Research On Path Optimization Problem In Cutting Machine Based On Improved Intelligent Optimization Algorithms

Cutting machine is a kind of numerical control system for cutting material segments such as leather,mechanical parts and clothing,which can greatly improve cutting efficiency and raw material utilization ratio compared with traditional manual cutting.In the process of cutting,the total path of cutting tool is composed of cutting path and non-cutting path between different segments.The length of the former is the sum of the contour length of each segment,which is fixed;while the length of the latter can be changed by cutting sequence of the tool passing through and the starting point of each segment.The path optimization problem for cutting machine is to optimize non-cutting path,by reducing the length of non-cutting path can effectively reduce the production costs,improve the production efficiency,and then improve the market competitiveness of enterprises.Therefore,the research of path optimization problem for cutting machine has important theoretical significance and application value.In general,intelligent optimization algorithm is used to solve the path optimization problem for cutting machine.Traditional intelligent optimization algorithms often have some problems such as easy to fall into local optimization and low accuracy.Although the intelligent optimization hybrid algorithm can improve the solution quality,the improvement of accuracy is not obvious enough and takes a long time.Therefore,from the perspective of global optimization and running time,this dissertation studies the non-cutting path optimization of holes and clothing for cutting machine.The main work of this dissertation is as follows:(1)Aiming at the path optimization problem for holes machining,an improved genetic algorithm with simulated annealing(IGA-SA)algorithm is proposed to avoid the traditional genetic algorithm falling into local optimization prematurely.Firstly,the path problem of holes machining is transformed into a traveling salesman problems.Secondly,a nearest neighbor algorithm is used to preprocess holes to obtain the initial cutting sequence.Then,in order to optimize the cutting sequence,the selection mechanism and variation strategy of traditional genetic algorithm are improved.The elite retention strategy and roulette selection method are introduced into the selection mechanism to accelerate the search of the optimal solution,and a linearly decreasing mutation rate is designed in the mutation strategy.Furthermore,simulated annealing algorithm is introduced to deal with the fitness function to adjust the difference of population evolution and then speed up the process of optimization.Finally,through application examples and comparative experiments,the optimization rate of the proposed algorithm is 6.7%,which is better than the contrast algorithm in most examples,the experimental results verify the feasibility of the proposed algorithm.(2)Aiming at the path optimization problem for clothing cutting,a modified variable neighborhood(MVNS)search based metaheuristic algorithm is proposed to improve the accuracy and running time of traditional variable neighborhood search.Firstly,we transfer the planar multi-contour path problem into a generalized traveling salesman problem.Secondly,to solve the cutting sequence,we modify the local search and shaking procedure in traditional variable neighborhood search.A 2-opt with insertion operator of neighborhood structure and an incremental calculation method are proposed for local search,which improve the solution quality and search efficiency.Combining genetic algorithm,some operators such as partition and reorganization are designed for shaking procedure,which avoid to prematurely fall into local optimum.Furthermore,a tabu search with dynamic programming algorithm is used to eliminate duplicate cutting sequences and determine the starting point of each segment.Finally,through application examples and comparative experiments,the solution accuracy and running time of the proposed algorithm perform well.The optimization rate is over 51%,the running time is within 1 minute,and the error is within 4%.The results demonstrate the effectiveness of the proposed algorithm.(3)A path optimization software system of cutting machine is designed and realized.Based on the three "golden rules" in UI design optimization,IGA-SA algorithm and MVNS algorithm are integrated into the software system,and functional modules such as user login,file import,and file processing are added.Finally,the system function is verified by using the cutting sheet layout file obtained by the enterprise,and the path optimization result is converted into the cutting instruction file that can be recognized by the cutting machine CAD/CAM software system.Finally,this dissertation is summarized,the existing problems and the direction of research are presented.