基于多目标进化的超限超重货物运输专列路径优化方法

Route Optimization Method of Out-of-Gauge and Overweight Freights Transportation Special Train Based on Multi-Objective Evolution

针对超限超重货物运输专列路径优化问题,引入限界改造和桥梁加固要素,以铁路运能损失、运输时间和运输费用的最小化为优化目标,构建铁路超限超重货物运输专列路径优化模型;考虑专列同时运送多件超限超重货物的特点,设计货物综合投影算法,并以Pareto支配关系框架下的NSGA-Ⅱ多目标进化算法为基础,结合基于移动不可行解的自适应惩罚函数的约束处理技术及超限超重货物运输专列开行特征,提出一种基于多目标进化的铁路超限超重货物运输专列路径优化算法。实例分析结果表明:该方法能高效合理地处理路径优化任务中的多个目标;与常用的3种约束多目标算法相比,设计的优化算法能使计算耗时降低0.24%~29.94%、超体积指标值提高4.25%~13.11%,并得到一组收敛性和多样性更好的相对最优方案。该方法克服了传统专列路径优化中单纯依靠经验或从备选方案中择优的弊端,可为铁路超限超重货物运输专列路径决策提供技术支持。

Aiming at the route optimization problem of out-of-gauge and overweight freights transportation special train(OOFTST),introducing factors of gauge modification and bridge reinforcement,the route optimization model of railway OOFTST is constructed with the optimization objective of minimizing railway capacity loss,transportation time and transportation cost.Considering the characteristics of the special train transporting multiple out-of-gauge and overweight freights at the same time,the freight comprehensive projection algorithm is designed.Based on NSGA-Ⅱmulti-objective evolutionary algorithm in the framework of Pareto domination relation,combined with the constraint-handling technique of adaptive penalty function based on the moving infeasible solution and the running characteristics of OOFTST,a route optimization algorithm based on multi-objective evolution for railway OOFTST is proposed.The results of case study show that the proposed method can deal with multiple objectives of route optimization tasks efficiently and reasonably.Compared with 3 kinds of common constrained multi-objective algorithms,the designed optimization algorithm can reduce computing time by 0.24%-29.94%,improve index of hypervolume by 4.25%-13.11%,and formulate a set of relatively optimal schemes with better convergence and diversity.It overcomes disadvantages of relying on experience or selecting the best from alternative schemes in traditional route optimization of special train,and provides technical support for route decision of railway OOFTST.