Benefits of spatiotemporal modeling for short‐term wind power forecasting at both individual and aggregated levels

Environmetrics - Tập 29 Số 3 - 2018
Amanda Lenzi1, Ingelin Steinsland2, Pierre Pinson3
1Department of Applied Mathematics and Computer Science, Technical University of Denmark, 2800 Kongens Lyngby, Denmark
2Department of Mathematical Sciences, Norwegian University of Science and Technology, N-7491 Trondheim, Norway
3Department of Electrical Engineering, Technical University of Denmark, 2800 Kgs, Lyngby, Denmark

Tóm tắt

The share of wind energy in total installed power capacity has grown rapidly in recent years. Producing accurate and reliable forecasts of wind power production, together with a quantification of the uncertainty, is essential to optimally integrate wind energy into power systems. We build spatiotemporal models for wind power generation and obtain full probabilistic forecasts from 15 min to 5 h ahead. Detailed analyses of forecast performances on individual wind farms and aggregated wind power are provided. The predictions from our models are evaluated on a data set from wind farms in western Denmark using a sliding window approach, for which estimation is performed using only the last available measurements. The case study shows that it is important to have a spatiotemporal model instead of a temporal one to achieve calibrated aggregated forecasts. Furthermore, spatiotemporal models have the advantage of being able to produce spatially out‐of‐sample forecasts. We use a Bayesian hierarchical framework to obtain fast and accurate forecasts of wind power generation not only at wind farms where recent data are available but also at a larger portfolio including wind farms without recent observations of power production. The results and the methodologies are relevant for wind power forecasts across the globe and for spatiotemporal modeling in general.

Từ khóa


Tài liệu tham khảo

10.1002/0470012684

10.1002/9781118950203

Breiman L., 1992, Submodel selection and evaluation in regression. The X‐random case, International Statistical Review/Revue Internationale de Statistique, 291

10.1002/we.107

10.1175/WAF966.1

10.1007/s10182-012-0196-3

10.1016/j.renene.2010.09.016

10.1016/j.atmosenv.2006.08.032

10.1111/j.1365-3121.1992.tb00605.x

Cressie N., 2015, Statistics for spatial data

Dowell J., 2016, Very‐short‐term probabilistic wind power forecasts by sparse vector autoregression, IEEE Transactions on Smart Grid, 7, 763

10.1002/env.1123

10.1016/S0167-6105(01)00222-7

Garver L. L., 1966, Effective load carrying capability of generating units, IEEE Transactions on Power apparatus and Systems, 8, 910, 10.1109/TPAS.1966.291652

10.1016/j.apenergy.2012.06.064

10.1016/j.ijforecast.2009.12.015

10.1111/j.1467-9868.2007.00587.x

Gneiting T., 2010, Continuous parameter spatio‐temporal processes, Handbook of Spatial Statistics, 97, 427, 10.1201/9781420072884-c23

10.1198/016214506000000456

10.1198/016214506000001437

10.1016/j.spasta.2013.06.002

R. Kohavi 1995 1137 1145

Lau A., 2010, Approaches for multi‐step density forecasts with application to aggregated wind power, The Annals of Applied Statistics, 1311

10.1093/biostatistics/kxj034

10.1111/j.1467-9868.2011.00777.x

Lindström J., 2013, SpatioTemporal: An R package for spatio‐temporal modelling of air‐pollution, Journal of Statical Software

10.1007/s10651-013-0261-4

10.1016/j.jweia.2008.03.013

10.1109/TPWRS.2010.2065818

10.1007/s00477-012-0652-3

Paciorek C. J., 2007, Bayesian smoothing with Gaussian processes using Fourier basis functions in the spectral GP package, Journal of Statistical Software, 19, nihpa22751, 10.18637/jss.v019.i02

10.1287/opre.2013.1174

10.1111/j.1467-9876.2011.01026.x

10.1109/TPWRS.2010.2045774

Pinson P. &Tastu J.(2013).Discrimination ability of the energy score(Technical Report). Lyngby Denmark: Technical University of Denmark.

Ross S., 2015, A first course in probability

10.1111/j.1467-9868.2008.00700.x

10.1175/MWR-D-14-00269.1

10.1002/we.401

10.1109/TEC.2009.2025431

10.1080/10618600.2014.977447

Toth Z., 2003, Forecast verification: A practitioners guide in atmospheric science

10.1016/j.apenergy.2011.04.011

10.1093/biomet/41.3-4.434

Whittle P., 1963, Stochastic‐processes in several dimensions, Bulletin of the International Statistical Institute, 40, 974

10.1093/biomet/86.4.815

Thông tin
Thông tin xuất bản

Environmetrics

Tập 29 Số 3

Thông tin tác giả