基于误差阈值通讯的网络控制系统控制方法研究

发布时间：2018-04-09 13:11

本文选题：网络控制系统　切入点：事件触发机制　出处：《杭州电子科技大学》2017年硕士论文

【摘要】：随着控制系统向分布化、智能化和综合化的快速发展,使得各网络节点的计算资源、通信资源以及能源条件受到一定程度的限制。在这种情况下,传统的周期采样(等周期时间触发)对各种资源的综合利用率很明显存在着一定的局限性。换句话说,这样得到的分布式信息通过网络互联传送到各个节点无疑将加重网络负载,特别在无线网络环境中频繁地周期性发送数据必将加重节点的计算负担和能耗损失。因此为了降低网络资源负载和能耗负荷,基于事件触发控制近几年来得到了国内外学者的重视和发展。事件触发控制即是在满足控制系统一定性能条件下,节点信息的发送将按预先设定的条件来决定是否执行,其中预先设定的条件即称之为事件触发机制或事件触发条件。已经得到证明在一阶随机系统中,相对传统周期采样系统,基于事件触发控制可以使系统输出的方差明显减小,这表明在降低网络资源时该方法具有很大的潜力。本文研究了基于误差阈值通讯的网络控制系统控制方法。首先,针对连续被控对象的网络控制系统建立了离散模型,利用李雅普诺夫泛函方法和线性矩阵不等式(LMI)方法得出了满足系统渐近稳定的一些方法和条件。实现了事件触发机制条件下系统的一致有界性,给出了系统收敛上界。通过调节参数可以调节系统的状态。然后,在同时考虑网络传输时延和事件触发机制影响下,网络控制系统与事件触发机制建立了网络诱导时延的离散时间模型。利用李雅普诺夫泛函和线性矩阵不等式方法,设计的状态反馈H?控制器保证了闭环系统是渐进稳定而且也保证了H?性能指标,从而有效的降低了节点的数据传输速率。最后,在前面研究的基础上,研究了一类基于绝对误差阈值的非线性网络控制系统的一致有界性。首先建立一种非线性连续被控对象的网络控制系统模型,通过离散化、线性化方法将非线性系统转换成一种离散化的模型。利用李雅普诺夫泛函和线性矩阵不等式方法,设计控制器使系统稳定。在系统包含事件触发机制的情况下,设计控制方法使系统一致有界。
[Abstract]:With the rapid development of distributed, intelligent and integrated control system, the computing resources, communication resources and energy conditions of each network node are restricted to a certain extent.In this case, the traditional periodic sampling (equal period time trigger) has some limitations on the comprehensive utilization of various resources.In other words, the distributed information will undoubtedly increase the network load, especially in the wireless network environment, the frequent periodic transmission of data will increase the computing burden and energy loss of the nodes.Therefore, in order to reduce the load of network resources and energy consumption, event-based trigger control has been paid attention to and developed by scholars at home and abroad in recent years.Event trigger control (EFC) is defined as the event triggering mechanism or event triggering condition, which is called event trigger mechanism or event trigger condition, when the node information is sent according to the pre-set conditions under certain performance conditions of the control system.It has been proved that in first-order stochastic systems, event-based trigger control can significantly reduce the variance of system output compared with traditional periodic sampling systems, which indicates that this method has great potential in reducing network resources.In this paper, the control method of networked control system based on error threshold communication is studied.First of all, a discrete model is established for the networked control system of a continuous controlled object, and some methods and conditions satisfying the asymptotic stability of the system are obtained by using the Lyapunov functional method and the linear matrix inequality (LMI) method.The uniform boundedness of the system under event triggering mechanism is realized, and the upper bound of system convergence is given.The state of the system can be adjusted by adjusting the parameters.Then, considering the influence of both network transmission delay and event trigger mechanism, the discrete-time model of network-induced delay is established by network control system and event trigger mechanism.Using Lyapunov Functionals and Linear Matrix inequality (LMI) methods, the state feedback H?The controller ensures that the closed-loop system is asymptotically stable and that H?Performance index, thus effectively reducing the data transmission rate of the node.Finally, the uniform boundedness of a class of nonlinear networked control systems based on absolute error threshold is studied.Firstly, a networked control system model of a nonlinear continuous controlled object is established, and the nonlinear system is transformed into a discrete model by means of discretization and linearization.By using Lyapunov functional and linear matrix inequality (LMI) method, the controller is designed to make the system stable.When the system contains event triggering mechanism, the control method is designed to make the system uniformly bounded.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273

【参考文献】