Cleantech Startup AutoGrid Tracks the Peaks of Austin’s Energy Use

Big Data has come to thermostats in Austin.

Austin Energy, along with Palo Alto, CA-based AutoGrid, is analyzing when and how much electricity is used by the city’s homes, businesses, and industrial users in order to make the power grid more efficient.

“We are crunching all the data and how we can drive efficiently throughout the electricity supply chain from generation to consumption,” says Amit Narayan, AutoGrid’s founder and CEO.

The startup’s pilot program with the Austin utility started last week and is designed to measure energy usage at about 50 thermostats and 15 electric vehicle stations. That’s a fraction of the utility’s approximately 450,000 residential, commercial, and industrial customers, but still, a sample size that can provide valuable information, says Russell Shaver, a consulting engineer who specializes in electric vehicles and emerging technologies at Austin Energy, the city-owned utility.

Most of us are familiar with some sort of energy management. We set our home thermostats higher when we are away during the hot summer days, to save energy and money on air conditioning. In Texas, where this past weekend, the heat index climbed as high as 111, some downtown skyscrapers in Dallas dimmed the lights in their lobbies, a small measure to reduce pressure on the electricity grid.

Often, however, those measures are not enough. Heat waves send us to the thermostat punching the temperature down. Tornadoes and hurricanes take out power infrastructure. These and other events put excessive strain on power grids that are already maxing out their capacity in the hot months, resulting in brownouts, power cuts, or power loss.

Increasingly, utilities and cities are implementing more formal programs of energy management, which is creating an opportunity for cleantech startups. In May, bought New York-based EnergyHub for an undisclosed amount, and a few days earlier, Nest, a smart thermostat company designed by former Apple engineers, purchased MyEnergy, a Boston-based energy data company.

The idea behind programs such as the partnership between Austin and AutoGrid is to balance electricity demand and supply in real time by using technology to predict times of peak demand. Then, smart thermostats can automatically reduce power usage for non-essential uses during those peak times. 

“There is very little storage in the electricity grid, so demand has to always match the supply,” Narayan added.

Normally, when demand shoots up during peak events, Austin Energy switches on “peaker plants,” which taps into the city’s electricity grid, as well as that of the state’s, the Electric Reliability Council of Texas, or ERCOT. And ERCOT charges a hefty premium for non-peak energy for “emergency” power. That’s a price utilities would like to avoid paying.

“The pilot can help in reducing our demand in order to reduce Austin’s contribution to the ERCOT grid,” says Austin Energy’s Shaver. “This reduces our costs.”

AutoGrid’s solution is “Demand Response Optimization and Management System,” or DROMS. Narayan says one advantage of AutoGrid is that it works with multiple thermostats and through a variety of reporting systems, whether over WIFI or through legacy thermostats that have only one-way communication. This makes AutoGrid attractive to smaller utilities, which usually can’t afford to purchase expensive hardware of their own, he added.

“It’s plug-and-click compatible,” he says.

The increasing adoption of smart meters, which can record and measure data as often as once a day or every hour, is making these programs more and more viable. Smart meters were used in about 23 percent of the U.S. market in 2011, more than double the amount in the year before, according to the Federal Energy Regulatory Commission’s 2012 Demand Response and Advanced Metering Survey. In Texas, smart meters are used by more than 30 percent of the market population.

AutoGrid says that in addition to compiling energy usage data it can also help utilities analyze that data to find ways to make usage more efficient. 

For Austin Energy, better management of its power grid means it has to buy less energy during peak use times—energy that costs five times the price of off-peak power, Shaver says.

The Austin pilot is partly funded through a $3.5-million ARPA-E grant from the Department of Energy, as part of its Green Electricity Network program.

AutoGrid, which was founded in 2011, raised $9 million in venture capital in October from Foundation Capital, Voyager Capital, and Stanford University, where Narayan was director of Smart Grid Research in Modeling & Simulation. 

AutoGrid’s other clients are mainly in California—the City of Palo Alto Utilities and the Sacramento Municipal Utility District— though Narayan says the company has two additional Texas utility clients, which he declined to name.

Shaver says the utility and AutoGrid will continue to monitor energy usage through August and September, the final two months of the typical peak-demand season, which begins in June in Texas. He says he’s most curious about the usage of energy at the electric vehicle stations they are monitoring, something that hasn’t been done before.

“That can be like plugging in another AC or another house, depending on the car,” he says. “If adoption of electric vehicles really takes hold, this will be another planning issue for us. We’ve got to make sure we can handle the increased load.”

Trending on Xconomy