Large Power Savings Found in Automated Demand Response Tests
During summer 2004, Lawrence Berkeley National Laboratory (Berkeley Lab) researchers used two different kinds of technology, a price signal sent over the internet to facility computers, and a hard-wired internet relay box, to test automated demand response. They found up to four megawatts (MW) of savings in 36 buildings located at 18 sites, according to a new report.
The research, published as "Findings from the 2004 Fully Automated Demand Response Tests in Large Facilities," took place under the auspices of the Demand Response Research Center, which is funded by California's Public Interest Energy Research Program and led by Berkeley Lab's Environmental Energy Technologies Division (EETD).
Demand Response Defined
Demand response (DR) is a set of time-dependent activities that reduce or shift electricity use to improve electricity grid reliability, manage electricity costs, and encourage load shifting or shedding when the grid is near its capacity or electricity prices are high.
Fully automated demand response does not involve human intervention but is initiated at a home, building, or facility by receipt of an external communications signal, which starts pre-programmed load-shedding strategies.
During summer 2004, researchers tested automated demand response at 36 buildings representing 10 million square feet of floorspace. The technologies involved were a price signal sent over the internet to the facilities' computers, and a hard-wired internet relay box. The facilities participating in the test included several office buildings, a supermarket, a cafeteria, industrial process sites, a university library, and a postal processing and distribution center. Facility staff at each site pre-programmed the site's Energy Managment and Control Systems to receive the signals.
The research team developed new technology to evaluate control and communications capabilities for automated demand response using energy management control systems (EMCS) and XML, the extensible Markup Language. The facilities participating in the test included several office buildings, a supermarket, a cafeteria, industrial process sites, a university library, and a postal processing and distribution center.
The team found that when the maximum amount of load had been shed at the study sites, a total of about four MW of demand response was available, as shown in Figure 1. Demand savings were more than one MW per site, with up to 42 percent of whole-building power saved. Maximum savings per site were 1.8 W/ft2, with an average of 0.5 W/ft2 and 14 percent of the whole-building load.
According to Mary Ann Piette, Director of the Demand Response Research Center and leader of the EETD research team, "This work has shown that today's control and communications technologies can be used to deploy broad-scale demand response that is safe and secure with minimal investment. Automating demand response helps reduce the need for facility staff to manually control equipment in response to utility communications currently based on email, phone calls, and pagers."
About one-third of California commercial building floor area is controlled by an EMCS that could be remotely accessed for automated demand response with this technology.
"The largest individual savings were observed from strategies that used a cooling zone set point increase," adds Piette. Lighting; anti-sweat heaters; and other heating, ventilation, and air conditioning strategies also contributed to the savings.
The research team concluded that "There is significant demand reduction potential in large buildings and commercial facilities during warm weather. No occupant complaints were registered even with these large reductions in whole-building power."
First Automated Demand Response Test Took Place 2003
EETD researchers completed a previous successful evaluation of automated demand response in 2003, at five large California buildings. This was the first test of the use of two-way internet-based communications to reduce electricity consumption in large buildings.
The 2003 test used a fictitious electricity price—a proxy for a critical peak price—to trigger a demand-response event over the internet; no one touched any control systems during the tests. When the price of electricity transmitted over the internet to the buildings reached 30 cents/kilowatt hour, the buildings automatically began to decrease demand by reducing lighting, air conditioning, and other energy-consuming uses.
Two-way communications were used to observe whether each site was receiving and responding to the price signal. When the internet indicated that the price reached 75 cents/kWh, the buildings automatically took additional preplanned actions to further reduce demand.
Current and Future Research
Encouraged by the success of the 2003 test, the research team initiated a new phase of testing during summer 2005, working with Pacific Gas and Electric Company (PG&E). In the 2005 test, demand response technology was used for the first time in a real utility demand-reduction program. About a dozen facilities are participating in this new, fully automated Critical Peak Pricing Program (CPP).
In exchange for a price break during off-peak hours, participating facilities curtail their loads in response to a price signal during peak summertime demand periods. Unlike in earlier tests, the CPP program is a formal utility-pricing program, so its use of automatic demand-response technology is the first test of automated demand response in an existing utility program.
Berkeley Lab collaborated with Itron and PG&E to develop the automated technology that connects PG&E's CPP notification system and the automated price server.
For more information, contact:
- Mary Ann Piette
- (510) 486-6286; Fax (510) 486-4089
A report titled Findings from the 2004 Fully Automated Demand Response Tests in Large Facilities was authored by Mary Ann Piette, David S. Watson, Naoya Motegi, and Norman Bourassa of Lawrence Berkeley National Laboratory (LBNL-58178). Download it here.
This research was supported by the California Energy Commission's Public Interest Energy Research Program.