networked control network is studied.On the basis of the nonlinearity in controller,the cooperative control is translated into an optimization about the stability of the closed-loop system via the Lyapunov stability theory and the Lurie absolute stability theory.It is proved that the proposed distributed cooperative control algorithm ensures the stability of the networked control system.2.The distributed sampled-data control is investigated for the cooperative control problem in cyber-physical system.According to the fixed relationship of subsystem in physical process,the topology in controller communication network is utilized to design the distributed sampled-data control by virtue of in-degree/out-degree.This distributed sampled-data control improves the utilization efficiency of resources properly in accordance with the pratical resources.3.The finite time consensus filtering algorithm based on event-triggered scheme is studied for wireless sensor network under the data injection cyber attack.The finite time consensus filtering is given via introducing the event-triggered scheme and the data injection attack model with data dropout.It is proved that the consensus of wireless sensor network system can be achieved in a finite time via the proposed filtering algorithm with less consumption on communication resourse.4.The cooperative control algorithm is investigated for relay-asisted wireless sensor/actuator network.By the transformation in algebraic graph theory,control problem in the the heterogeneous network is transformated to the stability issue in a unified framework.By algebraic graph theory and the Lyapunov stability theory,the cooperative distributed control is obtained.It is proved that the proposed cooperative control algorithm ensures stability of.relay-asisted wireless sensor/actuator network system.5.In order to reduce the occupation of the network communication resources,a distributed event-triggered scheme is investigated for the consensus of heterogeneous multi-agent systems.The consensus of the heterogeneous multi-agent systems is considered in two cases,respectively.a)For the fixed topology,by the local information at the event-triggered instants and the error between current state and the model-based estimation,an event-triggered distributed consensus is proposed.b)For the switching topology,the distributed event-triggered consensus is proposed by using a topology-depedent dwell time.Since the triggering condition is based on estimation,the proposed event-triggered method is good at enlarging the released time and alleviating the triggering number.6.In accordance with the non-fragilty of controller and the fault of actuator,a hybrid-triggering mechanism is proposed.This mechanism consists of event-triggered scheme and the period sampled-data scheme.In this hybrid-triggered mechanism,a non-fragile control is given by the virtue of Lyapunov-Krasovoskii functional and input delay method.It is proved that the proposed non-fragile control algorithm ensures the stability of NCS according to different triggering condition and actuator failure rate.