期刊
JOURNAL OF MATERIALS PROCESSING TECHNOLOGY
卷 196, 期 1-3, 页码 303-310出版社
ELSEVIER SCIENCE SA
DOI: 10.1016/j.jmatprotec.2007.05.057
关键词
optimisation; laser drilling; neural network; genetic algorithm
Nowadays, laser percussion drilling is finding increasingly widespread application in industry. Precise modeling has not yet been achieved due to the complexity of this process. The neural network has been used in this study for process modeling. Approximate experimental models of the process have been developed by the neural network (Generalized Regression Neural Network-GRNN) according to the results of the experiments. Then the optimum input parameters (peak power, pulse time, pulse frequency, number of pulses, gas pressure and focal plane position) were specified using the genetic algorithm (GA) method, the results of which are optimum output parameters. The output parameters include the hole entrance diameter, circularity of entrance and exit holes, hole exit diameter and taper angle of the hole. The tests were carried out on stainless steel 304 sheets with a thickness of 2.5 mm. A Nd:YAG laser machine was employed with a wavelength of 1.06 wm. Oxygen was used as an assist gas. Diameter of the central nucleus of laser beam was 600 wm. Considering the precision of the optimum numerical results and the high speed of the neural network in modeling, this method is reliable and economical and also confirms the qualitative results of the previous studies. Therefore, one can use this method to optimally adjust input parameters of the process in multipurpose and single purpose optimisation modes, which indicates substitute application of the method for optimising the laser percussion drilling process. (c) 2007 Published by Elsevier B.V.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据