[1]许贵芝.预测管线焊缝抗裂性指标的神经网络法[J].焊管,2012,35(7):68-71.[doi:1001-3938(2012)07-0068-04]
 Edited and Translated by XU Gui-zhi.Neural Network Method to Predict Cracking Resistance Index of Line Pipe Weld[J].,2012,35(7):68-71.[doi:1001-3938(2012)07-0068-04]
点击复制

预测管线焊缝抗裂性指标的神经网络法
分享到:

《焊管》[ISSN:1001-3938/CN:61-1160/TE]

卷:
35
期数:
2012年第7期
页码:
68-71
栏目:
国外焊管
出版日期:
2012-07-28

文章信息/Info

Title:
Neural Network Method to Predict Cracking Resistance Index of Line Pipe Weld
文章编号:
1001-3938(2012)07-0068-04
作者:
许贵芝
(南京航空附件厂,南京 210002)
Author(s):
Edited and Translated by XU Gui-zhi
(Nanjing Aviation Accessories, Nanjing 210002, China)
关键词:
关键词:管线钢管焊缝抗裂性指标神经网络变性处理
Keywords:
weld of line pipe cracking resistance index neural network denaturalization treatment
分类号:
TE973
DOI:
1001-3938(2012)07-0068-04
文献标志码:
B
摘要:
介绍了一种神经网络法的计算机信息工艺,它能在不同温度和不同变性剂含量的条件下,准确地预测出焊缝的若干抗裂性指标。通过神经网络法预测值与实际试验值的对比发现,变性剂中w(Ce)在0.025%~0.040%范围时,所得到的抗裂性指标处于最佳状态;当w(Ce)高于0.040%后,指标值开始下降。采用变性剂Ce进行焊缝金属的合金化,能够细化焊缝组织,净化晶粒边界,从而增强焊缝的抗裂性能。
Abstract:
In this article, it introduced a kind of computer information technology by adopting neural network method, which can accurately predict some cracking resistance indexes of weld on condition of different temperature and different content of denaturant. After comparing the predicted value from neural network method and actual experimental value, the results indicated that the cracking resistance value is the best when w(Ce) is in the range of 0.025%~0.040%;The cracking resistance value starts to reduce when w(Ce) is higher than 0.040%. Adopting denaturant Ce to conduct alloying for weld metal, which can refine weld structure, purify grain boundary, thus the weld cracking resistance will be strengthened.

参考文献/References:

译自:В. Д. МАКАРЕНКО . Нейросетевой метод анадиза показателей трещиностойкости сварных соединений трубных сталей . Сварочное производство. 2010(02):7-12 .
—————————

备注/Memo

备注/Memo:
译者简介:许贵芝,男,高级工程师,南京市焊接学会前委员,长期从事俄文的翻译工作。
更新日期/Last Update: