In modern bioenergy, the long-term storage of substantial quantities of biomass is necessary. A critical issue arises from the potential self-heating of biomass piles initialized by exothermic microbial activities, which could result in spontaneous ignition. To address this concern, we develop a comprehensive dynamic model to accurately simulate biomass degradation and microbial metabolism. The model encompasses four key processes: the conversion of a slowly-biodegradable fraction into an easily-biodegradable fraction (R1); the breakdown of the easily-biodegradable fraction (R2), consuming O₂, boosting microbial populations and releasing metabolites like CO₂ and H₂O; the decay of microbes into a non-biodegradable fraction (R3); and the biomass moisture migration (R4). In the context of R1, two biomass degradation modes (Tremier and Haug) are considered, while R2 is characterized using four microbial growth kinetics models (square-root-type model, polynomial model, generalized linear model, Cardinal temperature model with inflection). These models can adapt to varying temperature and moisture levels, affecting microbial growth rates. By combining different R1 and R2 pathways, eight modelling pathways are established, and their parameters are derived using genetic algorithm and particle swarm optimization techniques, with the aim of minimizing disparities between experimental outcomes and model predictions. The model’s performances are evaluated through several metrics such as CO₂ evolution rate, biomass dry matter loss, cumulative respiration, and overall prediction accuracy. The evaluation indicates that the Tremier model more accurately predicts biomass degradation across six types of biomass, and the Cardinal temperature model with inflection and square-root-type model offer superior predictions for microbial growth dynamics, especially for biomass types rich in cellulose and low in lignin content, e.g., wheat straw, corn stalk, rice straw and soybean hull. Our model dynamically refines the kinetic parameters for biomass decomposition and microbial metabolism based on changes in ambient temperature, humidity, and biomass moisture content over time. This feature enables real-time prediction of the decomposition state of biomass throughout extended storage periods. Furthermore, by incorporating an appropriate metabolic heat production model for the biomass, along with heat and mass transfer equations, our model can predict the internal temperature distribution within a biomass pile. This capability serves as a valuable tool for ensuring the safe storage of biomass, providing a robust means of managing the risks associated with biomass self-heating and self-ignition.
 

Biomass fuel,Biomass degradation,Biological respiration,Modelling,Biomass self-heating