Analysis of Wind Gust Data
Wind gust data were analyzed by statistical and mathematical procedures, developed for the bivariate gamma distribution. The results of the analysis are summarized.
Tubbs, J. D.↗
SEARCH · Engineering Papers
Results for “wind gust”
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.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Do Machine Learning Approaches Offer Skill Improvement for Short-Term Forecasting of Wind Gust Occurrence and Magnitude?
Abstract Wind gusts, and in particular intense gusts, are societally relevant but extremely challenging to forecast. This study systematically assesses the skill enhancement that can be achieved using artificial neural networks (ANNs) for forecasting of wind gust occurrence and magnitude. Geophysical predictors from the ERA5 reanalysis are used in conjunction with an autoregressive term in regression and ANN models with different predictors, and varying model complexity. Models are derived and assessed for the warm (April–September) and cold (October–March) seasons for three high passenger volume airports in the United States. Model uncertainty is assessed by deriving models for 1000 different randomly selected training (70%) and testing (30%) subsets. Gust prediction fidelity in independent test samples is critically dependent on inclusion of an autoregressive term. Gust occurrence probabilities derived using five-layer ANNs exhibit consistently higher fidelity than those from regression models and shallower ANNs. Inclusion of the autoregressive term and increasing the number of hidden layers in ANNs from 1 to 5 also improve the model performance for gust magnitudes (lower RMSE, increased correlation, and model standard deviations that more closely approximate observed values). Deeper ANNs (e.g., 20 hidden layers) exhibit higher skill in forecasting strong (17–25.7 m s −1 ) and damaging (≥25.7 m s −1 ) wind gusts. However, such deep networks exhibit evidence of overfitting and still substantially underestimate (by 50%) the frequency of strong and damaging wind gusts at the three airports considered herein. Significance Statement Improved short-term forecasting of wind gusts will enhance aviation safety and logistics and may offer other societal benefits. Here we present a rigorous investigation of the relative skill of models of wind gust occurrence and magnitude that employ different statistical methods. It is shown that artificial neural networks (ANNs) offer considerable skill enhancement over regression methods, particularly for strong and damaging wind gusts. For wind gust magnitudes in particular, application of deeper learning networks (e.g., five or more hidden layers) offers tangible improvements in forecast accuracy. However, deeper networks are vulnerable to overfitting and exhibit substantial variability with the specific training and testing data subset used. Also, even deep ANNs reproduce only half of strong and damaging wind gusts. These results indicate the need for future work to elucidate the dynamical mechanisms of intense wind gusts and advance solutions to their prediction.
54 ENVIRONMENTAL SCIENCES↗
Wind gust models derived from field data
Wind data measured during a field experiment were used to verify the analytical model of wind gusts. Good coincidence was observed; the only discrepancy occurred for the azimuth error in the front and back winds, where the simulated errors were smaller than the measured ones. This happened because of the assumption of the spatial coherence of the wind gust model, which generated a symmetric antenna load and, in consequence, a low azimuth servo error. This result indicates a need for upgrading the wind gust model to a spatially incoherent one that will reflect the real gusts in a more accurate manner. In order to design a controller with wind disturbance rejection properties, the wind disturbance should be known at the input to the antenna rate loop model. The second task, therefore, consists of developing a digital filter that simulates the wind gusts at the antenna rate input. This filter matches the spectrum of the measured servo errors. In this scenario, the wind gusts are generated by introducing white noise to the filter input.
Gawronski, W.↗
Short–Term Forecasting of Wind Gusts at Airports Across CONUS Using Machine Learning
Short–term forecasting of wind gusts, particularly those of higher intensity, is of great societal importance but is challenging due to the presence of multiple gust generation mechanisms. Wind gust observations from eight high–passenger–volume airports across the continental United States (CONUS) are summarized and used to develop predictive models of wind gust occurrence and magnitude. These short–term (same hour) forecast models are built using multiple logistic and linear regression, as well as artificial neural networks (ANNs) of varying complexity. A suite of 19 upper–air predictors drawn from the ERA5 reanalysis and an autoregressive (AR) term are used. Stepwise procedures instruct predictor selection, and resampling is used to quantify model stability. All models are developed separately for the warm (April–September) and cold (October–March) seasons. Results show that ANNs of 3–5 hidden layers (HLs) generally exhibit higher hit rates than logistic regression models and also improve skill with respect to wind gust magnitudes. However, deeper networks with more HLs increase false alarm rates in occurrence models and mean absolute error in magnitude models due to model overfitting. For model skill, inclusion of the AR term is critical while the majority of the remaining skill derives from wind speeds and lapse rates. A predictive ceiling is also clearly demonstrated, particularly for the strong and damaging gust magnitudes, which appears to be partially due to ERA5 predictor characteristics and the presence of mixed wind climates.
