GM Daewoo Introduces New Matiz Creative Global Mini-Car
Germany Targeting 1M Electric Cars by 2020

GM’s Approach to Integrating Smart Plug-ins to the Smart Grid Leverages Emerging Standards and OnStar

Bellino1
GM is embracing the use of communications standards that operate with optional data transports to enable smart charging and integration with the emerging smart grid. EVSE=Electric Vehicle Supply Equipment. PLC=Power Line Communications. Source: GM. Click to enlarge.

While the ultimate vision of plug-in electric vehicles (PEV) operating as smart nodes in the smart grid offers compelling benefits for both end users and utilities, achieving that vision will require coordinated development and a staged, evolutionary implementation involving both the utility and automotive industries, according to George Bellino, GM manager of strategic programs, in a presentation at the Plug-in 2009 conference in Long Beach, California last week.

Among the basic requirements for successfully implementing smart vehicle/smart grid combination are specific, standardized requirements and solutions for vehicle-to-utility communications, and compatible methods to bridge the many gaps in the charging and communications infrastructure until North American and global standards are in place.

The emerging smart grid system is characterized by three elements, Bellino noted:

  1. A sensor system to determine the current state of the grid;
  2. An intelligent power management system to determine and to communicate the desired state of the system to the smart loads; and
  3. A network of smart loads with communications to report their current state and act on smart grid commands.

Plug-in electric vehicles (i.e., plug-in hybrids, extended range electric vehicles and battery electric vehicles) are smart loads, and so fall into the third element. As such, PEVs must have the ability to communicate with as well as process control signals from the smart grid—PEV integration requires smart charging.

The basic concept of smart charging integrates three elements with different communications responsibilities:

  • A smart grid utility systems operation center. The operation center handles real-time rate information; TOU (time of use) rate information; critical peak pricing information; load control signals; demand response signals; and account authentication.
  • The emerging automated meter interface (AMI) infrastructure (smart meter/smart outlets). These are responsible for premise identification; revenue metering; vehicle identification binding; pass through; rate information; load control information; demand response signals; and customer response.
  • The smart PEV. The vehicle is responsible for communicating vehicle identification; energy desired; charging status; charger voltage; duration; demand response; customer preference; and opt-in/opt-out.

The basic intent, Bellino said, is that the vehicle—wherever it is—connect to the grid and intelligently communicate its ID, and tap into particular charging programs with the participation of the owner of the vehicle. There are a number of significant barriers to realizing this, however, including:

  • Diversity in AMI protocol implementation among utilities.
  • Legacy and near-term production PEVs with no communications ability.
  • Interconnectivity and communications with Home Area Networks and Energy Management Systems.
  • No communications connectivity between PEV and utility in Non-AMI utility environments. 85% of United States will not have AMI implemented for at least another 10 to 15 years, Bellino noted.
  • Derivative solutions that may impact compatibility and standardization.
Bellino3
Part of the challenge in implementing a smart charging/smart grid scenario is handling the multiple combinations of different basic capabilities. The top row is the only one representing full “smart” functionality in vehicle, outlet and grid. Source: GM. Click to enlarge.

In other words, while the optimal configuration is to have the smart functionality in all three elements—vehicle, outlet and grid—the reality is that such a condition will be fairly rare in the near- and medium-terms.

The fundamental requirement for successfully navigating this situation will be the development and application of standards—specifically SAE J2847 and SEP 2.0, Bellino said.

SAE J2847 is an emerging automotive industry standard that will specify communication between PEVs and the grid; the ZigBee Smart Energy Profile (SEP 2.0) is an application layer utility industry standard that focuses on communications related to efficiency, usage, price, and messaging.

There has to be harmonization between SAE and SEP.

—George Bellino

It will be important, Bellino noted, that any gap or derivative solutions developed by the auto and utility industries be forward compatible with the long-term direction provided in the standards. It is also important, he noted, to avoid allowing for incompatible alternatives within the standard.

Bellino2
GM’s forward-compatible, standards-compliant smart charging communications solution. Source: GM. Click to enlarge.

J2847 and SEP 2.0 support compatibility and interoperability among alternative message transport protocols—they do not specify the physical layer. This enables GM’s basic smart charging communications technology vision of a standard language that operates with optional communication transports—including the use of OnStar for a non-AMI communication path (cellular data as transport layer). (All Chevy Volts will come with OnStar.)

OnStar can provide smart charging communications using a J2847/SEP 2.0-compliant message structure directly between the PEV and the utility systems operations center in situations where there is no AMI communication path.

OnStar also offers the potential to aggregate and report aggregate charging data. Bellino said that GM is reviewing the technical and business requirements for an enhanced data mining capability that will be of potential value to AMI utilities.

This market is going to grow slowly,” said Sunil Chhaya, of the Electric Power Research Institute (EPRI) in another presentation at Plug-in 2009.

Automotive and AMI technology implementation has a lead time of 3-5 years for the maturation of standards, suppliers, technology and capacity. The challenge is, if standards are not ready today, what do the OEMs design to, knowing that the life of plug-ins introduced next year will overlap with mature [future] standards?

The answer is to use a forward compatible transport layer...and a version-controlled application layer.

