基于SMDP的高速动车组车轮镟修策略研究

发布时间：2018-08-30 07:54

【摘要】：近年来随着我国经济飞跃发展和社会需求的提高,高速铁路进入了全面发展时代,高铁运营里程和动车组数量不断增加。极大地提高了我国铁路客运能力,给社会带来了巨大的经济效益和社会效益。越来越多的动车组投入使用,对动车组在役安全和科学有效的维修管理提出了更高的要求和挑战。车轮作为动车组最重要的部件之一,更是维修研究中的重中之重。车轮在运行过程中,与轨道不断接触,造成轮缘和踏面难以避免的磨耗,对整个动车组的行车安全及稳定性产生威胁。为了恢复车轮的正常几何尺寸和确保动车组的运行安全,必须对磨耗车轮进行镟修或更换。然而,镟修会导致车轮材料的减少和维修成本的增加。如果镟修量过大会导致车轮过早报废,镟修量过少导致镟修成本增加。因此,"何时镟修,镟修量多少"成为车轮镟修策略制定中亟待解决的关键问题。针对该问题,本文提出一种基于车轮周期状态检测信息的动态镟修策略,降低车轮维修成本,延长其使用寿命。首先通过对某编组车轮测量数据的统计分析,确定踏面直径和轮缘厚度磨耗速率的分布,揭示动车组车轮踏面和轮缘磨耗规律。同时,根据镟修结果数据的统计和分析,确定镟修比例系数的分布规律,为后续车轮镟修策略的研究奠定基础。接着,在上述研究的基础上,将车轮踏面直径和轮缘厚度的退化过程离散成有限的退化状态,建立二维状态空间,以单位期望成本最低为目标,构建基于SMDP的镟修优化模型,并采用策略迭代算法对模型求解。然后,综合考虑磨耗和外部冲击对车轮退化过程的影响,构建了基于失效阈值冲击模型和SMDP的镟修优化模型。最后,在上述建立的磨耗分布和镟修比例系数分布基础上,根据某编组动车组车轮的实测数据,将本文提出的动态镟修策略与固定镟修策略作比较,验证了提出的动态镟修策略的有效性。
[Abstract]:In recent years, with the rapid development of economy and the improvement of social demand, high-speed railway has entered a comprehensive development era, and the number of high-speed rail operating mileage and EMU has been increasing. It has greatly improved the capacity of railway passenger transportation in China and brought great economic and social benefits to the society. More and more EMU are put into use, which puts forward higher requirements and challenges to the safety and scientific and effective maintenance management of EMU in service. As one of the most important parts of EMU, wheel is the most important part in maintenance research. In the course of running, the wheels are in constant contact with the track, resulting in the inevitable wear of the flange and the tread, which threatens the driving safety and stability of the whole EMU. In order to restore the normal geometry of the wheel and ensure the safety of the EMU, the wear wheel must be rotated or replaced. However, turning will result in less wheel material and higher maintenance costs. If overhauling results in premature scrapping of wheels, too little trimming results in increased cost. Therefore, "when turning and how much" becomes the key problem to be solved in the strategy of wheel turning. To solve this problem, this paper proposes a dynamic turning repair strategy based on the information of wheel periodic state detection, which can reduce the maintenance cost and prolong the service life of the wheel. Firstly, the distribution of tread diameter and flange thickness wear rate is determined by statistical analysis of the measured data of a certain marshalling wheel, and the wear law of wheel tread and flange of EMU is revealed. At the same time, according to the statistics and analysis of the result data, the distribution law of the turning scale coefficient is determined, which lays a foundation for the study of the subsequent wheel turning strategy. Then, on the basis of the above research, the degradation process of wheel tread diameter and flange thickness is discretized into a finite degenerate state, and a two-dimensional state space is established. With the lowest expected cost per unit as the goal, an optimization model of turning based on SMDP is constructed. And the strategy iterative algorithm is used to solve the model. Then, considering the impact of wear and external impact on wheel degradation process, a failure threshold impact model and a SMDP lathe optimization model are constructed. Finally, on the basis of the distribution of wear distribution and turn-out proportional coefficient, and according to the measured data of a marshalling EMU wheel, this paper compares the dynamic turning repair strategy with the fixed turning repair strategy. The effectiveness of the proposed dynamic turning repair strategy is verified.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U269

【参考文献】