Improving the accuracy of very-short-term (VST) photovoltaic (PV) power generation prediction can effectively enhance the quality of operational scheduling of PV power plants, and provide a reference for PV maintenance and emergency response. In this paper, the effects of different meteorological factors on PV power generation as well as the degree of impact at different time periods are analyzed. Secondly, according to the characteristics of radiation coordinate, a simple radiation classification coordinate (RCC) method is proposed to classify and select similar time periods. Based on the characteristics of PV power time-series, the selected similar time period dataset (include power output and multivariate meteorological factors data) is reconstructed as the training dataset. Then, the long short-term memory (LSTM) recurrent neural network is applied as the learning network of the proposed model. The proposed model is tested on two independent PV systems from the Desert Knowledge Australia Solar Centre (DKASC) PV data. The proposed model achieving mean absolute percentage error of 2.74–7.25%, and according to four error metrics, the results show that the robustness and accuracy of the RCC-LSTM model are better than the other four comparison models.
Owing to the clean, inexhaustible and pollution-free, solar energy has become a powerful means to solve energy and environmental problems. However, photovoltaic (PV) power generation varies randomly and intermittently with respect to the weather, which bring the challenge to the dispatching of PV electrical power. Thus, power forecasting for PV power generation has become one of the key basic technologies to overcome this challenge. The paper presents a grey relational analysis (GRA) and long short-term memory recurrent neural network (LSTM RNN) (GRA-LSTM) model-based power short-term forecasting of PV power plants approach. The GRA algorithm is adopted to select the similar hours from history dataset, and then the LSTM NN maps the nonlinear relationship between the multivariate meteorological factors and power data. The proposed model is verified by using the dataset of the PV systems from the Desert Knowledge Australia Solar Center (DKASC). The prediction results of the method are contrasted with those obtained by LSTM, grey relational analysis-back propagation neural network (GRA-BPNN), grey relational analysis-radial basis function neural network (GRA-RBFNN) and grey relational analysis-Elman neural network (GRA-Elman), respectively. Results show an acceptable and robust performance of the proposed model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.