A wide-ranging conversation yesterday with IBM's Jeff Katz, chief technology officer for energy and utilities, produced a few nuggets I thought I'd share today.
One of the themes of the conversation was that technology and systems implementations need to be fully exploited for their value, value that's not always originally anticipated. Conversely, without a holistic view of how systems affect each other, implementations can have adverse impacts that were not anticipated.
One example of the former, now fairly obvious, is that smart meters provide energy use data for billing but also serve in load profiling, smarter demand response programs, as line sensors for outage information and producing more accurate estimated times of restoration (ETR).
An example of the latter: as utilities add central and distributed intelligence to the grid, they need to understand the not-so-intelligent aspects of their grid and how intelligence-based actions or signals might elicit an unintended response from, say, an analog device.
"Smart systems need to have a model that accounts for where the 'dumber' parts are," Katz said. "Because all the consumer knows is that 'they put in a smart grid and we still had a blackout.' It's up to the utility and its vendors to minimize that."
Other potentially unanticipated system behaviors sometimes arise when scaling up an implementation, Katz said. On the IT infrastructure side, installing ten times the number of meters in a pilot may not mean ten times the server capacity. Sometimes server performance doesn't follow a linear progression but may instead have greater, even exponential impacts. System design should anticipate such apparent anomalies.
Katz referred to this subject as "systems of systems design issues" and the unanticipated behaviors of grid components as "emergent behavior."
For an industry that obviously has lurched ahead with interval meters and advanced metering infrastructure (AMI), it's healthy to remember that AMI is just one access point to grid modernization, Katz said.
"We see a lot of utilities enter smart grid at the smart meter phase, but we don't necessarily think that's the only access point," Katz said. "It really depends on the business case at a given utility—issues of performance, what issues their PUC (public utilities commission) may want them focusing on."
I suggested that we were hearing a lot about distribution automation as an application for "smarts."
"Well, distribution automation covers a lot of things," Katz noted. "We know utilities that are starting by putting in high-speed, bi-directional communications in transmission substations. They need to protect those [assets] against any centralized disturbances. There are vertically integrated utilities looking at 'what can we do between generation, transmission and distribution to optimize those links. Some of them are looking at the integration of renewables. How do you balance traditional solar, wind and storage capabilities to make the best use of green [power]—to not just say 'we're accommodating green [power]."
"Funny you mention 'distribution automation,'" Katz added. "One utility I spoke to said they saw a lot that could be done under 'distribution automation,' but they don't call it 'smart grid.' So there are many utilities doing very interesting work, they just don't call it 'smart grid.'"
That said, Katz said he didn't see any real hurdles to the largely residential and small business rollouts currently projected to reach some 65 million interval meters nationwide by 2015.
"We're seeing a greater enthusiasm among utilities for dealing with all of the security issues up front," Katz said. "Certain utilities have had some issues in that area and I think and vendors are improving their internal firmware and the backend, the meter data management systems are catching up to the promise of smart metering data.
Another important area of utility activity, of course, is "optimization," and Katz captured the amorphous nature of this term and its meaning to a specific utility with the observation that "optimization is in the eye of the beholder."
Finally, we chatted about cyber security.
I noted Katz' remark at Grid ComForum in 2010 when he said: "There are ways to deal with massively distributed risks in a reasonable way." (See "Small World.")
Yesterday he responded that the experience in other vertical industries, from finance to health care, from aerospace to chemicals and petroleum, can be brought to bear on the energy sector.
"There are ways to keep an eye on distributed intelligence, even when you don't know the provenance of all end points and their security," Katz said.
Smile of the day
Yesterday we noted author Daniel Yergin's view that the public's interest in energy issues tends to rise in a crisis. So I asked readers whether a crisis is the right time to sell the benefits of smart grid. We've seen a number of utilities point to those benefits following the extreme storms of the past year or two to justify infrastructure investments. ("If you gave us $100 million, those outages wouldn't have been so widespread or lasted as long." I.e., "We told you so.")
What could be more threatening to national security than a blackout (okay, two blackouts) at the stadium hosting Monday Night Football? According to The Washington Post, San Francisco Mayor Ed Lee called the incident at Candlestick Park "a national embarrassment." I think most felt the embarrassment should be focused a bit more locally.
Such an outage, though, is just one step away from a malevolent attack during the Super Bowl, right? Well, a vendor we know capitalized on the widely seen event to tout how its products could've predicted and prevented the transformer blowout that caused the first blackout. (The second remains under investigation.)
The Post said, in contrast, that the incident was only likely to spur action on the need for the San Francisco 49ers to find or build a new home. Candlestick Park is more than a half-century old. So much for a "teachable moment."
Intelligent Utility Daily