ISSN: 1003-6326
CN: 43-1239/TG
CODEN: TNMCEW

Vol. 11    No. 4    August 2001

[PDF]    [Flash]
Predicting effects of diffusion welding
parameters on welded joint properties
by artificial neural network
LIU Li-ming(刘黎明)1, ZHU Mei-li(祝美丽)1
NIU Ji-tai(牛济泰)2, ZHANG Zhong-dian(张忠典)2
(1. Department of Materials Engineering, Dalian University of
Technology, Dalian 116024, P.R.China;
2. National Key Laboratory of Advanced Welding Production Technology,
Harbin Institute of Technology, Harbin 150001, P.R.China
)
Abstract: The static model for metal matrix composites in diffusion welding was established by means of artificial neural network method. The model presents the relationship between weld joint properties and welding parameters such as welding temperature, welding pressure and welding time. Through simulating the diffusion welding process of SiCw/6061Al composite, the effects of welding parameters on the strength of welded joint was studied and optimal technical parameters was obtained. It is proved that this method has good fault-tolerant ability and versatility and can overcome the shortage of the general experiment. The established static model is in good agreement with the actual welding process, so it becomes a new path for studying the weldability of new material.
Key words: artificial neural network; diffusion welding; composite
Superintended by The China Association for Science and Technology (CAST)
Sponsored by The Nonferrous Metals Society of China (NFSOC)
Managed by Central South University (CSU) 湘ICP备09001153号-9