Stress Analysis in Double-Lap Adhesively Bonded FG Adherend Joints Using FEM and ANN

Articles in Press

Document Type : Original Research Paper

Authors

1 mechanical engineering department- university of kurdistan-pasdaran street- sanandaj- kurdistan-iran

2 Mechanical Engineering Department, Faculty of Engineering, University of Kurdistan

10.22084/jrstan.2025.31011.1269

Abstract

This study presents a comprehensive numerical investigation of stress distribution in double-lap adhesive joints between functionally graded material (FGM) adherends. Using three-dimensional finite element analysis (FEA) in ABAQUS, this study examined the effects of adhesive thickness, material gradation, and applied load on joint performance. To enhance computational efficiency and predictive capabilities, a feedforward artificial neural network (ANN) model was also developed and trained using simulation data. The results show that adhesive thickness has a significant influence on peak shear stress, with an optimal thickness of 0.2 mm minimizing stress concentrations under both 10 kN and 50 kN tensile loads. The inclusion of FGMs in the adherends improved stress distribution due to the gradual transition in material properties from ceramic to metal. The ANN model, trained on FEA outputs, achieved a high correlation (R² >0.99) and minimal error (MSE<0.001), validating its capability to provide rapid and accurate stress prediction. The proposed hybrid FEA-ANN framework provides a reliable and efficient tool for designing adhesive joints with graded materials. This approach can be extended to optimize joint configurations in aerospace, automotive, and structural applications where weight and stress management are critical.

Keywords

Journal of Stress Analysis

Articles in Press, Accepted Manuscript
Available Online from 18 November 2025

Files

Share

How to cite

Statistics