基于智能优化算法的机电产品拆卸过程规划研究

发布时间：2018-10-25 06:08

【摘要】：随着科技的发展,人们的生活水平不断提高,机电产品的更新换代也越来越快,报废处理问题随之而来,对报废产品进行回收不仅可以保护环境,还能节省资源。而拆卸是回收的前提,用合适的拆卸方案对废弃的机电产品进行拆卸,能提高拆卸效率,提升拆卸收益。拆卸序列规划是拆卸问题中的重要内容,拆卸序列规划问题在数学上属于NP-完全型,即随着产品复杂度的提高,对产品拆卸过程进行规划将变得异常复杂和困难。本研究针对复杂机电产品拆卸序列规划问题,利用智能优化算法进行求解,提高产品的拆卸效率。本文的主要研究内容包括以下四个方面:(1)本文首先介绍了拆卸的概念和类型、产品的拆卸信息、拆卸信息模型的分类等内容,在分析已有模型的基础上,对常规的拆卸信息模型进行改进,将拆卸对象、拆卸方向、约束类型、拆卸工具、拆卸成本等信息进行处理加入到改进后的拆卸信息模型中,极大的丰富了产品的拆卸信息。并对这些信息进行了数字化处理,以便于计算机程序读取,为接下来利用智能算法对拆卸序列的求解和优化打下基础。(2)介绍了智能优化算法的特点,具体介绍了蚁群算法的原理、数学模型,针对基本蚁群算法的不足,增加了蚂蚁路径的约束,改变了信息素更新机制,提出以拆卸优先约束矩阵为基础的改进蚁群算法,并使用MATLAB进行编码,以某型号洗衣机为例进行了验证。试例结果证明了改进蚁群算法的可行性。(3)介绍了标准遗传算法的原理和数学模型以及在拆卸序列规划方面的研究和应用,并针对其在早熟收敛和局部搜索方面的不足,对遗传算法的基因组编码方式、遗传算子和一般流程进行了改进,提出了基于拆卸优先约束矩阵的自适应遗传算法,并针对某型号洗衣机实例,使用MATLAB进行算法的具体编码和计算,取得了良好的结果,证明了该改进算法的可行性和有效性;并对蚁群算法和遗传算法改进前后的优化结果进行分析比较。(4)针对拆卸方案的评价问题,提出了基于理想度公式的多指标评价体系,综合考虑产品拆卸时的技术性、经济性、环境性等影响,建立拆卸序列评价指标体系。在用智能算法进行产品的拆卸过程规划的基础上,针对搜索到的少数几条可行拆卸序列进行分析,并进行多指标综合评价,从而得到综合指标优化的拆卸序列。为拆卸序列的评价问题提供了一种可行、有效的方法。
[Abstract]:With the development of science and technology, people's living standard is improving, and the upgrading of mechanical and electrical products is also faster and faster. The problem of disposal of end-of-life follows. Recycling of end-of-life products can not only protect the environment, but also save resources. Disassembly is the premise of recycling. Disassembly of discarded mechanical and electrical products with suitable disassembly scheme can improve the disassembly efficiency and increase the disassembly income. Disassembly sequence planning is an important part of disassembly problem. The disassembly sequence planning problem belongs to the complete type of NP- mathematically, that is, with the increase of product complexity, the planning of product disassembly process will become extremely complex and difficult. In order to improve the disassembly efficiency, intelligent optimization algorithm is used to solve the disassembly sequence planning problem of complex electromechanical products. The main contents of this paper include the following four aspects: (1) this paper first introduces the concept and type of disassembly, the disassembly information of products, the classification of disassembly information model, etc. The conventional disassembly information model is improved, and the information of disassembly object, disassembly direction, constraint type, disassembly tool and disassembly cost are added to the improved disassembly information model, which greatly enriches the disassembly information of the product. The information is digitally processed in order to facilitate the computer program to read, which lays the foundation for the solution and optimization of the disassembly sequence by using the intelligent algorithm. (2) the characteristics of the intelligent optimization algorithm are introduced. The principle and mathematical model of ant colony algorithm are introduced in detail. Aiming at the deficiency of basic ant colony algorithm, the constraints of ant path are added, the pheromone updating mechanism is changed, and an improved ant colony algorithm based on disassembly priority constraint matrix is proposed. MATLAB is used to code, and a washing machine is used as an example to verify. The results of an example show that the improved ant colony algorithm is feasible. (3) the principle and mathematical model of standard genetic algorithm and its research and application in disassembly sequence planning are introduced, and its shortcomings in premature convergence and local search are also discussed. An adaptive genetic algorithm based on disassembly priority constraint matrix is proposed to improve the genetic algorithm, genetic operator and general flow chart of genetic algorithm, and an example of a washing machine is given. MATLAB is used to code and calculate the algorithm, and good results are obtained, which proves the feasibility and effectiveness of the improved algorithm. The optimization results before and after the improvement of ant colony algorithm and genetic algorithm are analyzed and compared. (4) aiming at the evaluation problem of disassembly scheme, a multi-index evaluation system based on ideal degree formula is proposed. The evaluation index system of disassembly sequence was established. On the basis of planning the disassembly process of the product with intelligent algorithm, the paper analyzes a few feasible disassembly sequences which are searched, and evaluates the disassembly sequence with multiple indexes, and then obtains the disassembly sequence optimized by the synthetic index. It provides a feasible and effective method for evaluating disassembly sequence.
【学位授予单位】：内蒙古工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X705;TP18

【参考文献】