Physics-informed hybrid learning for bending analysis of multidirectional porous FG shells

Physics-informed: ML and GWO optimization for damping design of laminated structures

The design of laminated structures with enhanced damping capacity has become a major challenge in advanced engineering applications, where lightweight construction, vibration mitigation, and structural durability must be achieved simultaneously. In this work, a robust hybrid framework is proposed to predict and optimiz...

Design Optimization of Auxetic Composite Metamaterials Using Finite Element Simulations and Random Forest AI Models

The development of biomaterials with tunable mechanical properties is important for addressing stiffness mismatch between implants and bone, although the present study focuses on computational material design rather than implant‐level performance. In this work, a novel composite design is proposed in which a 3D re‐entr...

Tunability of Mechanical Behaviors in Multi-Morphology Functionally Graded Porous Beams with TPMS Architectures

Physics-Informed Neural Networks for Biot's Model via Fixed-Stress Splitting and Energy Natural Gradient Descent

Physics-Informed Neural Networks (PINNs) have recently gained considerable attention as a mesh-free framework for solving partial differential equations. Nevertheless, their performance deteriorates when applied to strongly coupled multiphysics systems, such as Biot's consolidation model, due to severely ill-conditione...

Hybrid Galerkin-machine learning framework for fluid-structure interaction and stability of immersed moving FG-CNTRC microplates under multiphysics effects

This study aims to develop a hybrid Galerkin-machine learning (ML) framework for the rapid and accurate prediction of the vibration characteristics and stability boundaries of bidirectionally moving functionally graded carbon nanotube-reinforced composite (FG-CNTRC) microplates immersed in a confined fluid under co...

Plasolver: Physics-Informed Neural Operators for Elastoplasticity

Plasolver, a physics-informed neural operator framework that combines the efficiency of operator learning with the accuracy and robustness of classical numerical solvers, provides an efficient, accurate, and discretization-invariant computational framework for nonlinear, path-dependent elastoplastic problems.