Description
Date depot: 12 avril 2023
Titre: AI-based simulation of hydrochars production from Lignocellulosic Residues for Green Electronics, Environmental Remediation and Agricultural Applications
Directeur de thèse:
Jérémie SUBLIME (LISITE)
Encadrante :
Andrea LANDAZURI (Univ. San Franscisco de Quito)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Intelligence artificielle
Resumé: Hydrothermal carbonization (HTC) of lignocellulosic residues (e.g.: peels, husks, hulls) has become an attractive method for the production of hydrochars from which valuable materials can be obtained for diverse applications. HTC is a lower energy consumption process compared to other conventional thermal treatments with other advantages such as a higher conversion rate and preparation of heteroatom-doped materials (Yang et al., 2023). However, predicting the outcome of HTC is a difficult task and a costly process due to the high number of parameters, as well as the time and energy consumption that would be required to test all possible cases. For this reason, HTC is a good candidate to be simulated using artificial intelligence and predictive Machine Learning methods. Indeed, hydrochar materials are easy to characterize and large amounts of data are already available on existing experiments, waiting to be exploited by AI methods.
Hydrochars can be characterized physically, chemically and electrically to gain understanding of their crystallographic structure, composition, morphology, surface area, vibrational modes, chemical state and dielectric/conductive properties. HTC temperature, HTC processing times and activation methods can influence the suitability of hydrochars for a potential application.