An official website of the United States government
ABSTRACT Accurate wind power forecasts are essential for energy management and resource allocation. However, because of complex weather dynamics and other nonlinearities, it is exceedingly difficult to forecast wind power on the multisite level for dozens of wind farms at once. This paper proposes a hybridized approach that leverages deep learning to predict future forecast errors from physics‐based numerical weather prediction (NWP) model estimates. Utilizing errors from NWP forecasts allows integration of critical atmospheric and meteorological dynamics into the forecasting model, and we demonstrate the importance of post‐calibration based on the physics versus pure data‐driven wind power prediction. This post‐calibration approach is enabled by the inverted transformer architecture, which efficiently and effectively learns meaningful wind farm variate representations, resulting in accurate spatiotemporal corrections to the forecasts. We also investigate modifying the iTransformer with a new embedding approach, named SpaceEmbed, that explicitly encodes spatial distance information into the network. The proposed approach is validated with a case study using real‐world data and forecasts from the Electric Reliability Council of Texas (ERCOT) in 2015 for 74 wind farms in Texas at different time scales. Using the high sustained limit as the metric for power generation, the iTransformer outperforms other state‐of‐the‐art deep learning forecasting methods, succeeding at the post‐calibration task by reducing NWP forecast error by up to 33% on average.
more »
« less
Genetic algorithm selection of the weather research and forecasting model physics to support wind and solar energy integration
To make a future run by renewable energy possible, we must design our power system to seamlessly collect, store, and transport the Earth's naturally occurring flows of energy – namely the sun and the wind. Such a future will require that accurate representations of wind and solar resources and their associated variability permeate power systems planning and operational tools. Practically speaking, we must merge weather and power systems modeling. Although many meteorological phenomena that affect wind and solar power production are well-studied in isolation, no coordinated effort has sought to improve medium- and long-term power systems planning using numerical weather prediction (NWP) models. One modern open-source NWP tool – the weather research and forecasting (WRF) model – offers the complexity and flexibility required to integrate weather prediction with a power systems model in any region. However, there are over one million distinct ways to set up WRF. Here, we present a methodology for optimizing the WRF model physics for forecasting wind power density and solar irradiance using a genetic algorithm. The top five setups created by our algorithm outperform all of the recommended setups. Using the simulation results, we train a random forest model to identify which WRF parameters contribute to the lowest forecast errors and produce plots depicting the performance of key physics options to guide energy researchers in quickly setting up an accurate WRF model.
more »
« less
- Award ID(s):
- 1751535
- PAR ID:
- 10432843
- Date Published:
- Journal Name:
- Energy
- Volume:
- 254
- Issue:
- Part B
- ISSN:
- 0360-5442
- Page Range / eLocation ID:
- 124367
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract The development of deep learning (DL) weather forecasting models has made rapid progress and achieved comparable or better skill than traditional Numerical Weather prediction (NWP) models, which are generally computationally intensive. However, applications of these DL models have yet to be fully explored, including for severe convective events. We evaluate the DL model Pangu‐Weather in forecasting tornadic environments with one‐day lead times using convective available potential energy (CAPE), 0–6 bulk wind difference (BWD6), and 0–3 km storm‐relative helicity (SRH3). We also compare its performance to the National Centers for Environmental Prediction (NCEP)'s Global Forecast System (GFS), a traditional NWP model. Pangu‐Weather generally outperforms GFS in predicting BWD6 and SRH3 at the closest grid point and hour of the storm report. However, Pangu‐Weather tends to underpredict the maximum values of all convective parameters in the 1–2 hr before the storm across the surrounding grid points compared to the GFS.more » « less
-
