Autoencoder-Based Anomaly Detection Model for Slab Thickness in Continuous Casting

Real-time anomaly detection of slab thickness is essential for ensuring quality in continuous casting. Existing methods face challenges in data preprocessing and real-time monitoring. This study addresses these issues by integrating Principal Component Analysis for dimensionality reduction and anomaly Detection with an autoencoder model. The proposed approach enables efficient and accurate real-time anomaly detection in slab thickness during the continuous casting process.