Compensation-Based Cooperative MFAILC for Multiple Subway Trains Under Asynchronous Data Dropouts
Compensation-Based Cooperative MFAILC for Multiple Subway Trains Under Asynchronous Data Dropouts
复制标题
DOI：
10.1109/tits.2022.3202911
复制
发表时间：
2022-12
影响因子：
8.5
通讯作者：
Qian Wang;S. Jin;Zhongsheng Hou
中科院分区：
文献类型：
--
作者：
Qian Wang;S. Jin;Zhongsheng Hou
研究了测量通道和下行通道同时存在异步数据丢失情况下的多列车协同控制问题。针对多城市地铁列车（MCST）系统由于数据丢失而导致控制性能下降的问题，提出了一种基于补偿的协作无模型自适应迭代学习控制（cCMFAILC）.首先，将非线性地铁列车系统转化为等效的动态线性化数据模型来描述地铁列车系统的输入输出动态特性。接下来，丢失的数据被最近可用迭代中相同时刻的对应数据替换。cCMFAILC的设计目的是保证MCST的速度跟踪误差沿着迭代轴有界，相邻地铁列车的间隔稳定在安全范围内。最后，理论分析和MCST仿真验证了cCMFAILC方案的有效性.
This paper researches the cooperative control problem for multiple subway trains under the asynchronous data dropouts in both measurement channel and downlink channel. The compensation-based cooperative model free adaptive iterative learning control (cCMFAILC) for the multiple city subway trains (MCSTs) is proposed to avoid deterioration of the control performance due to data dropouts. First, the nonlinear subway train system is transformed into an equivalent dynamic linearization data model to describe the input-output dynamics of the subway train system. Next, the lost data is replaced by the corresponding data of the same time instant in the latest available iteration. And the cCMFAILC is designed to guarantee that the speed tracking errors of MCSTs are bounded along the iteration axis and the headway of neighboring subway trains is stabilized in a safe range. Finally, theoretical analysis and MCSTs simulations verify the validity of the proposed cCMFAILC scheme.