Zusammenfassung
Mit der Energiewende in Deutschland steigt die Bedeutung der Vorhersage von Windenergie sowie Windgeschwindigkeit. Sowohl kurz- als auch mittelfristige genaue Vorhersagen von Windgeschwindigkeit und Windenergie spielen eine entscheidende Rolle in verschiedenen Wirtschaftsbereichen. Die Weiterentwicklung und Anwendung neuer Windvorhersagemodelle hilft dabei, den Nutzen aus Windenergie zu erhöhen. Der vorliegende Beitrag umfasst die Anwendung eines periodisch-saisonalen vektorautoregressiven Prognosemodells (VAR), um die mittlere Windgeschwindigkeit vorherzusagen. Darüber hinaus beinhaltet das betrachtete Modell autoregressive bedingte Heteroskedastizität basierend auf Schwellenwerten (TARCH). Die eingeführten Modelle werden mit Hilfe der Methode der kleinsten absoluten Minderung und Selektion (LASSO) mit iterativer Neugewichtung geschätzt. Dieser Ansatz wird wiederum mit der Methode der kleinsten Quadrate (OLS), der klassischen LASSO-Methode sowie weiteren gängigen Bezugsmethoden verglichen. Weiterhin sind die entsprechenden Vorhersagen der Windenergie zu ermitteln. Die Güte der betrachteten Modelle wird im Rahmen der Schätzung sowie anhand der Genauigkeit bei der Vorhersage beurteilt und diskutiert. Es werden Prognosen für die nachfolgenden 48 h bestimmt. Abschließ end werden die Windgeschwindigkeitsvorhersagen mit Hilfe der Nennleistungskennlinie in Windenergie transformiert. Hierbei wird ebenfalls die Genauigkeit der Vorhersage beurteilt.
Abstract
The importance of wind power as well as wind speed predictions increases with energy transition in Germany. Accurate short and medium term wind speed and wind energy predictions are essential in different fields of economy. Implementation and application of new wind forecasting models helps to increase the benefit of wind power. This paper deals with applications of the periodic seasonal vector autoregressive prediction model (VAR). Moreover, the threshold autoregressive conditional heteroscedasticity (TARCH) is considered. The introduced model is estimated using iteratively reweighted least absolute shrinkage and selection (LASSO). This method is compared to ordinary least squares (OLS) estimation, a pure LASSO approach as well as several benchmark models. In addition, wind energy predictions are computed. Moreover, both the in-sample performance and their out-of-sample accuracy are discussed. The findings are summarized in an overview of different time series wind speed prediction models and their accuracy. We provide the forecasting performance up to 48 h. Finally, we cover the problem of transforming wind speed to wind energy.
Notes
Das Azimut bzw. die Himmelsrichtung bezeichnet den Winkel im Koordinatensystem des Horizonts und in dem hier betrachteten Fall den geographischen Norden. Deshalb entspricht null Grad der Windrichtung Norden und folglich steigt der Winkel im Uhrzeigersinn.
Auf die Darstellung der Ergebnisse für die Windkraftanlage 2 bis 4 wird verzichtet, da die restlichen Abbildungen dieselben Ergebnisse aufzeigen.
References
Aguera-Perez A, Palomares-Salas JC, Gonzalez de la Rosa JJ, Moreno-Munoz A (2013) Spatial persistence in wind analysis. J Wind Eng Ind Aerod 119:48–52
Akaike H (1974) A new look at the statistical model identification. Automatic Control. IEEE Transactions on 6(19):716–723
Ambach D, Croonenbroeck C (2015) Space-time short- to medium-term wind speed forecasting. Statistical Methods and Applications. Published Online.
Ambach D, Schmid W (2015) Periodic and long range dependent models for high frequency wind speed data. Energy 82(0):277–293
Amjady N, Keynia F, Zareipour H (2011) Wind power prediction by a new forecast engine composed of modified hybrid neural network and enhanced particle swarm optimization. Sustainable Energy. IEEE Transactions on 2(3):265–276
Bazilevs Y, Hsu M, Scott M (2012) Isogeometric fluid-structure interaction analysis with emphasis on non-matching discretizations, and with application to wind turbines. Comput Methods Appl Mech Eng 249–252:28–41
Bentamy A, Croize-Fillon D (2014) Spatial and temporal characteristics of wind and wind power off the coasts of brittany. Renew Energ 66:670–679
Berkhout V, Faulstich S, Göorg P, Hahn B, Linke K, Neuschäfer M, Pfaffel S, Rafik K, Rohrig K, Rothkegel R, Zieße M (2013) Wind energy report germany 2013. Fraunhofer-Institut für Windenergie und Energiesystemtechnik-IWES-Kassel
Bouette JC, Chassagneux JF, Sibai D, Terron R, Charpentier A (2006) Wind in Ireland: long memory or seasonal effect? Stoch Env Res Risk A 20(3):141–151
Burton T, Sharpe D, Jenkins N, Bossanyi E (2001) Wind energy handbook. Wiley, New York
De Boor C (1978) A practical guide to splines. Mathematics of Computation Efron B, Hastie T, Johnstone I, Tibshirani R (2004) Least angle regression. Ann Stat 32(2):407–499
