Research and Application on Process Manufacturing industry with Artificial Intelligence Technologies: A Cement Process Case Study

This study examines challenges in heavy industry, focusing on cement manufacturing, and employs artificial intelligence to tackle issues like temporal delays and spatio-temporal data. We developed comprehensive models for time series forecasting, utilized deep learning for predictive models managing spatio-temporal complexities, and introduced a multimodal contrastive learning framework for combining sensor data and simulation outputs. Additionally, a pre-trained sensor feature representation network enhances operational efficiency with machine learning techniques. This innovative approach offers insights into industrial challenges, especially in cement production.