Google announced last week that it would use thermal storage to cool the $300 million data center it’s building on 15 hectares of land in Taiwan. The search giant has been experimental in its data center design with everything from seawater cooling to water recycling, and it’s the first time it’s tried thermal storage to cool its power hungry data centers.
Thermal storage typically involves using ice or a liquid coolant that is chilled or frozen at night when power is less expensive, and then deployed in a heat exchange system during the day to cool servers. From the perspective of Google, it’s able to draw and store the energy it needs overnight in a thermal form.
What’s interesting about Google’s choice of thermal storage is that while the technology has been around for a while, it signals interest from Google in using stored energy to cool its data centers rather than just drawing power direct from the grid when it needs it.
The holy grail for energy
Energy storage has long been the holy grail for those favoring renewable energy since the inherent intermittency of wind and solar mean that storing power when the sun is shining and the wind blowing is critical for the times that it isn’t. Battery storage tends to lead the conversation about storing energy, exhibited most recently by the massive 36 megawatt hour BYD made battery storage station in Zhangbei, China. (The ultimate energy storage solution is, of course, fossil fuels, which have been so successful because they are energy dense and portable.)
But there’s experimentation with energy storage beyond football field sized battery projects. At The Wall Street Journal’s ECO:nomics conference last month, energy investor Bill Gates briefly mentioned that he’d invested in an energy storage startup he was calling “gravel on ski lifts.” The company is Energy Cache and has a very simple idea–fill buckets with gravel on a “ski lift” at the bottom of a hill, haul them up a hill, and release them back down the hill to produce electricity when it’s needed. The process is essentially using gravity to store energy.
What it means to store energy is also expanding. As demand response gets more bidirectional with power customers able to both power down and power up depending on utilities’ needs, many are starting to view the heating or cooling of a building as a means of storing energy. What’s driving this is the fact that we’re starting to see cases of “renewable energy waste,” times when the wind is blowing hard and there isn’t sufficient demand for that power.
Being able to signal an industrial building to heat when renewable energy generation is strong allows that power to be captured and stored. Companies like Grid Mobility and Enbala Power Systems are working on ways to help power customers increase their use of renewable energy as well as build the software to make it easy for them to draw additional power when it’s abundant. (For more here, check out my research note, “The big data tsunami meets the next generation of smart grid companies,” subscription required.)
Energy storage in the home and car
On a residential level, many folks have pointed out that hot water heaters are great thermal stores of energy. If a quarter of America’s hot water heaters could communicate with the grid to turn off at peak hours (or take renewable energy when it’s available), it would save customers $420 million a year. Duke Energy has considered using water heaters as energy stores in order to balance out fluctuations in solar power generation.
But the other dark horse in the energy storage race is electrical vehicles. While vehicle to grid (V2G), allowing utilities to store power in EV’s batteries and draw it back when needed, remains a dream of some, the bigger issue down the line may actually be spikes in overnight demand. Thinking back to Google’s choice of thermal storage, the premise is that drawing off-peak overnight power and storing it thermally has cost advantages. The same is currently true of electric vehicles, which charge overnight.
But in 10-15 years the number of EVs on the road may be significant, so much so that charging them all at 2 a.m. overwhelms the grid. Which is why companies like Toyota are working on technology to put EV drivers in communication with utilities to incentivize them to schedule charges at the best time for the grid. If EVs really take off, the utilities might have to reconsider vehicle to grid technology because they’ll need to utilize electric vehicle batteries as energy stores.
A diversity of storage technologies is emerging, mostly because we haven’t seen the types of breakthroughs in battery technology that everyone has been hoping for (just look at the massive resources available to the mobile industry for R&D and still people are unhappy with the battery life in the iPhone 4s). There’s no silver bullet here but there’s some momentum toward testing new energy storage technology to help store energy when it’s cheapest and when it’s most available.