Page 14 - IJAMD-2-3

International Journal of AI for
Materials and Design AI applications in composite materials

A

B D

C
E

F

Figure 5. RNN, LSTM, and GRU architectures and research utilizing RNN in the field of composite materials. (A) Structure of RNN, LSTM, and GRU;
(B) Comparison of actual and predicted results of water absorption; (C) Comparison between the actual and predicted stress values; (D) Network
66
65
architecture and the transfer learning approach utilizing a large mean-field and a small full-field data set; (E) Overview of data-driven multiscale
67
framework using DRCN networks; (F) Architecture of PRNN for finite strain framework. 69
68
Abbreviations: BPNN: Backpropagation neural network; CHF: Corn husk fiber; DRCN: Decomposing residual convolutional neural; GRU: Gated
recurrent unit; LSTM: Long short-term memory; PP: Polypropylene; PRNN: Physically recurrent neural network; RF: Random forest; RNN: Recurrent
neural network; RVE: Representative volume element.

Volume 2 Issue 3 (2025) 8 doi: 10.36922/IJAMD025210016