Research on Photovoltaic Power Prediction Based on Deep Learning

• Main Authors: Xie, Chen, Wei, Yunbing, Wang, Sifang, Liu, Jiankang
• Format: Conference Proceeding
• Language: English
• Subjects: Accuracy; Data models; Deep learning; ensemble models; Forecasting; forecasting methods; Hyperparameter optimization; Meteorology; Optimization; Photovoltaic (PV) generation; Photovoltaic systems; power forecasting; Prediction algorithms; Predictive models
• ISSN: 2768-0525
• DOI: 10.1109/ICPRE62586.2024.10768392

The output of photovoltaic (PV) systems is significantly influenced by factors such as sunlight and weather conditions, leading to substantial variations. Ensuring stable electricity supply is crucial for meeting consumer demands. Accurate PV power forecasting enables grid operators to anticipate fluctuations in PV generation, facilitating timely scheduling and operational decisions to ensure grid stability and reliability. This article provides a comprehensive review of advanced deep learning methods applied in PV power forecasting, summarizing research developments from both domestic and international perspectives. It discusses key factors influencing PV power prediction, including solar irradiance intensity, temperature, and weather conditions. Additionally, it elaborates on prevalent forecasting techniques, with a particular focus on deep learning approaches and hybrid model predictions. The aim is to equip researchers and practitioners in the PV sector with a thorough understanding of how deep learning methods can innovate and advance power forecasting practices, thereby driving technological progress and industry development.