【作者】 杨杰；

【导师】 贾利民；

【作者基本信息】 北京交通大学 ， 安全科学与工程， 2017， 博士

【摘要】 轨道交通具有运量大、速度快、污染小、安全高效、准点性强等优势,正在逐渐成为国民经济的大动脉和现代城市交通运输的主体骨架。随着我国高速铁路网的快速建设,既有铁路的功能定位将逐渐向货物运输倾斜,形成更加高效、安全、快捷的货运通道。近年来,社会对能源安全和环境保护问题日益关注,轨道交通节能,特别是大功率电力机车牵引的货运列车的节能运行优化与控制问题,具有重要的学术意义和工程应用价值。货运列车是一个高度非线性、大惯性、运行环境恶劣、控制难度大的复杂机电系统,其运行过程是复杂系统与复杂环境相紧密耦合,形成的更为复杂的非线性、多约束、多目标、时变过程。本文围绕绿色、安全、智能的货运列车运行控制,开展了如下几个方面的工作:货运列车集总模型构建:论文在现有模型的基础上对列车系统的受力分析、运动方程、约束条件的建模进行凝练和拓展,结合被控对象特征,采用分段线性化、模糊逻辑、多模态建模等方法对模型进行了完善,构建了列车运行集总模型,较全面地体现了被控对象的关键特征,为后续的优化计算和智能控制提供更为坚实的模型基础。同时,集总建模是对传统建模方法的一种拓展,也是对复杂对象建模的一种有效尝试。列车运行能耗关键要素敏感度分析:论文对影响列车运行能耗的主要影响因素进行了仿真计算和对比分析,直观地体现各个因素对运行能耗的影响程度,以便在速度曲线优化和运行控制过程中对各种因素进行统筹考虑,并作为相关操纵策略选择的决策依据。论文通过仿真证实了最优惰行关系在一定的计算精度下,表现为一个区域,而非仅仅是一个点。为了便于程序对运行时间的精确控制,和统筹把握计算速度与计算精度之间的平衡关系,本文提出将能耗曲线的谷底区段按照能耗隶属度函数设置为基于置信区间的最优切换区域。从对能耗的影响程度来看,对于计划运行时分的充分利用才是速度曲线优化和运行控制过程中最为关键的要素。列车节能运行目标速度曲线优化:论文在列车运行过程四模态理论的基础上提出了六模态划分方法。基于理想条件下的最优速度曲线与实际工况的有机结合,以人类的先验知识对优化过程进行启发式引导,提出了基于步长自适应和双向迭代的速度曲线优化算法,可以快速确定模态序列及其切换点在时间和距离维度上的位置坐标,然后按照充分利用计划运行时分的原则,结合最优切换区域对切换点在给定邻域内进行精确调整。对于陡坡区段,论文按照分层优化的思路,提出了平均速度等效算法进行局部近似,在不影响能耗水平的前提下,将复杂的全局优化问题转化为相对简单的局部优化问题。列车运行速度跟踪控制:针对货运列车运行过程的非线性、大滞后、多目标、多约束的特点,在目标速度曲线给定的基础上,采用了牵引力前馈的方法,有效缓解了货运列车的大滞后特性带来的收敛速度慢的问题,较好地提升了控制系统的动态性能。根据列车运行过程多模态的特征,论文对应设置多个FPID子控制器构成MM-FPID控制器,各个子控制器根据所对应模态的需求特征进行差异化设计,较好地回避了控制器设计过程中不同要素的相互制约与平衡问题。针对车速超调在低限速区段引起的撞线制动问题,采用了牵引力自适应限幅的方法,在无需增加限速安全裕量的前提下,收到了满意的控制效果。并充分融入NN在学习和预测方面的优势,提出了 MM-FPID+NN的组合控制器,实现了准确、高效、稳定的列车速度跟踪控制。列车运行控制计算机仿真平台设计:论文在列车节能运行优化和车速智能跟踪控制的基础上,设计了一套列车运行控制计算机仿真软件,使得优化及控制的仿真效果能够得到直观展示。该仿真软件既可以为相关的科学研究提供仿真计算工具,也可以为实际铁路生产中的生产计划编制、优化、调整提供理论指导和参考依据。论文通过对上述问题的研究,揭示惰行范围对能耗的影响程度以及再生制动对最优惰行关系的影响规律,对现有列车节能运行基础理论起到了有效补充和完善,初步探索运行能耗与计划运行时分、线路条件、运行速度、牵引定数、牵引特性等要素之间的关系及其大致规律,提出简单、快捷、精确、高效,且易于工程化的速度曲线优化方法,为列车速度跟踪控制提供跟踪对象。车速跟踪控制器较好地克服了列车系统的大滞后、非线性特性,实现对目标速度精确、高效、稳定的跟踪控制。整个系统简洁、高效、鲁棒性好,在物理意义上具备很好的可实现性,将为实现货运列车的运行控制从以司机为主的半自动模式转向以机器为主的全过程智能控制模式起到积极的促进作用。更多还原

