时变时延网络控制系统的研究

发布时间：2018-05-29 16:51

  本文选题：网络控制系统 + 时变时延　； 参考：《哈尔滨理工大学》2017年硕士论文

【摘要】：网络控制系统是经网络中介连接传感器、控制器、执行器并进行信号传输的实时反馈系统,它是一种可以使在一定的区域内地理位置不同的用户之间实现资源共享的远程监控系统。由于各个环节中通信网络的加入使得信息传输过程中受到带宽限制以及网络负载变换等因素影响,发送的数据会随时间变化出现延迟甚至丢失。时变时延的存在,会使系统的稳定范围变小,降低系统的动态性能,有时会造成系统状态发散。稳定性是控制系统设计的基本要求,设计相应的控制器使得系统满足一定的性能指标亦是权衡系统优良的首要环节。因此,本文研究存在随机时变时延的网络系统的时滞依赖渐近稳定准则和反馈控制器设计方案,具备非常重要的理论价值和实际价值。其主要研究内容如下:针对一类具有区间时变时延的Markov随机跳变系统,研究其稳定性问题。基于一种新型的二次凸不等式边界方法,采用含交错项和二次项公式与高次常函数乘积的新型L-K泛函,选用含有较多项独立增广变量,利用交叉项估计方法,推导较少保守性的系统时滞依赖随机渐近稳定的充分条件。利用数值例子证明所提理论的可用性。针对非理想状况下含丢包和随机时延的网络系统,研究其稳定性和状态反馈控制器设计问题。基于定时触发信号传输方式下引起的网络时变时延,采用全新式L-K泛函和自由权矩阵,借助时延系统理论,利用二次凸组合技巧,推导闭环系统稳定性的充分条件和H_∞状态反馈控制器的设计步骤。并通过比较最大允许时延上界验证所提方法的优势。针对存在Markov随机跳变时变时延的网络系统,研究存在有界系统噪声的事件触发方式的量化控制问题。基于二次凸技术处理积分项和二次项的新型L-K泛函,借助于随机切换系统理论,采用对数量化器,推导出控制器的设计方法,并通过数值例子仿真验证新设计方法的有效性。
[Abstract]:The network control system is a real-time feedback system that connects sensors, controllers, actuators and signals through a network intermediary. It is a remote monitoring system which can realize resource sharing among users with different geographical location in a certain area. Due to the addition of communication network in each link, the information transmission process is affected by bandwidth constraints and network load transformation, so the data transmitted will be delayed or even lost with the change of time. The existence of time-varying delay will reduce the stable range of the system, reduce the dynamic performance of the system, and sometimes cause the state divergence of the system. Stability is the basic requirement of the control system design. The design of the corresponding controller to make the system meet certain performance indicators is also the first step to weigh the system excellence. Therefore, in this paper, we study the delay-dependent asymptotic stability criterion and feedback controller design scheme for network systems with stochastic time-varying delay, which has very important theoretical and practical value. The main research contents are as follows: for a class of Markov stochastic jump systems with interval time-varying delay, the stability problem is studied. Based on a new boundary method of quadratic convex inequality, a new L-K functional with the product of interlaced terms and quadratic formulas and higher order constant function is adopted. Sufficient conditions for stochastic asymptotic stability of delay dependent systems with less conservatism are derived. A numerical example is used to prove the availability of the proposed theory. In this paper, the stability and state feedback controller design of network systems with packet loss and random delay in non-ideal conditions are studied. Based on the time-varying delay caused by the timing trigger signal transmission mode, a new L-K functional and freedom matrix are adopted, and the theory of delay system is used to make use of the technique of quadratic convex combination. The sufficient conditions for the stability of the closed-loop system and the design steps of the H _ 鈭，

本文编号：1951665