论文部分内容阅读
在铝合金板材的激光弯曲成形中,工艺参数(激光功率、扫描速度和扫描次数)的不同组合使板材的弯曲角度也不同,而且从试验数据中很难得出激光工艺参数与弯曲角度之间的规律。本文用BP神经网络技术对铝合金板材的激光弯曲成形结果和工艺参数进行了预测。结果表明,预测值与真实值之间的误差都稳定在10%以内。精度比较高,说明运用BP神经网络对激光弯曲角度进行预测是可行的。
Different combinations of process parameters (laser power, scanning speed and number of scans) make the bending angle of the sheet different in the laser bending of the aluminum alloy sheet, and it is difficult to find the difference between the laser process parameters and the bending angle from the experimental data law. In this paper, BP neural network technology for aluminum alloy sheet laser bending forming results and process parameters were predicted. The results show that the error between the predicted value and the true value is stable within 10%. The accuracy is relatively high, which shows that using BP neural network to predict laser bending angle is feasible.