Research On Optimization Technology For Rail Maintenance And Renewal Decisions Of Beijing Subway Based On Approximate Dynamic Programming

With the continuous increase of the scale of the line network and the passenger traffic,prolonged use of the track under high load in Beijing subway cause the more rail defects,which brings challenge to the safe operation and the maintenance management of rail of Beijing subway.Research on optimization technology for maintenance and repair(M&R)decision of Beijing subway plays an important role in improving the safety management level of rail and control the M&R cost of rail.By analyzing the research status of the M&R decision of rail and Approximate Dynamic Programming,with the shortcomings of rail M&R management of urban rail transit,this paper constructs an Approximate Dynamic Programming-based Beijing subway rail M&R decision optimization model.Divided the line into multiple rail segments,by studying the deterioration characteristics and management demand of Beijing subway,with the intensive management method.The model is constructed by taking the rail network composed of multiple rail segments as the modeling object.Under the premise of considering the uncertainty of rail degradation,the model takes the minimum M&R costs of rail network in the planning period as the target.The model sets the M&R activity constraints,M&R budget constraints and network state constraints,and designed the model solution algorithm based on the Approximate Dynamic Programming.In order to verify the model,the author collected a total number of 36 rail abrasion inspection historical data from 2011 to 2019,the rail M&R data of 9 years from 2011 to 2019,and the facility account data of Beijing subway line 2 by participating in the Beijing Municipal Transportation Committee project “Big Data-Based Track Intensive Management” and the project of Beijing Subway Operation Co.,Ltd.“Beijing Subway Track Intelligent Operation and Maintenance Management Information System” undertaken by Beijing Jiaotong University.Based on the above data,taking the rail network composed of 136 curved rails of Beijing subway Line 2 as an example.Optimization the M&R decision of the rail network of Beijing subway line 2 for 5 years by the Approximate Dynamic Programming-based Beijing subway rail M&R decision optimization model,and the M&R decision obtained by the model is compared with the actual execution M&R decision.The results show that based on the M&R decision optimized by the model can arrange the M&R more pertinently,which have effectively reduced the maintenance cost while ensuring safety.The author has designed and developed the Beijing Subway Rail Maintenance and Repair Decision Optimization Subsystem based on the project “Beijing Subway Track Intelligent Operation and Maintenance Management Information System”.The author designs the function module and database of the Beijing Subway Rail Maintenance and Repair Decision Optimization Subsystem,and completes the system development work by Microsoft Visual Studio 2015 development tools.Finally,the function effect analysis of the system is carried out by using the actual data of Beijing subway Line 2,which verifies the effectiveness of the system function.