Surrogate-Based Design Optimization of a H-Darrieus Wind Turbine Comparing Classical Response Surface, Artificial Neural Networks, and Kriging

Document Type : Regular Article


Federal University of Itajubá, Itajubá, Minas Gerais, 37500-903, Brazil



Clean energy sources like wind energy have been receiving much attention, and great emphasis has been given to the design and optimization of horizontal axis wind turbines, but just as important are the vertical axis wind turbines that can be used for generating energy for small businesses, houses, and buildings. This article sought to study the optimal geometrical parameters of a H-Darrieus vertical axis wind turbine using surrogate-based optimization with three different types of surrogate models and compared them. Airfoil chord and thickness were chosen as the design variables and respective ranges set at 0.32-0.6 m and 0.04-0.16 m. All evaluations are carried out for a tip-speed ratio of 1.5. Three different surrogate models were used and compared, namely a quadratic polynomial response surface, an artificial neural network based on radial basis functions called Extreme Learning Machine and a Kriging interpolator. Surrogates were constructed based on an initial sample data distributed according to a full factorial design. A test set was designed to evaluate the accuracy of the surrogates. Both training and testing data sets were generated using 2D CFD modeling to reduce computational cost. From the test set, Extreme Learning Machine surrogate showed the smallest RMSE of 11.24%, followed by Kriging, at 17.64%, and Response Surface of 22.17%. For the optimal designs the same pattern ensued, with optimal power coefficient overestimated by 8.7% for the response surface surrogate, followed by 3.12% and 2.17% for the Kriging interpolator and the Extreme Learning Machine, respectively. Power coefficient curves comparing the three optimal geometries from each surrogate were calculated and plotted. Optimal turbine obtained from Kriging surrogate optimization process resulted in a 7.92% increase in the Cp, whilst Extreme Learning Machine and Response Surface resulted in a 7.86% and 4.29% increase, respectively, all when compared to baseline CFD model. Concluding guidelines are that the quadratic polynomial response surface may not be the best alternative when dealing with complex non-linear relationships as typically present in VAWT simulations. Superior techniques such as Extreme Learning Machine and Kriging could be more suitable for this application.


