Buildings account for 41% of annual energy consumption in the US, more than either the transportation (29%) or industrial (30%) sectors1. Interestingly, energy management in buildings is also the least optimized among these 3 sectors.
“More interestingly, in the near future buildings will interact with the electric grid to provide services that protect against blackouts, reduce electricity bills, and mitigate emissions.”
Such “smart” buildings are composed of mathematical models interacting with sensors and controllable heating, ventilation, air-conditioning (HVAC) and lights.
To develop the mathematical models, eCAL students assembled and deployed a 70-node wireless sensor network (left) to monitor temperature in offices and labs across the UC San Diego campus.
"The goal of the project was to take a data-driven approach toward understanding the performance of a building's HVAC system,” says CEE PhD student Eric Burger.
“By focusing on observable trends, we can gain insight into how the system actually behaves rather than how it was designed to behave.”
The nodes contain temperature sensors, ZigBee wireless radios, and Arduino microcontrollers. To construct the network, eCAL held a two-day “assembly party.” This involved soldering electronics, drilling plug ports in plastic enclosures, programming each node, and testing the system.
Loan Kim Pham, CEE undergraduate researcher, solders the sensors.
“Working together with the entire research group to assemble these sensor nodes was really fun," said CEE undergraduate researcher Loan Kim Pham. "It was nice seeing the final results after hours of assembling.”
Burger explains further, “I enjoyed the challenge of developing and deploying a sensor network on a large scale. It is one thing to find data on a website for an engineering project with no real understanding of where or how the data was collected.
"This project was completely different. We collected months of temperature data from around 60 points in a building. I know exactly how the data was collected.”
Buildings are engineering systems that we interact with every day, but we tend to give little attention to them. Unless something is going wrong, like a flickering light or a broken pipe, we have no real sense of how a building is performing.
“What excites me about this project,” says Burger, “and about smart buildings in general is that they provide the opportunity to better understand built environments through sensing and control. With this research, I see the chance to improve both the environmental and human health impacts of buildings, via real-time sensing and control.”
by Scott Moura
CEE Energy, Civil Infrastructure and Climate &
1US Dept. of Energy. (2010). Buildings Energy Data Book.