Spatial models for probabilistic prediction of wind power with application to annual-average and high temporal resolution data

• Published in: Stochastic environmental research and risk assessment Vol. 31; no. 7; pp. 1615 - 1631
• Main Authors: Lenzi, Amanda, Pinson, Pierre, Clemmensen, Line H., Guillot, Gilles
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01-09-2017; Springer Nature B.V
• Subjects: Approximation; Aquatic Pollution; Chemistry and Earth Sciences; Climatology; Computational Intelligence; Computer Science; Data processing; Earth and Environmental Science; Earth Sciences; Electric power generation; Environment; Environmental science; Epidemiology; Geostatistics; Math. Appl. in Environmental Science; Mathematical models; Original Paper; Physics; Predictions; Probability Theory and Stochastic Processes; Statistical analysis; Statistical models; Statistics for Engineering; Stochasticity; Temporal resolution; Waste Water Technology; Water Management; Water Pollution Control; Wind farms; Wind power; Wind power generation; Spatial prediction; Wind power; Latent Gaussian field; Integrated nested Laplace approximation
• DOI: 10.1007/s00477-016-1329-0
• EISSN: 1436-3259
• ISSN: 1436-3240

Producing accurate spatial predictions for wind power generation together with a quantification of uncertainties is required to plan and design optimal networks of wind farms. Toward this aim, we propose spatial models for predicting wind power generation at two different time scales: for annual average wind power generation, and for a high temporal resolution (typically wind power averages over 15-min time steps). In both cases, we use a spatial hierarchical statistical model in which spatial correlation is captured by a latent Gaussian field. We explore how such models can be handled with stochastic partial differential approximations of Matérn Gaussian fields together with Integrated Nested Laplace Approximations. We demonstrate the proposed methods on wind farm data from Western Denmark, and compare the results to those obtained with standard geostatistical methods. The results show that our method makes it possible to obtain fast and accurate predictions from posterior marginals for wind power generation. The proposed method is applicable in scientific areas as diverse as climatology, environmental sciences, earth sciences and epidemiology.