PSI - Issue 71

A Shivnag Sharma et al. / Procedia Structural Integrity 71 (2025) 469–476

475

(a)

(b)

Fig. 4: Original and predicted values of the slip system resistance (F P

22 ) with MPE (mean pointwise error) values. (a) at time increment, t=511 (b)

at time increment, t=560.

also highlights the difference in error magnitudes of the variables analysed, the comparisons were done for the 511 th and the 560 th increment. Literatures suggest with the increase in number of predicted frames, the MPE error starts propagating and this leads to limiting the no. of predicted increments up to a certain extent. In the present study, for S 22 and F P 22 minimal or no error propagation was observed whereas for g α the error propagation is of the order 10 -4 , this shows the prediction capability of the ConvLSTM for the evolution of CP variables. Now, the machine learning predicted time increments can be replaced with those obtained from the complex CP algorithm thereby reducing the computational cost. The future work of this study includes finding the optimum number of the machine learning predicted time increments to replace with the CP calculated time increments thereby eliminating the complex calculations of the CP algorithm and significantly reducing the computational time. References Ahmad, O., Kumar, N., Mukherjee R., Bhowmick, S., 2023. Accelerating microstructure modelling via machine learning: a new method combining Autoencoder and ConvLSTM. Physical review materials 7, 083802. Ali, U., Muhammad, W., Brahme, A., Skiba, O., Inal, K., 2019. Application of artificial neural networks in micromechanics for polycrystalline metals. Int. J. Plast. 120, 205–219. Asaro, R. J., Needleman, A., 1985. Texture development and strain hardening in rate dependent polycrystals. Acta Metallurgica 33, 923–953. Balasubramanian, S., 1998. Polycrystalline Plasticity: Application to Deformation Processing of Lightweight Metals. PhD thesis, Massachusetts Institute of Technology (MIT). Brahme, A., Winning, M., Raabe, D., 2009. Prediction of cold rolling texture of steels using an artificial neural network. Comput. Mater. Sci. 46 (4), 800–804. Evers, L., Brekelmans, W., Geers, M., 2004. Scale dependent crystal plasticity framework with dislocation density and grain boundary effects. Int. J. Solids Struct. 41 (18–19), 5209–5230. Ghaboussi, J., Garrett Jr, J., Wu, X., 1991. Knowledge-based modeling of material behavior with neural networks. J. Eng. Mech. 117 (1), 132–153.