A novel EFG meshless-ANN approach for static analysis of FGM plates based on the higher-order theory

• Published in: Mechanics of advanced materials and structures Vol. 31; no. 25; pp. 6501 - 6517
• Main Authors: K. P, Afsal, Swaminathan, K., Indu, N., H, Sachin
• Format: Journal Article
• Language: English
• Published: Abingdon Taylor & Francis 04-11-2024; Taylor & Francis Ltd
• Subjects: artificial neural network; Artificial neural networks; Computing time; Deformation analysis; Finite element method; functionally graded plates; Functionally gradient materials; meshless method; Meshless methods; moving least square approximation; Plates; Shape functions; Shear deformation; Static analysis
• ISSN: 1537-6494; 1537-6532
• DOI: 10.1080/15376494.2023.2231459

An Element Free Galerkin (EFG) meshless formulation and solutions using higher order shear deformation theory with nine degrees of freedom for the static analysis of Functionally Graded Material (FGM) plates are provided. This technique estimates the shape function using Moving Least Squares (MLS) method. The proposed method is validated by comparing the present findings with those in the literature. A novel Artificial Neural Network (ANN) model is developed to forecast the deflection of FGM plates within less computational time. Detailed parametric and convergent studies reveal that the proposed EFG solution and the ANN technique are more efficient than their conventional counterparts. The validation and comparison of the generated results in the present investigation with the other analysis methods revealed that the EFG method and ANN model give more accurate results than the FEM and other meshless methods. The current EFG-ANN model reduces computing time by 99.94% when compared to the EFG approach. Also, the accuracy is enhanced using the EFG approach with HSDT9 for the FGM plate.