基于增强模拟退火算法的动车所调车作业计划多目标优化方法OA
Multi-objective Optimization for Shunting Schedule of Electrical Multiple Unit Depot via Enhanced Simulated Annealing Algorithm
尽头式动车所调车作业存在多种可选作业模式,且在调车作业时需考虑作业总时间及转线复杂度等多性能指标,合理制定综合上述情况下的动车所调车作业计划,对提高动车所检修能力具有重要意义.为此,以最小调车作业时间及转线复杂度为目标,构建带咽喉区股道约束的动车所调车作业混合整数线性规划模型,并提出一种增强多目标模拟退火算法(EMOSA).该算法融合基于启发式规则的解码设计,以消除股道占用在时空上的冲突,面向调车作业进行股道合理分配;设计与问题规模相关的帕累托前沿解集重启机制,避免算法陷入局部最优.对不同规模的多个案例进行测试,验证了所提改进算子的有效性.最后,以某动车运用所的调车作业计划编制为例,验证了模型和算法的实用性和正确性.
There are many optional operation modes in the Stub-end electrical multiple units depot,and the performance indicators including total operation time and line transfer complexity should be considered in the shunting operation.Therefore,it is of great significance to reasonably formulate the shunting schedule under the above conditions to improving the maintenance capacity of the EMU depot.For this reason,a mixed integer linear programming model for shunting schedule in EMU depot with throat track constraints was constructed with a view to minimizing the total operation time and complexity of shunting schedule,with an enhanced multi-objective simulated annealing algorithm(EMOSA)proposed.The algorithm integrated a decoding approach based on heuristic rules to avoid the conflict of track occupancy in time and space,and allocates the track reasonably for shunting operation;and a restart mechanism of Pareto front solution set that was directly related to the problem scale was designed to avoid falling into local optimum.The effectiveness of the improved operator was verified by testing several cases of different scales.Finally,the practicability and correctness of the model and algorithm were verified by taking the shunting schedule of EMU depot as an example.
刘毅;唐秋华;何明
武汉科技大学 冶金装备及其控制教育部重点实验室,湖北 武汉 430081||武汉科技大学机械传动与制造工程湖北省重点实验室,湖北 武汉 430081
交通运输
尽头式动车所调车作业计划模拟退火算法多目标优化启发式规则
Stub-end Electrical Multiple Units DepotShunting ScheduleSimulated Annealing AlgorithmMulti-objective OptimizationHeuristic Rules
《铁道运输与经济》 2024 (002)
10-19 / 10
国家自然科学基金面上项目(52275504,51875421)
评论