A Study on Power Supply Prediction of Power System Based on Correlation Analysis and ARIMA Modeling

In order to predict the future structure of electricity supply and energy storage consumption, this study analyzes the correlation between China's electricity supply and energy storage consumption based on 30 factors such as economy, geography and policy. The ARIMA model was established. After verifying the stability and white noise of the model, the Auto-ARIMA function was used to find the optimal parameter as ARIMA (0, 1, 1). According to the forecast, China's electricity supply is growing linearly and is expected to reach 30,268.21 billion kWh by 2060. Then, the ARIMA (1, 2, 0) model is used to predict the proportion of photovoltaic power generation in China in 2060, and it is predicted that by 2060, China's non-fossil energy will account for 90.9%, more than 80%. The results show that by 2060, China will be able to replace coal-fired power generation with photovoltaic power generation. The predicted value of the model is basically consistent with the actual value, and the fit of the model is good.