【Abstract】 Rail transportation has the advantages of large transport capacity,high speed,small pollution,high efficiency,safety,and punctuality.It is gradually becoming the main channel of the economy and the basic framework of modern urban transportation.At present,China’s high-speed rail network construction has been gradually built.The function of the original railway will gradually transferred to the freight,to form a more efficient,safe and efficient freight channel.With the increasing attention to the issue of energy security and environmental protection,rail transportation energy saving is becoming more and more important,especially the speed profile optimization and the speed intelligent tracking control of freight trains with electric locomotive has important academic significance and engineering application value.Freight train is a complex system with high energy consumption,large inertia,poor operating environment,which is difficult to control.Its operation is closely coupled with the complex environment,and forms a more complicated nonlinear,multi constraint and multi objective time varying process.This thesis takes the freight train with electric locomotive as the research object,focusing on the task of tracktion energy saving,carry out the research work as the following:Freight train integrated model construction:Based on the existing models,the thesis makes a detailed analysis of the tracktion force,motion equations,and the modeling of the constraint conditions of the train system.According to the characteristics of the control object,the model is improved by using the method of piecewise linearization,fuzzy logic and multi modal modeling.An integrated model which can accurately reflect the characteristics of train operation is constructed.At the same time,it has enriched the traditional modeling methods,and has opened up a new approach for the modeling of complex systems.Energy consumption sensitivity analysis of train operation:This thesis carried out simulations and analysis on the main factors that affect the energy consumption of train operation,directly reflect the influence degree of various factors on energy consumption,so that to facilitate the overall evaluation of all the various factors as the basis for decision making.This thesis proves that the optimal coasting relationship in certain accuracy is a region rather than a point,and it put forward the concept of optimal switching regoin.The paper also draws a conclusion that the full use of the scheduled running time is the key to the optimization.Train Speed profile optimization algorithm design:Based on four phase’s operation strategy theory,the thesis put forward the six mode theory.In particular,the thesis proposes a speed profile optimization algorithm based on adaptive resolution and bidirectional iteration,which can quickly determine modal sequences and the switching points in the distance and time dimension,and then precisely adjust the switching points in the given neighbourhood of optimal switching region in accordance with the principle of make full use of the scheduled running time.For the not too long steep slope,the thesis proposes a local approximation approach to transform the complex global optimization problem into a relatively simple local optimization problem.Train running speed tracking control algorithm design:According to the freight train’s characteristics of large lag and nonlinear,based on given target profile,the thesis adopts the method of traction force feed forward,which effectively relieve the problem of slow convergence of train speed.It adopts the method of adaptive traction output limitation,so that restrains the speed overshoot effectively,and solves the problems of overspeed brake without increasing the safety speed margin.Combined with a multimodal combined control strategy,the thesis put forward the MM-FPID and MM-FPID+BPNN control algorithms to achieve accurate,efficient and stable train speed tracking control.Computer simulation software package design:Based on the energy-efficient train optimization and intelligent speed tracking control method,a computer simulation platform is designed,so that the effect of simulation and optimization can be visually displayed.The simulation platform can provide a helpfull simulation tools for the relevant scientific research,and can provide theoretical guidance and reference for the production planning,optimization and adjustment in the actual railway production.From the researches on the above problems,the thesis reveals the influence of coasting impact on energy consumption,and the influence of regenerative braking on the optimal coasting relationship.It plays an effective complement and improvement to the existing theory of energy-efficient train operation.It explores the relationship between the energy consumption and the scheduled running time,line condition,running speed,traction characteristic and so on.It presents a speed profile optimization method of simple,fast,accurate,efficient,and easy to engineering,and provides a tracking object for the train speed tracking control.The train speed tracking controller overcomes the large lag and nonlinear characteristics of the train system,and achieves accurate,efficient and stable speed tracking about the target speed profile.The system is simple,efficient and robust,and has a good performance in the engineering sense.It will play a positive role in promoting the operation of freight trains from the semi auto driving mode to an intelligent automatic driving mode.更多还原