Abstract
The Eastern Shore of Virginia (ESVA), which includes Accomack and Northampton counties, is facing a significant electricity reliability crisis. This peninsula is cut off from the mainland, relying on an old distribution grid that lacks sufficient transmission backup. The situation is exacerbated by increased electricity demand resulting from second-home development, widespread electrification, and intense heatwaves, particularly evident during the summer of 2025. Low-income residents, renters, elderly households, and shift workers are disproportionately affected by the reliability issues and associated costs. The primary challenge is to provide residents with more reliable and affordable electricity without requiring massive infrastructure upgrades immediately.
To tackle this issue, our team—comprising Danny Freedman, Avery Suriano, Walker Watson, and advisor Professor Henning Mortveit—created a detailed digital model of the Eastern Shore's electric grid. This model combines three elements: a synthetic household population of 28,221 households, yearly hourly electricity demand profiles adjusted for local demographics and climate, and a synthetic distribution network that connects homes to 21 substations. A logistic adoption model determines household participation in price-based demand response (PBDR) programs based on factors like age, income, and occupancy. We applied a Monte Carlo simulation to account for behavioral uncertainty in load outcomes at the substation level. We tested two PBDR policies: Time-of-Use (TOU) pricing, which encourages users to shift their energy use away from peak times in the evening, and Tempo-style tiered-day pricing, which prompts bigger usage shifts on high-demand “Red” days. Both approaches resulted in peak load reductions of 4-5%, consistent with existing research, and were validated at both the household and substation levels without compromising accuracy.
Implementing dynamic pricing changes the way grid management works, shifting who pays for it and who benefits from flexible incentives. Not all households can adjust their energy use easily when they cook, do laundry, or control heating and cooling. Higher-income households equipped with programmable thermostats, smart appliances, and electric vehicles are in a better position to benefit, while lower-income households, renters, elderly residents, and shift workers might face higher costs without much ability to adjust. If PBDR is built around an ideal consumer who has technology and flexible schedules, it risks increasing inequities within the grid system. Three frameworks in STS help clarify this issue. Langdon Winner’s idea that artifacts carry political meaning shows that TOU and Tempo pricing systems favor those with technological resources and flexible schedules, rather than being neutral. Sheila Jasanoff and Sang-Hyun Kim’s concept of sociotechnical imaginaries highlights how policy design creates an ideal energy consumer who is rational, informed, and tech-savvy, excluding many residents of the Eastern Shore. Lastly, energy justice frameworks, especially focusing on distributional and recognition justice, offer criteria to evaluate whether cost savings and benefits of flexibility are shared fairly and whether vulnerable households’ experiences influence policy design.
My STS research uses discourse and policy analysis. I investigate how PBDR policies appear in regulatory and utility planning documents, identifying which needs and behaviors are prioritized. I apply frameworks from Winner and Jasanoff to examine the underlying ideologies in smart grid design choices. I also analyze the distribution results from our digital twin simulations, and the STS analysis indicates that PBDR benefits primarily favor middle-to-higher-income households with three or more occupants, especially those aged 36 to 65. This demographic group is identified by our logistic adoption model as most likely to participate. Lower-income and elderly households, which have lower participation probabilities, are less likely to reap savings and may experience unintended cost shifts. The relative demographic similarity in the Eastern Shore reduces variation at the substation level, but the model reveals that in more diverse areas, concentrated demographic challenges can lead directly to unequal support from infrastructure.
This project underscores that modernizing the grid is more than just an engineering task; it is a sociotechnical one. The digital twin offers a detailed platform to simulate demand-side policies and assess their impact on load patterns at both the household and substation levels. This provides policymakers with a way to implement changes without waiting for expensive infrastructure upgrades. The STS analysis ensures that these policies are viewed through a fairness lens, highlighting whose flexibility is being used and who might be overlooked. The larger takeaway is that infrastructure modeling needs to consider equity and social inclusion from the start, not as an afterthought, to ensure that building a resilient grid is both effective and just.
