Enhancing Aluminum Extrusion Quality via AI-Driven Temperature Prediction in Cyberphysical Systems for Industry 4.0

Aluminum extrusion is pivotal in industrial manufacturing for ubiquitous daily products, and the extrusion process determines the profile quality. This study is the first to build a cyberphysical system (CPS) for aluminum extrusion that autonomously predicts profile temperatures by collecting sensor data, aiming to optimize the industrial process and boost productivity. The proposed artificial intelligence-driven CPS system excels at data collection from sensors, real-time preprocessing of extensive datasets, and leverages the eXtreme gradient boosting (XGBoost) method for accurate temperature prediction in real-world scenarios. This implementation demonstrates its practical adaptability across diverse extrusion methods. Experimental outcomes validate the model’s precise temperature identification. As a result, this system can significantly bolster profile temperature and quality maintenance in aluminum extrusion, notably enhancing efficiency and reducing cost.