论文部分内容阅读
采用不同工艺参数研究了激光熔凝K465高温合金的开裂行为,并采用BP神经网络模型描述了裂纹指数与工艺参数的关系。研究发现,激光熔凝K465合金的裂纹主要分布在熔池顶部和底部的界面处,并呈现典型的液化开裂特征。通过建立激光熔凝区裂纹指数的数学描述方法,以激光功率、扫描速度、光斑直径和预热温度作为输入参数,以裂纹指数为输出参数,发展了一个均方误差小于10-8的BP神经网络模型,并可对实验结果进行较好的预测。
The cracking behavior of laser fusion K465 superalloy was studied by using different process parameters. The relationship between crack index and process parameters was described by BP neural network model. It is found that the cracks in K465 laser alloy are mainly distributed at the interface between the top and the bottom of the weld pool and show typical characteristics of liquefaction and cracking. By establishing the mathematical description of the crack index in the laser melting zone, taking the laser power, scanning speed, spot diameter and preheating temperature as the input parameters and the crack index as the output parameter, a BP neural network with a mean square error of less than 10-8 Network model, and the experimental results can be better predicted.