SEARCH · Search NASA

Results for “Choice model”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

All sources NASA NTRS DOE OSTI

NASA NTRS2003

Air Cargo Transportation Route Choice Analysis

Using a unique feature of air cargo transshipment data in the Northeast Asian region, this paper identifies the critical factors that determine the transshipment route choice. Taking advantage of the variations in the transport characteristics in each origin-destination airports pair, the paper uses a discrete choice model to describe the transshipping route choice decision made by an agent (i.e., freight forwarder, consolidator, and large shipper). The analysis incorporates two major factors, monetary cost (such as line-haul cost and landing fee) and time cost (i.e., aircraft turnaround time, including loading and unloading time, custom clearance time, and expected scheduled delay), along with other controls. The estimation method considers the presence of unobserved attributes, and corrects for resulting endogeneity by use of appropriate instrumental variables. Estimation results find that transshipment volumes are more sensitive to time cost, and that the reduction in aircraft turnaround time by 1 hour would be worth the increase in airport charges by more than $1000. Simulation exercises measures the impacts of alternative policy scenarios for a Korean airport, which has recently declared their intention to be a future regional hub in the Northeast Asian region. The results suggest that reducing aircraft turnaround time at the airport be an effective strategy, rather than subsidizing to reduce airport charges.

Obashi, Hiroshi↗

NASA NTRSDate not supplied

A New Era of H-O-C-S Magma Solubility Modeling: Better, Faster, Stronger

H 2 O, CO 2 , and S are the most abundant volatiles in magmatic systems and are critical to understanding magma storage, phase equilibria, and volcanic eruptions. Models that consider all three of these components, however, may not allow for critical examination and adjustment of assumptions underlying the model, or provide benchmark testing or extensible interfaces. Thus, understanding why models produce different results can be challenging. We have gathered authors of established (D-Compress) and recent (VolFe, EVo, Sulfur_X, MAGEC) H-O-C-S volatile solubility models to work together to understand how and why our models diverge. We present a series of benchmark basalt degassing scenarios revealing that often understated model assumptions such as fO 2 buffer equations, fO 2 -Fe 3+ /ΣFe relationships, and even major element normalization routines have outsized effects on model results. All models consider S 2- and S 6+ melt species but with different approaches to sulfate/sulfide capacities, partition coefficients, and species fugacities, leading to divergence in the evolution of modeled gas compositions, melt S and Fe speciation, and fO 2 , with the extent of divergence depending on melt composition. Such scenarios enable meaningful intercomparison of existing models and lay the groundwork for a user-friendly yet powerful solubility modeling framework. Given our wealth of existing solubility literature, we suggest that the field of magmatic volatiles should focus now on the creation of modern tools and the modular implementation of existing model equations or methods, and that the evaluation of code usability, transparency, and benchmarking should be codified pillars of the peer-review process. As an example of such an endeavor, we present early work coupling these sulfur solubility models with VESIcal, an extensible and rigorously tested python library containing seven existing H 2 O-CO 2 solubility models. VESIcal includes the ability to extract, edit, and even interchange assumptions underlying any model. For example, users may combine or swap separately published H 2 O, CO 2 , and S models, as well as underlying model choices, such as Equations of State and redox models.

volatiles in magmas↗

NASA NTRS1984

On Choosing Between Two Probabilistic Choice Sub-models in a Dynamic Multitask Environment

An independent random utility model based on Thurstone's Theory of Comparative Judgment and a constant utility model based on Luce's Choice Axiom are reviewed in detail. Predictions from the two models are shown to be equivalent under certain restrictions on the distribution of the underlying random process. Each model is applied as a stochastic choice submodel in a dynamic, multitask, environment. Resulting choice probabilities are nearly identical, indicating that, despite their conceptual differences, neither model may be preferred over the other based solely on its predictive capability.