Hurricanes are among the costliest natural disasters in the United States and regularly inflict severe damage on urban infrastructure. Accurate forecasts are therefore essential for preparedness and limiting these extreme events' economic toll. Numerical weather‑prediction (NWP) models—such as the Weather Research and Forecasting (WRF) system—are powerful forecasting tools. However, some of their physical parameterizations were neither designed for nor tested with real hurricanes. This thesis addresses that gap by evaluating two key parameterizations in WRF: (i) subgrid‑scale (SGS) turbulence schemes and (ii) surface‑roughness and urban canopy treatments. The first part of the study investigates how SGS eddy‑viscosity choices affect hurricane intensity, turbulence, and wind profiles. Large‑eddy simulations (LES) of five major hurricanes were run with a 1.5‑order, three‑dimensional turbulent‑kinetic‑energy (TKE) SGS scheme. Each storm was simulated under three eddy‑viscosity settings— default, halved, and doubled—yielding 15 cases. A parallel set of 10 cases employed an alternative nonlinear backscatter and anisotropy (NBA) SGS scheme. Two idealized LES runs and one fine-grid (~80 m) nested simulation brought the total to 33. Reducing SGS stress intensified storms by raising boundary‑layer wind speeds and lowering the altitude of peak winds, improving surface‑wind forecasts by ~9 % and minimum sea‑level pressure by ~29 % relative to the default setting. These results reveal that standard SGS models are overly dissipative because they overlook the rotational suppression of turbulence, underscoring the need for SGS schemes tailored to hurricane dynamics. The second part assesses how aerodynamic roughness length (z0) and urban‑canopy schemes shape near‑surface winds over cities. For four land‑falling hurricanes affecting Houston and New Orleans, increasing z0 in the Single‑Layer Urban Canopy Model (SLUCM) reduced modeled wind speeds and cut mean absolute error (MAE) by ~20 %, whereas decreasing z0 introduced large positive biases. Additional experiments compared three urban options—Bulk (no‑urban), SLUCM, and the multi‑layer Building Energy Model (BEM). The Bulk scheme delivered the most accurate surface‑wind forecasts in every nested domain, while SLUCM slightly outperformed BEM in the limited vertical‑profile data. These findings highlight the need to recalibrate urban schemes and surface‑drag parameters when applying WRF to hurricane‑force winds.more » « less
-
Accurate and uncertainty-aware wind power forecasting is essential for reliable and cost-effective power system operations. This paper presents a novel probabilistic forecasting framework based on diffusion probabilistic models. We adopted a two-stage modeling strategy—a deterministic predictor first generates baseline forecasts, and a conditional diffusion model then learns the distribution of residual errors. Such a two-stage decoupling strategy improves learning efficiency and sharpens uncertainty estimation. We employed the elucidated diffusion model (EDM) to enable flexible noise control and enhance calibration, stability, and expressiveness. For the generative backbone, we introduced a time-series-specific diffusion Transformer (TimeDiT) that incorporates modular conditioning to separately fuse numerical weather prediction (NWP) inputs, noise, and temporal features. The proposed method was evaluated using the public database from ten wind farms in the Global Energy Forecasting Competition 2014 (GEFCom2014). We further compared our approach with two popular baseline models, i.e., a distribution parameter regression model and a generative adversarial network (GAN)-based model. Results showed that our method consistently achieves superior performance in both deterministic metrics and probabilistic accuracy, offering better forecast calibration and sharper distributions.more » « less
-
The intensity and frequency of wildfires in California (CA) have increased in recent years, causing significant damage to human health and property. In October 2007, a number of small fire events, collectively referred to as the Witch Creek Fire or Witch Fire started in Southern CA and intensified under strong Santa Ana winds. As a test of current mesoscale modeling capabilities, we use the Weather Research and Forecasting (WRF) model to simulate the 2007 wildfire event in terms of meteorological conditions. The main objectives of the present study are to investigate the impact of horizontal grid resolution and planetary boundary layer (PBL) scheme on the model simulation of meteorological conditions associated with a Mega fire. We evaluate the predictive capability of the WRF model to simulate key meteorological and fire-weather forecast parameters such as wind, moisture, and temperature. Results of this study suggest that more accurate predictions of temperature and wind speed relevant for better prediction of wildfire spread can be achieved by downscaling regional numerical weather prediction products to 1 km resolution. Furthermore, accurate prediction of near-surface conditions depends on the choice of the planetary boundary layer parameterization. The MYNN parameterization yields more accurate prediction as compared to the YSU parameterization. WRF simulations at 1 km resolution result in better predictions of temperature and wind speed than relative humidity during the 2007 Witch Fire. In summary, the MYNN PBL parameterization scheme with finer grid resolution simulations improves the prediction of near-surface meteorological conditions during a wildfire event.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- The DOI is not currently available.
