The rapid expansion of artificial intelligence (AI) infrastructure is driving unprecedented growth in electricity demand from data centers. Traditional power-system planning treats large computing facilities as inflexible peak loads, leading to costly infrastructure upgrades and long delays in grid interconnection. Recent work has shown that AI clusters can reduce electricity consumption during peak demand through software-based workload orchestration. This article explores how modern GPU-based AI data centers can operate as grid-interactive assets that respond dynamically to power system conditions. We describe an architecture integrating grid signals, workload scheduling, and power telemetry for fine-grained cluster power control. Experimental results from a real-world deployment on a 130 kW GPU cluster demonstrate multiple forms of flexibility, including rapid load reduction, sustained curtailment, and carbon-aware operation while preserving service levels for priority jobs. We further demonstrate performance-aware load shifting across geographically distributed clusters, enabling workloads to migrate toward regions with lower grid stress. Together, these capabilities transform AI infrastructure from static electricity consumers into flexible resources that support grid reliability, accelerate interconnection, and improve computing sustainability.
Measurements of 2-furfuraldehyde (2-FAL) in oil are routinely used to predict paper condition in mineral oil filled transformers. Since the level of 2-FAL in oil is influenced by moisture, the hydrophobicity of the oil, aging conditions and the presence of other materials, adjustments may be required before this methodology can be applied to ester filled transformers. The variables that influence the relationship between 2-FAL and degree of polymerization (DP) were investigated through thermal aging of oil/paper systems across four transformer insulating liquids; a synthetic ester (SE), a natural ester (NE), an ester based biofluid (BF) and a mineral oil (MO). Whilst a copper catalyst had a negligible effect, more 2-FAL was found in SE/paper systems and less in NE/paper systems compared with MO/paper systems. 2-FAL levels were depleted following long-term storage whilst the use of wetter paper had the opposite effect. With minor adjustments, correlations between 2-FAL and DP established for mineral oil filled transformers can be applied to ester filled plant.
As climate change intensifies, extreme weather events increasingly threaten communities and infrastructure. These shocks can act as a form of creative destruction, generating opportunities for sustainable reconstruction, while also reshaping risk perceptions and influencing investment in green technologies. This paper empirically examines how natural disasters and weather extremes affect the diffusion of green buildings across the United States. Using a panel data set of buildings certified under the U.S. Green Building Council's Leadership in Energy and Environmental Design (LEED) program at the city level from 2000 to 2020, we find that major flooding events increase the number of LEED-certified buildings within three to six years. Hurricanes initially reduce green building development, followed by a rebound in later years. We also find that extreme hot days are associated with increased green building adoption, whereas the effects of cold days and moderately hot days are limited and less consistent. These response patterns are driven primarily by private buildings rather than public properties. Overall, our findings highlight the heterogeneous and dynamic ways in which climate shocks shape green building adoption and provide insights for policymakers and planners seeking to promote sustainable urban development in a changing climate.
