摘要
在介绍人工神经网络模型特点的基础上,结合堆焊过程的复杂性、不确定性和随机性。
The characters of artificial neural network model have been introduced. The principle of three layers perception model and back propagation learning algorithm is given, and its application to predicting chemical composition of submerged arc tubular wires cladding is discussed.
出处
《武汉水利电力大学学报》
CSCD
1999年第6期78-81,共4页
Engineering Journal of Wuhan University
基金
武汉水利电力大学(宜昌)校青年教师科研基金
关键词
人工神经网络
堆焊
药芯焊丝
BP算法
artificial neural networks
cladding
flux cored wire
BP algorithm