arXiv subjects

Navid Zehtabiyan-Rezaie

Publications and source records attributed to Navid Zehtabiyan-Rezaie.

7 recordsLinked to original sources

arXiv2024

A progressive data-augmented RANS model for enhanced wind-farm simulations

The development of advanced simulation tools is essential, both presently and in the future, for improving wind-energy design strategies, paving the way for a complete transition to sustainable solutions. The Reynolds-averaged Navier-Stokes (RANS) models are pivotal in enhancing our comprehension of the complex flow within and around wind farms and, hence, improving their capacity to accurately model turbulence within this context is a vital research goal. The enhancement is essential for a precise prediction of wake recovery and for capturing intricate flow phenomena such as secondary flows of Prandtl's second kind behind the turbines. To reach these objectives, here, we propose a progressive data-augmentation approach. We first incorporate the turbine-induced forces in the turbulent kinetic energy equation of the widely used $k-ω\text{SST}$ model. Afterward, we utilize data from large-eddy simulations to progressively enhance the Reynolds-stress prediction of this baseline model, accurately capturing the evolution of eddy viscosity in the wake, as well as the emergence of secondary flows. We then apply the optimized model to two unseen cases with distinct layouts and conduct a comparative analysis focusing on the obtained quantities such as normalized streamwise velocity deficit, turbulence intensity, and power output. We also examine the success rate of the augmented model in predicting the secondary flows in the wake region. We also evaluate the performance of the augmented model in predicting wake characteristics by comparing it with wind-tunnel measurement data. Our comparisons and validations demonstrate the superior performance of the progressive data-augmented model over the standard version in all cases considered in this study.

physics.flu-dyn↗

arXiv2024

Wind-farm power prediction using a turbulence-optimized Gaussian wake model

In this study, we present an improved formulation for the wake-added turbulence to enhance the accuracy of intra-farm and farm-to-farm wake modeling through analytical frameworks. Our goal is to address the tendency of a commonly used formulation to overestimate turbulence intensity within wind farms and to overcome its limitations in predicting the streamwise evolution of turbulence intensity beyond them. To this end, we utilize high-fidelity data and adopt an optimization technique to derive an optimized functional form of the wake-added turbulence. We then integrate the achieved formulation with a widely used Gaussian wake model to study various intra-farm and farm-to-farm scenarios. The outcomes reveal that the new methodology effectively addresses the overestimation of power in both standalone wind farms and those impacted by upstream counterparts. Our new approach meets the need for accurate and lightweight models, ensuring the effective coexistence of wind farms within clusters as the wind-energy capacity rapidly expands.

physics.flu-dyn↗

arXiv2024

An extended $k-\varepsilon$ model for wake-flow simulation of wind farms

The Reynolds-averaged Navier-Stokes approach coupled with the standard $k-\varepsilon$ model is widely utilized for wind-energy applications. However, it has been shown that the standard $k-\varepsilon$ model overestimates the turbulence intensity in the wake region and, consequently, overpredicts the power output of the waked turbines. This study focuses on the development of an extended $k-\varepsilon$ model by incorporating an additional term in the turbulent kinetic energy equation. This term accounts for the influence of turbine-induced forces, and its formulation is derived through an analytical approach. To assess the effectiveness of the proposed model, we begin by analyzing the evolution of normalized velocity deficit and turbulence intensity in the wake region, and the normalized power of the waked turbines. This investigation involves a comparison of the predictions against results from large-eddy simulations in three validation cases with different layouts. We then simulate a wind farm consisting of 30 wind turbines and conduct a comparative analysis between the model-predicted normalized streamwise velocity and wind-tunnel measurements. Finally, to conclude our assessment of the proposed model, we apply it to the operational wind farm of Horns Rev 1 and evaluate the obtained normalized power with the results from large-eddy simulations. The comparisons and validations conducted in this study prove the superior performance of the extended $k-\varepsilon$ model compared to the standard version.

