A comparative study on phenomenon and deep belief network models for hot deformation behavior of an Al–Zn–Mg–Cu alloy

• Published in: Applied physics. A, Materials science & processing Vol. 123; no. 1; pp. 1 - 11
• Main Authors: Lin, Y. C., Liang, Ying-Jie, Chen, Ming-Song, Chen, Xiao-Min
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2017; Springer Nature B.V
• Subjects: Aluminum base alloys; Applied physics; Belief networks; Characterization and Evaluation of Materials; Comparative studies; Compression tests; Condensed Matter Physics; Constitutive models; Copper; Correlation coefficients; Deformation; Die forging; Heat resistant alloys; High temperature; Intelligent manufacturing systems; Machines; Magnesium; Manufacturing; Materials science; Nanotechnology; Optical and Electronic Materials; Physics; Physics and Astronomy; Processes; Strain rate; Surfaces and Interfaces; Thin Films; Zinc; True Stress; Restricted Boltzmann Machine; Hide Variable; Deformation Temperature; Hide Layer
• ISSN: 0947-8396; 1432-0630
• DOI: 10.1007/s00339-016-0683-6

The high temperature deformation behavior of an Al–Zn–Mg–Cu alloy is studied by isothermal compression tests at the temperature range of 573–723 K and strain rate range of 0.001–0.1 s −1 . Considering the coupled influences of deformation temperature, strain, and strain rate on hot deformation behavior, a deep belief network (DBN) model, as well as a phenomenological constitutive model, is developed for the studied alloy. In order to validate the developed models, the average absolute relative error and correlation coefficient are evaluated between the measured and predicted true stresses. The results show that the developed DBN model has the better predictability for the high temperature deformation behavior of the studied Al–Zn–Mg–Cu alloy. Moreover, the average absolute relative error and correlation coefficient of DBN model are 0.57% and 0.9997, respectively. In addition, the developed DBN model can be effectively applied in the intelligent manufacturing, such as intelligent isothermal die forging technology.