Soulsby, E. P.↗

NASA NTRS2015

The AgMIP GRIDded Crop Modeling Initiative (AgGRID) and the Global Gridded Crop Model Intercomparison (GGCMI)

Climate change is a significant risk for agricultural production. Even under optimistic scenarios for climate mitigation action, present-day agricultural areas are likely to face significant increases in temperatures in the coming decades, in addition to changes in precipitation, cloud cover, and the frequency and duration of extreme heat, drought, and flood events (IPCC, 2013). These factors will affect the agricultural system at the global scale by impacting cultivation regimes, prices, trade, and food security (Nelson et al., 2014a). Global-scale evaluation of crop productivity is a major challenge for climate impact and adaptation assessment. Rigorous global assessments that are able to inform planning and policy will benefit from consistent use of models, input data, and assumptions across regions and time that use mutually agreed protocols designed by the modeling community. To ensure this consistency, large-scale assessments are typically performed on uniform spatial grids, with spatial resolution of typically 10 to 50 km, over specified time-periods. Many distinct crop models and model types have been applied on the global scale to assess productivity and climate impacts, often with very different results (Rosenzweig et al., 2014). These models are based to a large extent on field-scale crop process or ecosystems models and they typically require resolved data on weather, environmental, and farm management conditions that are lacking in many regions (Bondeau et al., 2007; Drewniak et al., 2013; Elliott et al., 2014b; Gueneau et al., 2012; Jones et al., 2003; Liu et al., 2007; M¨uller and Robertson, 2014; Van den Hoof et al., 2011;Waha et al., 2012; Xiong et al., 2014). Due to data limitations, the requirements of consistency, and the computational and practical limitations of running models on a large scale, a variety of simplifying assumptions must generally be made regarding prevailing management strategies on the grid scale in both the baseline and future periods. Implementation differences in these and other modeling choices contribute to significant variation among global-scale crop model assessments in addition to differences in crop model implementations that also cause large differences in site-specific crop modeling (Asseng et al., 2013; Bassu et al., 2014).

climate↗

NASA NTRS2015

Simulation of Cold Flow in a Truncated Ideal Nozzle with Film Cooling

Flow transients during rocket start-up and shut-down can lead to significant side loads on rocket nozzles. The capability to estimate these side loads computationally can streamline the nozzle design process. Towards this goal, the flow in a truncated ideal contour (TIC) nozzle has been simulated for a range of nozzle pressure ratios (NPRs) aimed to match a series of cold flow experiments performed at the NASA MSFC Nozzle Test Facility. These simulations were performed with varying turbulence model choices and with four different versions of the TIC nozzle model geometry, each of which was created with a different simplification to the test article geometry.

Braman, Kalen↗

DOE OSTI2026

Fisher Forecasting for the DESC with $\texttt{Augur}$

The Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) has begun its ten-year survey of the entire visible southern hemisphere. To ensure robust cosmological measurements, computationally inexpensive investigations of modeling choices must be made to gauge the performance of proposed cosmological analyses. In this paper, we introduce the $\texttt{Augur}$ tool of the Dark Energy Science Collaboration (DESC), which provides Fisher forecasts for cosmological inference for the LSST using software frameworks designed for DESC science. We test the pipeline by comparing it to forecasts produced by external code and direct sampling of the posterior via nested sampling methods, finding good agreement between all methods. We additionally investigate a range of modeling and hyperparameter choices for a 3$\times$2pt investigation in harmonic space, providing users with diagnostics to obtain reliable forecasts. $\texttt{Augur}$ will be continually updated to be compatible with the other tools in the DESC software ecosystem as additional probes and functionality become available.

Rogozenski, Paul [Carnegie Mellon U.; Arizona U.] ↗

NASA NTRS2006

Gravity Modeling for Variable Fidelity Environments

Aerospace simulations can model worlds, such as the Earth, with differing levels of fidelity. The simulation may represent the world as a plane, a sphere, an ellipsoid, or a high-order closed surface. The world may or may not rotate. The user may select lower fidelity models based on computational limits, a need for simplified analysis, or comparison to other data. However, the user will also wish to retain a close semblance of behavior to the real world. The effects of gravity on objects are an important component of modeling real-world behavior. Engineers generally equate the term gravity with the observed free-fall acceleration. However, free-fall acceleration is not equal to all observers. To observers on the sur-face of a rotating world, free-fall acceleration is the sum of gravitational attraction and the centrifugal acceleration due to the world's rotation. On the other hand, free-fall acceleration equals gravitational attraction to an observer in inertial space. Surface-observed simulations (e.g. aircraft), which use non-rotating world models, may choose to model observed free fall acceleration as the gravity term; such a model actually combines gravitational at-traction with centrifugal acceleration due to the Earth s rotation. However, this modeling choice invites confusion as one evolves the simulation to higher fidelity world models or adds inertial observers. Care must be taken to model gravity in concert with the world model to avoid denigrating the fidelity of modeling observed free fall. The paper will go into greater depth on gravity modeling and the physical disparities and synergies that arise when coupling specific gravity models with world models.

