Considering that uncertain dwell disturbances often occur at metro stations, researchers have proposed many methods for solving the train timetable rescheduling (TTR) problem. This paper proposes a Modified Genetic Algorithm-Gate Recurrent Unit (MGA-GRU) method, which is a real-time TTR method based on deep learning. The proposed method takes the Gate Recurrent Unit (GRU) network as the decision network and uses the results produced by the Modified Genetic Algorithm (MGA) as the training set of the decision network. A well-trained decision network can provide effective solutions in real time after random disturbances occur, in order to optimize the net traction energy consumption of trains in metro systems. Based on the Shanghai Metro Line One (SML1) pilot network, this paper establishes a comprehensive model of the metro system as a training and testing environment to verify the energy-saving effect and real-time performance of the proposed method in solving the TTR problem. The experimental results show that in the two-train metro system, the three-train metro system, and the five-train metro system, the MGA-GRU method can save an average of energy by 4.45%, 6.16%, and 7.19%, while the average decision time is only 0.15 s, 0.27 s, and 0.33 s, respectively.

中文翻译：

---

基于深度学习的随机扰动地铁系统能量优化实时列车时刻表调度方法

考虑到不确定的居住干扰通常发生在地铁站，研究人员提出了许多解决列车时刻表重新安排（TTR）问题的方法。本文提出了一种改进的遗传算法-门递归单元（MGA-GRU）方法，它是一种基于深度学习的实时TTR方法。该方法以门递归单元（GRU）网络为决策网络，并将改进遗传算法（MGA）产生的结果作为决策网络的训练集。训练有素的决策网络可以在发生随机干扰后实时提供有效的解决方案，以优化地铁系统中火车的净牵引能耗。基于上海地铁一号线（SML1）的试验网络，本文建立了一个完整的地铁系统模型作为训练和测试环境，以验证所提出的方法解决TTR问题的节能效果和实时性能。实验结果表明，在两列地铁系统，三列地铁系统和五列地铁系统中，MGA-GRU方法可以平均节省4.45％，6.16％和7.19％的能源，而平均决策时间分别仅为0.15 s，0.27 s和0.33 s。