非线性系统的输出反馈分布式模型预测控制
发布时间：2018-08-31 09:59
【摘要】：分布式模型预测控制是解决大规模系统控制的一种有效方法,能在尽可能简单的系统通信方式和尽可能小的通信负担之下达到尽可能好的控制性能,同时保证算法的收敛性和系统的稳定性.由于实际过程往往是非线性的,且由于条件限制导致系统状态不易准确测量,甚至不可测,这种情况下状态反馈往往难以实现预期的目标,因此一般采用输出反馈控制,设计状态观测器对系统状态进行估计.本文针对一类状态不可测的非线性系统,以及受制于时滞和通信干扰的非线性系统,利用状态观测器提出了一种输出反馈分布式模型预测控制算法,保证了观测器估计状态的最终有界性以及估计状态和真实状态误差的有界性,进而得到原系统状态的最终有界性,使原系统最终稳定.本文的工作可以概括为以下两部分:第一,针对一类状态不可测的非线性系统,研究了一种输出反馈分布式模型预测控制算法.首先设计了状态观测器,在系统输出为异步测量的情况下证明了估计状态与真实系统状态的误差是有界的;然后设计了一个基于Lyapunov函数控制器,保证了标称观测器的原点是渐近稳定的;最后给出了输出反馈分布式模型预测控制算法,证明了观测器的估计状态是最终有界的,进而得到原系统状态的最终有界性.第二,研究了受制于时滞和通信干扰的非线性系统的输出反馈分布式模型预测控制问题.首先给出非线性系统,并对时滞和通信干扰模型进行了描述,设计了带有时滞的状态观测器,证明了估计状态与真实系统状态的误差是有界的;然后设计了一个基于Lyapunov函数控制器,保证了标称观测器的原点是渐近稳定的;最后提出了受制于时滞和通信干扰的分布式模型预测控制算法,为了处理通信干扰,在控制规则中解决一个约束可行性问题来确定通过通信信道得到的数据传递信息是否可用,证明了观测器的估计状态是最终有界的,进而得到原系统状态的最终有界性,保证了系统的稳定性.
[Abstract]:Distributed model predictive control is an effective method to solve large-scale system control. It can achieve the best control performance under the simplest possible communication mode and the smallest communication burden. At the same time, the convergence of the algorithm and the stability of the system are guaranteed. Because the actual process is often nonlinear and the system state is difficult to measure accurately or even undetectable due to the constraints of the condition, the state feedback is often difficult to achieve the desired goal, so the output feedback control is generally used. A state observer is designed to estimate the state of the system. In this paper, an output-feedback distributed model predictive control algorithm is proposed for a class of nonlinear systems with unmeasurable states and nonlinear systems subject to delays and communication disturbances. The ultimate boundedness of the observer estimation state and the boundedness between the estimated state and the real state error are guaranteed, and the ultimate boundedness of the original system state is obtained, which makes the original system finally stable. The work of this paper can be summarized as follows: first, an output-feedback distributed model predictive control algorithm is studied for a class of nonlinear systems with unmeasurable state. First, the state observer is designed, and the error between the estimated state and the real state is proved to be bounded when the output of the system is asynchronous, and then a controller based on Lyapunov function is designed. The origin of the nominal observer is asymptotically stable. Finally, an output feedback distributed model predictive control algorithm is presented, which proves that the estimated state of the observer is ultimately bounded, and the ultimate boundedness of the original system state is obtained. Secondly, the output feedback distributed model predictive control problem for nonlinear systems with time delay and communication disturbance is studied. First, the nonlinear system is given, and the delay and communication disturbance models are described. The state observer with time delay is designed, and the error between the estimated state and the real system state is proved to be bounded. Then a controller based on Lyapunov function is designed to ensure that the origin of the nominal observer is asymptotically stable. Finally, a distributed model predictive control algorithm is proposed to deal with the communication interference. A constrained feasibility problem is solved in the control rules to determine whether the data transfer information obtained through the communication channel is available. It is proved that the estimation state of the observer is ultimately bounded, and the ultimate boundedness of the original system state is obtained. The stability of the system is guaranteed.
【学位授予单位】：曲阜师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP13
本文编号：2214645
[Abstract]:Distributed model predictive control is an effective method to solve large-scale system control. It can achieve the best control performance under the simplest possible communication mode and the smallest communication burden. At the same time, the convergence of the algorithm and the stability of the system are guaranteed. Because the actual process is often nonlinear and the system state is difficult to measure accurately or even undetectable due to the constraints of the condition, the state feedback is often difficult to achieve the desired goal, so the output feedback control is generally used. A state observer is designed to estimate the state of the system. In this paper, an output-feedback distributed model predictive control algorithm is proposed for a class of nonlinear systems with unmeasurable states and nonlinear systems subject to delays and communication disturbances. The ultimate boundedness of the observer estimation state and the boundedness between the estimated state and the real state error are guaranteed, and the ultimate boundedness of the original system state is obtained, which makes the original system finally stable. The work of this paper can be summarized as follows: first, an output-feedback distributed model predictive control algorithm is studied for a class of nonlinear systems with unmeasurable state. First, the state observer is designed, and the error between the estimated state and the real state is proved to be bounded when the output of the system is asynchronous, and then a controller based on Lyapunov function is designed. The origin of the nominal observer is asymptotically stable. Finally, an output feedback distributed model predictive control algorithm is presented, which proves that the estimated state of the observer is ultimately bounded, and the ultimate boundedness of the original system state is obtained. Secondly, the output feedback distributed model predictive control problem for nonlinear systems with time delay and communication disturbance is studied. First, the nonlinear system is given, and the delay and communication disturbance models are described. The state observer with time delay is designed, and the error between the estimated state and the real system state is proved to be bounded. Then a controller based on Lyapunov function is designed to ensure that the origin of the nominal observer is asymptotically stable. Finally, a distributed model predictive control algorithm is proposed to deal with the communication interference. A constrained feasibility problem is solved in the control rules to determine whether the data transfer information obtained through the communication channel is available. It is proved that the estimation state of the observer is ultimately bounded, and the ultimate boundedness of the original system state is obtained. The stability of the system is guaranteed.
【学位授予单位】：曲阜师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP13
【参考文献】
相关期刊论文 前10条
1 刘美丽;;一类非线性系统的状态反馈预测控制[J];自动化技术与应用;2016年03期
2 黄静雯;李宏光;;基于切换控制的经济模型预测控制[J];控制与决策;2016年01期
3 李纹;穆文英;崔宝同;;切换非线性系统的分布式模型预测控制[J];控制与决策;2014年04期
4 张洁;张广辉;苏成利;;一种非线性系统的自适应无模型预测控制方法[J];工业仪表与自动化装置;2014年01期
5 Dewei LI;Yugeng XI;Yuanyuan ZOU;;Model predictive control for constrained uncertain piecewise linear systems[J];Journal of Control Theory and Applications;2013年02期
6 孔小兵;刘向杰;;双馈风力发电机非线性模型预测控制[J];自动化学报;2013年05期
7 蔡星;谢磊;苏宏业;古勇;;基于串联结构的分布式模型预测控制[J];自动化学报;2013年05期
8 蒋亚丽;邹媛媛;牛玉刚;;多速率分段线性系统预测控制的显式设计[J];自动化学报;2013年05期
9 席裕庚;李德伟;林姝;;模型预测控制——现状与挑战[J];自动化学报;2013年03期
10 邹涛;魏峰;张小辉;;工业大系统双层结构预测控制的集中优化与分散控制策略[J];自动化学报;2013年08期
，本文编号：2214645
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2214645.html
