摘要
列车优化调度是一个大规模、复杂的、具有非线性离散变量和多约束的多目标数学优化问题.在优化过程中,考虑了特快旅客列车中途离开时间和整个运行时间等因素.首次将粒子群优化(particle swarmoptimization,PSO)技术引入列车优化调度,克服了传统优化方法易陷入局部最优和维数灾难等弊端.通过一个工程实例验证了该算法的可行性和有效性.同时,与现存的列车优化调度方法相比,粒子群优化方法的搜索时间短而且优化结果更接近最优解.
The scheduling for high-speed passenger trains is in fact a complex multi-objective and multi-restrictive mathematical optimization problem involving nonlinear discrete variables, where the departing time of express trains in transit and the time for whole travel should be considered in optimization, To overcome the drawbacks of conventional mathematical optimization methods, such as easy to come into local optimization and dimensional disasters, the particle swarm optimization (PSO) is first introduced into the train scheduling with the intention of offering a better one. An example of train scheduling is given to illustrate the feasibility and effectiveness of the algorithm proposed, and the algorithm is compared with the existing optimal scheduling methods. The results show that the scheduling by PSO is shorter with the whole optimized closer to the ideal one.
出处
《东北大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2007年第2期176-179,192,共5页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(60274009)
关键词
列车调度
多目标优化
混沌粒子群优化算法
惩罚函数方法
train scheduling
multi-objective (CPSO) algorithm
penalty function approach optimization
chaotic particle swarm optimization