Assessment of machine learning techniques for deterministic and probabilistic intra-hour solar forecasts. (August 2018)

• Record Type: Journal Article
• Authors: Pedro, Hugo T.C.; Coimbra, Carlos F.M.; David, Mathieu; Lauret, Philippe
• Is Part Of: Renewable energy. Volume 123(2018)
• Page Start: 191
• Page End: 203
• Publication Date: 2018-08
• Subjects: Probabilistic solar forecasts -- Global irradiance -- Direct irradiance -- Machine learning -- Sky imagery; Renewable energy sources -- Periodicals; Power resources -- Periodicals; Énergies renouvelables -- Périodiques; Ressources énergétiques -- Périodiques; 333.794
• DOI: 10.1016/j.renene.2018.02.006
• Languages: English
• ISSNs: 0960-1481

Abstract

Abstract: This work compares the performance of machine learning methods ( k -nearest-neighbors ( k NN) and gradient boosting (GB)) in intra-hour forecasting of global (GHI) and direct normal (DNI) irradiances. The models predict the GHI and DNI and the corresponding prediction intervals. The data used in this work include pyranometer measurements of GHI and DNI and sky images. Point forecasts are evaluated using bulk error metrics while the performance of the probabilistic forecasts are quantified using metrics such as Prediction Interval Coverage Probability (PICP), Prediction Interval Normalized Averaged Width (PINAW) and the Continuous Ranked Probability Score (CRPS). Graphical verification displays like reliability diagram and rank histogram are used to assess the probabilistic forecasts. Results show that the machine learning models achieve significant forecast improvements over the reference model. The reduction in the RMSE translates into forecasting skills ranging between 8% and 24%, and 10% and 30% for the GHI and DNI testing set, respectively. CRPS skill scores of 42% and 62% are obtained respectively for GHI and DNI probabilistic forecasts. Regarding the point forecasts, the GB method performs better than the k NN method when sky image features are included in the model. Conversely, for probabilistic forecasts the k NN exhibits rather good performance. Highlights: Comparison of machine learning methods for intra-hour GHI and DNI forecasting. Predictors include irradiance data and sky image features. Forecasts and prediction intervals are evaluated qualitatively and quantitatively. Significant forecast skills are observed: 0.13–0.24 for GHI, and 0.14–0.30 for DNI. GHI prediction intervals (PIs) are consistent and reliable, DNI PIs less so.