基于神经网络的不锈钢板翅结构钎焊残余应力预测  

Predicting model for residual stress in brazed stainless steel plate-fin structure based on neural network

作　　者：赵亚琴[1] 

机构地区：[1]南京林业大学机械电子工程学院,江苏南京210037

出　　处：《焊接技术》2009年第11期14-16,共3页Welding Technology

摘　　要：由于影响不锈钢板翅结构在真空钎焊过程中产生的残余应力因素较多,这给残余应力的预测带来很大的困难,为此,本文提出一种基于神经网络的不锈钢板翅结构钎焊残余应力预测方法。通过对不锈钢板翅结构在真空钎焊过程中产生的残余应力进行有限元分析,获知影响残余应力的主要因素,以构造神经网络的输入向量,建立不锈钢板翅结构钎焊残余应力的预测模型,并采用该模型对不锈钢板翅结构钎焊工艺过程进行仿真,结果表明该预测方法较快速有效地预测不锈钢板翅结构钎焊残余应力。It was quite difficult to predict brazed residual stress in stainless steel plate-fin structure because of many factors considered.This paper presented a brazed residual stress predicting method based on neural network for stainless steel plate-fin structure.Main factors of bringing forth residual stress were obtained to construct input vectors for neural network by finite element analysis.Afterwards,a predicting model was established by neural network.To evaluate the predicting performance of the model,the brazed technologic course of stainless steel plate-fin structure were simulated by using the model.The results showed that the method could predict effectively and efficiently brazed residual stress in stainless steel plate-fin structure.

关 键 词：不锈钢板翅结构 钎焊 残余应力 神经网络 预测 

分 类 号：TG404[金属学及工艺—焊接] TP183[自动化与计算机技术—控制理论与控制工程]