54 ENVIRONMENTAL SCIENCES↗
Seismic Noise Induced by Wind Turbine Operation and Wind Gusts
Improved seismic noise characterization, due to varied sources, may benefit traditional applications. Some examples are earthquake detection, Earth structure research, and nuclear testing. This improvement could also permit use of seismic data in transdisciplinary research such as wind gust detection and wind turbine (WT) condition monitoring. However, it is a challenging task to unambiguously quantify relationships between potential sources of seismic noise and the actual seismic response. In this article, we analyze seismic and meteorological data (wind speed and pressure), measured at a remote site in a complex terrain area in eastern Portugal, to examine seismic signatures from WT operational status and wind gusts. Results presented herein show the following: (1) WT signatures in seismic data can be used to accurately determine WT-operational status. Attenuation of WT signatures in seismic data exhibits an exponential decay with distance, with attenuation coefficients that scale with frequency. (2) After WT signatures are removed from seismic power spectra, wind gust signatures remain. Analyses of these data further indicate that it may be possible to extract quantitative wind gust estimates from seismic data and decompose them into pressure and shear stress drivers of this coupling.
Geochemistry & Geophysics↗
Synchronization of wind turbine generators against an infinite bus under gusting wind conditions
Studies of synchronizing a wind turbine generator against an infinite bus are performed on a digital computer. In the digital simulation, wind gusts of different magnitudes and durations are hypothesized. Prior to the synchronization, differences of the frequency and phase position between voltages of the alternator and the bus are also included in the simulation. Solutions for rotor speed, generator power angle, electromagnetic torque, wind turbine torque, wind turbine blade pitch angle, and armature current are simulated and presented graphically. The ERDA-NASA 100-kW wind turbine is used as a case study. The results so obtained will thus have immediate applications.
Hwang, H. H.↗
Composite statistical method for modeling wind gusts for aircraft simulation
This paper discusses the application of three statistical methods in combination to model wind gusts for use in aircraft flight simulation. The approach combines principal components analysis, time series analysis and probability distribution model to analyze and simulate wind gust components. Comparisons are given between wind gust components generated by the model and components measured onboard an aircraft.
Schiess, J. R.↗
Low level wind gust amplitude and duration study
Wind velocity measurements of low level wind gust amplitude and duration and statistical gust shape characteristics
Camp, D. W.↗
Quantifying the compound hazard of freezing rain and wind gusts across CONUS
Abstract The co-occurrence of freezing rain, ice accumulation and wind gusts (FZG) poses a significant hazard to infrastructure and transportation. However, quantification of the frequency and intensity of FZG is challenged by the lack of direct icing measurements. In this work, we evaluate and then apply an energy balance model to high-frequency data collected during 2005–2022 to derive hourly ice accumulation at 883 stations across the contiguous USA. These estimates are combined with wind gust observations to compute time series of hourly FZG hazard magnitude using the Sperry–Piltz Ice Accumulation (SPIA) index. Results are evaluated using Storm Reports of damage and economic disruption. The hourly SPIA estimates are also used to (i) derive a geospatial atlas of the hazard including the 50 yr return period event intensities for each US state derived using superstations, and (ii) describe storylines of significant events in terms of meteorological drivers and socioeconomic impacts. The highest values of SPIA during the 18 yr study period occur in a region extending from the Southern Great Plains over the Midwest into the densely populated Northeast. States in these regions also have high 50 yr return period maximum radial ice accumulation of 3–5 cm and co-occurring wind gusts >30 ms −1 . These values are comparable to past estimates for the 500 yr event which may imply this hazard has been previously underestimated. This atlas can be used to inform optimal FZG hazard mitigation strategies for each state/region.
58 GEOSCIENCES↗
Using Buoy and Radar Data to Study Sudden Wind Gusts Over Coastal Regions
Significant sudden wind gusts can pose a threat to aviation near the coastline, as well as small (sailing) boats and commercial ships approaching the ports. Such cases can result in wind speed changes of more than an order of magnitude within 5 minutes, which can then last up to 20 minutes or more. Although the constellation of scatterometers is a good means of studying maritime convection, those sudden gusts are not easily captured because of the low time resolution. The National Data Buoy Center (NDBC) provides continuous measurements of wind speed and direction along the US coastal regions every 6 minutes. Buoys are platforms placed at specific places on the seas, especially along coastlines, providing data for atmospheric and oceanic studies. Next Generation Radars (NEXRADs), after the recent upgrade of the network to dual-pol systems, offer enhanced capabilities to study atmospheric phenomena. NEXRADs provide continuous full-volume scans approximately every 5 minutes and therefore are close to the time resolution of the buoy measurements. Use of single- Doppler retrievals might also provide a means of further validation.
