Chemistry-Informed Neural Networks modelling of lignocellulosic biomass pyrolysis. (July 2022)

• Record Type: Journal Article
• Authors: Xing, Jiangkuan; Kurose, Ryoichi; Luo, Kun; Fan, Jianren
• Is Part Of: Bioresource technology. Volume 355(2022)
• Publication Date: 2022-07
• Subjects: Lignocellulosic biomass -- TGA -- Machine learning -- Chemistry-Informed Neural Networks -- Pyrolysis kinetics; Biomass -- Periodicals; Biomass energy -- Periodicals; Bioremediation -- Periodicals; Agricultural wastes -- Periodicals; Factory and trade waste -- Periodicals; Organic wastes -- Periodicals; Bioénergie -- Périodiques; Déchets agricoles -- Périodiques; Déchets industriels -- Périodiques; Déchets organiques -- Périodiques; Déchets (Combustible) -- Périodiques; 662.88
• DOI: 10.1016/j.biortech.2022.127275
• Languages: English
• ISSNs: 0960-8524

Abstract

Graphical abstract: Highlights: Novel Chemistry-informed Neural Networks (CINNs) are developed for biomass pyrolysis. A biomass pyrolysis database is constructed from available published literature. Pyrolysis kinetics are derived using the developed CINNs from the database. The derived kinetics could well describe various biomass pyrolysis with R 2 > 0.95. Comparisons with the previous models prove the advantages of the derived kinetics. Abstract: Biomass pyrolysis is a complicated reaction process that involves complex components and reaction pathways. Due to measurement limitations, the intermediate components are difficult to be detected, therefore their detailed kinetics are still not well established. To address this issue, novel Chemistry-Informed Neural Networks (CINNs) were developed to derive the lignocellulosic biomass pyrolysis kinetics from the thermogravimetric analysis (TGA) measurements in published literature. The derived pyrolysis kinetics, involving eight species and eleven reactions, could accurately reproduce the pyrolysis process for both the seen and unseen samples with R 2 > 0.95. The comparisons with the CRECK multi-step and Bio-CPD models also demonstrated the advantages of the derived kinetics in predicting both the final volatiles yield and the pyrolysis process for various biomass types. This study explored a new tool for establishing solid fuel conversion kinetics from TGA measurements using chemistry-informed machine learning approaches.