Madden, Michael M.↗

NASA NTRS1994

Observation model and parameter partials for the JPL VLBI parameter estimation software MODEST, 19 94

This report is a revision of the document Observation Model and Parameter Partials for the JPL VLBI Parameter Estimation Software 'MODEST'---1991, dated August 1, 1991. It supersedes that document and its four previous versions (1983, 1985, 1986, and 1987). A number of aspects of the very long baseline interferometry (VLBI) model were improved from 1991 to 1994. Treatment of tidal effects is extended to model the effects of ocean tides on universal time and polar motion (UTPM), including a default model for nearly diurnal and semidiurnal ocean tidal UTPM variations, and partial derivatives for all (solid and ocean) tidal UTPM amplitudes. The time-honored 'K(sub 1) correction' for solid earth tides has been extended to include analogous frequency-dependent response of five tidal components. Partials of ocean loading amplitudes are now supplied. The Zhu-Mathews-Oceans-Anisotropy (ZMOA) 1990-2 and Kinoshita-Souchay models of nutation are now two of the modeling choices to replace the increasingly inadequate 1980 International Astronomical Union (IAU) nutation series. A rudimentary model of antenna thermal expansion is provided. Two more troposphere mapping functions have been added to the repertoire. Finally, corrections among VLBI observations via the model of Treuhaft and lanyi improve modeling of the dynamic troposphere. A number of minor misprints in Rev. 4 have been corrected.

Sovers, O. J.↗

DOE OSTI2025

Impacts of benchmarking choices on inferred model skill of the Arctic–Boreal terrestrial carbon cycle

Abstract Land surface models require continuous validation against observations to improve and reduce simulation uncertainty. However, inferred model performance can be heavily influenced by subjective choices made in the selection and application of observational data products. A key area often misrepresented by models is the Arctic–Boreal region, which is a potential tipping point region in Earth’s climate system due to large permafrost carbon stocks that are vulnerable to release with climate warming. We use the International Land Model Benchmarking (ILAMB) framework to evaluate how the model skill of TRENDY-v9 models varies based on the choice of observational-based benchmark and how benchmarks are applied in model evaluation. This analysis uses global datasets integrated into ILAMB and new, regionally-specific observational products from the Arctic–Boreal Vulnerability Experiment. Our results cover the overall time period of 1979–2019 and show that model scores can vary substantially depending on the data product applied, with higher model scores indicating better model performance against observations. The lowest model scores occur when benchmarked against regional, compared to global, datasets. We also evaluate observed and modeled functional relationships between ecosystem respiration and air temperature and between gross primary production and precipitation. Here, we find that the magnitude and shape of the responses are strongly impacted by the choice of observational dataset and the approach used to construct the functional relationship benchmark. These results suggest that model evaluation studies could conclude a false sense of model skill if only using a single benchmark data product or if not applying regional data products when performing a regional model analysis. Collectively, our findings highlight the influence of benchmarking choices on model evaluation and point to the need for benchmarking guidelines when assessing model skill.

Poe, Jeralyn (ORCID:0000000318495278)↗

DOE OSTI2025

Design Optimization of a Criticality Experiment for the Molten Chloride Reactor Experiment Facility

Neutronics simulations of Molten Chloride Fast Reactors have quantifiable biases that arise from nuclear data, modeling choices, or numerical methods. The multiphysics nature of molten salt reactors makes it challenging to disentangle neutronics modeling biases from biases originating from other physical phenomena. In comparison to a mock-up reactor, criticality experiments can specifically assess the neutronics modeling bias while limiting multiphysics effects. The criticality experiment must be neutronically representative of the full-scale reactor to be valuable. Here, in this paper, we describe the design of a criticality experiment to validate only the neutronics of TerraPower’s Molten Chloride Reactor Experiment (MCRE) and its criticality safety upset scenarios. The proposed experiment uses different chlorine-containing materials to maximize its similarity to the MCRE. The design process uses a constrained Bayesian optimization algorithm to investigate different objective functions that use covariance information for 35 Cl nuclear data. The experiments could reduce the nuclear data–induced uncertainty in k eff of the MCRE from 2161 to 886 pcm. They would also increase the upper subcritical limit of the MCRE criticality safety upset scenario from 0.94101 to 0.94476 when using the WHISPER analysis framework.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

