Are Smart Grids on the Verge of Viability?

Pacific Gas & Electric's detailed smart grid plans, laid out in a June 30 briefing, underscore the fact that California’s three largest utilities are planning to spend more than $5.5 billion on intelligent power networks over the next decade. PG&E worked with Southern Cal Edison and San Diego Gas and Electric to present a unified smart grid plan to the California Energy Commission in December 2010 -- which itself was a minor miracle.

The PG&E song-and-dance in late June was merely the show-stopper in a busy midsummer that included the UK's first significant utility buy-in for smart grids, an IBM-Tivoli effort to work with Sensus on smart grid security, and the first acquisition of an electrical contractor (Southern Power & Controls) by a smart grid platform provider (Infrax Systems). To cap it all off, the Department of Energy and the Federal Energy Regulatory Commission called July 6 for a National Action Plan to jump-start unified demand-response software.

Yet there's part of me that can't help but wonder if a universal smart grid is like nuclear fusion or gallium-arsenide chip technology, perennially remaining the technology of the future without ever quite getting there. This may be because I helped a smart grid software supplier work with the DoE a couple years ago, and I could see firsthand the difficulty in expanding an intelligent power grid beyond a municipality. It’s also the result of telltale signs from investors -- Smart Grid News reported in April that new funding for smart grids had collapsed, and M&A activity also took a nosedive in early 2011.

Let's start with the positives. The vision of providing both small power suppliers and end customers with real-time tools to optimize power usage and power supply for both green usage patterns and cost saving is a sound one. Smart grid technology is more than ready for primetime from a bottom-up perspective. Platform software can provide real-time weather updates for the insertion of solar and wind power from the supply side, and can integrate the input of smart meters from the customer side. Integration of smart grids with electric vehicle powering infrastructure lags a little bit, but, then, so does electric vehicle ubiquity. Deals like IBM-Sensus and Infrax-Southern Power show most point suppliers are thinking in terms of an end-to-end solution.

But the deal-killer for smart grids on a national and global basis is not unlike the thing that sunk T. Boone Pickens's plan to invest in large wind farms. Just as Pickens discovered that the transmission system would not support the centralized wind facilities he wanted to build, the disparate, pasted-together state of generating and transmission systems in North America is far from being ready for a unified smart grid software approach. Europe and portions of Asia are slightly ahead of the US here, but smart grid enablers are still local phenomena.

They say the most successful revolutions come from below, and smart grid technology certainly has the advantage of moving up from localized instantiations to region-wide power monitoring. Some of the tools available today can interface metering and demand-response across dissimilar utility networks. But we should not underestimate the task of rebuilding power-generation and transmission-point sources for an IP-connected, embedded-intelligence world. If we minimize the effort required, we may be left with islands of intelligence floating in a sea of dumb power networks, with few practical guides to converting the haphazard network into a unified smart grid.