Group Holes Machining Process Optimization Based On Differential Algorithm

This paper mainly studies the computer aided process planning of group holesmachining using the differential algorithm. Differential algorithm, which is based onpopulation difference, is a kind of heuristic search algorithm with fast convergencespeed, powerful global search abilities and few control parameters. Determining theholes machining process route is an important but tedious job, however, usingdifferential algorithm to optimize of holes machining processing technology caneffectively reduce workload and processing cost.Firstly, in this paper, the product process specification and characteristics forgroup holes machining are mainly discussed. The most important research work iscorrectly characterizing the part and the constraint, and arranging the processing route,so that the computer can be used to aid the process planning. Research purpose is tolower manufacturing cost, to improve the processing automation level. To facilitatethe research work of hole machining process optimization, machining center is chosenas the holes machining equipment, for it can realize various processing method andchange the tools automatically.Secondly, analyze the fundamental principles of differential algorithm (DE) andgenetic algorithm (GA), and it indicates that the difference between the twoalgorithms lies in the individual mutation ways, for DE applies differential strategy tomutate individuals. The different differential strategies for DE are expounded as well.The arrangement of holes machining processing route is similar to the travelingsalesman problem (TSP), so trying to solve a simple and intuitive TSP firstly with DEand GA can help to understand the features of the two algorithms. Comparing withGA, the characteristics and performance of DE will be obtained. In the light of weakerlocal search ability of DE,2-opt operator is added into DE in this paper in order tomodify the differential algorithm.Then establish a mathematical model for the holes machining process accordingto the process specification and characteristics. There are two objective functions, oneis minimizing the number of changing tools, and the other is minimizing the tool displacement. The fitness function is determined by using the linear weighting methodto change the multiple objective function optimization problems to single objectivefunction optimization problem. After repeatly application of the modified differentialalgorithm to optimize the holes machining process, results show that the solution fromDE is more stable and reliable than that from GA, and both of them can meet therequirements of actual production.At last, introduce the variance analysis of the parameters involved in themodified differential algorithm. According to the variance analysis results of zoomfactor F, it manifests that there is no significant difference among the average fitnessvalues in different levels of F. Then analyze the variance of weighting factors α and β.Multiple comparison is needed to acquire the significant difference performanceamong different levels of weighting factors α and β. Finally try to select a properparameter value and modify the fitness function according to the guidance of thesignificant difference performance.