An official website of the United States government
Abstract Previous studies have identified environmental characteristics that skillfully discriminate between severe and significant-severe weather events, but they have largely been limited by sample size and/or population of predictor variables. Given the heightened societal impacts of significant-severe weather, this topic was revisited using over 150 000 ERA5 reanalysis-derived vertical profiles extracted at the grid-point nearest—and just prior to—tornado and hail reports during the period 1996–2019. Profiles were quality-controlled and used to calculate 84 variables. Several machine learning classification algorithms were trained, tested, and cross-validated on these data to assess skill in predicting severe or significant-severe reports for tornadoes and hail. Random forest classification outperformed all tested methods as measured by cross-validated critical success index scores and area under the receiver operating characteristic curve values. In addition, random forest classification was found to be more reliable than other methods and exhibited negligible frequency bias. The top three most important random forest classification variables for tornadoes were wind speed at 500 hPa, wind speed at 850 hPa, and 0–500-m storm-relative helicity. For hail, storm-relative helicity in the 3–6 km and -10 to -30 °C layers, along with 0–6-km bulk wind shear, were found to be most important. A game theoretic approach was used to help explain the output of the random forest classifiers and establish critical feature thresholds for operational nowcasting and forecasting. A use case of spatial applicability of the random forest model is also presented, demonstrating the potential utility for operational forecasting. Overall, this research supports a growing number of weather and climate studies finding admirable skill in random forest classification applications.
more »
« less
Utilizing physics-based input features within a machine learning model to predict wind speed forecasting error
Abstract. Machine learning is quickly becoming a commonly used technique for wind speed and power forecasting. Many machine learning methods utilize exogenous variables as input features, but there remains the question of which atmospheric variables are most beneficial for forecasting, especially in handling non-linearities that lead to forecasting error. This question is addressed via creation of a hybrid model that utilizes an autoregressive integrated moving-average (ARIMA) model to make an initial wind speed forecast followed by a random forest model that attempts to predict the ARIMA forecasting error using knowledge of exogenous atmospheric variables.Variables conveying information about atmospheric stability and turbulence as well as inertial forcing are found to be useful in dealing with non-linear error prediction. Streamwise wind speed, time of day, turbulence intensity, turbulent heat flux, vertical velocity, and wind direction are found to be particularly useful when used in unison for hourly and 3 h timescales. The prediction accuracy of the developed ARIMA–random forest hybrid model is compared to that of the persistence and bias-corrected ARIMA models. The ARIMA–random forest model is shown to improve upon the latter commonly employed modeling methods, reducing hourly forecasting error by up to 5 % below that of the bias-corrected ARIMA model and achieving an R2 value of 0.84 with true wind speed.
more »
« less
- Award ID(s):
- 1921554
- PAR ID:
- 10313843
- Date Published:
- Journal Name:
- Wind Energy Science
- Volume:
- 6
- Issue:
- 1
- ISSN:
- 2366-7451
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Forecasting time series data is an important subject in economics, business, and finance. Traditionally, there are several techniques to effectively forecast the next lag of time series data such as univariate Autoregressive (AR), univariate Moving Average (MA), Simple Exponential Smoothing (SES), and more notably Autoregressive Integrated Moving Average (ARIMA) with its many variations. In particular, ARIMA model has demonstrated its outperformance in precision and accuracy of predicting the next lags of time series. With the recent advancement in computational power of computers and more importantly development of more advanced machine learning algorithms and approaches such as deep learning, new algorithms are developed to analyze and forecast time series data. The research question investigated in this article is that whether and how the newly developed deep learning-based algorithms for forecasting time series data, such as “Long Short-Term Memory (LSTM)”, are superior to the traditional algorithms. The empirical studies conducted and reported in this article show that deep learning-based algorithms such as LSTM outperform traditional-based algorithms such as ARIMA model. More specifically, the average reduction in error rates obtained by LSTM was between 84 - 87 percent when compared to ARIMA indicating the superiority of LSTM to ARIMA. Furthermore, it was noticed that the number of training times, known as “epoch” in deep learning, had no effect on the performance of the trained forecast model and it exhibited a truly random behavior.more » « less
-
null (Ed.)When forest conditions are mapped from empirical models, uncertainty in remotely sensed predictor variables can cause the systematic overestimation of low values, underestimation of high values, and suppression of variability. This regression dilution or attenuation bias is a well-recognized problem in remote sensing applications, with few practical solutions. Attenuation is of particular concern for applications that are responsive to prediction patterns at the high end of observed data ranges, where systematic error is typically greatest. We addressed attenuation bias in models of tree species relative abundance (percent of total aboveground live biomass) based on multitemporal Landsat and topoclimatic predictor data. We developed a multi-objective support vector regression (MOSVR) algorithm that simultaneously minimizes total prediction error and systematic error caused by attenuation bias. Applied to 13 tree species in the Acadian Forest Region of the northeastern U.S., MOSVR performed well compared to other prediction methods including single-objective SVR (SOSVR) minimizing total error, Random Forest (RF), gradient nearest neighbor (GNN), and Random Forest nearest neighbor (RFNN) algorithms. SOSVR and RF yielded the lowest total prediction error but produced the greatest systematic error, consistent with strong attenuation bias. Underestimation at high relative abundance caused strong deviations between predicted patterns of species dominance/codominance and those observed at field plots. In contrast, GNN and RFNN produced dominance/codominance patterns that deviated little from observed patterns, but predicted species relative abundance with lower accuracy and substantial systematic error. MOSVR produced the least systematic error for all species with total error often comparable to SOSVR or RF. Predicted patterns of dominance/codominance matched observations well, though not quite as well as GNN or RFNN. Overall, MOSVR provides an effective machine learning approach to the reduction of systematic prediction error and should be fully generalizable to other remote sensing applications and prediction problems.more » « less
-
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
-
As turbines continue to grow in hub height and rotor diameter and wind farms grow larger, consideration of stratified atmospheric boundary layer (ABL) processes in wind power models becomes increasingly important. Atmospheric stratification can considerably alter the boundary layer structure and flow characteristics through buoyant forcing. Variations in buoyancy, and corresponding ABL stability, in both space and time impact ABL wind speed shear, wind direction shear, boundary layer height, turbulence kinetic energy, and turbulence intensity. In addition, the presence of stratification will result in a direct buoyant forcing within the wake region. These ABL mechanisms affect turbine power production, the momentum and kinetic energy deficit wakes generated by turbines, and the turbulent mixing and kinetic energy entrainment in wind farms. Presently, state-of-practice engineering models of mean wake momentum utilize highly empirical turbulence models that do not explicitly account for ABL stability. Models also often neglect the interaction between the wake momentum deficit and the turbulence kinetic energy added by the wake, which depends on stratification. In this work, we develop a turbulence model that models the wake-added turbulence kinetic energy, and we couple it with a wake model based on the parabolized Reynolds-averaged Navier–Stokes equations. Comparing the model predictions to large eddy simulations across stabilities (Obukhov lengths) and surface roughness lengths, we find lower prediction error in both power production and the wake velocity field across the ABL conditions and error metrics investigated.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.5194/wes-6-295-2021
