基于多目标遗传算法的铁矿石码头泊位调度研究

发布时间：2018-07-16 08:49

【摘要】：自2003年开始,我国一直是世界铁矿石的进口第一大国,长期大批量的运输要求,促进了我国铁矿石专业化物流链的发展,沿海港口加速进行专业化铁矿石码头建设,但是,依然满足不了经济发展和钢铁工业的需要。那么只利用港口现有的资源,仅仅通过调度,不增加港口的设施设备,而能达到优化资源、加快船舶装卸、提高港口设备利用率和降低港口运营费用,具有重大的意义。论文以铁矿石码头泊位调度为研究对象,首先从我国是世界上第一大铁矿石进口国的现实,分析我国铁矿石港口接卸能力与进口总量的差距,引出论文研究的意义,结合国内外研究现状,提出论文的研究内容和方法。其次,分析铁矿石码头泊位调度的相关问题,对论文要建立的模型进行了简化,分析铁矿石船舶卸船的流程,确定两个优化目标函数。考虑船型不同的船舶,其待泊费不同,指出以往很多研究以船舶在港时间最小为目标的不足,确定船舶在港费用最小的目标函数；以船舶装卸时产生的泊位运营费用确定第二个目标函数,即最小化港口运营成本。再次,概述遗传算法的相关知识,论文对基本遗传算法的原理和流程作了相关综述,论述了多目标优化用遗传算法来实现的具体方法。最后,设计铁矿石码头泊位调度遗传算法的具体实现过程。基于论文研究排序问题,采用自然数编码；多目标优化通过权重系数法实现,适应度函数通过对目标函数的转换而得；选择算子采用无放回余数随机选择,使用两点交叉法进行交叉,变异操作为换位变异；选择相关控制参数,对约束条件进行处理；最后针对一个具体实例,运行遗传算法,验证算法有效性。
[Abstract]:Since 2003, China has been the largest import country of iron ore in the world. The long-term large quantity transportation requirements have promoted the development of the specialized logistics chain of iron ore in China, and the coastal ports have accelerated the construction of specialized iron ore wharves. Still can not meet the needs of economic development and iron and steel industry. Therefore, it is of great significance to utilize only the existing resources of the port, not to increase the facilities and equipments of the port, but to optimize the resources, speed up the loading and unloading of ships, improve the utilization ratio of port equipment and reduce the cost of port operation. The thesis takes the berth scheduling of iron ore terminal as the research object. Firstly, from the reality that our country is the largest iron ore importer in the world, this paper analyzes the gap between the iron ore port's unloading capacity and the total import quantity, and draws the significance of the paper's research. According to the present situation of domestic and international research, the research contents and methods of this paper are put forward. Secondly, the related problems of berth scheduling of iron ore terminal are analyzed, the model to be established is simplified, the process of unloading iron ore ship is analyzed, and two optimization objective functions are determined. Considering that different ship types have different berthing fees, it is pointed out that many previous researches have taken the minimum time of ship in port as the target to determine the objective function of the minimum cost of ship in port. The second objective function is to minimize port operation cost. Thirdly, the related knowledge of genetic algorithm is summarized, the principle and flow of basic genetic algorithm are summarized in this paper, and the specific method of realizing multi-objective optimization by genetic algorithm is discussed. Finally, the implementation process of genetic algorithm for berth scheduling of iron ore terminal is designed. Based on the research of sorting problem in this paper, natural number coding is adopted; multi-objective optimization is realized by weight coefficient method; fitness function is obtained by converting objective function; selection operator adopts random selection without retractive residue, Using the two-point crossover method, the mutation operation is transposition mutation; select the relevant control parameters to deal with the constraints; finally, for a specific example, run the genetic algorithm to verify the effectiveness of the algorithm.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F252;F552;TP18

【参考文献】