Forecasting Renewable Energy Generation Based on a Novel Dynamic Accumulation Grey Seasonal Model

With the increasing proportion of electricity in global end-energy consumption, it has become a global consensus that there is a need to develop more environmentally efficient renewable energy generation methods to gradually replace traditional high-pollution fossil energy power generation. Renewable energy generation has become an important method of supplying power across the world. Therefore, the accurate prediction of renewable energy generation plays a vital role in maintaining the security of electricity supply in all countries. Based on this, in our study, a novel dynamic accumulation grey seasonal model is constructed, abbreviated to DPDGSTM(1,1), which is suitable for forecasting mid- to long-term renewable energy generation. Specifically, to overcome the over-accumulation and old information disturbance caused by traditional global accumulation, a dynamic accumulation generation operator is introduced based on a data-driven model, which can adaptively select the optimal partial accumulation number according to the intrinsic characteristics of a sequence. Subsequently, dummy variables and a time trend item are integrated into the model structure, significantly enhancing the adaptability of the new model to the sample sequence with different fluctuation trends. Finally, a series of benchmark models are used to predict renewable energy generation in China, wind power generation in the United States, and hydropower generation in India. The empirical results show that the new model performs better than other benchmark models and is an effective tool for the mid- to long-term prediction of renewable energy generation.