Research on short-term forecasting method of photovoltaic power generation based on clustering SO-GRU method

Abstract

In order to enhance the stability of the power grid and increase the ability of the power grid to absorb photovoltaic energy, while enriching the prediction model of photovoltaic power generation, improve the precision reading of the prediction model. This paper proposes a short-term forecasting model of photovoltaic power generation based on K-means++clustering- Snake Optimizer algorithm (SO) - Gate Recurrent Unit(GRU). First, Pearson correlation coefficient analysis is used to screen out variables that are not related to the actual output of power generation.Then, select historical data with similar weather fluctuation degree as the training sample through K-means++clustering, use SO algorithm to optimize the parameters of GRU network, and use the optimized GRU network to predict photovoltaic power generation, Finally, through the simulation of a photovoltaic power station in Beijing, the results show that the prediction model proposed in this paper has improved the prediction accuracy compared with other traditional prediction models when the weather is almost no fluctuation, the weather has some fluctuations and the weather fluctuation is very severe. This research can provide reference for energy storage research of integrated energy system.

Keywords

• Short-term forecast of photovoltaic power generation
• K-means++clustering
• Snake optimization algorithm
• GRU neural network