Research Information System
Arabian Journal for Science and Engineering, vol.48, no.3, pp.2841-2850, 2023 (SCI-Expanded, Scopus)
Al alloys have low density, high strength and low production-maintenance costs. Thus, they are widely used in many sectors such as aerospace and automotive industry. The biggest problem of Al alloys is that their strength is lower than steels. For this reason, it was important to increase the strength of Al alloys in the studies, and heat treatment was applied to achieve this. The heat treatment applied to Al and its alloys is the aging heat treatment. In order to aging Al with heat treatment to be effective on the strength of Al alloys, it is important to determine the appropriate temperature and time values. Therefore, many experiments are needed to determine the optimum process parameters. Performing these experiments one by one is both costly and time consuming. For this reason, methods such as machine learning and ANN have been used recently to predict the mechanical properties of materials. In this study, it is aimed to determine the change in mechanical properties of AA 2024 material produced with different manufacturing methods after aging at different processing temperatures and times, using artificial intelligence methods. For this purpose, in the study, hardness and bending strength were tried to be predicted from existing experimental data by using other machine learning algorithms such as convolutional neural network (CNN) deep learning algorithm, artificial neural network (ANN) and random forest regression (RFR), which are artificial intelligence methods. It has been seen that the properties of Al alloys can be successfully determined using artificial intelligence methods. The best results for Powder Metal (PM) 2024 Al alloy were found as RMSE 0.09068, R-Squared 0.93476 and MAE 0.06734 by using the CNN algorithm. Also, the best results for Full Dense (FD) 2024 Al alloy were found as RMSE 0.08578, R-Squared 0.94166 and MAE 0.06212 by using the CNN algorithm.