【作者】 程涛；

【导师】 张永；

【作者基本信息】 东南大学 ， 交通运输工程， 2021， 硕士

【摘要】 煤炭是现代工业不可或缺的养料,中国富煤、贫油、少气的能源储量结构决定了中国的能源供给一直以煤炭为主。然而国内煤炭资源的分布不均,我国煤炭供需之间的空间地理矛盾产生了大量的煤炭铁路运输需求,因此对煤炭调运进行研究具有重要意义。目前大型能源企业的煤炭调运计划编制主要依靠粗放的人工方式实现,全面采用科学信息手段,实现人机共同决策制定调运计划将会是大型能源企业的未来发展方向。首先,本文对煤炭分级调运问题展开研究。文章对于煤炭调运问题的实质、构成要素及其特点进行了研究,将煤炭调运问题归纳为非传统运输问题;同时,对于煤炭调运问题的现有求解模型和求解算法进行分析,并结合算法的特性分析其适用情况。在剖析现有求解模型和算法的基础上,确定了本文研究的煤炭调运问题的建模类型和求解算法。其次,依据调运计划分解的逻辑,分别建立了煤炭调运优化年模型,月分解模型和逐日模型。以年模型为例,验证了遗传算法和Lingo求解煤炭调运问题的性能优劣。通过建立时空网络,将煤炭调运量、库存量和销售量的按月变化情况纳入月分解模型考虑范围,并以疫情前后煤炭价格变化为数据,验证模型在“黑天鹅”事件发生情况下的优化能力。在逐日模型中考虑煤运列车跨周期到达和装车平载的情况,将铁路天窗计划和装车站平载约束纳入模型考虑范围。最后,本文基于三级优化模型,建立了具有一致关联性的煤炭调运问题算例,依据年模型算例对求解算法性能的验证结论,采用Lingo对月分解模型和逐日模型的算例分别进行求解。通过计算发现,本文提出的求解模型和求解算法能够对煤炭分级调运优化问题进行有效的求解,且求解结果能够较好地为大型能源企业提供调运计划制定的决策依据。更多还原

【Abstract】 Coal is an indispensable nutrient for modern industry.China’s energy structure of rich coal,poor oil and little gas determines that China’s energy supply has always been dominated by coal.However,the distribution of domestic coal resources is uneven.The geographical contradiction between coal supply and demand has produced a large amount of coal transportation demand,so it is of great significance to study coal transportation.At present,the coal dispatching and transportation plan of large energy enterprises mainly relies on manual work.It will be the future development direction of large-scale energy enterprises to fully use scientific information means,realize man-machine joint decision-making and make dispatching plan.First of all,the problem of layered coal transportation is carried out.The essence,elements and characteristics of coal transportation is studied in this paper,and sums up the coal transportation problem is summed up as a non-traditional transportation problem.At the same time,the existing solution model and algorithm of coal transportation problem is analyzed,and its application combined with the characteristics of the algorithm is also analyzed in the paper.On the basis of analyzing the existing solving models and algorithms,the model types and solving algorithms of coal dispatching problem in this paper are determined.Secondly,according to the decomposition logic of coal transportation plan,three optimization models which contains Year model,Quarter-month decomposition model and Daily model is established respectively.Taking the Year model as an example,the performance of Genetic Algorithm and Lingo in solving coal dispatching problem is verified in the paper.Through the establishment of space-time network,the monthly changes of coal transportation,inventory and sales volume are taken into account in the Quarter-monthly decomposition model.Taking the coal price changes around the COVID-19 epidemic as the data,the optimization ability of the model in the case of sudden natural events was verified.The cross period arrival of coal trains and requirement of average-loading in the station is considered in the Daily model.The railway skylight plan and average-loading constraints are considered in the model.Finally,based on the three-level optimization model,an algorithm case of coal transportation problem with consistent correlation is established.Based on the results of the Year model,Lingo is used to solve the Quarter-month decomposition model and the Daily model.Through the calculation,the solution idea and algorithm are proved to be efficient to solve the optimization problem of hierarchical coal scheduling,and the results can better provide decision-making basis for large energy enterprises.更多还原