Big data driven multi-objective predictions for offshore wind farm based on machine learning algorithms

Abstract This paper explores the big data driven multi-objective predictions for offshore wind farm based on machine learning. A data-driven prediction framework is proposed to predict the wind farm power output and structural fatigue. Unlike the existing methods that are normally based on analytical models, mainly focus on single objective and ignore the control contributions, the proposed framework uses the turbine control inputs, inflow wind velocity and directions as the predictor variables. It is constructed by training five typical machine learning approaches: the general regression neural network (GRNN), random forest (RF), support vector machine (SVM), gradient boosting regression (GBR) and recurrent neural network (RNN). The assessment of these approaches is based on the FLOw Redirection and Induction in Steady State (FLORIS) under 6 different scenarios. The test results in different cases are highly consistent with each other and validate that very minor accuracy differences exist among these approaches and they all can achieve the relative accuracy of around 99% or more, which is sufficiently accurate for practical applications. The RNN and SVM exhibit the best accuracy, and particularly the RNN has the best accuracy in thrust predictions. The results also demonstrate that the GRNN has the best computational efficiency.

[1]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[2]  Chih-Jen Lin,et al.  A Study on SMO-Type Decomposition Methods for Support Vector Machines , 2006, IEEE Transactions on Neural Networks.

[3]  F. Porté-Agel,et al.  A new analytical model for wind-turbine wakes , 2013 .

[4]  Junwei Wang,et al.  Representing conditional preference by boosted regression trees for recommendation , 2016, Inf. Sci..

[5]  Grzegorz Litak,et al.  Analytical analysis of the vibrational tristable energy harvester with a RL resonant circuit , 2019, Nonlinear Dynamics.

[6]  Andreas Bechmann,et al.  Evaluation of the wind direction uncertainty and its impact on wake modeling at the Horns Rev offshore wind farm , 2014 .

[7]  Yi-Shih Chung,et al.  Factor complexity of crash occurrence: An empirical demonstration using boosted regression trees. , 2013, Accident; analysis and prevention.

[8]  Miao He,et al.  A Spatio-Temporal Analysis Approach for Short-Term Forecast of Wind Farm Generation , 2014, IEEE Transactions on Power Systems.

[9]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[10]  Hui Liu,et al.  Comparison of two new ARIMA-ANN and ARIMA-Kalman hybrid methods for wind speed prediction , 2012 .

[11]  Ji Feng,et al.  Multi-Layered Gradient Boosting Decision Trees , 2018, NeurIPS.

[12]  Zhi Liu,et al.  Personalized Variable Gain Control With Tremor Attenuation for Robot Teleoperation , 2018, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[13]  Chae-Joo Moon,et al.  The study on prediction of power generation of offshore wind farm of western and southern coast utilizing offshore buoy meteorological observations data , 2012, 2012 IEEE Vehicle Power and Propulsion Conference.

[14]  Fernando Porté-Agel,et al.  A new analytical model for wind farm power prediction , 2015 .

[15]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[16]  Aoife Foley,et al.  Current methods and advances in forecasting of wind power generation , 2012 .

[17]  Tsutomu Kaizuka,et al.  The benefits of an asymmetric tri-stable energy harvester in low-frequency rotational motion , 2019, Applied Physics Express.

[18]  Donald F. Specht,et al.  Probabilistic neural networks and the polynomial Adaline as complementary techniques for classification , 1990, IEEE Trans. Neural Networks.

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

[20]  Andrew Kusiak,et al.  Wind farm power prediction: a data‐mining approach , 2009 .

[21]  Junyi Cao,et al.  Broadband tristable energy harvester: Modeling and experiment verification , 2014 .

[22]  Vojislav Kecman,et al.  Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised, and Unsupervised Learning , 2006, Studies in Computational Intelligence.

[23]  Wang Liang,et al.  Study of wind farm power output predicting model based on nonlinear time series , 2011, 2011 International Conference on Electrical Machines and Systems.

[24]  Jennifer Annoni,et al.  A tutorial on control-oriented modeling and control of wind farms , 2017, 2017 American Control Conference (ACC).

[25]  Gregor Giebel,et al.  The State-Of-The-Art in Short-Term Prediction of Wind Power. A Literature Overview , 2003 .

[26]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[27]  Stefan Krauter,et al.  Short term wind and energy prediction for offshore wind farms using neural networks , 2015, 2015 International Conference on Renewable Energy Research and Applications (ICRERA).

[28]  P.M.O. Gebraad.  Data-driven wind plant control , 2014 .

[29]  T. F. Pedersen,et al.  On wind turbine power performance measurements at inclined airflow , 2004 .

[30]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[31]  Sen Guo,et al.  A hybrid annual power load forecasting model based on generalized regression neural network with fruit fly optimization algorithm , 2013, Knowl. Based Syst..

[32]  A. Kusiak,et al.  Short-Term Prediction of Wind Farm Power: A Data Mining Approach , 2009, IEEE Transactions on Energy Conversion.

[33]  J Elith,et al.  A working guide to boosted regression trees. , 2008, The Journal of animal ecology.

[34]  S. Nahavandi,et al.  Prediction Intervals for Short-Term Wind Farm Power Generation Forecasts , 2013, IEEE Transactions on Sustainable Energy.