多目标遗传算法在船闸调度中的应用

摘要

针对传统水运调度效率低下的现象提出一种基于优化多目标遗传算法的船闸调度方法.借鉴生物进化过程,将船闸排档结果定义为进化对象的个体,进行染色体编码和种群初始化;根据不同调度指标构建目标函数,对初始种群进行个体适应度计算;反复按序进行选择操作、交叉操作和变异操作,从全局的角度进行评估决策,输出满足调度需求的最优排档方案,并使用长洲船闸的实际数据进行实验,结果表明该方法既提高船闸利用率又缩短调度时间.%Aiming at the inefficiency of traditional water transport scheduling, a ship lock scheduling method based on improved multi-objective genetic algorithm is proposed. Based on the biological evolution process, the results of ship gearings are defined as individuals of evolutionary objects, and chromosome coding and population initialization are performed. The objective function is constructed according to different scheduling indexes, and individual fitness is calculated for the initial population. Selection opera-tions, crossover operations and variance operations are used to evaluate and make decisions from a global perspective and to output the optimal gear schedule that meets the scheduling requirements. Experiments are carried out using the actual data of Changzhou Ship Lock, which shows that the method not only improves the utilization rate of lock but also shortens the scheduling time.