|
|||||||
‘Smart’ sensors promise savings in electricity costs
By Diane Ainsworth, Public Affairs
updated 15 June 2001 | A network of tiny, wireless sensors capable of monitoring the lighting and temperature of a building has the potential to cut California energy costs by as much as $7 billion — and reduce the Berkeley campus energy bill by about $900,000 a year. The
campus could realize those savings if all of its 250-plus buildings were
outfitted with the new sensors, a group of engineers, environmental experts
and facilities managers predicted at a recent demonstration of the latest
“smart” sensor technology to be installed in Cory Hall.
“Very, very small changes in our ability to monitor and control energy,
and in how we consume it, can make dramatic changes in the cost to the
state and, ultimately, in our ability to make it through these (black)
outs,” Newton said. “It’s a very leveraged approach to energy management.”
For the demonstration, researchers tucked the diminutive sensors into
the corners of offices, placed them along hallways and mounted them in
conference rooms throughout the fourth and fifth floors of Berkeley’s
main electrical engineering building, Cory Hall, to demonstrate how the
network operates. Berkeley engineers and experts in the College of Environmental
Design are working with campus facilities managers and industry partners
to build and test the array of “smart dust motes,” which trimmed 2 megawatts
(10 percent) off Cory Hall’s energy consumption during a 1.5-hour practice
test of energy conservation in late May.
"We have two key new technologies,” Newton said. “One is on the
hardware side: very small sensors that we think we can get down to the
size of a penny. Imagine sticking a penny on the wall; having 50 of them
that will all communicate with each other automatically, wirelessly, and
form a network that would control conditions in a room or building.”
The other key technology development is software architecture — in this
case a small platform with a tiny operating system that gives the sensors
computing power and their own “smarts” — that allows technicians to monitor
and control lighting and temperatures.
“Sensor networks are not a new idea, but the technology that is used
today is very large and clunky, expensive, and difficult to deploy in
large infrastructures like this one (Cory Hall),” Newton said. “There
is no software architecture that is integrated and tied into the (energy
distribution) network right now that allows the utilities people and those
making policy decisions to manage the relationship very well.”
The new smart sensor information technology also packs the potential
for myriad applications. Smart sensors can be designed to monitor all
sorts of environmental conditions, such as traffic congestion, air pollution
or magnetic fields; be used for disaster mitigation by monitoring the
motion of buildings during an earthquake; monitor the pulse, blood pressure
or movements of elderly people; and be used in new distance education
networks.
“Our researchers have developed a prototype energy management technology
that can eventually be deployed for less than a dollar per sensor,” Newton
said, “but the goal of CITRIS research is to develop technologies like
this that can readily be put to use to solve large-scale societal and
quality-of-life problems.”
In preparation for the demonstration, Associate Professor Kris Pister
and Professor Jan Rabaey, both of electrical engineering and computer
sciences, installed more than 50 solar and battery-powered smart dust
motes throughout Cory Hall, an older building that consumes large amounts
of power — about $1 million worth each year. The sensors were outfitted
with wireless radio transceivers and their own miniature operating systems
to transmit continuous lighting and temperature readings to facilities
managers controlling the building’s electricity usage.
“This web of sensors is providing huge reams of data about what’s actually
happening at any moment,” Pister said. “We were able to drop 50,000 to
80,000 watts of power in this building during a practice trial. On a critical
power shortage day in California, we could drop the load for one, two
or three hours with no impact at all to people’s comfort. That’s as much
electricity as it takes to power my street.”
Miniaturization is the centerpiece of Pister’s research. With the wireless
sensors already developed, now it’s time to make the matchbox-sized sensors
even smaller — about the size of a button — and reduce the cost of manufacturing
them to less than $1 each. Researchers are only two to five years away
from commercializing that miniature sensor technology, Pister said.
With proposed funding for CITRIS —slated to receive $33 million in matching
state funds from next year’s budget — further miniaturization and mass
production of the technology will be possible, dropping the cost of the
sensors down to pennies, Pister said.
Buildings, which use about one-third of all energy consumed in the state
and nation, could be outfitted with tens of thousands of tiny sensors,
all tied into a central computer that would regulate energy usage, said
Edward Arens, professor of architecture and director of Berkeley’s Center
for the Built Environment. During peak hours of energy consumption, air
conditioners, ventilating systems, lights and computer networks would
be automatically shut off, then turned on again when the demand had subsided.
Once buildings have a bit of brains, Pister said, passive sensors will
be designed to perform more intelligently. The next generation of smart
dust motes could be sophisticated enough to cut power automatically to
certain building systems during specific times of the day.
“We would see a savings of about $900,000 a year on the Berkeley campus
if all of the buildings were outfitted with these sensor networks,” he
said. “The network would pay for itself in a year.” |
|||||||
|
|||||||
|
|||||||
Home | Search | Archive | About | Contact | More News Copyright 2000, The Regents of the University of California. Produced and maintained by the Office of Public Affairs at UC Berkeley. Comments? E-mail berkeleyan@pa.urel.berkeley.edu. |