A comparative study of phenomenological, physically-based and artificial neural network models to predict the Hot flow behavior of API 5CT-L80 steel. (December 2020)

• Record Type: Journal Article
• Authors: Ahmadi, H.; Rezaei Ashtiani, H.R.; Heidari, M.
• Is Part Of: Materials today communications. Volume 25(2020)
• Publication Date: 2020-12
• Subjects: 5CT-L80 steel -- Hot working -- Phenomen -- Ological model -- Physically-Based model -- Artificial neural network model; Materials science -- Periodicals; 620.11
• DOI: 10.1016/j.mtcomm.2020.101528
• Languages: English
• ISSNs: 2352-4928

Abstract

Highlights: Hot compression tests on API 5CT-L80 Steel alloy at elevated temperatures were done. The flow behavior is greatly affected by strain, strain rate and deformation temperature. The suitability of the three phenomenological, Physically-Based and ANN models was compared. Predictability of the developed models is evaluated in terms of R, AARE and RMSE. The ANN model gives a precise estimate for hot deformation behavior of API 5CT-L80 Steel alloy. Abstract: The hot compressive deformation behavior of as-cast 5CT-L80 medium-carbon steel was investigated under the different elevated temperatures range from 1173 K to 1373 K and diverse strain rate range from 0.001 s −1 to 1 s −1 . The multi-peak flow stress curves were observed during deformation, indicating a complex behavior of this steel due to metallurgical phenomena like dynamic recrystallization. Therefore, the modified Johnson-Cook (J-C) as a phenomenological model, modified Zerilli-Armstrong (Z-A) as a physically-based model, and feed-forward back propagation artificial neural network (BP-ANN) model were purposed to describe the high-temperature flow behavior of the studied material. So, the predictability of the developed models is compared by the standard statistical parameters. The extraordinary and accurate performance was observed in the BP-ANN model to predict the complexities of the compressive behavior of 5CT-L80 steel, whereas the modified J–C model has more precise than the modified Z-A model.