Research On Parameter Optimization And Holes Machining Path Planning With GA And TSP

In manufacturing in 21 Century to meet the need of agile manufacturing and green manufacturing, machining parameter optimization and path Optimization become highly important to make manufacturing realize high quality, high efficiency, low cost , to full use the costly automatic machining tools. But so many insufficiencies in choice of the independent variable, constraints, integrating optimizing mathematic model, and multi-goal intelligent optimizing method, especially, the lack of the holes machining path planning mathematic model, ask researchers to focus on the modeling and optimization of parameter and path.Genetic Algorithms is a predominant search method, it can fleetly seek the optimum point under complicated functions and restrictions conditions by probability select, and widely used in optimizing.Though the research on the choice of machining parameters and paths with GA, the embodiment of the contributions in the paper may be summarized as follows:Firstly, by researching the standard fundamental of Genetic Algorithms(GA), the multi-goal intelligent optimizing models and objective function for minimum production cost and maximum profit of turning and milling parameters were deducted and built up.Secondly, single-goal intelligent optimizing mathematic model for holes machining path planning of minimum production cost, multi-goal intelligent optimizing mathematic model of shortcut tools path and least inversions and holes machining path planning of injection mold were set up by Traveling Saleman Problem(TSP) mathematics models and path optimum models.Lastly, optimum cutting parameters of turning and milling and holes machining path models were ascertained by means of Genetic Algorithms. the study on the models of machining parameter and path includes: parameter sitting, program list, objective function, overall evaluation and three-dimensional diagram and tracking drawing, the veracity and practicability of the above models and algorithm were validated though the contrast between the fact and the optimum results.