Forecasting the annual household electricity consumption of Chinese residents using the DPSO-BP prediction model

In recent years, global climate change caused by carbon dioxide emissions has attracted more and more attention. Adjusting the energy mix by predicting energy demands is currently a more effective way to address climate issues and energy supply issues. Based on the panel data from 1999 to 2018 in China, this paper designed a new hybrid prediction model to predict the future electricity consumption of Chinese residents by double improvement of particle swarm optimization. By comparing with the BP neural prediction model without mixing and several BP neural prediction models with other improved and mixed forms, the results show that the BP neural network hybrid prediction model with DPSO-BP is more suitable for forecasting the electricity consumption of Chinese residents. At the same time, the prediction results of the DPSO-BP prediction model show that the annual electricity consumption of Chinese residents will increase from 9685 (100 million kWh) in 2018 to 13,171 (100 million kWh) in 2025 in the next 7 years. The research results provide a reference for future scholars in the design of algorithms and provide suggestions for the government to adjust energy and avoid severe power shortages or surpluses. Graphical abstract In this paper, a new hybrid prediction model is established to predict the annual electricity consumption of Chinese residents. To achieve the research purposes, a brief flowchart of the work of this study is shown in Fig. 1.