人工神经网络在Al_2O_3陶瓷激光铣削中的应用研究  

Application of artificial neural network in Al_2O_3 ceramics laser milling

作　　者：许兆美[1,2] 周建忠[1] 黄舒[1] 孙全平[3] 

机构地区：[1]江苏大学机械工程学院,江苏镇江221013 [2]淮阴工学院机械工程学院,江苏淮安223003 [3]淮阴工学院数字化制造技术重点实验室,江苏淮安223003

出　　处：《红外与激光工程》2013年第11期2957-2961,共5页Infrared and Laser Engineering

基　　金：国家自然科学基金(51075173);江苏省自然科学基金(BK2010288);江苏省苏北科技发展计划(BC2011437)

摘　　要：为了有效地控制Al2O3陶瓷激光铣削层质量,以人工神经网络(ANN)技术为基础,以MATLAB软件作为开发平台,建立了Al2O3陶瓷激光铣削层质量与铣削参数之间的关系模型。并以激光功率、扫描速度和离焦量作为输入参数,激光铣削层深度和宽度作为输出参数,对激光铣削层质量进行了预测。结果表明,该模型的平均误差小,拟合精度高。并在训练样本之外,选取了5组工艺参数来检验网络模型的可靠性,检验输出值和实验样本值的最大相对误差为7.06%。说明运用该模型可以方便、准确地选择激光工艺参数,提高Al2O3陶瓷激光铣削层的加工质量。In order to control the quality of Al2O3 ceramics, based on the artificial neural network （ANN）, a model was established to describe the relation between the laser milling quality of Al2O3 ceramics with the ceramics parameters. The milling quality of Al2O3 ceramics were predicted with the model in which the input parameters consisted of laser power, scanning speed and defocus amount and the output parameters included the milling depth and width. The results show that the mean error is small, and the model has good verifying precision and excellent ability of predicting. Five group process parameters were chosen to test the reliability of the neural network model out of the train samples. The maximum relative error of the output test value and the experiment sample value was 7.06%. The laser process parameters can be chosen easily and accurately to improve the processing quality of Al2O3 ceramics.

关 键 词：激光铣削 人工神经网络 陶瓷 工艺参数 

分 类 号：TN249[电子电信—物理电子学]