Adaptive modelling and forecasting of offshore wind power fluctuations with Markov‐switching autoregressive models

Journal of Forecasting - Tập 31 Số 4 - Trang 281-313 - 2012
Pierre Pinson1, Henrik Madsen2
1European Centre for Medium‐range Weather forecasts and Technical University of Denmark
2Technical University of Denmark, DTU Informatics, 2800 Kgs. Lyngby, Denmark

Tóm tắt

ABSTRACTWind power production data at temporal resolutions of a few minutes exhibit successive periods with fluctuations of various dynamic nature and magnitude, which cannot be explained (so far) by the evolution of some explanatory variable. Our proposal is to capture this regime‐switching behaviour with an approach relying on Markov‐switching autoregressive (MSAR) models. An appropriate parameterization of the model coefficients is introduced, along with an adaptive estimation method allowing accommodation of long‐term variations in the process characteristics. The objective criterion to be recursively optimized is based on penalized maximum likelihood, with exponential forgetting of past observations. MSAR models are then employed for one‐step‐ahead point forecasting of 10 min resolution time series of wind power at two large offshore wind farms. They are favourably compared against persistence and autoregressive models. It is finally shown that the main interest of MSAR models lies in their ability to generate interval/density forecasts of significantly higher skill. Copyright © 2010 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1002/env.753

10.1260/030952407780811384

Chatfield C., 2004, The Analysis of Time‐Series: An Introduction

CollingsIB RydénT.1998.A new maximum likelihood gradient technique algorithm for on‐line hidden Markov model identification. InProceedings of IEEE International Conference on Acoustics Speech Signal Processing Seattle WA.

10.1109/78.340791

10.1111/j.1468-2354.2007.00418.x

10.1016/j.rser.2007.01.015

10.1002/hec.654

10.1023/A:1013722215208

10.1214/009053604000000021

Ea Energy Analyses.2007.50% wind power in Denmark by 2025: English summary. Technical Report Ea Energy Analyses Copenhagen Denmark. Available:http://www.windpower.org[16 July 2010].

10.1016/S0167-6105(01)00222-7

GiebelG KariniotakisG BrownswordR.2003. The state of the art in short‐term prediction of wind power: a literature overview. Technical report ANEMOS EU project deliverable report D1.1. Available:http://www.anemos‐project.eu[16 July 2010].

10.1198/016214506000000456

10.1111/j.1467-9868.2007.00587.x

10.1111/1467-9868.00308

10.2307/1912559

10.1002/we.82

10.1198/jasa.2009.ap08117

Hoerl AE., 1962, Application of ridge analysis to regression problems, Chemical Engineering Progress, 58, 54

10.1111/j.1467-9892.1994.tb00206.x

10.1016/S0005-1098(96)00168-9

10.1260/030952403322770959

10.1007/978-3-642-51684-9

LeGlandF MevelL.1997.Recursive estimation in hidden Markov models. InProceedings of IEEE Conference on Decision and Control San Diego CA.

Ljung L, 1983, Theory and Practice of Recursive Estimation

Madsen H., 2007, Time‐Series Analysis, 10.1201/9781420059687

10.1260/030952405776234599

10.1109/MSPEC.2002.1038567

10.1016/j.csda.2007.06.027

NielsenHAA NielsenTS MadsenH BadgerJ GiebelG LandbergL SattlerK VoulundL TøftingJ.2006.From wind ensembles to probabilistic information about future wind power production: results from an actual application. InProceedings of IEEE PMAPS Conference: Probabilistic Methods Applied to Power Systems Stockholm Sweden.

PinsonP.2006.Estimation of the uncertainty in wind power forecasting. PhD thesis Ecole des Mines de Paris. Available:http://pastel.paristech.org[16 July 2010].

10.1002/we.309

10.1002/we.230

10.1016/j.jweia.2008.03.010

10.1080/00207390500433103

10.1016/S0304-4149(96)00114-7

10.1016/j.ijforecast.2005.05.003

10.1016/j.ijforecast.2008.08.008

10.1198/016214502753479464

10.1198/jasa.2009.ap08615

10.1109/TPWRS.2007.901615

10.1002/we.246

10.1109/97.774870

Stone M., 1974, Cross‐validation and assessment of statistical predictions (with discussion), Journal of the Royal Statistical Society B, 36, 111

Tikhonov AN, 1977, Solutions of Ill‐Posed Problems

Toth Z, 2003, Forecast Verification: A Practitioner's Guide in Atmospheric Science, 137

10.1175/2009JAMC2058.1