Virtual fatigue diagnostics of wake-affected wind turbine via Gaussian Process Regression

Abstract We propose a data-driven model to predict the short-term fatigue Damage Equivalent Loads (DEL) on a wake-affected wind turbine based on wind field inflow sensors and/or loads sensors deployed on an adjacent up-wind wind turbine. Gaussian Process Regression (GPR) with Bayesian hyperparameters calibration is proposed to obtain a surrogate from input random variables to output DELs in the blades and towers of the up-wind and wake-affected wind turbines. A sensitivity analysis based on the hyperparameters of the GPR and Kullback-Leibler divergence is conducted to assess the effect of different input on the obtained DELs. We provide qualitative recommendations for a minimal set of necessary and sufficient input random variables to minimize the error in the DEL predictions on the wake-affected wind turbine. Extensive simulations are performed comprising different random variables, including wind speed, turbulence intensity, shear exponent and inflow horizontal skewness. Furthermore, we include random variables related to the blades lift and drag coefficients with direct impact on the rotor aerodynamic induction, which governs the evolution and transport of the meandering wake. In addition, different spacing between the wind turbines and Wohler exponents for calculation of DELs are considered. The maximum prediction normalized mean squared error, obtained in the tower base DELs in the fore-aft direction of the wake affected wind turbine, is less than 4%. In the case of the blade root DELs, the overall prediction error is less than 1%. The proposed scheme promotes utilization of sparse structural monitoring (loads) measurements for improving diagnostics on wake-affected turbines.

[1]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[2]  Patrick Guillaume,et al.  C ONTINUOUS FATIGUE ASSESSMENT OF AN OFFSHORE WIND TURBINE USING A LIMITED NUMBER OF VIBRATION SENSORS , 2015 .

[3]  J. Kaimal,et al.  Spectral Characteristics of Surface-Layer Turbulence , 1972 .

[4]  Nikolay Krasimirov Dimitrov,et al.  Wind farm layout optimization with load constraints using surrogate modelling , 2020, Journal of Physics: Conference Series.

[5]  Andrew Kusiak,et al.  Prediction, operations, and condition monitoring in wind energy , 2013 .

[6]  Haitao Liu,et al.  When Gaussian Process Meets Big Data: A Review of Scalable GPs , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[7]  Nikolay Krasimirov Dimitrov,et al.  Surrogate models for parameterized representation of wake‐induced loads in wind farms , 2019, Wind Energy.

[8]  Paola Annoni,et al.  Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index , 2010, Comput. Phys. Commun..

[9]  Gunner Chr. Larsen,et al.  On atmospheric stability in the dynamic wake meandering model , 2014 .

[10]  Eleni Chatzi,et al.  Multivariate GP-VAR models for robust structural identification under operational variability , 2020, Probabilistic Engineering Mechanics.

[11]  A surrogate model approach for associating wind farm load variations with turbine failures , 2020 .

[12]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[13]  Anand Natarajan,et al.  Model of wind shear conditional on turbulence and its impact on wind turbine loads , 2015 .

[14]  J. Jonkman,et al.  Definition of a 5-MW Reference Wind Turbine for Offshore System Development , 2009 .

[15]  Junyoung Park,et al.  Predicting Wind Turbine Power and Load Outputs by Multi-task Convolutional LSTM Model , 2018, 2018 IEEE Power & Energy Society General Meeting (PESGM).

[16]  Deep Unsupervised Learning for Condition Monitoring and Prediction of High Dimensional Data with Application on Windfarm SCADA Data , 2019, Model Validation and Uncertainty Quantification, Volume 3.

[17]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[18]  Keith Worden,et al.  Fault diagnosis of wind turbine structures using decision tree learning algorithms with big data , 2018, Safety and Reliability – Safe Societies in a Changing World.

[19]  Konstantinos Tatsis,et al.  RELIABILITY PREDICTION OF FATIGUE DAMAGE ACCUMULATION ON WIND TURBINES SUPPORT STRUCTURES , 2017 .

[20]  John Dalsgaard Sørensen,et al.  Impact of uncertainty in airfoil characteristics on wind turbine extreme loads , 2015 .

[21]  David Infield,et al.  Online wind turbine fault detection through automated SCADA data analysis , 2009 .

[22]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[23]  Keith Worden,et al.  A Performance Monitoring Approach for the Novel Lillgrund Offshore Wind Farm , 2015, IEEE Transactions on Industrial Electronics.

[24]  J. Zico Kolter,et al.  Optimal Planning and Learning in Uncertain Environments for the Management of Wind Farms , 2015, J. Comput. Civ. Eng..

[25]  Keith Worden,et al.  Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo , 2017, Front. Built Environ..

[26]  Federico Tagliatti.  Investigation of Wind Turbine Fatigue Loads under Wind Farm Control: Analysis of Field Measurements , 2019 .

[27]  David Charles Maniaci,et al.  Dynamic Wake Meandering Model Comparison with Varying Fidelity Models for Wind Turbine Wake Prediction , 2015 .

[28]  Hyungki Shin,et al.  A study of the wake effects on the wind characteristics and fatigue loads for the turbines in a wind farm , 2015 .

[29]  H. F. Veldkamp,et al.  Chances in wind energy: A probalistic approach to wind turbine fatigue design , 2006 .

[30]  James M. W. Brownjohn,et al.  Gaussian Process Time-Series Models for Structures under Operational Variability , 2017, Front. Built Environ..

[31]  B. Sudret,et al.  Hierarchical Kriging for multi-fidelity aero-servo-elastic simulators-application to extreme loading on wind turbines , 2017 .

[32]  Rick Damiani,et al.  The Unsteady Aerodynamics Module For FAST8 , 2019 .

[33]  Matthew A. Lackner,et al.  Implementing the Dynamic Wake Meandering Model in the NWTC Design Codes , 2014 .

[34]  Dmitri Tcherniak,et al.  Gaussian process models for mitigation of operational variability in the structural health monitoring of wind turbines , 2020 .

[35]  Jason Jonkman,et al.  FAST User's Guide , 2005 .

[36]  John Aasted Sørensen,et al.  Wind turbine site-specific load estimation using artificial neural networks calibrated by means of high-fidelity load simulations , 2018, Journal of Physics: Conference Series.

[37]  Jens Jakob Wedel-Heinen,et al.  A consistent turbulence formulation for the dynamic wake meandering model in the atmospheric boundary layer , 2013 .

[38]  Anand Natarajan,et al.  From SCADA to lifetime assessment and performance optimization: how to use models and machine learning to extract useful insights from limited data , 2019 .

[39]  Helge Madsen Aagaard,et al.  Dynamic wake meandering modeling , 2007 .

[40]  Søren Juhl Andersen,et al.  Analytical model for the power–yaw sensitivity of wind turbines operating in full wake , 2020 .

[41]  Nicolai Cosack,et al.  Fatigue load monitoring with standard wind turbine signals , 2010 .

[42]  Torben J. Larsen,et al.  Mapping Wind Farm Loads and Power Production - A Case Study on Horns Rev 1 , 2016 .

[43]  Jinkyoo Park,et al.  Physics-induced graph neural network: An application to wind-farm power estimation , 2019, Energy.

[44]  Jacob Berg,et al.  From wind to loads: wind turbine site-specific load estimation with surrogate models trained on high-fidelity load databases , 2018, Wind Energy Science.

[45]  J. Sørensen,et al.  Determination of Wind Turbine Near-Wake Length Based on Stability Analysis , 2014 .

[46]  Gunner Chr. Larsen,et al.  In search for a canonical design ABL stability class for wind farm turbines , 2016 .