Abstract
Accurate photovoltaic (PV) power forecasting has become one of the key basic technologies to improve the operation quality of power system and reduce the reserve capacity. Accurate Numerical Weather Prediction (NWP) is the important factor for short-term PV power forecasting. However, the accuracy of NWP is sometimes unsatisfactory. Meanwhile, due to various reasons, there were many kinds of anomalies in the measured data of photovoltaic stations, which will lead to the difficulty in obtaining accurate enough forecasting results. In this paper, a NWP correction method considering the coupling correlation of meteorological factors and a short-term photovoltaic power forecasting model based on the correction results are proposed to further improve the prediction accuracy. Firstly, the anomalies are detected and eliminated. Secondly, the optimal forecasting model is trained with the measured meteorological data and power data, meanwhile the optimal combination of meteorological factors is determined. Finally, a short-term PV power forecasting model based on convolutional long-short term memory network (CNN-LSTM) is established. The simulation results show that, compared with the method without optimal meteorological factors determination and NWP correction, the accuracy can be significantly improved by proposed method in this paper. The necessity and effectiveness of NWP correction are also verified.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.