Refined Accommodation Analysis Method for New Energy Based on VMD-GRU Deep Learning

As the national "dual-carbon" strategy is being implemented, the construction of new power system with new energy as the mainstay is accelerating. It is of great guiding significance to scientifically and comprehensively evaluate the level of new energy accommodation and utilization on aspects such as power supply planning and layout, grid construction and coordination of grid operation. A refined accommodation analysis method for new energy based on VMD-GRU deep learning is proposed. Firstly, the VMD-GRU deep learning method is used to identify and correct massive data from the station side, improving the quality of source-side data. Then, a multi-level nested analysis method for refined accommodation of new energy is proposed with the analysis granularity precisely down to the single station level. Finally, a multi-level, multidimensional, multi-data source, minute-level and customized online integration analysis function is realized based on the support system of the new generation scheduling technology. Taking the new energy operation and accommodation data for the first half of 2023 in a certain province as an example, the effectiveness and practical value of the proposed analysis method are verified.