Ahmad, M., A. Shahzad, F. Akram, F. Ahmad and S. I. A. Shah (2022). Design optimization of Double-Darrieus hybrid vertical axis wind turbine. Ocean Engineering 254, 111171.##
ANP, Oil and NGL National Production in cubic meters. Available at: <https://
tou-52-71-em-relacao-a-2010>. Access on: January 14th, 2022.##
Bianchini, A., F. Balduzzi, P. Bachant, G. Ferrara and L. Ferrari (2017). Effectiveness of two-dimensional CFD simulations for Darrieus VAWTs: a combined numerical and experimental assessment. Energy Conversion and Management 136, 318-328.##
Bravo, R., S. Tullis and S. Ziada (2007). Performance testing of a small vertical-axis wind turbine. In Proceedings of the 21st Canadian Congress of Applied Mechanics, Mechanical Engineering Department, McMaster University, Canada.##
Cheng, B. and Y. Yao (2022). Design and optimization of a novel U-type vertical axis wind turbine with response surface and machine learning methodology. Energy Conversion and Management 273, 116409.##
Chou, P. Y. (1945). On the velocity correlations and the solution of the equations of turbulent fluctuation. Quarterly of Applied Mathematics 3(1), 38-54.##
Davidov, B. I. (1961). on the Statistical Dynamics of an Incompressible Fluid. Doklady Akademiya Nauk SSSR 136, 47.##
Elsakka, M. M., D. B. Ingham, L. Ma, M. Pourkashanian, G. H. Moustafa and Y. Elhenawy (2022). Response surface optimisation of vertical axis wind turbine at low wind speeds. Energy Reports 8, 10868-10880.##
Gosselin, R., G. Dumas and M. Boudreau (2016). Parametric study of H-Darrieus vertical-axis wind turbines using CFD simulations. Journal of Renewable and Sustainable Energy 8(5), 053301.##
GWEC, Global Wind Report 2021. Available at: <>.
 Access on: January 14th, 2022.##
Hansen, J. T., M. Mahak and I. Tzanakis (2021). Numerical modelling and optimization of vertical axis wind turbine pairs: A scale up approach. Renewable Energy 171, 1371-1381.##
Harlow, F. H. and P. I. Nakayama (1968). Transport of Turbulence Energy Decay Rate. Alamos Science Lab, University of California Report.##
Hashem, I. and M. H. Mohamed (2017). Aerodynamic performance enhancements of H-rotor Darrieus wind turbine. Energy 142, 531-545.##
Hashem, I. and B. Zhu (2021). Metamodeling-based parametric optimization of a bio-inspired Savonius-type hydrokinetic turbine. Renewable Energy 180, 560-576.##
Haykin, S. (1999). Neural Networks a comprehensive foundation. McMaster University. Pearson Education.##
Huang, G. and C. K. Siew (2004) Extreme Learning Machine: a new scheme of feedforward neural networks. IEEE International Joint Conference on Neural Networks 2, 985-990.##
Jang, H., Y. Hwang, I. Paek and S. Lim (2021). Performance evaluation and validation of h-darrieus small vertical axis wind turbine. International Journal of Precision Engineering and Manufacturing-Green Technology 8, 1687-1697.##
Jones, D. R. (2001). A taxonomy of global optimization methods based on response surfaces. Journal of Global Optimization 21, 345-383.##
Jones, W. P. and B. E. Launder (1972). The prediction of laminarization with a two-equation model of turbulence. International Journal of Heat and Mass Transfer 15, 301-314.##
Kim, C. K., S. Ali, S. M. Lee and C. M. Jang (2020). Blade optimization of a small vertical-axis wind turbine using the response surface method. Renewable Energy and Sustainable Buildings 801-812.##
Launder, B. E. and B. I. Sharma (1974). Application of energy dissipation model of turbulence to the calculation of flow near a spinning disc. Letters in Heat and Mass Transfer 1(2), 131-137.##
Lee, S. L. and S. Shin (2020). Wind turbine blade optimal design considering multi-parameters and response surface method. Energies 13(7).##
Ma, N., H. Lei, Z. Han, D. Zhou, Y. Bao, K. Zhang, L. Zhou and C. Chen (2018). Airfoil optimization to improve power performance of a high-solidity vertical axis wind turbine at a moderate tip speed ratio. Energy 150, 236-252.##
Manwell, J. F., J. G. McGowan and A. L. Rogers (2009) Wind Energy Explained, Theory, Design and Application. Wiley.##
Meana-Fernández, A., L. Díaz-Artos, J. M. Fernández Oro and S. Velardez-Suárez (2020). An optimized airfoil geometry for vertical-axis wind turbine applications. International Journal of Green Energy 17, 181-195.##
Montgomery, D. G. (2009). Design and Analysis of Experiments. Arizona State University. John Wiley & Sons, Inc. Ninth edition.##
Oh, S. (2020). Comparison of a response surface method and artificial neural network in predicting the aerodynamic performance of a wind turbine airfoil and its optimization. Applied Sciences 10(18), 6277.##
Pan, L., H. Xiao, Y. Zhang and Z. Shi (2020). Research on aerodynamic performance of j-type blade vertical axis wind turbine. In Chinese Control and Decision Conference, IEEE.##
Raul, V. and L. Leifsson (2021). Surrogate-based aerodynamic shape optimization for delaying airfoil dynamic stall using Kriging regression and infill criteria. Aerospace Science and Technology 111, 106555.##
Rezaeiha, A., I. Kalkman and B. Blocken (2017). CFD simulation of a vertical axis wind turbine operating at a moderate tip speed ratio: Guidelines for minimum domain size and azimuthal increment. Renewable Energy 107, 373-385.##
Wilcox, D. C. (2006). Turbulence Modeling for CFD. DCW Industries, Inc., La Canada, CA.##