Urban Sensing and Modeling to Inform Building Energy Efficiency

Urbanization is one of the great challenges of this century, with linkages to climate change and the need to develop sustainable use of energy and other natural resources. In the U.S., more than two-thirds of the population lives in urban areas that are trying to manage growth and build for resilience in the face of more extreme weather events, while much of America's aging urban infrastructure - including buildings, transmission systems, gas pipelines and the electricity grid -need to be repaired or replaced.

Buildings consume 40% of the primary energy in U.S. and up to 70% of cities' primary energy. To achieve cities' energy and climate goals, it is thus a key strategy to reducing energy use and associated GHG emissions in buildings through energy conservation and efficiency improvements. Urban sensing (e.g., AoT node), using large scale low-cost and connected sensors, can monitor environmental conditions, e.g., outdoor air temperature, humidity, pressure, CO2, PM2.5, PM10, noise, illuminance. The collected data can inform urban building energy performance through advanced data analytics such as machine learning and computational modeling and simulation. Computational tools empowered with rich urban datasets can model performance of buildings at urban scale to provide quantitative insights for stakeholders and inform their decision making on urban energy planning as well as building energy retrofits at scale to achieve efficiency, sustainability and resilience of urban buildings. This talk introduces LBNL's research on urban buildings using data collected from urban sensing network as well as cities' public datasets. LBNL's tool CityBES, an open data and computing platform for urban buildings, will be introduced with a case study.