A Solution For PCB Punching Machine's Path Planning Problem Based On An Improved Genetic And Ant Colony Algorithm

As the prosperity of handheld electronic devices and Industrial automation,the Printed circuit board(PCB)has been widely applied.Drilling is one of the most important processes during the PCB manufacturing.Drilling refers to using different tools to drill the PCB on a numerically controlled machine tool and this process directly influences the cost and efficiency.According to search,the movement of the tools cost the most time of the whole manufacturing.So it's important to make a path planning for the tools as to improve the drilling efficiency.After referring to how the numerical numerically controlled machine works,the path planning problem for the tools turns out to be a typical Traveling Salesman Problem(TSP).On the basis of how the TSP being solved home and abroad,a core algorithm is proposed in this paper,an improved mixture algorithm of Genetic and Ant Colony Algorithm.The improved mixture algorithm mainly has improved the following 3 parts compared to the traditional mixture algorithm:(1)a dynamic integration strategy: at the beginning of the search,fully take advantage of the Genetic Algorithm's Global search ability to find out a better solution,then apply the dynamic integration strategy to switch to the Ant Colony Algorithm,as the positive feedback of the Ant Colony Algorithm helps to accelerates the search speed;(2)an improved Ant Colony Algorithm: as for the Ant Colony Algorithm's weak searching ability at the beginning and the algorithm is easy to fall into local optimal solution,an improved Ant Colony System Algorithm is proposed in this paper.The improved Ant Colony Algorithm has better strategy to update the pheromone to make sure that the pheromone better reflects the route information.This strengthens the global search ability of the Ant Colony Algorithm and improved the quality of the solution;(3)adaptive control of the parameters: the parameters of the Ant Colony Algorithm and the Genetic Algorithm have always been set by human according to the experiences,this results in the instability of the algorithm.Cloud association rules is introduced in this paper to adaptively control the parameters,making the comprehensive performance of the algorithm more stable.To verify the performance of the algorithm,this paper compares and analyses the performance of different algorithms.It turns out that the improved algorithm has a relatively better performance.At last,the algorithm is used in the real path planning of the PCBs,the results shows that the solution actually gains a shorter route,this algorithm is of help to increase the efficiency of the PCB manufacturing.