Blog Post

EPRI’s Erfan Ibrahim: 5 Myths About the Smart Grid Buildout

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Panel discussions at technology conferences are often pretty dry affairs, but smart grid expert Erfan Ibrahim of the influential group Electric Power Research Institute (EPRI) proved today at a conference in San Francisco that even topics like home area networks can rile a crowd. Speaking at The Networked Grid conference hosted by Greentech Media, Ibrahim — who heads EPRI’s work on communications, systems management and cyber security for the smart grid — dished out a sort of grab bag of opinions on a range of hot smart grid topics. Here are five myths, misunderstandings and truths he raised about how the smart grid should be rolled out.

The electric grid is not antiquated but it needs connectivity:  There is a rumor that the electric grid is antiquated that is “spreading from Silicon Valley” by people who don’t really understand the industry. The grid is made up of many intelligent devices that are widely dispersed. The challenge, generally speaking, is not to improve the devices that are already deployed but to cost-effectively network them together.

Don’t put too much smarts into the meter: If you do, five years down the line you’ll regret it as technology advances. Then your meter will be obsolete. Instead, the intelligence around services like home area networks should be “in a box” that you place in the home and that can easily be swapped like a home computer as technology advances. In a related comment, Ibrahim said meters aren’t the only route for the smart grid to extend into the home. Other options could evolve, though he didn’t offer any examples.

Grid security requires comprehensive security, not just a big wall: Utilities can’t just build a big wall and expect to keep everyone out. A comprehensive approach to security will include technological advancement but also personnel training, process control and open digital security standards.

Go slow with the adoption of interoperability standards: Smart grid communication standards should be based on best practices. So the industry should develop the science first and then put the standard in place, not the other way around (Pacific Gas & Electric’s (s PCG) Kevin Dasso, while speaking on the same panel, appeared to disagree, however, when he said that you can either let the standards control you or you can work to control the standards) . Ibrahim also said the smart grid industry shouldn’t think it can take Internet Protocol and quickly shift it over to the electric grid. There are several reasons, one of them being that the electric grid must be more secure than IP currently enables.

Home energy management systems need to impress the neighbors: Many Americans won’t pay a dollar for a soda out of a machine, but put them at a bar next to someone they want to impress and they’ll easily spend three. So home energy management solutions that just show you how much energy you use and when won’t get much market penetration. The solution has to have some bells and whistles that will allow customers to impress their neighbors, say by linking it in some way with the family flat screen TV or by controlling appliances automatically. In a related comment, Ibrahim said he expects the private sector demand-response players, such as EnerNOC or Tendril, to increasingly offer customer-centric solutions for lowering their power usage. These solutions may be tied to home energy management systems.

7 Responses to “EPRI’s Erfan Ibrahim: 5 Myths About the Smart Grid Buildout”

  1. MR Ibrahim is more wright than wrong. As soon as commentators talk about the problems of peer to peer power arrangements, they are confusing physics (electricity flows in the direction of a potential difference), with contractual and regulatory arrangements. If the smart grid is about anything at all, it is about information management and knowledge about what is happening on the grid at all points at all times. Anything beyond that confuses “what should be done” with “who should do it”. And myth #2 goes to the heart of that confusion.

  2. waltinseattle

    I almost inverted the “myth” #1 and thought he meant to say it was a myth that the grid was not antiquated. I was thinking heres a reasonable mind amond the hoopla. The grid IS old and antiquated, this is true. Yes? but he says lets just give it some paint and a tune up and let it rip? hmmmmm

    About that time pricing- I’m sure, TIm, that can be massaged to the utilities advantage, and that it will….from their side a few dollars upgrade a bunch new recievables. From the customers perspective a meter to take out a mortgage on and a few dollars monthly….will it be one of those non-amortizing affairs like some recent home payment scams? Considering what some will pay as a “premium” for green cars- I bet , ah, how to say it except yep old P.T. Barnum is still right.

  3. “You can either let the standards control you or you can work to control the standards.”

    This sounds like something the Sphinx would say in the movie, “Mystery Men”.


