Research on Biomass Energy High-Resolution Prediction Methods Driven by Multi-Source Data

This paper proposes a high-resolution prediction method for biomass energy resources driven by multi-source data and applies it to Sichuan Province as a case study. By integrating biomass resource potential data such as agricultural residues, forestry residues, and energy crops, along with multi-source environmental and socio-economic data, the XGBoost algorithm is used to accurately predict biomass resources. The results show that the model's coefficient of determination (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) exceeds 0.9, and the root mean square error (RMSE) is as low as 0.1, demonstrating superior prediction performance under complex spatiotemporal distribution variations. The study indicates significant regional differences in the spatial distribution of biomass resources in Sichuan Province, with higher development potential in the central and southeastern regions and relatively lower potential in the northwest. Temporal analysis reveals a fluctuating trend in biomass resources after 2018, which may be related to environmental policy adjustments, climate change, and market demand fluctuations. This study provides new insights and scientific evidence for the efficient management and sustainable development of biomass energy resources, with important application value and development prospects.