摘要
针对基于固定增益迭代学习的交通子区边界控制方法收敛速度慢、迭代次数过多及控制精度差的问题。提出了一种迭代学习结合改进狼群算法的交通子区边界控制方案。该方案首先根据宏观基本图理论建立交通子区路网的车辆平衡方程,设计出系统的迭代学习控制律。其次分析了迭代学习控制对宏观基本图的影响,引入自适应步长的狼群算法,该算法以上一批次的宏观基本图为模型,离线对迭代学习控制器的比例和微分增益系数进行寻优,再将最优结果代入下一控制周期迭代学习控制中,进而改善收敛速度与精度。最后,对该方案的收敛性提供了数学证明,而仿真实验结果也表明该算法相较于具有固定增益的迭代学习控制器,收敛速度得到提升,对系统期望轨迹也具有较好的跟踪精度,具有较强的可行性与有效性。
Aiming at the problems of slow convergence speed,too many iterations and poor control accuracy of traffic subregion boundary control method based on fixed gain iterative learning,this paper proposed a traffic subarea boundary control scheme based on iterative learning and improved wolf pack algorithm.In this scheme,it established the vehicle balance equation of traffic subarea network based on macroscopic fundamental diagram theory,and designed the iterative learning control law of the system.Secondly,it analyzed the influence of iterative learning control on the macroscopic fundamental diagram,and introduced the adaptive step size wolf pack algorithm to optimize the scale and differential gain coefficient of the iterative learning controller offline,and then put the optimal results into the next control cycle iterative learning control,so as to improve the convergence speed and accuracy.Finally,the convergence of the algorithm was proved mathematically,and the simulation results show that compared with the iterative learning controller with fixed gain,the proposed algorithm improves,the convergence speed and has better tracking accuracy of the expected trajectory of the system,and it has strong feasibility and effectiveness.
作者
贾光耀
闫飞
张添翼
Jia Guangyao;Yan Fei;Zhang Tianyi(School of Electrical&Power Engineering,Taiyuan University of Technology,Taiyuan 030024,China)
出处
《计算机应用研究》
CSCD
北大核心
2023年第9期2775-2780,共6页
Application Research of Computers
基金
国家自然科学基金资助项目(61703300)
中国博士后科学基金资助项目(2019M651082)
山西省应用基础研究项目(201801D221191)
关键词
交通子区
边界控制
宏观基本图
迭代学习控制
狼群算法
traffic area
perimeter control
macroscopic fundamental diagram
iterative learning control
wolf pack algorithm