physics.flu-dyn↗

arXiv2023

A short note on turbulence characteristics in wind-turbine wakes

Analytical wake models need formulations to mimic the impact of wind turbines on turbulence level in the wake region. Several correlations can be found in the literature for this purpose, one of which is the formula proposed in A. Crespo, J. Hernandez, Turbulence characteristics in wind-turbine wakes, Journal of Wind Engineering and Industrial Aerodynamics 61 (1) (1996) 71 - 85, which relates the added turbulence to the induction factor of the turbine, ambient turbulence intensity, and normalized distance from the rotor through an equation with one coefficient and three exponents for the effective parameters. Misuse of this formula with an incorrect exponent for the ambient turbulence intensity is propagating in the literature. In this note, we implement the original and the incorrect formulation of turbine-induced added turbulence in a Gaussian wake model to quantify its impact by studying the Horns Rev 1 wind farm. The results reveal that the turbulence intensity and the normalized power of the waked turbines predicted by the wake model with the correct and the incorrect implementation of turbine-induced added turbulence correlation have a difference equal to 1.94% and 3.53%, respectively, for an ambient turbulence intensity of 7.7%. For an ambient turbulence intensity of 4%, these discrepancies grow to 2.7% and 4.95%.

physics.flu-dyn↗

arXiv2023

Physics-guided machine learning for wind-farm power prediction: Toward interpretability and generalizability

With the increasing amount of available data from simulations and experiments, research for the development of data-driven models for wind-farm power prediction has increased significantly. While the data-driven models can successfully predict the power of a wind farm with similar characteristics as those in the training ensemble, they generally do not have a high degree of flexibility for extrapolation to unseen cases in contrast to the physics-based models. In this paper, we focus on data-driven models with improved interpretability and generalizability levels that can predict the performance of turbines in wind farms. To prepare the datasets, several cases are defined based on the layouts of operational wind farms, and massive computational fluid dynamics simulations are performed. The extreme gradient boosting algorithm is used afterward to build models, which have turbine-level geometric inputs in combination with the efficiency from physics-based models as the features. After training, to analyze the models' capability in generalization, their predictions for the unseen cases with different operating conditions, inflow turbulence levels, and wind-farm layouts are compared to the Park model and an empirical-analytical Gaussian wake model. Results show that the physics-guided machine-learning models outperform both physics-based models showing a high degree of generalizability, and the machine is not sensitive to the choice of the physics-based guide model.

physics.flu-dyn↗

arXiv2022

Data-driven quantification of model-form uncertainty in Reynolds-averaged simulations of wind farms

Computational fluid dynamics using the Reynolds-averaged Navier-Stokes (RANS) remains the most cost-effective approach to study wake flows and power losses in wind farms. The underlying assumptions associated with turbulence closures are one of the biggest sources of errors and uncertainties in the model predictions. This work aims to quantify model-form uncertainties in RANS simulations of wind farms at high Reynolds numbers under neutrally stratified conditions by perturbing the Reynolds stress tensor through a data-driven machine-learning technique. To this end, a two-step feature-selection method is applied to determine key features of the model. Then, the extreme gradient boosting algorithm is validated and employed to predict the perturbation amount and direction of the modeled Reynolds stress toward the limiting states of turbulence on the barycentric map. This procedure leads to a more accurate representation of the Reynolds stress anisotropy. The data-driven model is trained on high-fidelity data obtained from large-eddy simulation of a specific wind farm, and it is tested on two other (unseen) wind farms with distinct layouts to analyze its performance in cases with different turbine spacing and partial wake. The results indicate that, unlike the data-free approach in which a uniform and constant perturbation amount is applied to the entire computational domain, the proposed framework yields an optimal estimation of the uncertainty bounds for the RANS-predicted quantities of interest, including the wake velocity, turbulence intensity, and power losses in wind farms.

physics.flu-dyn↗

arXiv2022

Data-driven fluid mechanics of wind farms: A review

With the growing number of wind farms over the last decades and the availability of large datasets, research in wind-farm flow modeling - one of the key components in optimizing the design and operation of wind farms - is shifting towards data-driven techniques. However, given that most current data-driven algorithms have been developed for canonical problems, the enormous complexity of fluid flows in real wind farms poses unique challenges for data-driven flow modeling. These include the high-dimensional multiscale nature of turbulence at high Reynolds numbers, geophysical and atmospheric effects, wake-flow development, and incorporating wind-turbine characteristics and wind-farm layouts, among others. In addition, data-driven wind-farm flow models should ideally be interpretable and have some degree of generalizability. The former is important to avoid a lack of trust in the models with end-users, while the most popular strategy for the latter is to incorporate known physics into the models. This article reviews a collection of recent studies on wind-farm flow modeling covering both purely data-driven and physics-guided approaches. We provide a thorough analysis of their modeling approach, objective, and methodology, and specifically focus on the data utilized in the reviewed works.

physics.flu-dyn↗