Based on BP neural network , network model used for predicting laser fuse cladding appearance and size has been established, further correspondence between characteristic signals of laser fuse cladding (ultraviolet emission, infrared emission and audible sound ) and laser fuse cladding appearance (fuse cladding height&width) has been re-searched.The result showed that the network model prediction was provided with lower mean error and higher test pre -cision and possessed better prediction capability .%基于BP神经网络,建立网络模型对激光熔覆层形貌尺寸进行预测,研究激光熔覆特征信号(蓝紫光信号、红外辐射信号、可听声信号)和激光熔覆形貌(熔覆层高、宽)之间的对应关系。结果表明,该网络模型预测平均误差小,检验精度高,具有较好的预测能力。
展开▼