—Sunil Chhaya

Resources

Comments

ai_vin

Remember yesterday when I said everybody's jumping on the bandwagon?

GM makes it 5.

sulleny

"The answer is to use a forward compatible transport layer...and a version-controlled application layer. "

Looks like they're resigned to slow evolution of the layers dependent on growth of the installed base. Beyond the need of car makers to gather metrics data (GM via OnStar) Mr. Bellino's vision for smart-grid suffers from central-engineering.

Specifically the positioning of EVs as "smart loads." Bellino sees the world through the eyes of a utility. His first allegiance is to the "desired state of the grid." He envisions smart loads as taking commands from the utility. While this may be important to the utility, it is our contention that the grid must respond to real world demands of citizens - not the other way around.

None of this has any significance if the vast majority of EV owners simply plug in at night and set the timer to charge after rates drop (typically 11PM.) Rather than applying resources to complicating and expanding the grid with "smart" features - civic planners should be pressing to downsize the grid. Of course this means the adoption of Residential Power Units which immediately cut the demand for new power plants, lower energy costs for consumers and provide jobs and security domestically.

At this point GM should focus on data gathering of charge/discharge behavior. This is readily communicated via OnStar and logic already built in to the Voltage drivetrain firmware.

citizen

The other part of this that someone should be working on is an ability for the plug-in to not require human attention. If, for example, we needed to check the oil at the end of every trip we would be tired of that in a couple of days. The same will be true of plugging in. It needs to be automatic. Any ideas?

Mannstein

And you all thought owning an EV will free you from the clutches of big business.

Think again.

doggydogworld

Yeah, that's why everyone I know quit using cell phones. Plugging in all the time was just too much hassle.

Ken

I like using OnStar or other wireless communication. And GM would love to lock in OnStar as a standard for information to the grid and to PEVs.

As I said about the Ford announcement, avoid new wired infrastructures when possible.

What is the drivers incentive to send identity information the other way, from the vehicle to OnStar? Certainly law enforcement could use it. And marketing. But how does that benefit the driver?

Whether people will like their vehicles constantly reporting on their activities is problematic. But it is coming.

Alain

@ Sulleny
Right now, rates drop after 11PM because demand is low at night. If, as you say, everybody would plug in at 11PM, the rates wouldn't drop anymore since demand would not drop anymore.
What must be done is a continuously changing price, depending on supply/demand, where the actual price of one kWh is sent as a data-signal within the grid. That price will depend on supply - with an increasing fraction renewables it will fluctuate more - and demand. When for some reason the supply/demand would become >>1, the price will increase dramatically, so many cars will start selling their kW if they are 'allowed' to do so. That's the 'smart' part of the smart grid. With a lot of smart suppliers and a lot of smart customers, everybody should have the optimal price at any moment. Nobody controlls the price, since everybody can sell at any price, (as long as there are enough suppliers and there is a free trade). It's alomst exactly the same principle as buying/selling stakes : as a seller, you set a minimal price you are prepared to sell for ; as a buyer you set a maximal price you are prepared to buy for. That price may change continuously, depending on any factors (hour of the day, wind speed in Montana, sunshine in Texas, temperature in California, ..., and personal need for a fully-charged battery) just like in stake-trading.
If there start to appear predictable moments of high prices, smart algorithms will detect it within a few days and will preserve kW to sell at that very moment...

sulleny

Alain, your example is interesting but not rooted in a real world yet. Yes, rates will likely increase (utils will claim "demand" is causing this) as EVs come online. Remember however that the largest consumers of electricity are business, government and residences - all of which are offline after 5-6PM (business & gov) and 11PM for residences. So the rates at 11PM-7AM should remain the lowest regardless of new loads.

One of the benefits of EVs is the potential to charge them independent of grids. For those in the sunbelts of the world - a PV/storage system MAY prove economically viable, dependent on incentives, scale. Better, is offloading some residential demand via CHP Residential Power Units or gensets like MicroFueler's Gridbuster burning alcohol/water mixtures.

People who claim to be green should be comfortable with the idea of expanding distributed energy. Removing demand from the grid (via RPUs) rather than increasing grid size is second only to conservation in constraining energy consumption.

sulleny

Forgot to mention - with current li-ion chemistry in EV battery packs, adding discharge cycles is an economic loser. Depleting the battery to sell energy is nice for utilities - but damaging for the consumer. At .15-.20 cents a kWh how smart is it to sell even half your available charge ~4kWh??

And you must then replace that energy so you can drive the vehicle to or from work. "Selling" energy from your battery EV is not anywhere near ready for prime time.

fred schumacher

The grid using BEV battery packs as a storage, buffering medium is not workable in the real world. It means that vehicle owners will never be sure of the charge state of their battery packs. It also means more charge/discharge cycles which will shorten battery life.

We've been using dual-fuel electric heating for 30 years. Our primary heat, at the farm in North Dakota and our home in northern Minnesota, is electric, supplied at an off-peak rate of around 3.8 cents per kwh. At both places, power supply is remote-controlled by our electric co-op and is shut off during peak-load periods. Back up heating is wood, at the farm, and propane at our northern Minnesota home. Back up heating is a requirement for getting the lower dual fuel rate. What's the back-up for a BEV?

The comments to this entry are closed.