This project integrates data science and machine learning into engineering and social science frameworks to address challenges in creating low-carbon power systems with large shares of renewable energy sources.

Research includes dynamic electricity pricing based on supply and demand; reinforcement learning to control home appliances; federated learning at the grid edge for enhanced security and privacy; and policies to implement equitable energy transition and assist low-income households in adopting decarbonization technology, where machine learning is instrumental.

Hawaii has high penetration of renewable energy and an exceptionally high share of households with rooftop solar systems in the United States. Thus Hawaii and the UHM campus form a natural `living laboratory’ to conduct experiments to evaluate and validate the research.


Decarbonization pathways

Descartes applies data science to address critical needs in the pathway to decarbonization.


Efficiency and equity of renewables integration

Descartes investigates the socioecomic impacts of renewables integration.