Distributed Control Research Of Networked Control Systems Based On Event-Triggering Transmission Mechanism

NCS are spatially distributed systems for which communication between plant, sensors, actuators and controllers is connected by a shared communication networks. In recent years, NCS have brought many innovative impacts on control systems. They are becoming increasingly important in industrial processes for many advantages, such as low installation and maintenance costs, high reliability, increased system flexibility, and decreased wiring. As such, network-based analysis and design have many industrial applications in, for example, aircrafts,.manufacturing plants, robots, automobiles, remote surgery and so on.However, because of the physical properties, NCS also has some shortcomings in operation process, such as network transmission delays, data packet dropouts, disorder, which can deteriorate the performance of the NCS and even destabilize the systems. Furthermore, in the traditional control strategy, the control task is executed periodically; however, periodically executing control task will cause the waste of communication resources but also increased the network load when the whole system is in good operation performance. Since the networks’ bandwidth is limited, the way of periodically triggering control tasks can affect the performance and stability of the control system. And in general, the scale of the networked control system is relatively large and the plants, sensors, actuators and controllers module are distributed in the space, the centralized control algorithm doesn’t work, so we need to introduce the distributed control mechanism.Motivated by the above observations, in recent years, many experts and scholars have focused their attention on the control strategy based on event-triggering transmission mechanism about the networked control system. In this mechanism, the control task is performed by the occurring of a given event, namely control tasks executed on demand. The research about this field has been achieved some good results and has important theoretical significance, but it still has some shortcomings. So, in order to reduce the unnecessary waste of bandwidth resources, improve the performance of the networked control system, and also make up the lack of previous studies, this paper proposes a new control strategy of event-triggering transmission mechanism. In the networked control systems, a subsystem broadcast its state information to its neighbors only when the subsystem’s local state error exceeds a given threshold. Compare with previous studies, the events triggering transmission mechanism of this paper is distributed, only need to use the subsystem’s local state information. Through the study of the paper, this control method can well ensure the entire networked control system’s performance and stability, reduce the network bandwidth requirements, improve the efficiency of the network transmission.First of all, in an ideal world, we build the model of networked control systems contain the subsystems of nonlinear and linear respectively, after that we use the theory of Lyapunov stability to analyze the asymptotic stability of the entire networked control system. Through the analysis of networked control system stability we established the distributed triggered events, it can be seen from the simulation analysis that the triggered event designed in this paper guaranteed the asymptotic stability of the networked control system commendably, and the triggered control task number is less than the traditional control strategy significantly.In the actual situation, the phenomenon of network transmission delays and data packet dropouts will appear in the networked control systems. Aiming at these two problems, the distributed event-triggering transmission scheme in this paper can locally predict the maximum allowable network transmission delays and the maximal allowable number of successive data packet dropouts(MANSD). We show that if the number of each subsystem’s successive data packet dropouts is less than its MANSD and the network transmission delays are zero, the resulting system is finite-gain pL stable. If the delays are bounded by given deadlines, the entire system is asymptotically stable.In this paper, an event-triggering H? filter design method is addressed for a class of networked control systems consists of continuous linear subsystems in chapter V. A new event-triggering mechanism is proposed and a novel filtering error system model is established to account for both the event-triggering condition and network transmission delay in a unified framework. At last, based on this model, we can use Lyapunov functional approach together with LMI technique to analyze both the stability and the H? performance of the event-triggering filtering error system, then design the H? filter through the result of a group of LMI.Finally, we summarize the results of the dissertation and propose some problems that are worth further study.