Ph.D. Thesis Defense Announcement

CHARACTERIZING AND MODELING ENERGY FLEXIBILITY AND DECARBONIZATION POTENTIAL THROUGH METRICS, HEURISTICS, AND FORECASTING IN CRYPTOCURRENCY AND HIGH-PERFORMANCE DATA CENTERS

By Veronica Margot Paez

Advisor:

Dr. John E. Taylor

Committee Members:

Dr. John E. Taylor (CEE), Dr. Susan Burns (CEE), Dr. Joe F . Bozeman III (CEE), Dr. Neda  Mohammadi (CEE), Dr. Troy Cross (Reed College)

Date and Time:  April, 20, 2026, 1PM

Location:  SEB 122 / https://shorturl.at/slOVK

Across three complementary studies, an empirically grounded, multi-scale
framework is developed to measure and reduce the energy and emissions impacts
of flexible high-performance computing (HPC) and cryptocurrency data center
(CDC) loads. Using novel hourly electricity data from 21 North American
cryptocurrency data centers matched to locational marginal emissions, the analysis
shows that emissions outcomes depend not simply on whether facilities curtail, but
on when, how often, and how deeply they do so, quantified through engineered
metrics of curtailment dynamics and emissions alignment. A data-light behavioral
signature—a Kneedle-derived knee-threshold—is then introduced that separates
both HPC and CDC facilities into statistically distinct operational types and provides a practical proxy for benchmarking flexibility and mitigation performance using only time-series energy data. Finally, site-level behavior is connected to system-level drivers through a block-level cryptocurrency network simulator that endogenizes miner economics, hardware profitability, difficulty adjustment, and halving dynamics to hindcast and forecast hashrate, revenue, and energy under alternative scenarios. While the empirical case studies focus on cryptocurrency data centers, the broader framework speaks to a wider class of flexible compute loads, including emerging HPC and AI facilities whose power demand is increasingly shaping grid planning, emissions accounting, and infrastructure investment. Together, the studies link operational heterogeneity, behavioral signatures, and protocol-driven market incentives, showing that the energy and climate impacts of large-scale compute are not fixed properties of demand, but emergent outcomes of facility behavior, grid conditions, and system design. This framework provides a stronger foundation for emissions measurement, scenario analysis, and policy design for lower-carbon flexible compute loads.