Efron B, Hastie T, Johnstone I, Tibshirani R (2004) Least angle regression. Ann Stat 32(2):407–499
Eilers PH, Marx BD (1996) Flexible smoothing with B-splines and penalties. Stat Sci 11:89–102
EuropeanWind Energy Association EWEA (2014) Wind in power, 2013 european statistics, 2014. EWEA, Brussels
Ewing BT, Kruse JB, Schroeder JL (2006) Time series analysis of wind speed with timevarying turbulence. Environmetrics 17(2):119–127
Giebel G, Brownsword R, Kariniotakis G, Denhard M, Draxl C (2011) The state-of-theart in short-term prediction of wind power. Tech. rep., ANEMOS.plus, Ris DTU, Wind Energy Division
Glosten L, Jagannathan R, Runkle D (1993) On the relation between the expected value and the volatility of the nominal excess return on stocks. J Finance 48:1779 1801
Gneiting T, Larson K, Westrick K, Genton MG, Aldrich E (2006) Calibrated probabilistic forecasting at the stateline wind energy center: the regime-switching space-time method. J Amer Stat Assoc 101(475):968–979
Gneiting T, Balabdaoui F, Raftery AE (2007) Probabilistic forecasts, calibration and sharpness. J Royal Stat Soc Series B 69(2):243–268
Hamilton J (1994) Time Series Analysis. Princeton University Press, Princeton
Haslett J, Raftery AE (1989) Space-time modelling with long-memory dependence: assessing ireland’s wind power resource. Appl Stat 30(1):1–50
Hering AS, Genton MG (2010) Powering up with space-time wind forecasting. J Amer Stat Assoc 105(489):92–104
Jung J, Broadwater RP (2014) Current status and future advances for wind speed and power forecasting. Renew Sust Energ Rev 31:762–777
Lange M, Focken U (2006) Physical approach to short-term wind power prediction. Springer, Berlin
Le Guyader C, Apprato D, Gout C (2014) Spline approximation of gradient fields: applications to wind velocity fields. Math Comput Simul 97:260–279
Lei M, Shiyan L, Chuanwen J, Hongling L, Zhang Y (2009) A review on the forecasting of wind speed and generated power. Renew Sust Energ Rev 13:915 920
Madsen H, Pinson P, Kariniotakis G, Nielsen HA, Nielsen TS (2005) Standardizing the performance evaluation of shortterm wind power prediction models. Wind Eng 29(6):475–489
Mbamalu G, El-Hawary M (1993) Load forecasting via suboptimal seasonal autoregressive models and iteratively reweighted least squares estimation. Power Systems. IEEE Transactions on 8(1):343–348
Nielsen TS, Joensen A, Madsen H, Landberg L, Giebel G (1998) A new reference for wind power forecasting. Wind Eng 1(1):29–34
Ren Y, Zhang X (2010) Subset selection for vector autoregressive processes via adaptive lasso. Stat Probab Lett 80(23):1705–1712
\(\hat{S}\)altyte Benth J, \(\hat{S}\)altyte L (2011) Spatial-temporal model for wind speed in lithuania. J Appl Stat 38(6):1151–1168
Santos-Alamillos F, Pozo-Vzquez D, Ruiz-Arias JA, Lara-Fanego V, Tovar-Pescador J (2014) A methodology for evaluating the spatial variability of wind energy resources: application to assess the potential contribution of wind energy to baseload power. Renew Energ 69:147–156
Soman SS, Zareipour H, Malik O, Mandal P (2010) A review of wind power and wind speed forecasting methods with different time horizons. In: North American Power Symposium (NAPS), 2010, IEEE, S 1–8
Taylor JW, McSharry PE, Buizza R (2009) Wind power density forecasting using ensemble predictions and time series models. Energy Conversion, IEEE Transactions on 24(3):775–782
Thapar V, Agnihotri G, Sethi V (2011) Critical analysis of methods for mathematical modelling of wind turbines. Renew Energ 36:3166–3177
Whittle P (1951) Hypothesis testing in time series analysis. Almquist and Wicksell, Uppsala
World Wind Energy Association WWEA (2014) 2014 half year report. WWEA, Bonn
Wu YK, Hong JS (2007) A literature review of wind forecasting technology in the world. In: Power Tech, 2007 IEEE Lausanne, IEEE, S 504–509
Zhu X, Bowman KP, Genton MG et al (2014a) Incorporating geostrophic wind information for improved space-time short-term wind speed forecasting. Ann Appl Stat 8(3):1782–1799
Zhu X, Genton MG, Gu Y, Xie L (2014b) Space-time wind speed forecasting for improved power system dispatch. Test 23(1):1–25
Ziel F (2015) Iteratively reweighted adaptive lasso for conditional heteroscedastic time series with applications to AR-ARCH type processes. arXiv preprint arXiv:150206557
Ziel F, Steinert R, Husmann S (2015) Efficient modeling and forecasting of electricity spot prices. Energy Econ 47:98–111
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ambach, D., Garthoff, R. Vorhersagen der Windgeschwindigkeit und Windenergie in Deutschland. AStA Wirtsch Sozialstat Arch 10, 15–36 (2016). https://doi.org/10.1007/s11943-016-0177-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11943-016-0177-1