Skip to main content
placeholder image

Photovoltaic Power Forecasting in Dunhuang based on Recurrent Neural Network

Conference Paper


Abstract


  • Photovoltaic power generation is a new energy mode in modern smart city, and photovoltaic power forecasting plays an important role in the development and regulation of smart city, especially for emerging cities with very different geographical, meteorological and environmental factors. It is still a challenging subject to develop a high-precision photovoltaic power prediction system with little or even missing historical data. Recurrent neural network (RNN) is the mainstream time series forecasting model in deep learning, but it still has defects in the application scenario of photovoltaic power prediction, such as complex calculation, large sample demand and so on. Aiming at this situation, this paper studies the photovoltaic power forecasting method in the case of small samples, and the prediction effects of different RNN models are compared and analyzed.

Publication Date


  • 2022

Citation


  • Lv, Q., Ma, M., Jiang, X., Jiang, M., Yong, B., Shen, J., & Zhou, Q. (2022). Photovoltaic Power Forecasting in Dunhuang based on Recurrent Neural Network. In 2021 IEEE 23rd International Conference on High Performance Computing and Communications, 7th International Conference on Data Science and Systems, 19th International Conference on Smart City and 7th International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021 (pp. 1346-1352). doi:10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00203

Scopus Eid


  • 2-s2.0-85132434763

Web Of Science Accession Number


Start Page


  • 1346

End Page


  • 1352

Abstract


  • Photovoltaic power generation is a new energy mode in modern smart city, and photovoltaic power forecasting plays an important role in the development and regulation of smart city, especially for emerging cities with very different geographical, meteorological and environmental factors. It is still a challenging subject to develop a high-precision photovoltaic power prediction system with little or even missing historical data. Recurrent neural network (RNN) is the mainstream time series forecasting model in deep learning, but it still has defects in the application scenario of photovoltaic power prediction, such as complex calculation, large sample demand and so on. Aiming at this situation, this paper studies the photovoltaic power forecasting method in the case of small samples, and the prediction effects of different RNN models are compared and analyzed.

Publication Date


  • 2022

Citation


  • Lv, Q., Ma, M., Jiang, X., Jiang, M., Yong, B., Shen, J., & Zhou, Q. (2022). Photovoltaic Power Forecasting in Dunhuang based on Recurrent Neural Network. In 2021 IEEE 23rd International Conference on High Performance Computing and Communications, 7th International Conference on Data Science and Systems, 19th International Conference on Smart City and 7th International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021 (pp. 1346-1352). doi:10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00203

Scopus Eid


  • 2-s2.0-85132434763

Web Of Science Accession Number


Start Page


  • 1346

End Page


  • 1352