Predictive modeling of hybrid-layer friction drill joining of cast aluminum via integrated finite element and deep learning techniques

Abstract

Joining cast aluminum alloys remain challenging due to their inherent brittleness and susceptibility to cracking under thermo-mechanical loading. This limitation is particularly pronounced in A380, which exhibits severe petal formation and unstable bushing development during friction drilling. To investigate potential improvements, this study examines the friction drill joining (FDJ) of A380 using three distinct upper-sheet materials—Cu, Al6061, and AISI304—focusing on their influence on heat distribution, material flow, and bushing formation. A combined numerical and data-driven approach was adopted. Friction drilling tests under a single calibrated process condition were performed to ensure the appropriateness of input parameters for the ABAQUS simulations, which evaluated temperature evolution, von Mises stress, and bushing formation within the joint. Numerical results indicate that the presence of an upper sheet modifies heat generation and stress distribution, promoting more uniform material flow and improved bushing geometry under the investigated condition. A multilayer feed-forward artificial neural network (ANN) was further employed to predict bushing length, peak temperature, and stress response across different process scenarios. The study provides insight into the thermo-mechanical behavior of hybrid-layer FDJ assemblies and establishes a framework for guiding future numerical investigations and material selection in lightweight multi-material applications.