【作者】 周华；

【导师】 杜志达；

【作者基本信息】 大连理工大学 ， 土木工程管理， 2006， 硕士

【摘要】 机械配置是施工组织设计的一项重要内容，特别是对大型工程项目的施工尤为重要。合理的机械配置方案能保证工程按期完成，并且节约施工成本，从而给施工单位带来良好的经济效益。所以对施工机械配置方案进行优化研究，不仅具有很好的理论意义，而且有一定的实际意义。 长期以来，人们对机械配置优化问题的研究主要是基于排队论、程序化领域专家法、计算机仿真法以及线性规划法等方法，并且取得了一定的研究成果。本文在前人研究的基础上，选取土石方工程中装运机械配置问题作为研究对象，并采用粒子群算法对配置模型进行了求解。 粒子群算法是一种群体智能优化方法，它具有对目标问题要求不高、收敛速度快、全局搜索能力强等特点。针对土石方工程中装运机械配置问题，在算法的应用中设计了以机械的型号和数量为自变量，以基于动态权重系数的时间—费用目标为评价函数的求解过程，最后通过一个算例验证了算法求解该问题的可行性。 具体研究工作和成果如下： 1．在分析土石方装运系统特点的基础上，构建基于排队论的土石方装运系统过程仿真模型，通过模型的运行能得到每个配备方案的总费用、总工期等指标。 2．利用多目标优化理论，改进了以时间—费用为目标函数的机械配置模型，模型中引入动态赋权法，在求解模型时实现随机分配目标权重系数。 3．利用改进的粒子群算法求解模型，结果显示该算法具有较好的优化效果，并且能提供多种优化配备方案。 4．建立机械配备管理系统，该系统为仿真工程机械配备的初始化方案提供依据。更多还原

【Abstract】 Deployment of construction equipment is considered as a highly important content to constructional organization design, especially in the big-scale project. The optimal deployment of Construction Equipment not only ensures the project finishes on time and Saves the cost, but also Brings many benefits for the builder during the construction. Therefore, it is a Significant topic about optimal deployment of construction equipment, because research to it not only has very good theory significance, but also has very big usability.The ressarch of this problem in the early stage is mainly based on methods such as queuing theory, expert system, computer simulating, linear programming, etc., and has made some achievements. This paper chooses the topic on optimal deployment of construction equipment in earthwork as the research object, which is basis on the forefathers’ study. A algorithms named particle swarm optimization is adopted to solve the deployment model.Particle swarm optimization algorithm is a new optimization approach that developed recently. Compared with other algorithms, the particle swarm optimization algorithm has a quicker rate of convergence, higher accuracy and the better global convergence ability. Furthermore it has little requirement to the goal questions. First of all, regard mechanical type and quantity as the independent variable in this algorithm, secondly, regard the time-cost goal which is based on dynamic weight coefficient as the evaluation function, finally, through of verify the feasibility of the algorithm by recalculating an example which is gaven in another paper.The concrete research work and achievement in this paper are as follows: Firstly, construct the loading system model on the basis of analyzing the characteristic of loading system in the earthwork, which used in earthwork procedure simulation using queueing theory, then through the operation of the model can receive index of each scheme such as total time, total cost,etc.Secondly, construct deployment model by utilizing multi-objective optimization and take time-cost goal as its evaluateion function, then, introduce dynamic weight coefficient in the model, which is able to distribute the weight coefficient of the goal at random.Thirdly, utilize particle swarm optimization algorithm to solve the model, and the result reveals this algorithm not only is effective, but also can offer many kinds of scheme.更多还原