Computational Decarbonization and Sustainable Energy Systems
Buildings, transportation, the electric grid, and computing itself account for a large share of global carbon emissions. As part of the NSF Expeditions in Computing project on Computational Decarbonization of Societal Infrastructure (CoDec), we develop computational methods to measure, reduce, and manage this carbon footprint (vision paper). Our work includes SpiderGen, which uses generative AI to produce procedures for carbon life cycle assessments, and studies of cost-effective incentives for siting solar deployments. We also look at the energy and carbon cost of AI: ECO-LLM orchestrates domain-specific language models across edge and cloud to balance accuracy against resource and energy use. This direction builds on the lab's long history of energy-efficient computing, from low-power PC sleep states to energy-aware mobile systems and smart buildings.
Projects
A nationwide agent-based model of rooftop-solar adoption to design cost-effective incentives as the 30% federal tax credit ends.
An LLM workflow that generates the process flow graphs behind carbon life cycle assessments, in minutes and for under $1.
Jointly optimizes the whole edge–cloud LLM serving pipeline per query, cutting cost by 60% and latency by up to 6x.
A research agenda for cutting lifecycle carbon emissions across interdependent computing, transportation, building, and power-grid infrastructure.
A 12-week field study with 46 university staff where social-norm energy feedback emails cut computer energy use by 10–11%.
Shifting content-delivery load among data centers by hourly electricity prices can cut both operator costs and health and environmental damages from emissions.
Splits a dormitory's single smart-meter reading into per-room energy use by fusing meter events with occupants' smartphone WiFi and audio data.
Fuses home electricity meter data with smartphone WiFi and microphone sensing to infer which appliance is used, where, when, and by whom.