Across the United States, power grids are increasingly under strain from extreme weather events, such as heatwaves, high winds, and heavy precipitation, that result in frequent, long-duration and widespread power outages. The strain is intensified when these events are compounded, which amplifies their impact and exacerbates the risk of disruptions. To identify weather variables driving outages and cluster regions based on these variables, we employed a self-organizing map (SOM) approach using county-level outage data from 2015 to 2022, obtained using the U.S. Department of Energy’s EAGLE-I platform, and weather data from the ASOS weather station observations. The findings revealed substantial regional differences in outages resulting from variability in heat, wind, and precipitation. In California, for example, high heat coupled with strong wind gusts led to the most severe outages, while in Texas, the primary contributors were high heat followed by heavy precipitation. This research provides a nuanced understanding of weather-induced power outages, offering a data-driven approach for infrastructure planning and resilience efforts. This study underscores the crucial role of continental weather variability in shaping outage patterns and highlights the necessity for region-specific adaptation strategies to enhance grid resilience, considering the rise in heatwaves and compounding weather events.
The environmental issues we are currently facing require long-term prospective efforts for sustainable growth. Renewable energy sources seem to be one of the most practical and efficient alternatives in this regard. Understanding a nation's pattern of energy use and renewable energy production is crucial for developing strategic plans. No previous study has been performed to explore the dynamics of power consumption with the change in renewable energy production on a country-wide scale. In contrast, a number of deep learning algorithms demonstrated acceptable performance while handling sequential data in the era of data-driven predictions. In this study, we developed a scheme to investigate and predict total power consumption and renewable energy production time series for eleven years of data using a Recurrent Neural Network (RNN). The dynamics of the interaction between the total annual power consumption and renewable energy production are investigated through extensive Exploratory Data Analysis (EDA) and a feature engineering framework. The performance of the model is found satisfactory through the comparison of the predicted data with the observed data, visualization of the distribution of the errors and Root Mean Squared Error (RMSE) value of 0.084. Higher performance is achieved through the increase in the number of epochs and hyperparameter tuning. The proposed framework can be used and transferred to investigate the trend of renewable energy production and power consumption and predict the future scenarios for different communities. Incorporation of the cloud-based platform into the proposed pipeline may lead to real-time forecasting.
The environmental issues we are currently facing require long-term prospective efforts for sustainable growth. Renewable energy sources seem to be one of the most practical and efficient alternatives in this regard. Understanding a nation’s pattern of energy use and renewable energy production is crucial for developing strategic plans. No previous study has been performed to explore the dynamics of power consumption with the change in renewable energy production on a country-wide scale. In contrast, a number of deep learning algorithms have demonstrated acceptable performance while handling sequential data in the era of data-driven predictions. In this study, we developed a scheme to investigate and predict total power consumption and renewable energy production time series for eleven years of data using a recurrent neural network (RNN). The dynamics of the interaction between the total annual power consumption and renewable energy production were investigated through extensive exploratory data analysis (EDA) and a feature engineering framework. The performance of the model was found to be satisfactory through the comparison of the predicted data with the observed data, the visualization of the distribution of the errors and root mean squared error (RMSE), and the R2 values of 0.084 and 0.82. Higher performance was achieved by increasing the number of epochs and hyperparameter tuning. The proposed framework has the potential to be used and transferred to investigate the trend of renewable energy production and power consumption and predict future scenarios for different communities. The incorporation of a cloud-based platform into the proposed pipeline to perform predictive studies from data acquisition to outcome generation may lead to real-time forecasting.
Power grids across the United States are increasingly exposed to weather-related disruption, yet reproducible analyses are limited by differences in the spatial and temporal structure of outage and meteorological records. This Data Descriptor presents a harmonized, event-based archive linking county-level customer power-outage records from the U.S. Department of Energy EAGLE-I platform with National Weather Service watches, warnings and advisories encoded through the Valid Time Event Code system. The archive spans the contiguous United States from 2015 through 2024. The primary released file contains hourly indexed, positive processed outage observations that overlap at least one VTEC warning and include event identifiers, matched warning types, and source-record availability diagnostics. Companion files provide the consolidated warning archive, daily county-level EAGLE-I source-record availability and sensitivity summaries evaluating missing-record treatment, temporal aggregation, and recovery persistence. Technical validation quantifies the consequences of preprocessing assumptions and documents spatial, temporal and warning-matching consistency. The dataset supports reproducible analysis of spatiotemporal co-occurrence between reported power disruptions and operational weather warnings while explicitly retaining limitations associated with the underlying outage record.