NASA NTRS2003

On the Computation of Space-Time Correlations by Large-Eddy Simulation

Numerical comparisons in decaying isotropic turbulence suggest that there exist discrepancies in time correlations evaluated by DNS and LES using eddy-viscosity-type SGS models. This is consistent with the previous observations in forced isotropic turbulence. Therefore, forcing is not the main cause of the discrepancies. Comparisons among different SGS models in the LES also indicate that the model choice affects the time correlations in the LES. The multi-scale LES method using the dynamic Smagorinsky model on the small scale equation is the most accurate of the all models, the classic Smagorinsky model is the least accurate and the dynamic Smagorinsky model and spectral eddy viscosity model give intermediate results with small differences. The generalized sweeping hypothesis implies that time correlations in decaying isotropic turbulence are mainly determined by the instantaneous energy spectra and sweeping velocities. The analysis based on the sweeping hypothesis explains the discrepancies in our numerical simulations: the LES overpredicts the decorrelation time scales because the sweeping velocities are smaller than the DNS values, and underpredicts the magnitudes of time correlations because the energy spectrum levels are lower than the DNS ones. Since the sweeping velocity is determined by the energy spectra, one concludes that an accurate prediction of the instantaneous energy spectra guarantees the accuracy of time correlations. An analytical expression of sound power spectra based on Lighthill's theory and the quasi-normal closure assumption suggests that the sound power spectra are sensitive to errors in time correlations. Small errors in time correlations can cause significant errors in the sound power spectra, which exhibit a sizable drop at moderate to high frequencies accompanied by a shift of the peaks to lower frequencies. Based on the above analysis, two possible ways to improve the acoustic power spectrum predictions can be considered. The first is to construct better SGS models to improve the LES accuracy for time correlations. The second is to remedy the temporal statistics of the Lighthill stress tensor in order to "recover" the contribution from the unresolved scales in LES to time correlations.

He, Guo-Wei↗

NASA NTRS2014

Natural Ocean Carbon Cycle Sensitivity to Parameterizations of the Recycling in a Climate Model

Sensitivities of the oceanic biological pump within the GISS (Goddard Institute for Space Studies ) climate modeling system are explored here. Results are presented from twin control simulations of the air-sea CO2 gas exchange using two different ocean models coupled to the same atmosphere. The two ocean models (Russell ocean model and Hybrid Coordinate Ocean Model, HYCOM) use different vertical coordinate systems, and therefore different representations of column physics. Both variants of the GISS climate model are coupled to the same ocean biogeochemistry module (the NASA Ocean Biogeochemistry Model, NOBM), which computes prognostic distributions for biotic and abiotic fields that influence the air-sea flux of CO2 and the deep ocean carbon transport and storage. In particular, the model differences due to remineralization rate changes are compared to differences attributed to physical processes modeled differently in the two ocean models such as ventilation, mixing, eddy stirring and vertical advection. GISSEH(GISSER) is found to underestimate mixed layer depth compared to observations by about 55% (10 %) in the Southern Ocean and overestimate it by about 17% (underestimate by 2%) in the northern high latitudes. Everywhere else in the global ocean, the two models underestimate the surface mixing by about 12-34 %, which prevents deep nutrients from reaching the surface and promoting primary production there. Consequently, carbon export is reduced because of reduced production at the surface. Furthermore, carbon export is particularly sensitive to remineralization rate changes in the frontal regions of the subtropical gyres and at the Equator and this sensitivity in the model is much higher than the sensitivity to physical processes such as vertical mixing, vertical advection and mesoscale eddy transport. At depth, GISSER, which has a significant warm bias, remineralizes nutrients and carbon faster thereby producing more nutrients and carbon at depth, which eventually resurfaces with the global thermohaline circulation especially in the Southern Ocean. Because of the reduced primary production and carbon export in GISSEH compared to GISSER, the biological pump efficiency, i.e., the ratio of primary production and carbon export at 75 m, is half in the GISSEH of that in GISSER, The Southern Ocean emerges as a key region where the CO2 flux is as sensitive to biological parameterizations as it is to physical parameterizations. The fidelity of ocean mixing in the Southern Ocean compared to observations is shown to be a good indicator of the magnitude of the biological pump efficiency regardless of physical model choice.

