In June, Andreas Breuer, vice president of new technologies for German utility RWE, told an audience at GTM's Grid Edge Live conference that a team from the giant utility was searching Silicon Valley for new distributed energy technologies to invest in. It looks like RWE has found its first -- behind-the-meter battery startup Stem. 

On Thursday, Stem announced it had closed a $30 million investment led by RWE Supply & Trading, the energy trading arm of Germany-based RWE Group, It’s the final installation of Stem’s $45 million Series C round, which began in April with $12 million from Japan's Mitsui, and came in about 50 percent oversubscribed, CEO John Carrington said.

Some of Stem's previous investors also joined in Thursday's round, which brings the company’s total VC investment to about $75 million. The company has raised money from Angeleno Group, Iberdrola (Inversiones Financieras Perseo), GE Ventures, Constellation New Energy, and Total Energy Ventures. 

RWE will also take a seat on Stem’s board of directors to “support the company’s evaluation of new markets,” according to Thursday’s announcement. While Carrington wouldn’t say just how much RWE invested, he did call it a "substantial lead" in an interview this week. 

“They’re one of the largest electricity and gas providers in Europe, and they have operations in a lot of important markets for us, including Germany, the Netherlands and the United Kingdom,” Carrington said. “The financing model we’ve put together, and the value we provide to the utilities and the C&I customer, is really interesting to them. And the software we’ve developed was a big, big part of their decision to partner with us.”

Stem is just one of a number of companies connecting behind-the-meter batteries with emerging business models. Green Charge Networks, Coda Energy, Sunverge, Advanced Microgrid Solutions and the dynamic duo of Tesla and SolarCity are significant competitors in California, the epicenter for behind-the-meter energy storage, as well as in states like Hawaii and New York.

Stem uses lithium-ion batteries and power converters/inverters from a variety of partners, and integrates them with its own control hardware and software, which meters and analyzes second-by-second building energy data in search of spikes or peaks in energy use. When Stem catches one of these spikes in progress -- or even better, predicts one coming -- it injects battery power to feed the buildings circuits and prevent that peak from occurring.

These consumption spikes are bad, because they trigger demand charges, a fee that's based not on how much energy a customer consumes over the course of a month, but on their highest single point of usage, usually measured in 15-minute increments. In states like California, Hawaii and New York, where almost all of this type of energy storage has been deployed to date, demand charges can add up to nearly half a commercial customer’s utility bill.

Of course, batteries can do a lot more than hunt down and subdue building energy spikes. In California, Stem has been participating in a Pacific Gas & Electric pilot program that allows it to aggregate its customers’ energy shifting capacity as a demand-response resource. It’s also contracted to provide 85 megawatts of aggregated capacity to Southern California Edison as part of the utility’s long-range plan for managing the grid impacts of losing the San Onofre nuclear power plant. 

European markets are a bit different. Right now, the U.K.’s competitive energy markets and capacity-constrained grid might make for the best fit with this demand-charge reduction model, Carrington said. In the Netherlands and Germany, however, the primary driver might be more focused on managing the new energy-market conditions that have arisen with their massive expansion of wind and solar power.

Stem is currently working with Kyocera and SunPower to combine solar and batteries for commercial buildings. To date, Germany’s behind-the-meter energy storage market has largely focused on residential, where falling feed-in tariffs and rising retail energy prices are making storing rooftop solar more lucrative than feeding it back to the grid.

In the larger scheme of things, RWE and other German utilities have been facing billion-dollar losses in recent years, as the rising scale and falling cost of renewables have undercut energy sales and revenues at their fossil-fuel-fired power plants. In response, RWE and fellow utilities like E.ON have been making moves into the distributed energy and customer energy services fields -- and energy storage is likely to be a key part of this transition.

On the commercial front, RWE is white-labeling residential solar-storage systems from German startup Sonnenbatterie, and has built a large-scale virtual power plant with Siemens that aggregates combined-heat-and-power systems, backup generators and small-scale hydropower resources as a grid market resource. It’s also fielding about a dozen pilot projects across Europe that combine large-scale renewable energy, customer-sited energy resources and local control systems, with the goal of finding cheaper and more reliable ways to integrate green energy into its operations. (Here’s a cool interactive map of RWE’s projects across Europe.)

RWE has developed a “Smart Operator” system, which autonomously regulates energy flows and grid conditions on a local level, using a shoebox-sized control box that links individual homes to a local substation. About 250 homes have been testing the system over the past year or so, with in-home devices hooked up to reduce power use when local grids are under stress.

The changes could go deeper. According to an internal RWE document revealed two years ago, the utility plans to transform its business from dependence on large-scale power generation to a capital-light, distributed energy model. That would be following in the footsteps of E.ON, which last year announced a radical plan to spin off its traditional energy business units into a separate company and refocus its core efforts on renewable energy and distributed customer-facing project development.