Temperature Prediction of Friction Stir Welding Based on Bayesian Neural Network

Abstract:

This paper established a neural network based on Bayesian regularization, which was used for temperature prediction of feature points in friction stir welding metal. Sample data included rotation speed, welding speed, distance, depth, time and temperature of feature points. The representative sample data was obtained through the orthogonal experiments. Neural network was improved by Bayesian algorithm, and the constraint term representing complexity of network was introduced into the objective function. Then generalization ability of neural networks has been enhanced. Prediction accuracy, training time and stability of network are better than in the traditional neural network. Test proved that this smaller network model operates faster, and it can accurately predict the temperature of feature points of friction stir welding.

• Periodical: Applied Mechanics and Materials (Volumes 48-49)
• Pages: 1208-1212
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.48-49.1208
• Online since: February 2011
• Authors: Shu Jin Chen, Zhong Ke Zhang
• Keywords: Bayesian Regularization, Friction Stir Welding (FSW), Neural Network (NN), Prediction of Temperature

References

[1] Dong C H. Matlab Neural Net & Application [M]. Beijing: National Defense Indus try Press, (2005).

[2] Zong Lijiang. Introduction to Artificial Neural Network [M], Beijing: Higher Education Press, (2001).

[3] Fratini, Livan and Buffa. Metallurgical phenomena modeling in friction stir welding of aluminum alloys: Analytical versus neural network based approaches [J]. Journal of Engineering Materials and Technology, 130 (3): 310011-310016.

DOI: 10.1115/1.2931142

[4] Wang Qing, Na Yue, Sun Dongli, etal. Prediction of tensile property of TIG welding joints in GH99 alloy by artificial neural network [J], Transactions of the China Welding Institution. 2010, 31(3): 77-80.

[5] Wang Min, Dong Zhibo, Wei Yanhong. etal. Mechanical property prediction of welding joint based on artificial neural network for titanium alloy [J], Aeronautical Manufacturing Technology. 2009, 16: 79-81.

[6] Buffa G, Fratini L, Micari F. A neural network based approach for the design of FSW processes [J]. Key Engineering Materials, 2009, 410(7): 413-420.

DOI: 10.4028/www.scientific.net/kem.410-411.413

[7] Dong Junhui, Zhang Yanfei, Tang Zhengkui. Mechanical property prediction of welding joint based on fuzzy-artificial neural network [J]. Transactions of the China Welding Institution, 2008, 29(7): 29-33.

[8] Jiang Zongli. Introduction to artificial neural network [M], Beijing: Higher Education Press, (2001).