【作者】 王洋；

【导师】 刘树安；

【作者基本信息】 东北大学 ， 系统工程， 2015， 硕士

【摘要】 直线滚动导轨(简称线轨)具有摩擦力小、运动平稳、精度高、在预紧条件下工作刚性好等特点,广泛应用于数控机床、精密机械、电子加工装置等方面。目前,直线滚动导轨已系列化、规格化和模块化,而制造商生产的原材料线轨的长度固定的,根据客户多种多样的产品需求,合理安排切割计划,对充分利用原材料线轨,提高企业经济效益,具有重要意义。线轨切割问题属于一维下料问题。本文通过查阅大量国内外相关资料,对一维下料问题、可变尺寸装箱问题的相关理论及研究进行了综述,详细分析了线轨的结构、特点以及线轨切割的主要性能指标、工艺流程和工艺要求,在此基础上,综合考虑线轨的孔间距、端距、切割损耗、切割方向、余料的利用等因素,分别以使用原材料线轨根数最少和产生废料最少为目标,建立了线轨切割的多目标数学规划模型。线轨切割问题属于组合优化问题,具有可变尺寸装箱问题的特性,但比一般的装箱问题更复杂。针对模型求解的复杂性,本文分别设计了问题求解的余料优先利用启发式算法和遗传模拟退火算法。余料优先利用启发式算法从线轨孔距、端距和自身长度的特点出发,以优利用较短余料为贪婪准则,以期获得使用较少的原材料线轨和较高的原材料线轨利用率。遗传模拟退火算法以遗传算法为主,结合线轨切割的特性,设计了以产品排列为主同时考虑切割方向的染色体编码,以产品为单位的顺序交叉,以及针对产品的插入变异和针对切割方向的位变异,在适值函数中综合考虑使用原材料线轨最少和产生废料最少来完成遗传迭代。为克服遗传算法因编码冗余问题而导致的早熟现象,在遗传进化过程中引入模拟退火算法改善进化过程,在遗传算法生成的子代中选取e适值较小的染色体作为模拟退火的初始解并进行模拟退火操作,从而获得较好的子代,以此改善算法的收敛性,以期获得更好的线轨切割方案。最后,基于Excel,利用VBA技术,开发了具有较强实用性的仿真程序,仿真结果验证了两种算法的可行性、有效性与实用性。对比分析结果表明遗传模拟退火算法优于启发式算法。更多还原

【Abstract】 Linear sliding guidway(linear guidway)has advantage of least resistance,smooth motion and high precision,so it widely used in CNC machine,precision machinery and electronic processing device,etc.Currently,the linear guideway has been serialized,standardized.For the various needs of customers and the fixed length of raw materials,a reasonable arrangement for cutting plan of raw materials should be made.This arrangement can improve economic efficiency and be significant.Linear guideway cutting issues belong to one-dimensional cutting problem.By consulting a large number of domestic and foreign relevant information on the one-dimensional cutting stock problem,the problem of variable size packing theory and research were reviewed.Considering the hole spacing,loss of cutting,the structure,characteristics and key performance indicators of the linear guideway cutting processes,the process requirements need to analyse.The goal is using least raw materials and producing minimal waste and establish a multi-objective mathematical program of linear guideway cutting.Linear guideway cutting issue is combinatorial optimization problem with variable size packing characteristics,and it is more complex than the general packing.The solutions for the complexity of this paper are designed more expected first heuristic algorithm and genetic simulated annealing algorithm.More than expected first heuristic algorithm pitchs from the end of the length and its own characteristics,taking the excellent use of shorter expected as greedy guidelines.This can get less use of raw material and higher raw material.Genetic simulated annealing algorithm,combined with the characteristics of linear guideway cutting,designed mainly to product arrangements which considering cutting direction of chromosome coding,product unit sequence crossover,and mutation insert for the product.Considering the use of raw materials and the least waste of the allelic variation in fitness minimum function,genetic iteration should be made.To overcome premature for the genetic algorithm coding redundancy problems which caused by the introduction of simulated annealing algorithm,the process of evolution in the evolutionary process and select the appropriate value of smaller echromosome in the genetic algorithm to generate progeny should be improved.Finally,based on Excel,use VBA technology,develop simulation program which has a strong utility,and the simulation results show the feasibility of the two algorithms,effectiveness and practicality.Comparative analysis showed that the genetic simulated annealing algorithm outperforms the heuristic algorithm.更多还原