Abdulrahman Abbadi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Abdulrahman Abbadi has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Ali Qabur
Jazan University
3 shared publications - Khaled Aati
Department of Civil and Architectural Engineering, Jazan University, Jazan 45142, Saudi Arabia
2 shared publications Abdullah Alshahrani
1 shared publicationAdeyemi Adesina
1 shared publication- Ahmed Ahmed Abdou Elabbasy
Civil and Architectural Engineering Department, College of Engineering and Computer Sciences, Jazan University, P.O. Box 706, Jazan 45142, Saudi Arabia
1 shared publication Ali Qabur
1 shared publication
Latest Publications
- Open Access
- 22 September 2026
- Published by Elsevier in Materials Today Communications
- Vol. 56
Role of SCM type and dosage on the performance, microstructure and environmental impact of concrete mixtures
- 22 September 2026
- Published by Elsevier in Materials Today Communications
- Vol. 56
Open AccessRole of SCM type and dosage on the performance, microstructure and environmental impact of concrete mixtures
- 2 August 2026
- Published by Springer Nature
- p. 206-213
Low-Carbon GGBS-Fly Ash Concrete: Fresh and Mechanical Performance
- 2 August 2026
- Published by Springer Nature
- p. 206-213
Low-Carbon GGBS-Fly Ash Concrete: Fresh and Mechanical Performance
- 15 October 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Applied Mechanics
- Vol. 17 (12)
Physics-Aware Mesh-Free Deep Learning for Rectangular Elastic Plate Deformation with Support-Dependent Boundary Enforcement
This study introduces a physics-aware, mesh-free deep learning framework for modeling the bending behavior of elastic plates under complex and arbitrary support conditions. The proposed method serves as a data-free surrogate model that generalizes across a wide range of boundary configurations, addressing key limitations of conventional numerical techniques and earlier PINN-based approaches. The formulation is based on minimizing the total potential of the system, which incorporates internal strain energy, virtual external work and support-specific boundary penalty terms. Distinct constraints on each plate edge are enforced through edge-dependent penalties, enabling automatic adaptation to changing boundary conditions without remeshing or problem reformulation. The neural network architecture consists of three decoupled subnetworks to approximate transverse and in-plane displacements. Automatic differentiation is employed to compute curvature and slope terms essential for higher-order plate theory. The training process combines structured and randomized sampling across both domain interiors and boundaries to ensure accuracy and stability. Benchmark comparisons against finite element method (FEM) results demonstrate that the proposed model achieves mean relative errors below 3% across diverse support configurations. The framework captures both out-of-plane and in-plane displacement fields with high fidelity and robust convergence. These results highlight the method’s potential as a flexible, generalizable solution for elastic plate deformation analysis in engineering applications involving complex geometries and support conditions.
- 15 October 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Applied Mechanics
- Vol. 17 (12)
Physics-Aware Mesh-Free Deep Learning for Rectangular Elastic Plate Deformation with Support-Dependent Boundary Enforcement
This study introduces a physics-aware, mesh-free deep learning framework for modeling the bending behavior of elastic plates under complex and arbitrary support conditions. The proposed method serves as a data-free surrogate model that generalizes across a wide range of boundary configurations, addressing key limitations of conventional numerical techniques and earlier PINN-based approaches. The formulation is based on minimizing the total potential of the system, which incorporates internal strain energy, virtual external work and support-specific boundary penalty terms. Distinct constraints on each plate edge are enforced through edge-dependent penalties, enabling automatic adaptation to changing boundary conditions without remeshing or problem reformulation. The neural network architecture consists of three decoupled subnetworks to approximate transverse and in-plane displacements. Automatic differentiation is employed to compute curvature and slope terms essential for higher-order plate theory. The training process combines structured and randomized sampling across both domain interiors and boundaries to ensure accuracy and stability. Benchmark comparisons against finite element method (FEM) results demonstrate that the proposed model achieves mean relative errors below 3% across diverse support configurations. The framework captures both out-of-plane and in-plane displacement fields with high fidelity and robust convergence. These results highlight the method’s potential as a flexible, generalizable solution for elastic plate deformation analysis in engineering applications involving complex geometries and support conditions.
- 15 July 2025
- Published by Elsevier in Construction and Building Materials
- Vol. 490
Ultra-fine organophilic nano clay in asphalt: A laboratory-based performance study
- 15 July 2025
- Published by Elsevier in Construction and Building Materials
- Vol. 490
Ultra-fine organophilic nano clay in asphalt: A laboratory-based performance study
- Open Access
- 30 March 2022
- Published by Elsevier in Materials Today: Proceedings
- Vol. 65, 681-687
Effect of hybrid fibres on the static load performance of concrete beams
- 30 March 2022
- Published by Elsevier in Materials Today: Proceedings
- Vol. 65, 681-687
Open AccessEffect of hybrid fibres on the static load performance of concrete beams