By Jeff St. John
Smart meters aren't just remote-controlled utility cash registers. They can also be valuable sensors for utilities, telling them about voltage, frequency, power factor and other such data that's needed to manage the modern power grid. There's a catch, however — and it lies in the utility back office.
Simply put, utilities aren't ready for the "big data" flood that's coming from their new smart meters. Many may be struggling with the transition from billing once a month to every hour or 15 minutes, let alone trying to manage other meter functions, like outage detection and anti-tampering functions. Asking them to comb through reams of other smart meter data may be too much.
At least, that's how Varun Nagaraj, senior vice president of product management at Echelon (ELON), explains the company's latest variation on power sensing for smart meters. On Tuesday, the San Jose, Calif.-based company announced a new set of software for its smart grid platform that separates grid power quality data from billing data, and makes it available on an as-needed basis.
Echelon, which reported last month that it planned to lay off employees and restructure over the next two quarters of 2012, also named Danish utility NRGi as a first customer. In fact, Nagaraj told me in a phone interview that Echelon developed the capability in partnership with NRGi, which was looking for ways to get more useful power data out of their meters without being subjected to a "data deluge" for the effort.
It's an interesting way to tackle a problem that may just be emerging for smart meter vendors and the utilities they serve. Lots of smart meter companies can provide power quality data and integrate it with distribution grid management and automation. Silver Spring Networks does it for utility customers like American Electric Power Company (AEP) and FPL, for example, and Elster does it for Entergy (ETR) and Dominion Virginia Power.
Likewise, meter data management software from the likes of eMeter (now part of Siemens), Integral Analytics (now part of Toshiba's Landis+Gyr) and others is built with the goal of supporting such functions.
But when it comes to actually leveraging smart meters for grid power management, there's a big disconnect between potential and current practice, Nagaraj said.
"Any self-respecting AMI infrastructure vendor should be able to send a query when required to the meter to retrieve a certain piece of data," he said. "That's probably table stakes." But being able to get a piece of data doesn't necessarily equate to knowing what to do with it, he said.
To translate data to meaning, utilities need to put together load profiles, or combinations of data that are useful for utilities, he said. But most smart meter platforms today are limited in the number of separate load profiles they can run at once, Nagaraj said. They're also hard to reset outside of the once-a-month window in which billing cycles are programmed into the system, he said.
Echelon's new capability, to be upgradeable for its MTR 1000 series single phase (residential) and MTR 3000 series polyphase (commercial and industrial) meters, is meant to overcome both limitations, Nagaraj said. Essentially, it puts the distributed intelligence into each meter to compile as many as 16 load profiles at a time.
From there, the meters can send the profile data back to central office in simple, manageable bits of data, rather than flooding it with all the raw data that goes into each calculation, he said.
That's important, Nagaraj said, because utilities today are struggling to manage the flood of data they already have coming at them. Indeed, they may have more power quality sensor data than they know what to do with, he said.
"Echelon has long had the bragging rights for the number of power quality measurements it makes," he said. "But it was a hollow bragging right, because there wasn't much that utilities were doing with it." Reducing the amount of data, yet making it more relevant, can allow utilities to actually put smart meter functions to use.
"If utilities don't compact the data the way we're doing it, I think it will force utilities to move to a completely new data infrastructure they don't have today," he said. That, in turn, could push back the time it takes for them to utilize the functions they got when they bought the smart meters in the first place, perhaps by years, he said.
Whether or not these issues are affecting how utilities are using the power quality measurement capabilities of other smart meters is far from clear. But there's no doubt that the "big data" problem will continue to be a major focus for the industry.
Part of the solution could come from IT giants like IBM (IBM), which is working with Itron, SAP and Teradata to manage smart meter data, and Oracle (ORCL), a big MDMS vendor that's been pushing the envelope in massive processing of billing data.
At the same time, Oracle and grid giants like Siemens (SI), ABB (ABB), General Electric (GE) and Schneider Electric (GM:SBGSF) (via its acquisition of Telvent) are building distribution grid management systems that need that smart meter data in a form they can digest, so to speak, without too much big data crunching. On that front, Echelon also announced a new smart meter on Tuesday, the MTR 3500 Series, that can measure power quality at the distribution transformer level.