LSTM-based neural network architecture for predicting the nonlinear dynamic behavior of functional gradient viscoelastic porous plates

Journal: Materials Today Communications · Elsevier · Quartile: Q1

Authors: M. JANANE ALLAH, et al.

Abstract

This work develops a Long Short-Term Memory (LSTM) recurrent neural network architecture for predicting the nonlinear dynamic behavior of viscoelastic functionally graded porous plates. The approach combines the Third-Order Shear Deformation Theory (TSDT) with deep learning, offering a computationally efficient alternative to classical numerical methods. Results demonstrate remarkable prediction accuracy with significant computational time reduction.

Keywords: LSTM · FGM · Viscoelasticity · Nonlinear dynamics · Deep Learning · TSDT

Google Scholar Profile · ORCID

Recommended citation: JANANE ALLAH, M., et al. (2024). "LSTM-based neural network architecture for predicting the nonlinear dynamic behavior of functional gradient viscoelastic porous plates." Materials Today Communications. Elsevier. Q1.
Download Paper