Local linear regression with adaptive orthogonal fitting for the wind power application

Abstract Short-term forecasting of wind generation requires a model of the function for the conversion of meteorological variables (mainly wind speed) to power production. Such a power curve is nonlinear and bounded, in addition to being nonstationary. Local linear regression is an appealing nonparametric approach for power curve estimation, for which the model coefficients can be tracked with recursive Least Squares (LS) methods. This may lead to an inaccurate estimate of the true power curve, owing to the assumption that a noise component is present on the response variable axis only. Therefore, this assumption is relaxed here, by describing a local linear regression with orthogonal fit. Local linear coefficients are defined as those which minimize a weighted Total Least Squares (TLS) criterion. An adaptive estimation method is introduced in order to accommodate nonstationarity. This has the additional benefit of lowering the computational costs of updating local coefficients every time new observations become available. The estimation method is based on tracking the left-most eigenvector of the augmented covariance matrix. A robustification of the estimation method is also proposed. Simulations on semi-artificial datasets (for which the true power curve is available) underline the properties of the proposed regression and related estimation methods. An important result is the significantly higher ability of local polynomial regression with orthogonal fit to accurately approximate the target regression, even though it may hardly be visible when calculating error criteria against corrupted data.

[1]  H. Madsen,et al.  From wind ensembles to probabilistic information about future wind power production -- results from an actual application , 2006, 2006 International Conference on Probabilistic Methods Applied to Power Systems.

[2]  Ben J. A. Kröse,et al.  A k-segments algorithm for finding principal curves , 2002, Pattern Recognit. Lett..

[3]  Risø National Laboratory Wind Energy Department , 2001 .

[4]  D. Ruppert,et al.  Nonparametric regression in the presence of measurement error , 1999 .

[5]  H. Aa,et al.  PREDICTION OF WIND POWER USING TIME-VARYING COEFFICIENT-FUNCTIONS , 2002 .

[6]  Sven-Erik Thor,et al.  Long‐term research and development needs for wind energy for the time frame 2000–2020 , 2002 .

[7]  A. J. Geurts.  On numerical stability , 1984 .

[8]  W. Cleveland,et al.  Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting , 1988 .

[9]  Gabrielle E. Kelly.  Robust Regression Estimators—The Choice of Tuning Constants , 1992 .

[10]  R. Tibshirani.  Principal curves revisited , 1992 .

[11]  J. Holst,et al.  Tracking Time-Varying Coefficient-Functions , 2000 .

[12]  Gregor Giebel,et al.  Wind Power Prediction using Ensembles , 2005 .

[13]  Gerhard Tutz,et al.  Local principal curves , 2005, Stat. Comput..

[14]  Gene H. Golub,et al.  An analysis of the total least squares problem , 1980, Milestones in Matrix Computation.

[15]  Jonathan D. Cryer,et al.  Time Series Analysis , 1986 .

[16]  Peter J. Huber,et al.  Robust Statistics , 2005, Wiley Series in Probability and Statistics.

[17]  T. Hastie,et al.  Principal Curves , 2007 .

[18]  P. Delicado.  Another Look at Principal Curves and Surfaces , 2001 .

[19]  L. Ljung,et al.  Fast calculation of gain matrices for recursive estimation schemes , 1978 .

[20]  Jianqing Fan,et al.  Nonparametric regression with errors in variables , 1993 .

[21]  B. Jonsson,et al.  Prediction with a linear regression model and errors in a regressor , 1994 .

[22]  Gene H. Golub,et al.  Matrix computations , 1983 .

[23]  Kai-Bor Yu,et al.  Recursive updating the eigenvalue decomposition of a covariance matrix , 1991, IEEE Trans. Signal Process..

[24]  Pierre Pinson,et al.  Estimation of the uncertainty in wind power forecasting , 2006 .

[25]  Phillip A. Regalia,et al.  On the numerical stability and accuracy of the conventional recursive least squares algorithm , 1999, IEEE Trans. Signal Process..

[26]  James D. Hamilton.  Time Series Analysis , 1994 .

[27]  B. Ripley,et al.  Robust Statistics , 2018, Wiley Series in Probability and Statistics.

[28]  Yaron A. Felus.  Application of Total Least Squares for Spatial Point Process Analysis , 2004 .

[29]  Bernard D. Flury,et al.  Self-Consistency: A Fundamental Concept in Statistics , 1996 .

[30]  J. Holst,et al.  Recursive forgetting algorithms , 1992 .

[31]  Ismael Sánchez,et al.  Short-term prediction of wind energy production , 2006 .

[32]  Henrik Madsen,et al.  Robust estimation of time-varying coefficient functions - Application to the modeling of wind power production , 2007 .

[33]  Jianqing Fan,et al.  Robust Non-parametric Function Estimation , 1994 .

[34]  P. D. de Groen.  An Introduction to Total Least Squares , 1996 .

[35]  P. D. Groen.  An Introduction to Total Least Squares , 1998, math/9805076.

[36]  M. Lange,et al.  Physical Approach to Short-Term Wind Power Prediction , 2005 .

[37]  Pierre Pinson,et al.  Standardizing the Performance Evaluation of Short-Term Wind Power Prediction Models , 2005 .

[38]  A. U.S..  Nonparametric regression in the presence of measurement error , 1999 .