论文部分内容阅读
微孔表面质量是激光微孔加工的核心问题之一。本文采用粗糙度对孔的表面质量进行表征,将激光打孔表面粗糙度与其各项加工参数匹配起来,建立起相应的预测模型。首先对 304 不锈钢试样进行激光打孔试验,然后使用形貌仪得到了孔截面粗糙度参数表,并通过 BP 神经网络,建立了激光功率、脉冲频率和离焦量三个工艺参数与孔表面粗糙度之间的神经网络预测模型。通过大量试验数据对样本进行网络训练,并采用试验数据验证该预测模型,最终得到了预测效果精度较好的神经网络模型,并使误差控制在 6.00%左右,且最大误差不超过 8.08%。通过该模型,可以准确的预测激光打孔孔表面的粗糙度大小,从而有效的缩短了激光打孔作业的准备周期。
Micropore surface quality is one of the core issues in laser micropore processing. In this paper, the surface quality of the hole is characterized by roughness, the laser drilling surface roughness and its processing parameters match, to establish the corresponding prediction model. First of all, the stainless steel samples were drilled by laser, and then the roughness parameters of the holes were obtained by using a profilometer. The three parameters of laser power, pulse frequency and defocus were established by BP neural network. Neural network prediction model between roughness. A large number of experimental data were used to train the samples. The experimental data were used to verify the model. Finally, a neural network model with good prediction accuracy was obtained. The error was controlled at 6.00% and the maximum error was less than 8.08%. The model can accurately predict the roughness of the surface of the laser drilling hole, which effectively shortens the preparation period of the laser drilling work.