摘要
针对具有模型不确定的城市路网系统,研究了使闭环系统渐近稳定的分布式鲁棒预测状态反馈控制器设计问题.首先,将由"存储与向前"路段线性模型扩展得到的城市路网拓扑模型分解为分布式模型.同时考虑实际情况中路况变化、测量误差等因素引起的饱和流量、车辆数的不确定性对城市路网系统的影响,得到了具有不确定参数的城市路网分布式模型.然后,提出了一种合作式的分布式鲁棒预测控制算法,并利用线性矩阵不等式(linear matrix inequality,LMI)处理方法在线求解多个凸优化问题,给出了子系统控制器求解方法.最后通过Matlab仿真验证了分布式鲁棒预测控制是一种能有效缓解城市交通拥堵和优化信号灯的控制策略.
The distributed robust predictive state feedback controller design problem is investigated in this paper for an urban road network with model uncertainty. Firstly, the urban road network topology model obtained from the extending of the ＂store and forward＂ linear model is divided into a distributed model. Considering the influence of uncertainty on urban road network systemun from road conditions change and the error of measurement, the distributed model with parameter uncertainty is derived for the urban road network system. Furthermore, a cooperative distributed robust predictive control algorithm is proposed, where the state feedback controller is designed by solving convex optimization problems online with LMIs. Finally, an example is given to demonstrate the effectiveness of the proposed method to alleviate the urban traffic congestion and optimize the light split.
出处
《浙江工业大学学报》
CAS
北大核心
2016年第6期633-638,共6页
Journal of Zhejiang University of Technology
基金
国家自然科学基金资助项目(61273117
61403344)