Antarctic Ocean↗

DOE OSTI2024

A Model for Optimally Allocating Curbside Space Among Competing Uses

The emergence of various new forms of urban mobility services in recent years is leading to new pressures on curbside space. Municipalities, the entities typically responsible for managing the curbside, are in many instances handling these growing pressures by reallocating portions of the curbside away from traditional uses (such as metered and residential parking) in favor of uses such as ridehailing, scooter and bike-share corrals. As yet, however, such actions are being undertaken on an ad-hoc basis, due to the rapidly growing complexity of the curbside and the lack of standard analytical approaches. This lack of analytical capability is due to the traditional focus of transportation network modeling being focused predominantly on the interaction of supply and demand on links and nodes, with limited focus on link edges (the curbside). In this paper we address this research need by proposing a framework for modeling inter-modal competition for curbside space, inspired by the classical Bid-Rent Model of urban land use, intended to support curb managers to move towards maximizing the aspects of economic welfare that relate to curb access. In the bi-level model, choices made by the curbside manager impact travelers’ mode choices, and vice versa. We then present a simple numerical case study to demonstrate the properties of the proposed model, showing its tractability, flexibility, and intuitive sensitivity to systematic variation in inputs. The framework demonstrates the type of adaptive and evolving approach needed to maximize benefits from increasingly dynamic curb management strategies. The paper concludes with a brief discussion of future research needs to advance this line of inquiry.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2026

Dispersive and nondispersive 𝐾-matrix formalisms

The modeling of coupled-channel effects has become increasingly important due to the availability of highly precise data for a large variety of hadronic (re)scattering processes. The 𝐾-matrix is a powerful, yet comparatively simple, method to describe scattering amplitudes, including coupled-channel effects, with the aim of interpreting experimental data. Throughout the literature, a range of dispersive and nondispersive 𝐾-matrix methods are employed. Here, we compare the dispersive and nondispersive formulations in the context of the N/D method. It is shown that the methods are equivalent in the physical region under 𝐾-matrix reparametrization. Differences away from the physical region are examined. Applications to synthetic data are used to illustrate the effects of model choices concerning form factors and the application of dispersion relations, with the goal of clarifying best practices. We find no clear preference with regard to dispersive modeling. In contrast, we find that interpretational ambiguity of the bare model parameters—and even of the form of the bare model—is endemic, and recommend a thorough sampling of data and model spaces to assess conclusion robustness.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2026

TDCOSMO XXIII. Measurement of the Hubble constant from the doubly lensed quasar HE 1104−1805

Time-delay cosmography leverages strongly lensed quasars to measure the Universe’s current expansion rate, H 0 , independently from other methods. The latest TDCOSMO milestone measurement primarily used quadruply lensed quasars for their mass profile constraints. However, doubly lensed quasars, being more abundant and offering precise time delays, could expand the sample by a factor of 5, significantly advancing towards a 1% precision measurement of H 0 . We present the first TDCOSMO analysis of a doubly imaged source, HE 1104−1805, including the measurement of the four necessary ingredients. First, by combining 17 years of data from the SMARTS, Euler, and WFI telescopes, we measured a time delay of 176.3 +11.4 −10.3 days. Second, using MUSE data, we extracted stellar velocity dispersion measurements in three radial bins with 5% to 13% precision. Third, employing F160W HST imaging for lens modelling and marginalising over various modelling choices, we measured the Fermat potential difference between the images. Fourth, using wide-field imaging, we measured the convergence added by objects not included in the lens modelling. By combining these four ingredients, we measured the time delay distance and the angular diameter distance to the deflector, favouring a power-law mass model over a baryonic and dark matter composite model. The measurement was performed blindly to prevent experimenter bias and resulted in a Hubble constant of H 0 = +5.8 −5.0 × λ int km s −1 Mpc −1 , where λ int is the internal mass sheet degeneracy parameter. This is in agreement with the TDCOSMO-2025 milestone and its precision for λ int = 1 is comparable to that obtained with the best-observed quadruply lensed quasars (4–6%). This work is a stepping stone towards a precise measurement of H 0 using a large sample of doubly lensed quasars, supplementing the current sample. The next TDCOSMO milestone paper will include this system in its hierarchical analysis, constraining λ int and H 0 jointly with multiple lenses.

