基于正点和节能的铁路机车实时控制模型与算法

The energy consumption in China mainline railway is very huge. Optimizing locomotive control is proved to be an effective way to reduce the energy consumption. However, the mainline railway operations in China are characterized by the high volume of traffic and service delay becomes common. Previous studies mainly focuses on the energy-efficient locomotive control, regardless of the need to reduce train delays. In this study, an online model of locomotive control to reduce both energy consumption and train delays will be built. To improve the accuracy of proposed model, the calibration methods on critical parameters of the proposed model will be developed. The fundamental principles on optmizing locomotive control to reduce energy consumption and train delays will be investigated from the aspects of theory and practice. Considering the intrinsic relationships among adjacent trains' movements, a real-time fuzzy predictive locomotive control algorithm to reduce train delays and energy consumption will be developed, taking into account the changes of traffic flow states. Simulation experiments will be conducted to facilitate a practical evaluation of effects and reliability of the proposed method. This study aims to reveal the mechanism of energy-efficient train movement with the minimal train delays, and to propose an enforceable algorithm on real-time locomotive control, and to provide foundation for developing a train automatic control system or a driver support system, which is helpful to reduce railway energy consumption.

我国铁路运输能耗总量较大，优化机车控制是实现节能减排的有效途径。然而，运输密度较高状态下，列车运行面临较高的晚点风险。既有的机车节能优化控制研究限于能耗最小单一目标的实现，对如何减少列车晚点风险与程度考虑不够充分，成为制约其成果应用的主要瓶颈。本项目将构建基于正点和节能双目标的机车实时控制模型，采集实际列车运行数据对关键参数进行滚动修正以提高模型精度；采用实验数据分析和最优控制理论相结合的方法研究机车正点节能优化控制策略；分析追踪列车运行耦合关系,提出基于列车流状态推演的机车模糊预测控制实时优化算法；设计仿真案例验证算法的效果与可靠性，分析其应用前景。本项目旨在揭示列车正点节能运行机理，提出具有可实施性的机车控制实时优化方法，为研究列车自动控制系统或司机操纵辅助决策支持系统、实现我国铁路运输的节能减排提供方法依据。

我国铁路运输任务繁重，牵引能耗较高且呈继续增长的趋势。另一方面，行车间隔较小，列车运行晚点现象较为常见。因此，研究机车正点与节能操纵优化方法具有重要的现实和理论意义。为此，本项目基于实际测试数据分析了传统列车运行计算误差及其成因，提出了牵引计算参数动态修正方法，建立了较为准确的列车运动方程；调研了我国铁路运输列车操纵的规程与要求，建立了基于实际操纵约束的单列车牵引节能操纵实时优化模型，提出了基于优秀司机操纵经验和专家规则的模糊预测控制求解算法；在实际运行线路的测试验证了本项目模型与算法的效果；考虑前后列车可能存在的相互干扰，提出了列车追踪运行协同最优控制策略；针对晚点列车提出了基于正点和节能双目标的优化模型与算法，案例分析表明该方法可以在尽快恢复列车晚点的同时实现牵引节能。本项目的研究成果揭示了给定时分下列车节能运行的机理，为未来研究列车自动控制系统或司机操纵辅助决策支持系统、实现我国铁路运输的节能减排提供了方法依据。