Answers are grounded in this paper and cite their sources.
At present, the construction of distributed photovoltaic power stations in China lacks systematic and comprehensive preliminary planning; The construction cost exceeded the estimated estimate. After the completion of the project economic benefits cannot reach the expected income, project operating costs exceed expectations and other problems. In order to solve these problems, it is urgent to reasonably forecast the electricity consumption data of distributed photovoltaic power stations. Therefore, in order to solve these problems, a reliable model is established to predict the electricity consumption data of distributed photovoltaic power stations, and the indirect prediction method is used to forecast, that is, the irradiance of medium and long-term time scales is predicted by historical meteorological data, and then the system electricity consumption data is obtained. Among them, the model used is the Long short-term memory (LSTM) neural network model. Under the effect of this model, the electricity consumption data prediction of distributed photovoltaic power stations is carried out. The result shows that the MAPE of monthly prediction is 3.5%, and the annual prediction is 1.1%, which has ideal prediction accuracy and can achieve better prediction effect. This indirect forecasting method breaks the shackles of traditional forecasting methods, avoids the problems of data collection and other aspects, and is a new development trend and the performance of scientific and technological progress, which is conducive to the development of distributed photovoltaic power stations. Doi: 10.28991/HIJ-2024-05-04-05 Full Text: PDF
No takes yet. Share an insight, caveat, or question.
Yao et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: