Abstract
With the fast expansion of renewable energy systems during recent years, the stability and quality of smart grids using solar energy have been challenged because of the intermittency and fluctuations. Hence, forecasting photo-voltaic (PV) power generation is essential in facilitating planning and managing electricity generation and distribution. In this paper, the ultra-short-term forecasting method for solar PV power generation is investigated. Subsequently, we proposed a radial basis function (RBF)-based neural network. Additionally, to improve the network generalization ability and reduce the training time, the numbers of hidden layer neurons are limited. The input of neural network is selected as the one with higher Spearman correlation among the predicted power features. The data are normalized and the expansion parameter of RBF neurons are adjusted continuously in order to reduce the calculation errors and improve the forecasting accuracy. Numerous simulations are carried out to evaluate the performance of the proposed forecasting method. The mean absolute percentage error (MAPE) of the testing set is within 10%, which show that the power values of the following 15 min. can be predicted accurately. The simulation results verify that our method shows better performance than other existing works.
Highlights
With the growing scarcity of fossil energy, the use of sustainable energy for power generation has increasingly attracted the attention of global power industry
Neural networks on MATLAB, which means fewer hidden layer neurons than the number of input data sets are required for training
The low error of the testing set fully reflects the high generalization ability of the model, and it is indicated that the forecasting task for the unlearned samples is finished accurately
Summary
With the growing scarcity of fossil energy, the use of sustainable energy for power generation has increasingly attracted the attention of global power industry. The accurate prediction of PV power generation is an effective way to provide guidance for grid operation scheduling and regulation of the PV system itself, and to reduce the instability of PV grid-connected operation These are strong motivations for ultra-short-term forecasting of the output of PV power plants. The average daily root mean square error (RMSE) and R2 of the HKGE model are 4.3210 kW and 0.9953, which outweighed other eight compared models These existing indirect forecasting methods are generally employed in a large forecasting time span, ranging from several hours to one day, and the model structure or the data processing method was complex, while the forecasting accuracy is limited. The problem of how to accurately predict the PV power in ultra-short-term, reduce the complexity of the network structure, and improve the calculation speed and forecasting accuracy, has not been well solved.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
