Machine learning-based process parameter optimisation and strip plate shape improvement in steel production

Accurate control of steel rolling parameters and strip shape is crucial for industrial competitiveness. However, traditional methods are limited by the process's nonlinear, dynamic, and multi-objective nature. While machine learning (ML) offers promising solutions, a systematic review of its application in this domain is lacking. This paper addresses this gap by providing a comprehensive review of ML-based process parameter optimization and strip shape improvement. We analyse the key technical frameworks, main application scenarios, and typical solutions. Furthermore, we explore potential future research directions and strategies. This work aims to provide a valuable reference for academic and industrial professionals, promoting the advancement of intelligent steel production and enhancing process efficiency and strip quality.