An official website of the United States government
Power grid operators rely on solar irradiance forecasts to manage uncertainty and variability associated with solar power. Meteorological factors such as cloud cover, wind direction, and wind speed affect irradiance and are associated with a high degree of variability and uncertainty. Statistical models fail to accurately capture the dependence between these factors and irradiance. In this paper, we introduce the idea of applying multivariate Gated Recurrent Units (GRU) to forecast Direct Normal Irradiance (DNI) hourly. The proposed GRU-based forecasting method is evaluated against traditional Long Short-Term Memory (LSTM) using historical irradiance data (i.e., weather variables that include cloud cover, wind direction, and wind speed) to forecast irradiance forecasting over intra-hour and inter-hour intervals. Our evaluation on one of the sites from Measurement and Instrumentation Data Center indicate that both GRU and LSTM improved DNI forecasting performance when evaluated under different conditions. Moreover, including wind direction and wind speed can have substantial improvement in the accuracy of DNI forecasts. Besides, the forecasting model can accurately forecast irradiance values over multiple forecasting horizons.
more »
« less
A Multi-variable Stacked Long-Short Term Memory Network for Wind Speed Forecasting
Precisely forecasting wind speed is essential for wind power producers and grid operators. However, this task is challenging due to the stochasticity of wind speed. To accurately predict short-term wind speed under uncertainties, this paper proposed a multi-variable stacked LSTMs model (MSLSTM). The proposed method utilizes multiple historical meteorological variables, such as wind speed, temperature, humidity, pressure, dew point, and solar radiation to accurately predict wind speeds. The prediction performance is extensively assessed using real data collected in West Texas, USA. The experimental results show that the proposed MSLSTM can preferably capture and learn uncertainties while output competitive performance.
more »
« less
- Award ID(s):
- 1737634
- PAR ID:
- 10128846
- Date Published:
- Journal Name:
- 2018 IEEE International Conference on Big Data (Big Data),
- Page Range / eLocation ID:
- 4561-4564
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Obtaining reliable estimates of aerodynamic roughness is necessary to interpret and accurately predict aeolian sediment transport dynamics. However, inherent uncertainties in field measurements and models of surface aerodynamic properties continue to undermine aeolian research, monitoring, and dust modeling. A new relation between aerodynamic shelter and land surface shadow has been established at the wind tunnel scale, enabling the potential for estimates of wind erosion and dust emission to be obtained across scales from albedo data. Here, we compare estimates of wind friction velocity (u*) derived from traditional methods (wind speed profiles) with those derived from the albedo model at two separate scales using bare soil patch (via net radiometers) and landscape (via Moderate Resolution Imaging Spectroradiometer [MODIS] 500 m) data sets. Results show that profile‐derived estimates ofu*are highly variable in anisotropic surface roughness due to changes in wind direction and fetch. Wind speed profiles poorly estimate soil surface (bed) wind friction velocities necessary for aeolian sediment transport research and modeling. Albedo‐based estimates ofu*at both scales have small variability because the estimate is integrated over a defined, fixed area and resolves the partition of wind momentum between roughness elements and the soil surface. We demonstrate that the wind tunnel‐based calibration of albedo for predicting wind friction velocities at the soil surface (us*) is applicable across scales. The albedo‐based approach enables consistent and reliable drag partition correction across scales for model and field estimates ofus*necessary for wind erosion and dust emission modeling.more » « less
-
In recent years, the global transition towards green energy, driven by environmental concerns and increasing electricity demands, has remarkably reshaped the energy landscape. The transformative potential of marine wind energy is particularly critical in securing a sustainable energy future. To achieve this objective, it is essential to have an accurate understanding of wind dynamics and their interactions with ocean waves for the proper design and operation of offshore wind turbines (OWTs). The accuracy of met-ocean models depends critically on their ability to correctly capture sea-surface drag over the multiscale ocean surface—a quantity typically not directly resolved in numerical models and challenging to acquire using either field or laboratory measurements. Although skin friction drag contributes considerably to the total wind stress, especially at moderate wind speeds, it is notoriously challenging to predict using physics-based approaches. The current work introduces a novel approach based on a convolutional neural network (CNN) model to predict the spatial distributions of skin friction drag over wind-generated surface waves using wave profiles, local wave slopes, local wave phases, and the scaled wind speed. The CNN model is trained using a set of high-resolution laboratory measurements of air-side velocity fields and their respective surface viscous stresses obtained over a range of wind-wave conditions. The results demonstrate the capability of our model to accurately estimate both the instantaneous and area-aggregate viscous stresses for unseen wind-wave regimes. The proposed CNN-based wall-layer model offers a viable pathway for estimating the local and averaged skin friction drag in met-ocean simulations.more » « less
-
Abstract Atlantic tropical cyclones (TCs) can cause significant societal and economic impacts, as 2019's Dorian serves to remind us of these storms' destructiveness. Decades of effort to understand and predict Atlantic TC activity have improved seasonal forecast skill, but large uncertainties still remain, in part due to an incomplete understanding of the drivers of TC variability. Here we identify an association between the East Asian Subtropical Jet Stream (EASJ) during July–October and the frequency of Atlantic TCs (wind speed ≥34 knot) and hurricanes (wind speed ≥64 knot) during August–November based on observations for 1980–2018. This strong association is tied to the impacts of EASJ on a stationary Rossby wave train emanating from East Asia and the tropical Pacific to the North Atlantic, leading to changes in vertical wind shear in the Atlantic Main Development Region (80–20°W, 10–20°N).more » « less
-
Abstract The wind‐blown flux of sand generates dunes, wind erosion, and mineral dust aerosols. Existing models predict sand flux using the wind friction velocity that characterizes near‐surface turbulent momentum fluxes. However, these models struggle to accurately predict sand fluxes. Here we analyze root causes of these model discrepancies using high‐frequency field measurements of winds and sand fluxes. We find that friction velocity is only predictive of sand fluxes on long timescales, when it correlates with horizontal wind speed. On shorter timescales, and for non‐ideal surface conditions, friction velocity is much less predictive, likely because the near‐surface wind momentum budget is dominated by other, less predictable terms. We furthermore find that variability in 30‐min averaged sand fluxes at a given friction velocity is not driven by changes in turbulence but by changes in surface conditions, raising a challenge for models. These findings can improve sand flux models and clarify their limitations.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
