应用科学学报 ›› 2002, Vol. 20 ›› Issue (3): 309-312.

• 论文 • 上一篇    下一篇

用于铸铁力学性能快速预测的人工神经网络

夏善木, 李大勇   

  1. 哈尔滨理工大学材料学院, 黑龙江哈尔滨 150080
  • 收稿日期:2000-11-09 修回日期:2001-11-03 出版日期:2002-09-30 发布日期:2002-09-30
  • 作者简介:夏善木(1975-),男,湖北武汉人,助教,硕士;李大勇(1958-),男,黑龙江哈尔滨人,教授,博导.
  • 基金资助:
    黑龙江省自然科学基金资助项目(E9926)

The Network for Fast Prediction of Mechanical Properties of Cast Iron

XIA Shan-mu, LI Da-yong   

  1. College of Material, Harbin University of Science and Technology, Harbin 150080, China
  • Received:2000-11-09 Revised:2001-11-03 Online:2002-09-30 Published:2002-09-30

摘要: 对人工神经网络用于铸铁力学性能预测进行了系统的研究,并利用Matlab建立了一个基于热分析的预测网络,同时采用了多种有效的方法以提高网络的性能.测试结果表明网络具有良好的泛化能力,能准确快速地预测铸铁力学性能.

关键词: 热分析, 人工神经网络, 力学性能

Abstract: The artificial neural network for fast prediction of mechanical properties of cast iron has been studied. Matlab is used to create and train a predicting network. Meanwhile some efficient methods have been adopted to promote the performance of the network. The testing result shows that the network has a fine ability of generalization and can predict mechanical properties of cast iron precisely and quickly.

Key words: mechanical property, thermal analysis, ANNs

中图分类号: