Development and Validation of Hierarchical Methodology for the Numerical Simulation of the Flow Field around and in the Wake of Horizontal Axis Wind Turbines

A hierarchy of computational methods for Horizontal Axis Wind Turbine (HAWT - NREL Phase VI) flow field is proposed, focusing on rotor models for Reynolds- Averaged Navier-Stokes (RNS) simulations. Three models are systematically compared to determine their adequacy to capture performance and wake dynamics, and the trade-offs between accuracy and computational cost. The Moving Reference Frame (MRF) model is prescribed for detailed flow field studies, specifically at the root and blade tip. The Blade Element Model (BEM) does not reproduce the near wake region but successfully predicts the velocity deficit in the axisymmetric far wake, and the power and torque coefficients. The Actuator Disk Model (ADM) underestimates the velocity deficit in the far wake, but can be corrected to perform simulations of large wind farms. This methodology is applied to a canonical rotor and compared against experimental data. These models span three orders of magnitude in computational time and cost for the study of HAWT aerodynamic performance and wake interaction.

Wind Eng. Journal Paper M.Sc. Thesis (Ch. 2-3)

Experimental Validation of the General Numerical Methodology for Performance Analysis and Flow Field Characterization of a Full-Scale (i.e. DOE RM1) and a Lab-Scale (i.e. re-design of the DOE RM1) Horizontal Axis Marine Hydrokinetic (MHK) Turbine

This work is an experimental/numerical study of a full- and lab.-scale Horizontal Axis Marine Hydrokinetic (MHK) Turbine. The full-scale turbine is called the DOE Reference Model 1 (DOE RM1), which is a refrence MHK turbine model developed and published at National Renewable Energy Laboratory (NREL) in US. The lab.-scale model turbine is re-designed based on the DOE RM1, with a modified geometry to reproduce performance at the flume scale Reynolds numbers. These modifications were necessary to overcome the strong Reynolds number effect on the NACA 6 airfoil family used on the design, and therefore on the device performance in experimental analysis. The performance and wake structure of a the lab.-scale model was analyzed with measurements conducted on a 45:1 scale physical model of the modified design of the DOE RM1 rotor. The details of the rotor flow field and wake evolution are analyzed from numerical solution of the RANS equations solved around a computational model of both the full- and lab.-scale models turbine. A comparison between the experimental and numerical results is presented. These comparisons highlight the strengths as well as limitations of the experimental and numerical analysis for these types of MHK turbine characterizations. On a more general sense, these comparisons provide useful guidelines for developing a set of experimental flume scale data and to use it to validate numerical tools, and as pilot projects start to go in the water in the US, to perform a similar type of analysis and design validation of full scale devices.

METS Paper (Full-Scale DOE RM1) METS Paper (Lab.-Scale DOE RM1) Ph.D. Dissertation (Ch. 2)

Numerical characterization of Horizontal Axis Hydrokinetic Turbines Array

Hydrokinetic turbines, similarly to wind turbines, should be placed in the form of an array to have the potential for electricity production necessary to connect to the electrical grid and feed it at a commercial scale. Due to the confined nature of tidal sites, however, the spacing of turbines in an array of Marine Hydrokinetic (MHK) turbines is much more critical to the energy production and economic viability of the turbine farm than in wind energy installations, and needs to be highly optimized. This optimization process must maximize the efficiency of power generation while minimizing the capital cost of the infrastructure, the potential environmental effects of the array and the fatigue load on the turbines' structure. The initial step to perform this optimization process is to understand the flow field in the array and how the devices perform while interacting with this complex velocity field. This initial step also provides an in-depth understanding of the potential constraints for the array optimization process. A numerical methodology can then be developed to model the essential physics of the flow field needed for the array computations, in an time-efficient way compatible with the optimization of turbine array performance. This numerical methodology can be used to examine the effect of various constraints in this optimization process, such as the number of turbines per surface area, the number of turbine rows, ... .
In this part of my research the Blade Element Model (BEM), one of the models from the previously developed methodology in the context of the characterization of a single MHK turbine, is used for the flow field simulation in an array of turbines. The BEM of several turbines in different spatial configurations and operating at different set points is implemented to match experimental conditions in two- and three-turbine arrays in a research flume. The goal of this process is to further validate the application of the BEM model to turbine wake and performance characterization in an array.

Ph.D. Dissertation (Ch. 3)

Development and Validation of a Calibration Methodology for the Discrete Random Walk (DRW)

Simulation of heavy particles or sediment in turbulent flows has practical use in fields of industrial and environmental Multiphase Flows. These simulations rely on numerical models to simulate the interaction of these particles with turbulent eddies. The Discrete Random Walk (DRW) model is commonly used for simulation of particle dispersion in various turbulent flow fields and applications, such as interaction of particles with wake of a Marine Hydrokinetic (MHK) turbines. The previous studies confirmed and reported that there are still sources of uncertainty in the use of the DRW model in conjunction with RANS simulations of the turbulent carrier flow. One of the key issues for the DRW model in various applications is to determine the characteristic eddy lifetime scale, which controls the length of time over which a particle interacts with an eddy in the turbulent flow. We develop a simple methodology, based on G.I. Taylor’s classical dispersion theory, for determination of the characteristic eddy lifetime value required in the engineering predictions of particle transport for which detailed experimental data is not available. We experimentally validated our methodology and then applied it to a novel ocean engineering problem for which experimental data is lacking over a wide range of param- eters: transport of suspended sediment in a high Reynolds number channel flow with a MHK Turbine.

Ocean Eng. Journal Ph.D. Dissertation (Ch. 4)

Investigation on Potential Environmental Effect of Marine Hydrokinetic Turbines (e.g. MHK Turbine-Sedimentation Interaction)

In this paert of my research, the calibrated Discrete Random Walk (DRW) model is coupled with the RANS model, which provides the simulated flow field around and in the wake of the Marine Hydrokinetic (MHK) turbine (modeled via BEM) in the channel. The DRW model uses the RANS velocity field and turbulence variables to simulate the trajectory of particles in the flow field. The goal of this modeling is to understand the effect of the HAHT wake on the sedimentation process of suspended particles in the tidal channel.
These injected particles travel along the channel and interact with the velocity deficit in the turbine wake and the accelerated flow around the blades. The decelerated and accelerated flows affect the particle paths and the overall sedimentation process. In order to investigate the effect of the turbine wake on the sedimentation of the modeled particles, each of the particles is tracked from their initial injection position at the inlet as it is convected and sediments towards the bottom of the tidal channel. The sedimentation of particles with different Stokes number (i.e. sizes) in a channel with and without MHK turbine effect was compared against each other to reveals physics behind the MHK turbine-sedimentaion interaction.

Ocean Eng. Journal Ph.D. Dissertation (Ch. 4)