Proceedings of the 5th Management Science Informatization and Economic Innovation Development Conference, MSIEID 2023, December 8–10, 2023, Guangzhou, China

Research Article

A Neural Network-Based Model for Predicting Electricity Revenue in the Context of New Energy

Download37 downloads
  • @INPROCEEDINGS{10.4108/eai.8-12-2023.2344761,
        author={Li  Wu and Yongjun  Chen and Bo  Li and Liangyu  Zou},
        title={A Neural Network-Based Model for Predicting Electricity Revenue in the Context of New Energy},
        proceedings={Proceedings of the 5th Management Science Informatization and Economic Innovation Development Conference, MSIEID 2023, December 8--10, 2023, Guangzhou, China},
        publisher={EAI},
        proceedings_a={MSIEID},
        year={2024},
        month={4},
        keywords={big data; electricity fee prediction; medium and long-term electricity market; spot market; assisting decision-making},
        doi={10.4108/eai.8-12-2023.2344761}
    }
    
  • Li Wu
    Yongjun Chen
    Bo Li
    Liangyu Zou
    Year: 2024
    A Neural Network-Based Model for Predicting Electricity Revenue in the Context of New Energy
    MSIEID
    EAI
    DOI: 10.4108/eai.8-12-2023.2344761
Li Wu1, Yongjun Chen1,*, Bo Li1, Liangyu Zou1
  • 1: State Grid Huitongjincai (Beijing) Information Technology CO., LTD
*Contact email: chenyongjun@sgec.sgcc.com

Abstract

Through big data analysis and modeling research on short-term and long-term electricity bill prediction in the power market, the results show that exponential smoothing method, ARMA model, maximum information entropy model, and "association rule+ARMA" models are suitable for short-term electricity bill prediction; State space models and neural network models are more suitable for long-term electricity bill prediction. By using predictive models, it has a certain guiding role in providing transaction assistance decision-making for the medium to long term electricity market and spot electricity energy market.