Empowering engineering with data, machine learning and artificial intelligence: Application to reactor modeling

IFPEN is a leading actor in the development of hydrocracking processes ranging from the hydro-treatment of petroleum products to the production of biofuels. In order to properly scale-up the catalytic sections of these processes robust and evaluated simulation tools are needed and developed by IFPEN. The development of accurate and robust performances predicting models for catalytic reactions with complex feedstocks is challenging because of the scarcity of data and extreme complexity of the reaction networks involved (1000s of reactions). The objective of this thesis is to combine the traditional kinetic modeling approach, based on system of equations of physics/chemistry with Machine Learning and Data Science methods. 
Machine Learning has been applied to a wide variety of problems over the last years, however, the challenge of integrating physical constraints and expert knowledge is an active research problem. The integrated approach to be developed in this thesis has the potential to leverage the capabilities of both, kinetic and Machine Learning models, by retaining the strengths and compensating for the respective weaknesses. 
This thesis offers the opportunity to contribute to opening Machine Learning methods to the realm of real-world industrial application, working with datasets from both pilot plant tests and industrial units, and to interact with experts in both process modeling (IFPEN) and Machine Learning experts.

Keywords: Kinetic model, Hybrid Model, Machine Learning, Neural Network, PINs

• Academic supervisor    Prof. Jean Marc COMMENGE, LRGP, ORCID : 0000-0003-2792-9357
• Doctoral School   ED608 - SIMPPE (Sciences et Ingénierie des Molécules, des Produits, des Procédés et de l’Energie), Université de Lorraine
• IFPEN supervisor    Dr Benoit CELSE, ORCID : 0000-0002-2503-6734
• PhD location    IFPEN, Solaize, France
• Duration and start date    3 years, starting in the fourth quarter 2024 (Novembre 4)
• Employer    IFPEN
• Academic requirements    University Master degree in Machine Learning, or Chemical Engineering 
• Language requirements    English level B2 (CEFR)    

To apply, please send your cover letter and CV to the IFPEN supervisor indicated below.