基于Pareto人工鱼群算法的多目标斗链式拆卸线平衡特性研究

发布时间：2018-08-06 20:40

【摘要】：科技的快速发展加快了产品更新换代的步伐,由此产生大量废旧机电产品。实现废旧机电产品的拆卸回收再利用不仅符合绿色制造的要求,还能缓解日益紧张的资源需求局面,且为拆卸企业带来巨大的经济效益。拆卸线是实现规模化、自动化拆卸作业的必然选择,拆卸线平衡问题的研究具有重要的理论研究意义和广泛的实际应用价值。传统生产线中固定工作站节拍时间的作业方式很难实现生产线的平衡,而斗链生产组织方式具有高度的平衡性,通过理论分析与数值仿真验证了考虑回取速度时斗链系统依然具有平衡性。进而建立了基于斗链生产作业方式的拆卸线平衡问题数学模型,该模型包含多个优化目标,不仅考虑了生产线的平衡特性,还需尽早拆卸需求高、有危害的零部件,并尽可能减少拆卸方向改变次数,以满足对拆卸生产系统经济高效、安全环保等先进制造工艺的要求。在求解多目标拆卸线平衡问题时,为克服传统求解方法不能很好的均衡存在多个冲突的目标、求解结果单一等不足,设计了一种基于Pareto解集的多目标人工鱼群算法。为提高人工鱼的寻优能力,采用随机交叉操作指导人工鱼向最优拆卸方向觅食。通过拥挤距离不断筛选觅食、聚群、追尾行为过程中的非劣解,实现了各行为结果的多样性,并将外部档案中的非劣解添加到算法迭代种群中,加快了算法的收敛速度。通过实例验证了所设计算法具有较好的收敛性与分布性,进而对不同规模任务的算例进行求解,得到多种平衡方案,对比已有算法结果,验证了所设计算法的有效性与优越性。将所建立的模型和所设计的算法应用于实际拆卸线中,在求解回取速度相同的3人作业的发动机斗链式拆卸线平衡问题中得到16种平衡方案,在求解回取速度不同的5人作业的打印机斗链式拆卸线平衡问题中得到20种平衡方案,均实现了生产作业的平衡,提高了有效作业时间,为决策者提供了广泛的决策空间。因此,本文所建立的模型和所设计的算法能够较好的解决拆卸线平衡问题,有效改善了线平衡问题,具有较强的适用性和广阔的应用前景。
[Abstract]:The rapid development of science and technology has accelerated the pace of product upgrading, resulting in a large number of used mechanical and electrical products. The realization of disassembly and recycling of used mechanical and electrical products not only meets the requirements of green manufacturing, but also alleviates the increasingly tight demand for resources, and brings enormous economic benefits to disassembly enterprises. Disassembly line is the inevitable choice to realize large-scale and automatic disassembly operation. The research of disassembly line balance has important theoretical significance and extensive practical application value. In the traditional production line, it is difficult to achieve the balance of the production line with the fixed workstation beat time, but the organization mode of the bucket chain production has a high balance. The theoretical analysis and numerical simulation show that the hopper chain system is still balanced when the retrieval speed is considered. Furthermore, the mathematical model of disassembly line balance problem based on bucket chain production mode is established. The model includes several optimization objectives, which not only considers the balance characteristics of production line, but also needs to disassemble parts with high demand and harm as soon as possible. In order to meet the requirements of advanced manufacturing processes such as economy, efficiency, safety and environmental protection, the times of disassembly direction change are reduced as much as possible. In order to solve the problem of multi-objective disassembly line balance, a multi-objective artificial fish swarm algorithm based on Pareto solution set is designed in order to overcome the problems of multiple conflicting targets in the traditional method. In order to improve the optimization ability of artificial fish, the random cross operation was used to guide the artificial fish to forage in the direction of optimal disassembly. The non-inferior solutions in the process of foraging, clustering and rear-end behavior are continuously selected by crowding distance. The diversity of the results of each behavior is realized, and the non-inferior solutions in the external files are added to the iterative population of the algorithm, thus speeding up the convergence of the algorithm. The proposed algorithm is proved to be convergent and distributed by an example, and various balancing schemes are obtained by solving the examples of different scale tasks. Compared with the results of the existing algorithms, the validity and superiority of the proposed algorithm are verified. The established model and the designed algorithm are applied to the actual disassembly line, and 16 balancing schemes are obtained in solving the engine bucket chain disassembly line balance problem with the same retrieval speed. In solving the balance problem of printer bucket disassembly line with different retrieval speed, 20 kinds of balancing schemes were obtained, which realized the balance of production operation, increased the effective working time, and provided a wide range of decision-making space for decision makers. Therefore, the model and the algorithm designed in this paper can solve the problem of disassembly line balance, effectively improve the line balance problem, and have a strong applicability and broad application prospects.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP278;TP18

【相似文献】