三峡升船机与船闸梯级枢纽联合调度算法研究

发布时间：2018-07-05 17:50

本文选题：联合调度 + 遗传算法　；参考：《中南林业科技大学》2017年硕士论文

【摘要】：三峡升船机的投入运行以及多线梯级联合调度环境的改变,给三峡升船机与船闸梯级枢纽联合调度问题提出了新的挑战。需要对升船机投入后的梯级联合调度问题进行分析研究,主要研究内容如下:(1)建立两坝间联合调度问题的数学模型。通过对两坝间联合调度问题分析,找出船舶航运约束、船闸运行约束和其他约束条件,提出以最大化闸室面积利用率与最小化船舶的待闸时间为优化目标,确立了两坝间联合调度问题的数学模型。(2)建立船舶过坝优先级模型。分析了影响船舶过闸的因素和调度规则,提出了静态权重与动态权重相结合的船舶优先级模型。静态权重考虑船舶类型、过坝方式和船舶货种三个因素,并采用二进制编码方法计算静态权重。动态权重以船舶静态权重和船舶待闸时间为参数进行计算。(3)基于遗传算法的闸室编排算法。闸室编排是求解三峡升船机与船闸梯级枢纽联合调度问题的重要组成部分。提出了闸室编排问题的数学模型,该模型属于NP问题,然后提出基于遗传算法的三峡—葛洲坝船闸闸室编排算法,包含编码表示、选择策略、杂交算子和变异操作,优化了闸室面积利用率与船舶待闸时间,提出的算法有效地解决了闸室面积利用率和过闸船舶优先级之间的矛盾。(4)两坝间联合调度问题的混合智能算法。采用基于闸次时间表优化进行联合调度,提出了基于粒子群优化算法和模拟退火算法的两坝间联合调度智能算法。采用两层循环对变量进行搜索,外层循环利用粒子群优化算法对船闸闸次总数和闸次航向的值进行变换,内层循环在由船闸闸次总数和闸次航向划定的子空间内利用模拟退火算法对闸次开始时刻进行搜索。选取三峡—葛洲坝2015年4月船舶流数据进行测试,比较混合智能算法与实际人工编排的平均闸室面积利用率以及船舶平均待闸时间结果,并在梯级枢纽联合调度验证软件上进行试验,验证了算法的有效性、实用性。
[Abstract]:The operation of the three Gorges ship lift and the change of the multi-line cascade combined dispatching environment bring new challenges to the joint operation of the three Gorges ship lift and the shiplock cascade hub. It is necessary to analyze and study the cascade joint dispatching problem after the ship lift is put into operation. The main research contents are as follows: (1) the mathematical model of the joint dispatch problem between two dams is established. Based on the analysis of the joint dispatching problem between the two dams, the ship shipping constraints, shiplock operation constraints and other constraints are found out. The optimal objectives are to maximize the utilization ratio of the lock chamber area and minimize the waiting time of the ship. The mathematical model of joint dispatching between two dams is established. (2) the priority model of ship crossing dam is established. This paper analyzes the factors that affect the ship crossing and the dispatching rules, and puts forward a ship priority model combining static weight and dynamic weight. The static weight takes into account three factors: the type of ship, the way of crossing the dam and the type of ship cargo, and the binary coding method is used to calculate the static weight. The dynamic weight is calculated by the static weight and the waiting time of the ship. (3) the gate layout algorithm based on genetic algorithm. Lock chamber arrangement is an important part of solving the joint scheduling problem of three Gorges ship lift and Lock Cascade. In this paper, a mathematical model of gate layout problem is proposed, which belongs to NP problem. Then, a genetic algorithm based on genetic algorithm is proposed for the layout of three Gorges Gezhouba lock, which includes coding representation, selection strategy, hybrid operator and mutation operation. The efficiency of area utilization of sluice chamber and the waiting time of ship are optimized. The proposed algorithm solves the contradiction between the area utilization ratio of gate chamber and the priority of ship crossing gate effectively. (4) A hybrid intelligent algorithm for joint dispatching between two dams. In this paper, a joint scheduling intelligent algorithm based on particle swarm optimization (PSO) and simulated annealing (SA) is proposed. Using two-layer cycle to search variables, the outer loop uses particle swarm optimization algorithm to transform the total number of locks and the course of lock. In the subspace defined by the total number of locks and the course of the lock, the inner loop uses simulated annealing algorithm to search for the starting time of the lock. The ship flow data of TGP Gezhouba in April 2015 were selected and compared with the average area utilization ratio of the gate chamber and the average waiting time of the ship. The validity and practicability of the algorithm are verified by experiments on the verification software for joint dispatch of cascade hubs.
【学位授予单位】：中南林业科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U641.7;U642.7;TP18

【相似文献】