Grid data analytics is an expansive field, but for simplicity’s sake, let’s break it into two broad categories: the visionary (and expensive) and the mundane (and necessary).

Last week’s Grid Edge Live conference in San Diego provided a series of examples of both kinds of data analytics efforts, from the grand-scale, million-plus-customer data analytics projects like Baltimore Gas & Electric’s engagement with C3 Energy, to cloud-based data aggregation and analysis apps for municipal and rural utilities.

It also offered a nice example of how San Diego Gas & Electric, a noted leader in applied grid technology, is investing in data analytics at both the grand scale and in a more remedial fashion. The first, SDG&E’s multi-million-dollar weather data collection, analysis and information platform, represents the high-end model, suitable for the biggest utilities and available as a resource to many more.

The second, SDG&E’s Smart Meter Operations Center, is a more workaday effort to merge smart meter, grid sensor, SCADA control system and customer service and social media data to find out why a smart meter isn’t working. That seemingly simple problem underscores a broad market need for fine-tuning AMI networks, and perhaps a commensurate potential for scale for the software vendors that can provide the right mix of solutions.

The visionary: A world-class weather platform to fight wildfires

Jeff Nichols, information security and management director for SDG&E parent company Sempra Energy, described both projects during a data analytics and software panel at last week’s conference. While the day’s weather was typically mild and Mediterranean, those cloudless skies also mean tinder-dry foothills that can spring into devastating fires -- and for SDG&E, the value of investing in IT to fight them is incalculable, Nichols said.

SDG&E has also suffered real-world losses from fires, most notably a 2007 blaze in which the utility’s power lines were found to be one of the causes. The utility embarked on its weather data project in 2008, and has since built long-range wireless data collection and standards-based interfaces to share its weather data with federal, state and local firefighters and emergency responders, who can access the information via handheld or tablet device.

SDG&E also employs onsite meteorologists to interpret the data, and has proven its acumen to the extent that the U.S. National Oceanographic and Atmospheric Administration (NOAA) has started using the utility’s data, Nichols said. Even so, SDG&E has struggled to quantify the cost-benefit equation of investing in prevention of this kind under the state’s current regulatory structure, he noted.

Still, the value in preventing accidents, coordinating with firefighters and keeping the public informed outweighs those kinds of regulatory concerns, he said. Utilities around the country are focusing on finding economic and regulatory structures to boost their spending on grid reliability and resiliency in the wake of extreme weather events like Western droughts and East Coast superstorms. Grid giants like Schneider Electric/Telvent and ABB/Ventyx are building weather forecasting into their grid outage restoration and workforce management platforms, and IT giants like IBM are crunching massive weather data sets to predict the impact on energy infrastructure.

The mundane: Fine-tuning the ubiquitous smart meter network

SDG&E’s smart meter operations center (SMOC) project is trying to solve a different unknown variable, not as complex as the weather, but still limited by the technology the utility has today. That’s the seemingly simple task of figuring out why a smart meter or series of smart meters in SDG&E’s 2.1-million-customer network isn’t working, he said.

In day-to-day reality, that’s a hard question to answer, he said. The wireless network itself might be breaking down, or it could be a software problem -- or it could be a customer removing the meter from its socket, or a downed power line. The AMI network head-end software can answer some questions, but the more information that can be drawn from alternative sources, the more likely the information coming out the other end is to be correct, he said.

Networking giants including Cisco and South Korea’s SK Telecom (through its U.S. subsidiary GridMaven) have staked a claim in a growing business for this kind of smart meter network management system (NMS) expertise. Most of the smart meter makers and grid networking vendors are beefing up their own NMS capabilities, as are meter data management software vendors like Siemens’ eMeter. One startup landing notable contracts on this front is Bit Stew, a Vancouver, B.C.-based company that’s helping hometown utility BC Hydro and networking partners Cisco and Itron collect, analyze and interpret large amounts of data to find and fix network faults.

The tricky part comes in identifying the value of fixing problems that are as yet undetected, and thus unmeasured. That’s always been part of the SMOC’s goals, which was designed in collaboration with U.S. and Canadian utilities to create a system architecture and formal business case for this kind of network monitoring and visualization, according to the utility’s 2008 smart grid update.

Notably, SDG&E says that the SMOC is just the precursor to a separate proposed effort for “applied data analytics, exception management, asset management, and predictive modeling,” which makes for a good short list of the kinds of tasks lots of data analytics providers are promising out of their AMI networks.

Nichols noted that most utilities are still in more of a wait-and-see mode when it comes to investing heavily in these kinds of projects. But as more utilities deploy smart meters, they’re likely to draw on efforts like SDG&E’s SMOC for guidance on how to optimize them for their core functions like meter reading, remote shutoff and turn-on, and tamper detection -- and maybe then, they can start on the more advanced analytics applications.