AI-Assisted Modelling and Parameter Optimization for Hot-Press Joining of Aluminium–Polymer Hybrid Structures via RSM, ANN, and GA
Main Article Content
Abstract
Poly Butylene Terephthalate (PBT) was joined with Aluminum-Magnesium-Silicon alloy (AA6082) using a hot-press
joining technique. The experiments have been carried out per the established setup through the joining process parameters (time, pressure, and temperature) and the succeeding surface modification with three different plies of glass fiber between the two substrates. The experimental design has been carried by Design of Experiment (DOE) method through Box Behnken approach. Subsequently, the optimization of joining process parameters has been done through Response Surface Methodology (RSM), machine learning via Artificial Neural Network (ANN) and Genetic Algorithm (GA). The optimum joining parameter was found with maximum joining strength at pressure of 5bar, temperature of 270oC and duration of 3min as recorded through RSM. The optimum shear strength of the joint was found 0.9508 kN at a process parametric combination of temperature, pressure, and time at 270.3oC, 5.2 bar, and 3.03 minutes respectively with grounded surface modifications.