An Analog Technique to Improve Storm Wind Speed Prediction Using a Dual NWP Model Approach

Yang, Jae-Mo; Astitha, Marina; Monache, Luca Delle; Alessandrini, Stefano

doi:10.1175/mwr-d-17-0198.1

Cited by 29 publications

(13 citation statements)

References 47 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This type of method is constantly employed for forecasting weather data, such as wind speed, wind direction, temperature, and so on [8]. Basically, numerical weather prediction (NWP) models [9] are used for predicting future data through applying current physical data and topographical information [10]. Dong et al proposed a comprehensive review of NWP models and put forward an effective hybrid model to predict day-ahead wind power [11].…”

Section: B Literature Review 1) Physical Methodsmentioning

confidence: 99%

Wind Speed Forecasting Based on Extreme Gradient Boosting

et al. 2020

View full text Add to dashboard Cite

As the integration of wind power into electrical energy network increasing, accurate forecast of wind speed becomes highly important in the case of large-scale wind power connected into the grid. In order to improve the accuracy of wind speed forecast and the generalization ability of the model, Extreme Gradient Boosting (XGBoost) as an improvement from gradient boosting decision tree (GBDT) is trained and deployed in the cheaper central processing unit (CPU) devices instead of graphics processing unit (GPU) devices, thus, a wind speed forecast model based on Extreme Gradient Boosting is proposed in this paper. Firstly, the historical data is taken as a part of the input vectors for the model. Moreover, considering the monthly change of wind speed characteristics, the dataset of wind power is divided into four parts by month so that the models are constructed in different complexity by month. Finally, compared with back propagation neural networks (BPNN) and linear regression (LR) models, the experimental results show that the improved XGBoost model can promote the forecast accuracy effectively. INDEX TERMS Short-term wind speed forecasting, XGBoost, time series, historical characteristics, power grid.

show abstract

Section: B Literature Review 1) Physical Methodsmentioning

confidence: 99%

Wind Speed Forecasting Based on Extreme Gradient Boosting

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The probabilistic forecast itself may have systematic errors, which can be removed by statistical postprocessing, or by an approach like the Analog Ensemble approach, which builds an ensemble by using a set of past observations that correspond to the best analogs of NWP forecasts. This approach has been used to improve the skill of both deterministic and probabilistic forecasting for both wind and solar power (Yang, Astitha, Delle Monache, & Alessandrini, 2018). Recent developments in renewable energy forecasting have focused on probabilistic forecasts and can be divided into two groups: univariate and multivariate probabilistic forecasting.…”

Section: Postprocessing Numerical Weather Predictionsmentioning

confidence: 99%

The future of forecasting for renewable energy

Sweeney

Bessa

Browell

et al. 2019

WIREs Energy & Environment

136

View full text Add to dashboard Cite

Forecasting for wind and solar renewable energy is becoming more important as the amount of energy generated from these sources increases. Forecast skill is improving, but so too is the way forecasts are being used. In this paper, we present a brief overview of the state‐of‐the‐art of forecasting wind and solar energy. We describe approaches in statistical and physical modeling for time scales from minutes to days ahead, for both deterministic and probabilistic forecasting. Our focus changes then to consider the future of forecasting for renewable energy. We discuss recent advances which show potential for great improvement in forecast skill. Beyond the forecast itself, we consider new products which will be required to aid decision making subject to risk constraints. Future forecast products will need to include probabilistic information, but deliver it in a way tailored to the end user and their specific decision making problems. Businesses operating in this sector may see a change in business models as more people compete in this space, with different combinations of skills, data and modeling being required for different products. The transaction of data itself may change with the adoption of blockchain technology, which could allow providers and end users to interact in a trusted, yet decentralized way. Finally, we discuss new industry requirements and challenges for scenarios with high amounts of renewable energy. New forecasting products have the potential to model the impact of renewables on the power system, and aid dispatch tools in guaranteeing system security. This article is categorized under: Energy Infrastructure > Systems and Infrastructure Wind Power > Systems and Infrastructure Photovoltaics > Systems and Infrastructure

show abstract

“…Slingo and Palmer showed that the root mean square error of the ensemble mean anomaly forecast grew from less than 1 to 70 for forecast lead times ranging from hours to decades [30]. Yang et al compared the forecast wind speed for 146 storms based on the Weather Research and Forecasting (WRF) and Integrated Community Limited Area Modeling System models, showing that the RMSE for both models increased from zero hours to 54 forecast hours [31]. Using the gridded Bayesian linear regression to improve the deterministic wind speed prediction with the NCAR's Real-Time Ensemble Forecast System, the authors showed that R-square decreased by 28% and centered root mean square error increased by 38% for lead times ranging from 0 to 48 h [32].…”

Section: Introductionmentioning

confidence: 99%

The Effect of Lead-Time Weather Forecast Uncertainty on Outage Prediction Modeling

2021

View full text Add to dashboard Cite

Weather-related power outages affect millions of utility customers every year. Predicting storm outages with lead times of up to five days could help utilities to allocate crews and resources and devise cost-effective restoration plans that meet the strict time and efficiency requirements imposed by regulatory authorities. In this study, we construct a numerical experiment to evaluate how weather parameter uncertainty, based on weather forecasts with one to five days of lead time, propagates into outage prediction error. We apply a machine-learning-based outage prediction model on storm-caused outage events that occurred between 2016 and 2019 in the northeastern United States. The model predictions, fed by weather analysis and other environmental parameters including land cover, tree canopy, vegetation characteristics, and utility infrastructure variables exhibited a mean absolute percentage error of 38%, Nash–Sutcliffe efficiency of 0.54, and normalized centered root mean square error of 68%. Our numerical experiment demonstrated that uncertainties of precipitation and wind-gust variables play a significant role in the outage prediction uncertainty while sustained wind and temperature parameters play a less important role. We showed that, while the overall weather forecast uncertainty increases gradually with lead time, the corresponding outage prediction uncertainty exhibited a lower dependence on lead times up to 3 days and a stepwise increase in the four- and five-day lead times.

show abstract

An Analog Technique to Improve Storm Wind Speed Prediction Using a Dual NWP Model Approach

Cited by 29 publications

References 47 publications

Wind Speed Forecasting Based on Extreme Gradient Boosting

Wind Speed Forecasting Based on Extreme Gradient Boosting

The future of forecasting for renewable energy

The Effect of Lead-Time Weather Forecast Uncertainty on Outage Prediction Modeling

Contact Info

Product

Resources

About