Performance assessment of five MCP models proposed for the estimation of long-term wind turbine power outputs at a target site using three machine learning techniques

Abstract Various models based on measure-correlate-predict (MCP) methods have been used to estimate the long-term wind turbine power output (WTPO) at target sites for which only short-term meteorological data are available. The MCP models used to date share the postulate that the influence of air density variation is of little importance, assume the standard value of 1.225 kg m−3 and only consider wind turbines (WTs) with blade pitch control. A performance assessment is undertaken in this paper of the models used to date and of newly proposed models. Our models incorporate air density in the MCP model as an additional covariable in long-term WTPO estimation and consider both WTs with blade pitch control and stall-regulated WTs. The advantages of including this covariable are assessed using different functional forms and different machine learning algorithms for their implementation (Artificial Neural Network, Support Vector Machine for regression and Random Forest). The models and the regression techniques used in them were applied to the mean hourly wind speeds and directions and air densities recorded in 2014 at ten weather stations in the Canary Archipelago (Spain). Several conclusions were drawn from the results, including most notably: (a) to clearly show the notable effect of air density variability when estimating WTPOs, it is important to consider the functional ways in which the features air density and wind speed and direction intervene, (b) of the five MCP models under comparison, the one that separately estimates wind speeds and air densities to later predict the WTPOs always provided the best mean absolute error, mean absolute relative error and coefficient of determination metrics, independently of the target station and type of WT under consideration, (c) the models which used Support Vector Machines (SVMs) for regression or random forests (RFs) always provided better metrics than those that used artificial neural networks, with the differences being statistically significant (5% significance) for most of the cases assessed, (d) no statistically significant differences were found between the SVM- and RF-based models.

[1]  Kenneth J. Berry,et al.  A Chronicle of Permutation Statistical Methods , 2014 .

[2]  Stefan Emeis,et al.  Wind Energy Meteorology , 2013 .

[3]  A. Schlosser,et al.  Characterizing wind power resource reliability in southern Africa , 2016 .

[4]  Massoud Tabesh,et al.  A comparison between performance of support vector regression and artificial neural network in prediction of pipe burst rate in water distribution networks , 2014, KSCE Journal of Civil Engineering.

[5]  Stefan Emeis.  Wind energy meteorology : atmopsheric physics for wind power generation , 2013 .

[6]  Ying Zhi Liu,et al.  Effect of Air Density on Output Power of Wind Turbine , 2013 .

[7]  Nurulkamal Masseran,et al.  Evaluating wind power density models and their statistical properties , 2015 .

[8]  Bernhard Schölkopf,et al.  Learning with kernels , 2001 .

[9]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[10]  Martin Raubal,et al.  Assessment of the wake effect on the energy production of onshore wind farms using GIS , 2014 .

[11]  Moncho Gómez-Gesteira,et al.  Sensitivity of the WRF model wind simulation and wind energy production estimates to planetary boundary layer parameterizations for onshore and offshore areas in the Iberian Peninsula , 2014 .

[12]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[13]  Max Kuhn,et al.  Building Predictive Models in R Using the caret Package , 2008 .

[14]  P. Good.  Permutation, Parametric, and Bootstrap Tests of Hypotheses , 2005 .

[15]  Barbara Hammer,et al.  A Note on the Universal Approximation Capability of Support Vector Machines , 2003, Neural Processing Letters.

[16]  José M. Matías,et al.  Use of Bayesian networks classifiers for long-term mean wind turbine energy output estimation at a potential wind energy conversion site , 2011 .

[17]  José M. Matías,et al.  Influence of the input layer signals of ANNs on wind power estimation for a target site: A case study , 2011 .

[18]  William H. Press,et al.  Numerical recipes , 1990 .

[19]  J. A. Carta,et al.  Comparison of several measure-correlate-predict models using support vector regression techniques to estimate wind power densities. A case study , 2017 .

[20]  Ali Lahouar,et al.  Hour-ahead wind power forecast based on random forests , 2017 .

[21]  Hossam Faris,et al.  A Comparison between Regression, Artificial Neural Networks and Support Vector Machines for Predicting Stock Market Index , 2015 .

[22]  Maria Grazia De Giorgi,et al.  Comparison Between Wind Power Prediction Models Based on Wavelet Decomposition with Least-Squares Support Vector Machine (LS-SVM) and Artificial Neural Network (ANN) , 2014 .

[23]  James F. Manwell,et al.  A new method for improved hub height mean wind speed estimates using short-term hub height data , 2010 .

[24]  Fernando Castellano,et al.  Comparison of feature selection methods using ANNs in MCP-wind speed methods. A case study , 2015 .

[25]  A. Zeileis.  Econometric Computing with HC and HAC Covariance Matrix Estimators , 2004 .

[26]  Tamer Khatib,et al.  A novel hybrid model for hourly global solar radiation prediction using random forests technique and firefly algorithm , 2017 .

[27]  Michael C. Brower,et al.  Wind Resource Assessment: A Practical Guide to Developing a Wind Project , 2012 .

[28]  Cha Zhang,et al.  Ensemble Machine Learning: Methods and Applications , 2012 .

[29]  G. Stenchikov,et al.  Wind resource characterization in the Arabian Peninsula , 2016 .

[30]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[31]  Jack Chin Pang Cheng,et al.  Identifying the influential features on the regional energy use intensity of residential buildings based on Random Forests , 2016 .

[32]  J. A. Carta,et al.  Comparison between ANNs and linear MCP algorithms in the long-term estimation of the cost per kWh produced by a wind turbine at a candidate site: A case study in the Canary Islands , 2011 .

[33]  J. A. Carta,et al.  Use of finite mixture distribution models in the analysis of wind energy in the Canarian Archipelago , 2007 .

[34]  Sungmoon Jung,et al.  Weighted error functions in artificial neural networks for improved wind energy potential estimation , 2013 .

[35]  Bri-Mathias Hodge,et al.  A hybrid measure-correlate-predict method for long-term wind condition assessment , 2014 .

[36]  J. A. Carta,et al.  A review of measure-correlate-predict (MCP) methods used to estimate long-term wind characteristics at a target site , 2013 .

[37]  J. A. Carta,et al.  A review of wind speed probability distributions used in wind energy analysis: Case studies in the Canary Islands , 2009 .

[38]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[39]  Mostafa Modiri-Delshad,et al.  Development of an enhanced parametric model for wind turbine power curve , 2016 .

[40]  A. Immanuel Selvakumar,et al.  A comprehensive review on wind turbine power curve modeling techniques , 2014 .

[41]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[42]  Erwan Scornet,et al.  A random forest guided tour , 2015, TEST.