Pacific NW smart grid project called a success as it concludes

Source: Colin Sullivan, E&E reporter • Posted: Monday, July 13, 2015

A five-year regional smart grid project spanning five states in the Pacific Northwest recently ended with a healthy note of optimism from the consultant that ran the pilot.

Project leader Battelle, which oversaw the experiment for the Energy Department, praised the demonstration as one of the nation’s largest rollouts of a complex smart grid to date, noting a number of lessons learned on the ground both good and bad.

The project was a $178 million affair involving 11 utilities; the Bonneville Power Administration; two universities; and multiple companies in Washington, Oregon, Idaho, Montana and Wyoming. It was funded under the American Reinvestment and Recovery Act through DOE along with matching funds from participants.

In sum, a report from Battelle — which has managed DOE’s Pacific Northwest National Laboratory in Richland, Wash., since 1965 — found that smart meters, automated control of power distribution assets and “intelligent” control systems can improve efficiency and reduce power costs if correctly implemented.

At the same time, Battelle noted some blips in the system and said more research and development is needed before utility-led smart grids become widespread reality throughout the United States.

The project evaluated 55 “smart grid” technologies and found that energy reductions and cost savings varied with each technology and each test location.

On the up side was the case of Avista Utilities, according to the report. The utility gained the ability to start and stop power service remotely in Pullman, Wash., eliminating nearly 3,000 service calls per year and saving about $235,000 in the process. Avista was also able to reduce the voltage of its distribution system by 2.1 percent, possibly saving as much as $500,000 in annual costs associated with feeder distribution power lines.

Another case cited by the report was Portland General Electric having attained a 5-megawatt battery as part of a “high-reliability zone” in Salem, Ore., that could save the utility $146,000 annually by providing an alternative power source during periods of peak demand. And NorthWestern Energy reported that a “fault detection and isolation” system significantly reduced two power outages that otherwise would have lasted much longer.

Problems identified were broader in nature. For instance, participants were not necessarily prepared to handle “the onslaught of data and sometimes mislabeled data” that come from new smart grid equipment. Battelle also found that smart grid products from different vendors have to be better coordinated to make sure they work together.

“Some manufacturers went out of business or stopped servicing their products, while some equipment simply failed,” a summary of the report said.

Perhaps most significant was the progress made on the development of “transactive control” under which automatic, electronic transactions between energy providers and users determine whether providers should buy or sell power. IBM, a project participant, is one of the companies at the forefront of developing such systems to help old-style utilities in their transition to a less centralized grid (EnergyWire, July 7).

Transactive controls were updated every five minutes with the data sent to participating utilities on a test grid that did not directly influence regional power generation. When these signals predicted peak power demand, and therefore high costs, the smart grid devices in play were designed to theoretically cut electricity use.

The test grid, built by Alstom Grid, was run in parallel with the actual grid and showed “transactive signals would have correctly advised smart grid equipment to alter their operations during two critical moments on the actual regional grid,” the report says.

These moments were during an unexpected nuclear power outage in Washington and a sudden decrease in winds, both in February 2014.

IBM created another model that looked at different simulated situations on the regional grid and found that the Northwest’s peak power demand could be reduced by 7.8 percent if 30 percent of the traditional grid used transactive demand-response equipment.

The legacy of the demonstration project will most likely be felt in the installed equipment left behind. The Battelle report estimated that more than 30,000 smart meters, more than 12,000 “fault location” monitors and $80 million in equipment were installed, 88 percent of which remains in place.

Ron Melton, the project’s director, had this to say in a message attached to the report: “Like any demonstration project of this size and nature, we worked our way through unanticipated and perplexing technical issues and challenges. But we also experienced rewarding accomplishments, learned a lot, and achieved some really great outcomes.”

Click here to see a 32-page “highlights” version of the report.