cosmological parameters↗

DOE OSTI2026

SPT-3G D1: A Measurement of Secondary Cosmic Microwave Background Anisotropy Power

We report new measurements of millimeter-wave temperature power spectra in the angular multipole range $1700 \le \ell \le 11,000$ (wavelengths $13^\prime \gtrsim λ\gtrsim 2^\prime$). We use two years of data in three observing bands centered near 95, 150, and 220 GHz from the SPT-3G receiver on the South Pole Telescope that cover a 1646 deg$^2$ region of the Southern sky. Using the measured power spectra, we present constraints on the thermal and kinematic Sunyaev-Zel'dovich (SZ) effects, radio galaxies, and cosmic infrared background (CIB). We find that inferred SZ powers are dependent on the detailed modeling of the thermal SZ-CIB correlation, and to a lesser extent on the assumed angular dependence of the SZ spectra. We report constraints for simulation-based model templates as well as fits where the angular dependencies of the SZ and CIB power spectra are allowed to vary. In the latter case at $\ell=3000$, we find thermal SZ power at 143 GHz of $D_{3000}^{\rm tSZ} = 4.91\pm0.37\, μ{\rm K}^2$ and kinematic SZ power of $D_{3000}^{\rm kSZ} =1.75\pm0.86\, μ{\rm K}^2$. We use the measured kinematic SZ power to estimate the duration of reionization, noting that the reionization inferences are sensitive to the model choices and assumed level of homogeneous kinematic SZ power from the late-time universe. We find a 95% limit on the duration from an ionization fraction of 25% to 75% of $Δ^{50} z_{\rm re} <\,3.8$ based on a semi-analytic model, or a limit on the duration from an ionization fraction of 5% to 95% of $Δ^{90} z_{\rm re} <\,6.1$ based on the AMBER simulations.

Chaubal, P. [Melbourne U.]↗

NASA NTRS1984

Dynamics of the Venus ionosphere - A two-dimensional model study

A two-dimensional model of the ionosphere of Venus which simulates ionospheric dynamics by self-consistently solving the plasma equations of motion, including the inertial term, in finite difference form has been constructed. The model, which is applied over the solar zenith angle range extending from 60 to 140 deg and the altitude range 100 to 480 km, simulates the measured horizontal velocity field quite satisfactorily. The ion density field is somewhat overestimated on the dayside, because of the choice model neutral atmosphere, and underestimated on the nightside, because of setting the ionopause height at too low an altitude. It is concluded that solar photoionization on the dayside and ion recombination on the nightside are the processes mainly responsible for accelerating the plasma to the observed velocities. The plasma flow appears to be sufficient to maintain the nightside ionosphere at or near the observed median level of ion densities.

Whitten, R. C.↗

DOE OSTI2026

Benchmarking of three DWM-based wake models at below-rated wind speeds

Wind turbine wake models are essential tools for predicting power losses and structural loads in wind farms. Among these, the dynamic wake meandering (DWM) model, included as a recommended approach in the International Electrotechnical Commission design standard, is a widely used engineering-fidelity method that balances accuracy and computational cost. This study compares the performance of three DWM-based wake model implementations (from the Technical University of Denmark, the National Renewable Energy Laboratory, and the Institute for Energy Technology) under below-rated wind speed conditions. Model predictions of wake flow, power output, and structural loads for a four-turbine row are evaluated across different ambient turbulence levels and wind-direction misalignments and compared against high-fidelity large-eddy simulation results. All three models captured the overall wake evolution and mean turbine performance with reasonable accuracy; their predicted time-averaged thrust and power were typically within 5 %–10 % of the large-eddy simulation benchmark. However, notable differences emerged in wake structure and unsteady load predictions, with discrepancies increasing for turbines further downstream. These differences highlight the importance of modelling choices such as wake summation and turbulence treatment, which strongly influence power-deficit and fatigue-load predictions. Comparison with large-eddy simulations reveals each approach's strengths and weaknesses, indicating where improvements are needed. Overall, the findings point to specific refinements for DWM models to improve their fidelity, ultimately enabling more robust wake predictions for wind farm design and operation.

17 WIND ENERGY↗