Research on Short-Term Prediction of Photovoltaic Power Generation Capacity Based on Big Data Technology
[Purposes]Through the short-term prediction technology of photovoltaic power generation,the output power of photovoltaic power generation can be grasped in real time,which helps the power grid dispatch department to coordinate the coordination between conventional power sources and photovoltaic power generation,adjust the dispatch plan reasonably,effectively reduce the adverse impact of photovol-taic power generation system access on the power grid,and ensure the safe and stable operation of the power grid.[Methods]By extracting the influencing factors of photovoltaic power generation,this paper analyzes the impact of solar irradiance,temperature,and meteorological factors on photovoltaic power generation to avoid the problem of short-term prediction errors.This paper builds a short-term predic-tion model for photovoltaic power generation based on big data technology,and utilizes long short-term memory neural networks to extract short-term features of photovoltaic power generation to ensure the ac-curacy of power prediction.And then,this paper generates a short-term non-stationary prediction se-quence for photovoltaic power generation,captures the time series characteristics of photovoltaic power generation,and updates sand correct the prediction results in a timely manner to obtain accurate predic-tion results.[Findings]The power prediction fluctuation of the design method is consistent with the ac-tual fluctuation,and the deviation between the predicted value and the actual value is small,which can adapt to the subsequent power scheduling and operation requirements.[Conclusions]This study can lay an important foundation for formulating reasonable scheduling plans,reduce resource waste,and play an important role in improving the economic benefits of power plants.
big data technologyphotovoltaic power stationpower generation capacityshort-termpre-diction methods
NETL
NSTL
万方数据
