Characterizing forecastability of wind sites in the United States

Feng, Cong; Sun, Mucun; Cui, Mingjian; Chartan, Erol Kevin; Zhang, Jie

doi:10.1016/j.renene.2018.08.085

Cited by 22 publications

(11 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In power systems, forecasting has grown in importance in light of recent advances in AI. AI is used to predict the future load, RES generation, and energy prices [23]. For example, the authors in [24] implemented a cloud-based machine learning platform for predicting electricity demand within buildings in Malaysia, considering voltage, current, and power factors as candidate features, however, environmental impact such as weather is neglected in their model.…”

Section: A Literature Reviewmentioning

confidence: 99%

Human Mobility-Based Features to Analyse the Impact of COVID-19 on Power System Operation of Ireland

Zarbakhsh

Misaghian

McArdle

2022

IEEE Open J. Power Energy

View full text Add to dashboard Cite

are changing human mobility patterns; however, the effects on power systems remain unclear. Previous loads and timings along with weather features are often used in literature as input features in load forecasting, but these may be insufficient during COVID-19. As a result, this paper proposes an analytical framework to assess the impact of COVID-19 on power system operation as well as day-ahead electricity prices in Ireland. To improve peak demand forecasting during pandemics, we incorporate mobility, NPIs, and COVID-19 cases as complementary input features and representative of human behaviour changes. By defining different combinations of these explanatory features, several Machine Learning (ML) algorithms are applied and their performance is compared with the baseline scenario currently used in the literature. Using SHapley Additive Explanations (SHAP), we interpret the best performing model, Light Gradient Boosted Machine, to determine the influence of each feature on the predicted outcomes. We discover that typical load forecasting features still influence ML outcomes the most, but mobility-related changes are also significant. Our finding shows that NPIs impact human behaviour and electricity consumption during times of crisis and can be used in the context of load forecasting to assist policymakers and energy distributors.

show abstract

Section: A Literature Reviewmentioning

confidence: 99%

Human Mobility-Based Features to Analyse the Impact of COVID-19 on Power System Operation of Ireland

Zarbakhsh

Misaghian

McArdle

2022

IEEE Open J. Power Energy

View full text Add to dashboard Cite

show abstract

“…Feng et al in [7] analyse the characterisation of the time series structure by applying decomposition, linearity analysis of entropy. This approach is used in a subsequent work [8] where the defined characterisations are applied to wind sites in North America, analysing the relationship of the uncertainty in forecastability to spectral entropy using regression approaches.…”

Section: Previous Work In Forecastabiltymentioning

confidence: 99%

Forecastability Measures that Describe the Complexity of a Site for Deep Learning Wind Predictions

Manero

Béjar²

2021

JSFI

View full text Add to dashboard Cite

The application of deep learning to wind time series for multi-step prediction obtains good results at short horizons. The accuracy of a wind forecast is highly dependent on the specific structure of wind in the specific location, as many local features influence wind behaviour. The characterization of the complexity of a site for wind prediction is defined as forecastability or predictability and can be obtained from the inner structure of the meteorological time series observations from a site. We analyze the time series structure searching for properties that have a high correlation with the prediction result, properties that can create measures that have the potential to describe the forecastability of a site. The best measures will show a high correlation with the accuracy of the predictions. In this work, we analyze wind time series from 126,692 wind locations in the US, where we apply several deep learning methods first, and then we verify several forecastability descriptors with the accuracy deep learning results. We require High-Performance Computing (HPC) resources for this task as the deep learning algorithms have sensible resource requirements and are applied to a large set of data. The measures defined and explored in this work are based on several techniques that decompose or transform the wind time-series. By combining several of these measures, we can obtain better predictors of the site complexity, which will allow us to evaluate the future error of a prediction on this site. Forecastability measures can contribute to a wind site multi-dimensional description, becoming a valuable tool for wind resource analysts and wind forecasters.

show abstract

“…This shows the WMA model accurately models the scaled correlations while there are greater errors introduced from the NWP model. K( j) = f loor T j (13) where K are the power time-series segment, T is the total time, and j is the scale.…”

Section: Validationmentioning

confidence: 99%

“…For this study, power forecasts are made assuming persistence, as standard wind forecasting methods provide little additional benefit for horizons of <6 h [12]. Recent developments in wind power forecasting have greatly improved results but these are often site‐specific and tailored to specific applications [13], whereas a generic method is required. A dependability metric is also used, describing how much power can be relied upon at any time and is measured using the standard deviation of power.…”

Section: Introductionmentioning

confidence: 99%

Quantifying benefits of wind power diversity in New Zealand

McQueen

Wood

2019

IET Renewable Power Generation

View full text Add to dashboard Cite

Wind integration studies often focus on the capacity value of wind power without considering Unit Commitment and Economic Dispatch or resolving requirements for ancillary services. Here, a novel method for simulating wind power time series with sufficient temporal span to support capacity studies and temporal resolution to support UCED studies is developed. Wind speed time series (WSTS), with 6 h temporal and 0.7 × 0.7 degree spatial resolutions, are extracted from the ECMWF-interim reanalysis, interpolated, scaled, and imputed so that they are representative of a point wind speed measurement with a 5 min temporal resolution. Imputation is made using a wavelet multi-resolution analysis approach that ensures temporally consistent correlations while accounting for heteroskedasticity. WSTS are transformed to power using wind power plant power curves, lowpass filters, and a Markov Chain model of operational efficiencies. The wind power model is validated using a set of measurements made at wind power plants (WPPs) in New Zealand and used to simulate power time series for 2 GW portfolios of WPPs representing compact, disperse, diverse, and business-as-usual portfolios. Metrics for dependability, variability, and predictability are applied to quantify the benefits of spatial diversification.

show abstract

Characterizing forecastability of wind sites in the United States

Cited by 22 publications

References 36 publications

Human Mobility-Based Features to Analyse the Impact of COVID-19 on Power System Operation of Ireland

Human Mobility-Based Features to Analyse the Impact of COVID-19 on Power System Operation of Ireland

Forecastability Measures that Describe the Complexity of a Site for Deep Learning Wind Predictions

Quantifying benefits of wind power diversity in New Zealand

Contact Info

Product

Resources

About