Fueling the EV revolution with IT innovation

Fueling the EV revolution with IT innovation

Vehicles outfitted with a battery, electric motor and a plug have gone from curiosities to commonplace in just a few years. Dozens of new electric vehicle models (EVs) are slated to hit markets this year and next.

As prices keep falling, and range and mileage keep rising, sales of plug-in EVs are expected to grow by nearly 40 percent per year over the remainder of the decade, predicts Navigant Research, roughly 20 times faster than overall auto sales.

Yet to make the leap beyond early adopters and appeal to mainstream drivers, tomorrow's EVs face a hurdle. They must be as easy to refuel as today's petrol-powered cars. Hitting that goal won't be easy. It will demand integrating today's atomized network of vehicles, utilities, fueling points, retailers, banks and others into an interconnected web that relies on advanced analytics to keep an emerging ecosystem of EVs humming smoothly.

The refueling challenge

Today, filling up your tank is arguably easier than ordering fast food. At one of the country's 120,000 gas stations, a driver pulls in, picks regular or premium, fills up, swipes a card and goes -- all in a matter of minutes.

Behind that apparently simple exercise is a century's worth of process evolution. Invisible to the driver are decades of chemistry research to standardize how gasoline is refined. Also unseen is the distribution infrastructure -- pipelines, trucks and tanks -- that cost billions and took decades to build. Today, there are more gas stations in the U.S. than the top 10 fast food chains combined, including Subway, McDonald's and Starbucks.

The intangible aspects of this transaction also have taken years to perfect. In seconds, credit card readers swap data between a web of financial companies via secure, high-speed data networks. Now consider just a few of the challenges EVs face matching this convenience. Because electricity can't yet be stored on site, as can gasoline, utilities must be able to dynamically track and respond to demand from EV recharging points in real time, or risk blackouts. After all, recharging as quickly as possible using today's technology, a single EV briefly can pull the same amount of juice as an average-sized supermarket.

Tomorrow's EV driver might also opt for different "flavors" of power. Instead of regular electricity produced from fossil fuels, some might pay a premium for 100 percent renewable power.

Electric vehicle charging photo by ruigsantos on Shutterstock.

Nor will recharging be confined to gas stations. A driver might prefer to refuel while parked at work, using energy generated only from on-site solar panels. Others might opt for a quick top up from a charge point at their supermarket.

Wherever the juice comes from, the payment will be settled between a complex web of players, including:

  • The site host, which could be a gas station, parking garage or retail parking lot;
  • The charging post owner, who is likely to be different from the site host;
  • The utility that owns the wires through which the power is delivered;
  • The generator of the preferred power, perhaps a wind farm;
  • Loyalty programs and coupons, likely to come into play;
  • And in some cases, transactions may be routed back to an EV owner's home utility account.

IT delivers smarter charging

It's a complex puzzle, but it's one that utilities, EV manufacturers, retailers and key technology players are already beginning to solve using advanced analytics. Unlike yesterday's petrol stations, tomorrow's EV infrastructure is being built from the get-go with state-of-the-art information technology at its core.

In California, for example, Honda, utility PG&E and IBM are collaborating to learn and predict how EV drivers behave. To make charging simpler and more reliable, data streaming from a fleet of Honda Fit EVs -- being driven in real world conditions -- is analyzed by cloud computing systems. By digesting millions of data points, advanced analytics is teasing out patterns from this information, which in turn is allowing utilities to plan more effectively and address some of the Smart Grid challenges.

As an example, EVs can help overcome the variability challenges common with renewable energy -- we can charge EVs when the sun is shining and curtail them if cloud cover limits solar generation. But this is no easy feat – in order to do so, we need data that is onboard the vehicle, data from the grid, weather feeds and even anticipated EV usage patterns to optimize recharging, thereby helping to facilitate renewable energy.

Simply look at those behind the wheel -- EV drivers tend to top off their batteries whenever possible, even if they have plenty of charge to get home. Analyzing this data can enable planners to experiment to help refine recharging capacity. For instance, variable peak pricing can help encourage more off-hour charging, or retailers and offices might aim to lure EV drivers by offering free charging as an amenity.

Further out, smarter charging systems also promise to help utilities move a step closer to the holy grail of turning EVs' batteries into a form of grid storage. By 2020, Navigant Research expects that some 200,000 EVs will be able to feed power back to buildings. If so, EVs could help prevent blackouts and also boost renewables by offering a way to mop up excess output from solar panels and wind mills.

As more EVs hit the road and more charge points pop up on the grid, the charging process must grow steadily more intelligent. Analytics is poised to provide utilities the smarts they need to manage the surge of information that the EV ecosystem is generating.

The better utilities understand today's early stages of recharging behavior, the sooner tomorrow's drivers will have a true alternative to petrol. Technology's challenge is to make the recharging process as easy as possible, by making invisible all the complexity going on behind the scenes.