Dr. Jian Du | Machine Learning in Physics | Best Scholar Award 

Mr. Jian Du is a fourth-year Ph.D. candidate in Petroleum and Natural Gas Engineering at China University of Petroleum–Beijing, and a visiting Ph.D. researcher at the Department of Energy, Politecnico di Milano, Italy. His research focuses on the integration of physics-based knowledge and advanced machine learning techniques to address complex industrial challenges in liquid and multi-product pipeline systems. His core interests include explainable machine learning for pipeline process monitoring, physics-informed neural networks (PINNs) for efficient simulation of complex fluid dynamics, and knowledge-embedded data science frameworks for intelligent pipeline management. Through these efforts, he aims to bridge the gap between traditional physical modeling and data-driven approaches, improving reliability, interpretability, and real-time applicability in energy transportation systems. Jian Du has made significant research contributions in the areas of contamination tracking, hydraulic transient simulation, batch tracking, corrosion prediction, and energy system forecasting. He has authored or co-authored more than 30 peer-reviewed publications, with over 17 papers as first or second author, published in leading journals such as Energy, Engineering Applications of Artificial Intelligence, Journal of Industrial Information Integration, Renewable and Sustainable Energy Reviews, and Chemical Engineering Research and Design. His cumulative journal impact factor exceeds 95, and his work includes an ESI Hot Paper and Highly Cited Paper ranked in the top 1% of the engineering field. A recurring theme in his research is the development of the “DeepPipe” framework—a series of theory-guided, physics-enhanced, and multi-modal neural networks tailored for real-time pipeline monitoring and decision support.

View Scopus Profile

Featured Publications