Research Team: Clifford Federspiel, Edward Arens, David Auslander, Craig Lin, Shan Tang, Danni Wang

The project is a collaborative effort between four research centers at the University of California: The Center for the Built Environment (CBE), the Berkeley Sensor and Actuator Center (BSAC), the Berkeley Wireless Research Center (BWRC), and the Integrated Manufacturing Lab (IML), in the Department of Mechanical Engineering. These centers have expertise in environmental control for buildings, micro-sensor technology, ultra-low power radio technology, energy scavenging, and packaging, respectively. The project will produce multi-modal wireless sensing and communication technology suitable for buildings as well as many other applications.

Research under this project at CBE is focused on two areas, airflow measurement technology and the use of sensor networks for controlling indoor temperature:

AIRFLOW MEASUREMENT TECHNOLOGY
We have developed a new algorithm for using ultrasonic chirp signals for measuring air velocity in open areas of buildings. The technology enables us to measure air velocity over arbitrarily long path lengths. Ultimately we plan to use networks of these sensors for flow visualization indoors. This work is important because air motion affects thermal comfort, indoor air quality, and energy consumption in buildings. It is also important because currently available technology is expensive and cannot produce spatially averaged readings.

Download Matlab code associated with this work. (zip, 6K)

MULTI-SENSOR, SINGLE-ACTUATOR CONTROL OF TEMPERATURE
We have developed an algorithm for using information from a sensor network to control multiple spaces in a building with a single actuator. Usually a single sensor placed in one of the spaces is used to control this type of system. Our results indicate that we can simultaneously reduce energy consumption and improve comfort by replacing the single sensor with a sensor network that has at least one sensor in each space. The performance improvement is achieved without changing the actuation, making the strategy ideal for retrofits in existing buildings.

Download a paper describing the results of this work. (pdf, 86K)
Download Matlab code associated with this work. (zip, 309K)

MULTIPLE SENSOR CONTROL OF UFAD SYSTEMS
We have also studied the energy implications of using sensor networks to control systems that are designed to intentionally produce a temperature gradient indoors. These systems are called underfloor air distribution (UFAD) systems. They are becoming increasingly popular in commercial buildings. These systems are commonly controlled with a single temperature sensor. Our results indicate that we could significantly improve energy performance by using a sensor network with two or more sensors in each space to control such a system.

Download Matlab code associated with this work. (zip, 13K)
Download a paper describing the results of this work. (pdf, 236K)