An official website of the United States government
Energy justice advocates for the equitable and accessible provision of energy services, mainly focusing on marginalized communities. Adopting machine learning in analyzing energy-related data can unintentionally reinforce social inequalities. This perspective highlights the stages in the machine learning process where biases may emerge, from data collection and model development to deployment. Each phase presents distinct challenges and consequences, ultimately influencing the fairness and accuracy of machine learning models. The ramifications of machine learning bias within the energy sector are profound, encompassing issues such as inequalities, the perpetuation of negative feedback loops, privacy concerns regarding, and economic impacts arising from energy burden and energy poverty. Recognizing and rectifying these biases is imperative for leveraging technology to advance society rather than perpetuating existing injustices. Addressing biases at the intersection of energy justice and machine learning requires a comprehensive approach, acknowledging the interconnectedness of social, economic, and technological factors.
more »
« less
Unfairness Detection within Power Systems Through Transfer Counterfactual Learning
Energy justice is a growing area of interest in interdisciplinary energy research. However, identifying systematic biases in the energy sector remains challenging due to confounding variables, intricate heterogeneity in treatment effects, and limited data availability. To address these challenges, we introduce a novel approach for counterfactual causal analysis centered on energy justice. We use subgroup analysis to manage diverse factors and leverage the idea of transfer learning to mitigate data scarcity in each subgroup. In our numerical analysis, we apply our method to a large-scale customer-level power outage data set and investigate the counterfactual effect of demographic factors, such as income and age of the population, on power outage durations. Our results indicate that low-income and elderly-populated areas consistently experience longer power outages, regardless of weather conditions. This points to existing biases in the power system and highlights the need for focused improvements in areas with economic challenges.
more »
« less
- Award ID(s):
- 1938106
- PAR ID:
- 10565887
- Publisher / Repository:
- nergy-transition NeurIPS 2023 Workshop on Causal Representation Learning.
- Date Published:
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The U.S. Department of Energy (DOE) Communities LEAP (Local Energy Action Program) pilot partners with low-income, energy-burdened communities experiencing either environmental justice challenges or direct economic impacts due to a shift away from historical reliance on fossil fuels. In Alachua County, Florida, Communities LEAP partnered with the community-led Project EMPOWER (Energy Modernization for People, Opportunity, Work, Equity, and Renewables) to provide technical assistance in the areas of community engagement, solar power, weatherization and energy efficiency, green jobs, and fund development. This final report summarizes work accomplished during the Communities LEAP technical assistance and provides potential next steps for the EMPOWER team to consider as it continues pursuing its goal of a fair and equitable clean energy transition for Alachua County.more » « less
-
The growing integration of distributed energy resources (DERs) in distribution grids raises various reliability issues due to DER's uncertain and complex behaviors. With large-scale DER penetration in distribution grids, traditional outage detection methods, which rely on customers report and smart meters' “last gasp” signals, will have poor performance, because renewable generators and storage and the mesh structure in urban distribution grids can continue supplying power after line outages. To address these challenges, we propose a data-driven outage monitoring approach based on the stochastic time series analysis with a theoretical guarantee. Specifically, we prove via power flow analysis that dependency of time-series voltage measurements exhibits significant statistical changes after line outages. This makes the theory on optimal change-point detection suitable to identify line outages. However, existing change point detection methods require post-outage voltage distribution, which are unknown in distribution systems. Therefore, we design a maximum likelihood estimator to directly learn distribution pa-rameters from voltage data. We prove the estimated parameters-based detection also achieves optimal performance, making it extremely useful for fast distribution grid outage identifications. Furthermore, since smart meters have been widely installed in distribution grids and advanced infrastructure (e.g., PMU) has not widely been available, our approach only requires voltage magnitude for quick outage identification. Simulation results show highly accurate outage identification in eight distribution grids with 17 configurations with and without DERs using smart meter data.more » « less
-
Abstract Major disasters such as wildfire, tornado, hurricane, tropical storm, and flooding cause disruptions in infrastructure systems such as power and water supply, wastewater management, telecommunication, and transportation facilities. Disruptions in electricity infrastructure have negative impacts on sectors throughout a region, including education, medical services, financial services, and recreation. In this study, we introduced a novel approach to investigate the factors that can be associated with longer restoration time of power service after a hurricane. Considering restoration time as the dependent variable and using a comprehensive set of county-level data, we estimated a generalized accelerated failure time (GAFT) model that accounts for spatial dependence among observations for time to event data. The model fit improved by 12% after considering the effects of spatial correlation in time to event data. Using the GAFT model and Hurricane Irma’s impact on Florida as a case study, we examined: (1) differences in electric power outages and restoration rates among different types of power companies—investor-owned power companies, rural and municipal cooperatives; (2) the relationship between the duration of power outage and power system variables; and (3) the relationship between the duration of power outage and socioeconomic attributes. The findings of this study indicate that counties with a higher percentage of customers served by investor-owned electric companies and lower median household income faced power outage for a longer time. This study identified the key factors to predict restoration time of hurricane-induced power outages, allowing disaster management agencies to adopt strategies required for restoration process.more » « less
-
Household power outage experiences vary based on outage characteristics and the household's ability to cope with a disruption. While disaster management scholarship has produced methods to predict where the most significant impacts of a hazard may occur, these methods do not anticipate secondary effects, such as those from power outages. This research is necessary as the expected risks associated with power outages will increase in the United States due to climate change, increasing electricity demand, and aging infrastructure. To understand households' power outage experiences, we collected 896 surveys from three cities in the United States: Detroit, MI; Miami, FL; and Phoenix, AZ. Participants were recruited through Amazon's Mechanical Turk (MTurk) service to complete a survey. We hypothesized that racial/ethnic minority groups, specifically non-white households and lower-income households experienced more frequent and prolonged power outages. We also hypothesized that the same groups were more likely to have experienced more significant adverse effects, such as throwing away perishable food and not receiving assistance. We found that non-white households in Phoenix and Detroit were more likely to experience longer outages than white households; however, this association was not present in Miami and was not statistically significant in any city. Income was not a major factor in predicting food waste or assistance received during the longest self-reported outage. Further assessments in varying geographical contexts and more generalizable samples are necessary to increase understanding of how households experience power outages.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