Winds↗
Vector wind gust model
The development of a vector wind gust model through statistical analysis is described. Wind perturbation statistics which include location, altitude, season, and wavelength range are used in the synthesis of detailed wing profiles. These profiles provide the basis for the establishment of improved discrete gust design criteria guidelines for ascending launch vehicles.
Adelfang, S. I.↗
Tracking error of 100-m antenna due to wind gust
A procedure is shown to derive root-mean-square tracking error for an antenna system in response to wind gust loading. Example calculations are illustrated for the design of a proposed 100 m antenna. The effect of wind gust correlation is also considered.
Massoudi, M.↗
Effect of wind gusts on the motion of a balloon-borne observation platform
The effect of wind gusts on the magnitude of the pendulation angles of a balloon-borne observation platform is determined. A system mathematical model is developed and the solution of this model is used to determine the magnitude of the observation platforms pendulation angles.
Nigro, N. J.↗
Effect of Wind Gusts on Local Run-of-River Hydro Plants
In this work, we focus on the integration of wind farms in a hydro dominated power system, and analyze the challenges and solutions associated with this integration. Due to sudden increase in wind power, the system frequency can increase, causing excessive reduction in the run-of-river (ROR) system’s power. This can result in damage to the physical structure due to water level violations and cause issues for downstream customers. To prevent excessive and prolonged ROR power deviations, and to prevent tie-line deviations, we are proposing a wind-hydro centralized coordination scheme, which is not existing currently. The controller showed promising results by reducing the system frequency extrema by 0.8Hz.
13 HYDRO ENERGY↗
Mitigating the Effect of Wind Gusts on Local Run-of-River Hydropower Plants Using Cooperative Control
This paper focuses on the challenges and solutions associated with the integration of wind farms in a hydro- dominated power system. Due to sudden increases in wind power, the system frequency can increase, causing excessive reduction in the run-of-river (ROR) system’s power. Considering their small pondage, this can result in damage to the physical structure due to water level violations which are also penalized by stakeholders. To prevent excessive ROR power deviations, ROR governor wear and tear, and to prevent tie-line deviations, a centralized, distributed, and hybrid scheme for coordinating the response from a cluster of ROR plants is examined. Both centralized and distributed control as well as their hybrids can reduce frequency peaks. However, the inclusion of decentralized cooperative controls leads to reduced governor actions from fast- acting plants while enhancing net response from the hydro unit cluster.
13 HYDRO ENERGY↗
Composite statistical method for modeling wind gusts
Explore the source record for details and available documents.
Schiess, J. R.↗
BOREAS AES MARSII Surface Meteorological Data
Canadian AES personnel collected several data sets related to surface and atmospheric meteorological conditions over the BOREAS region. This data set contains 15-minute meteorological data from six MARSII meteorology stations in the BOREAS region in Canada. Parameters include site, time, temperature, dewpoint, visibility, wind speed, wind gust, wind direction, two cloud groups, precipitation, and station pressure. Temporally, the data cover the period of May to September 1994. Geo-graphically, the stations are spread across the provinces of Saskatchewan and Manitoba. The data are provided in tabular ASCII files, and are classified as AFM-Staff data.
Atkinson, G. Barrie↗
Statistical wind profile gust model
A statistical wind profile gust model for the Space Transportation Operations and Trade Studies is developed by using 1800 Jimsphere wind profile data collected at Cape Kennedy during 1965 to 1972. Wind profiles from the surface to 20 km in component form, i.e., zonal and meridional are processed through the digital filters of different wave length ranges bases on the Martin-Graham cosine rolloff model. The residuals obtained from the filtering processes for the data base for the statistical analysis. For each wind component the gust and gust length at a specified reference altitude in a residual profile are defined. A two parameter gamma probability marginal distribution seems to fit the component gust amplitude and the gust length when redefined. The problem of finding an appropriate bivariate joint distribution of the gust amplitude and length remains to be solved. The probability distribution of the modulus of the gust amplitudes was derived under the assumption that they are independently distributed as gamma variates.
Doss, D. C.↗