Analytics for Solar and Energy Storage: Show Me the Money

As solar PV and grid battery markets mature, so will data analytics.

Software that analyzes the complex interplay between solar PV, energy storage systems and customer energy use sounds like a great idea -- but is it something that companies are ready to pay for?

This depends on the level of maturity of the industry in question, and just what financial stake they see in capturing information that is now lost in the sea of data they’re collecting every day. That’s the view laid out by some experts on the topic during a Wednesday panel at Greentech Media’s Soft Grid 2014 conference, entitled "Intelligent Analytics for Distributed PV and Storage.”

“We are starting to see an increased focus, not just on wanting analytics, but wanting to pay for analytics -- which is an important distinction. What’s driving this is that the stakes are high,” Michael Herzig, president and founder of solar monitoring company Locus Energy, said during Wednesday’s panel.

Locus has built its business on metering and managing about 40,000 solar systems around the country for companies like Clean Power Finance, NRG Energy and Sungevity. But it’s also built a suite of analytics tools, including a “virtual irradiance” tool that converts weather and satellite data to allow customers to check system performance, and its PVIQ Waterfall software, set for commercial release next year, which can identify and differentiate between soiling, shading, panel degradation, inverter clipping and other reasons why solar systems aren’t performing as well as expected, all from raw data.

“Your basic monitoring system will tell you whether your system is up or down,” he said in an interview. “But this gets at the last 5 to 10 percent” of system inefficiencies. Back in the gold-rush days of solar, when installers were just putting up as many panels as possible, that wasn’t such a big deal. But now that solar is being aggregated, financed and securitized, it’s becoming a pressing issue, he said.

“The value of the securitization is a clear driver for paying for the data,” he said. “One of our customers literally guarantees one of its customers millions of dollars per system” based on the performance they predict for the system. “If analytics can give them 1 percent more accuracy, that’s a lot of money.”

While utility-scale solar is well monitored and controlled in terms of grid integration, distributed solar is spread out and disconnected from utility monitoring and control, making it a particular focus for data analytics, he noted. Knowing what all that customer-owned rooftop PV is doing to the grid is going to become increasingly important as it becomes a bigger part of the energy mix.

Grid-scale energy storage is still in its infancy compared to solar, noted Rogers Weed, vice president of product management for grid battery software startup 1Energy. Most of the big batteries, flywheels and other storage systems deployed on the grid to date have been one-off projects, and “certainly the first generation of software controlling them wasn’t able to take any analytics from the grid,” he said.

Behind-the-meter energy storage systems, by contrast, require a certain level of analytics to provide the demand charge mitigation, load balancing and other tasks they perform for the buildings they’re installed in, Weed added.

That work could help behind-the-meter battery providers like Stem, Green Charge Networks, Sunverge and SolarCity better manage their goals of aggregating and controlling these distributed resources for grid needs.

“Stem’s behind-the-meter storage requires a substantial amount of analytics,” said John Carrington, Stem CEO. That’s because each system is monitoring its building’s energy use, predicting power spikes, and even providing insight into inefficient use of HVAC systems or other loads, he added. Of course, not every distributed energy storage system needs this kind of sophistication -- “If you’re using it for emergency backup for a power outage, that’s a different context,” Carrington said.

But more and more distributed storage is being paired with rooftop solar, plug-in electric vehicles and home and building energy management systems, which makes it a useful tool for helping to balance the disruptions from those grid-edge systems. Stem is bidding aggregated batteries into a California demand response program and just announced a 1-megawatt pilot with Hawaiian Electric to help the island utility manage the effects of rooftop solar.

“You really have to look at coupling these together,” said Ajay Madwesh, vice president of utilities industry business for Space-Time Insight, a startup providing data visualization and analysis platforms for utilities and grid operators across North America.

Last week, Space-Time Insight landed a new customer -- Japan’s NEC, which is tapping STI's software for its demand response, microgrid and advanced metering infrastructure work with utilities across Asia. NEC is also the new owner of A123 Energy Solutions, the grid-scale energy storage business of the bankrupt U.S. lithium-ion battery maker bought by China’s Wanxiang two years ago.

“NEC bought A123 awhile back, and now they’ve got this great storage platform that’s generating a lot of data,” Steve Ehrlich, Space-Time Insight’s senior VP of marketing and product management, said in a Wednesday interview. “Now, what do they do with the data?”