Enabling Innovation in Wind Turbine Design Using Artificial Intelligence

DOE OSTI · 1869434

Enabling Innovation in Wind Turbine Design Using Artificial Intelligence

View original source ↗Read source document ↗Export metadata ↓

Abstract

The Inverse Network Transformations for Efficient Generation of Robust Airfoil and Turbine Enhancements (INTEGRATE) project is developing a new inverse-design capability for wind turbine rotors using invertible neural networks. This artificial intelligence (AI)-based technology can capture complex nonlinear aerodynamic effects 100 times faster than alternative design approaches.

Explore related subjects

aerodynamics airfoil blade inverse design machine learning MATHEMATICS AND COMPUTING,WIND ENERGY neural networks wind energy

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

2022-05-17. Enabling Innovation in Wind Turbine Design Using Artificial Intelligence. https://www.osti.gov/biblio/1869434

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2024

Heliostat Consortium: Gap Analysis on State of the Art in Wind Load Design

Wind loads are a major driver of heliostat cost. Standardized methods and tools are needed for a more detailed understanding of the static and dynamic loads of a heliostat design. This will enable cost reduction of wind-dependent heliostats to avoid unnecessarily conservative heliostat designs and increase field efficiency and reliability to reduce the risk of component failures due to high-wind events. Gaps related to wind load include lack of site characterization for wind measurements, insufficient critical load cases for heliostat design, insufficient understanding of turbulence impacts on heliostat tracking error, lack of knowledge on wind load under various heliostat array configurations, and underexplored heliostat field wind-load reduction and operating strategies. Recommended pathway forward is to develop wind load and site characterization guidelines for heliostat design and develop heliostat field wind-load models with optical performance impacts.

aerodynamics↗

DOE OSTI2023

Aerodynamic Sensitivity of a Novel Data-Driven Airfoil Shape Representation Framework

We explore the aerodynamic implications of a novel data-driven separable shape tensor framework used to represent discrete airfoil shapes. In this study, we construct a data-driven parameter space defined by separable shape tensors and informed by tens of thousands of distinct airfoils. We use this design space to generate new airfoil designs to study parametric sensitivities with respect to various aerodynamic responses. We use a HAM2D RANS solver to approximate the lift, drag, and moment coefficients for the generated airfoils at two different angles-of-attack. We analyze the robustness and sensitivities of using the separable shape tensor design space by examining the coverage of the aerodynamic response space, uncovering low-dimensional polynomial ridge approximations, and computing various sensitivity metrics. The results show that the data-driven design space produce significant variation in target aerodynamic quantities and facilitate highly accurate approximations (R^2 > 0.96) of one- and two-dimensional structures in each aerodynamic response. This further reduces the effective dimension to enable simplified design and optimization tasks.

aerodynamics↗

NASA NTRS2023

Computational Study of Vortex-Induced Separation for a 5-Bladed Rotor

This work computationally investigated the rotor blade vortex-induced separation recently observed during an aerodynamic rotor test campaign in the NASA Langley Research Center 14- by 22-Foot Subsonic Tunnel. Two separate approaches (i.e., airfoil modification and blade tip modification) were studied to mitigate the vortex-induced separation. Low-fidelity tools based on blade element momentum theory were shown to mispredict the rotor inflow and were also shown to not capture the vortex-induced separation caused by perpendicular blade-vortex interaction. This misprediction was exploited to isolate the aerodynamic thrust deficit caused by the vortex-induced separation (i.e., 20%) from the thrust deficit due to inflow variation (i.e., 31%). High-fidelity tools were shown to reasonably predict aerodynamic forces within 13% and flow separation when compared to experimental results. The modified airfoil variant of the baseline rotor effectively mitigated the vortex-induced separation while the blade tip modified variant still showed separation, though the size and strength of the vortex was reduced. Acoustic predictions were underpredicted by 10 dB from preliminary measurements taken in the untreated wind tunnel. Broadband noise contributions from different rotor blade sections showed that self-noise due to flow separation and other turbulent boundary layer mechanisms was the dominant noise source for all three rotor cases, followed by blade-wake interaction noise caused by perpendicular blade-vortex interactions.

aerodynamics↗