[vc_empty_space][vc_empty_space]
Multi-objective kriging-based optimization for high-fidelity wind turbine design
Zuhal L.R.a, Faza G.A.a, Palar P.S.a,b, Shimoyama K.b
a Faculty of Mechanical and Aerospace Engineering, Bandung Institute of Technology, Indonesia
b Institute of Fluid Science, Tohoku University, Sendai, 980-8577, Japan
[vc_row][vc_column][vc_row_inner][vc_column_inner][vc_separator css=”.vc_custom_1624529070653{padding-top: 30px !important;padding-bottom: 30px !important;}”][/vc_column_inner][/vc_row_inner][vc_row_inner layout=”boxed”][vc_column_inner width=”3/4″ css=”.vc_custom_1624695412187{border-right-width: 1px !important;border-right-color: #dddddd !important;border-right-style: solid !important;border-radius: 1px !important;}”][vc_empty_space][megatron_heading title=”Abstract” size=”size-sm” text_align=”text-left”][vc_column_text]© 2019 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.In this paper, we present the implementation of multi-objective Kriging-based optimization for high-fidelity wind turbine design. Specifically, a multi-objective Bayesian optimization (MOBO) technique based on expected hypervolume improvement and high-fidelity computational fluid dynamics are utilized to solve the wind turbine design optimization problem. The primary aim is to solve multi-objective wind turbine design optimization problem using a high-fidelity CFD solver without the need to obtain gradient information; although such information can be incorporated into MOBO if available. The radial basis function-based mesh deformation technique is applied to simultaneously deform the mesh and wind turbine geometry. This set of methodologies is then applied to the optimization of NREL Phase VI wind turbine where we applied 70 mesh deformation control points. The multi-objective optimization aims to maximize the torque production and minimize the blade volume of the NREL Phase VI wind turbine. By using this procedure, we obtained a set of non-dominated solutions that dominate the baseline design in terms of both volume and torque production. From the results, we observe that the torque-optimized and volume-optimized geometry yields 6% increase in torque and 7% decrease in blade volume, respectively.[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Author keywords” size=”size-sm” text_align=”text-left”][vc_column_text]Baseline design,Bayesian optimization,Gradient informations,Mesh deformation,Nondominated solutions,Optimized geometries,Radial basis functions,Turbine design optimizations[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Indexed keywords” size=”size-sm” text_align=”text-left”][vc_column_text][/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Funding details” size=”size-sm” text_align=”text-left”][vc_column_text]Part of this work was funded through Riset P3MI ITB program, which we gratefully acknowledge. Part of the work was also carried out under the Collaborative Research Project of the Institute of Fluid Science, Tohoku University.[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”DOI” size=”size-sm” text_align=”text-left”][vc_column_text]https://doi.org/10.2514/6.2019-0539[/vc_column_text][/vc_column_inner][vc_column_inner width=”1/4″][vc_column_text]Widget Plumx[/vc_column_text][/vc_column_inner][/vc_row_inner][/vc_column][/vc_row][vc_row][vc_column][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][/vc_column][/vc_row]