Predicting bioenergy power generation structure using a newly developed grey compositional data model: A case study in China

Accurate short-term prediction of bioenergy power generation structure can optimize the bioenergy structure and help achieve carbon neutrality. However, there are currently few related studies, and most of them present long-term predictions on the development potential of bioenergy, which cannot meet the modeling requirements of predicting structure. Hence, the Fractional-order-accumulation grey Compositional data Model with Particle swarm is proposed in this paper (PFCM (1,1)) for forecasting bioenergy power generation structure. The proposed model satisfies the modeling requirements by introducing the fractional accumulation operator to ensure the prediction accuracy, and constructing the spherical mapping space to reduce the data dimension. The empirical studies prove that the newly developed model performs better than other models, which is successfully employed to predict bioenergy power generation structure of China for 2020–2024. The results show that the share of renewable municipal waste in bioenergy power generation will exceed that of solid biofuels by 2023 and the share of biogas power generation has not changed much. Furthermore, although the total amount of bioenergy power generation in China is growing rapidly, unbalanced development and small share of power are two important challenges.