    The smart grid is really of benefit to the utilities, not the consumers. As a result, unless we are careful, we will end up with a solution that likely will be utility oriented, at the expense of the consumer. All that’s really needed is real-time pricing, to which the consumer can react to with either manual or automatic controls guiding his/her instantaneous consumption. But it’s gotta be KISS, because the typical consumer would likely save less than $20 per month at best, with any such scheme. That doesn’t pay for too many smart appliances.

  4. In the din of ‘Smart’ grid talk, this is one of the rare occasion to hear from someone who knows what they are talking about. #1 should be clearly understood by anyone who want to be successful in Home Energy Management(yes, I prefer that over Smart Grid) business.

    I’m not 100% convinced about the overall economic impact of home energy management and its viability in the market place but if a communication network build out happens anyway, communication infrastructure security has to be the top consideration for any network architecture and protocol designs.

  5. Don’t let the EPRI talk fool you…because Ibrahim is most definitely wrong about myth #1. Sure, there are some newer technologies that ISO operators have at their disposal for some transmission level power flows, but the picture is quite different at sub-transmission and especially at the distribution levels. Further, with all the advanced controls that ISO operators apparently do have at their disposal, one wonders why transmission losses continue in the 8-10% range.

    The underlying electro-mechanical technology at every substation, radial network, spot network, service transformer and circuit recloser (and everything in between) has remained largely unchanged since the 1960s.

    Grid operators have been able to get away with using this type of technology largely because of the network topology of the grid is simple: top-down, one-way power flows. Electricity travels from the generating stations through transmission lines, to substations, to distribution wires and finally to a service drop with essentially no room for interference from sideways or upstream power flows.

    The problem that looms just as large as the information flows that smart meters will provide is interconnection. Simply put, the grid in its current form is unable to manage any type of peer-to-peer power flow from any kind of distributed generation. So, utilities place strict penetration limits on the amount of power that can be present on part of a feeder that isn’t originating at the substation

    Understandably, utilities are averse to risks to safety and grid stability, but there are technical solutions to the interconnection issue that don’t require a complete rebuilding of the electricity grid and its control schemes.

    The problem with the antiquated technology (and 19th century thinking that went into determining how grid power flows were to be managed) is that it necessarily limits the amount of clean, reliable, and economically viable distributed generation that can exist. Plenty of room for massive wind farms, concentrated solar and other big central plants. All of those, of course, require massive (think trillions of dollars) investments in transmission lines…the very same ones that lose up to 10% of the power being delivered over them.

    • Gosh, this sounds so good, and yet it’s so wrong. “Understandably, utilities are averse to risks to safety and grid stability, but…”

      No, reliability is job number 1. It’s not risk aversion, it’s called the primary responsibility and government mandate. That regulation is the price they pay for having a monopoly. At the same time I’m not going to apologize for slow uptake on technology and poor communication models from corporate offices. That’s the downside of monopoly. But if you had electricity delivery interrupted as much as internet service, there would be an armed uprising. Reliability and safety are super important.

      Distributed generation sounds easy and wonderful, but there are some fundamentals built into the existing grid that need to be re-engineered. And that takes time, plain and simple. Connectivity is definitely one of the issues. Overcurrent protection devices are what protects the distribution network, but there are real issues setting those up for distributed sources. Fault current is a nasty bugger to work with when it varies according to the source. Even if you can get these new devices installed, they do no good if you can’t communicate with them in real-time.

      It takes bandwidth to do real-time communication with devices that respond within one wave cycle to ever changing conditions. And so communication has to be carefully planned and engineered. It’s estimated that there is 300,000 times as much data to consider when operating all of these smart nodes, so that’s a serious problem.

      Wherever smartgrid is going, it’s going to take time to get there. Microgrids have a much larger chance of succeeding in the short-run until utilities re-engineer their systems.

      • I’m not sure how you see utility aversion to risk as wrong…you yourself just out and say that its job number 1…be it government mandate, bottom line, or otherwise…utilities just don’t like risk. And yes they are painfully slow to move on anything and everything – ever seen how long an interconnection request can take?. So you lost me on the whole “sounds so good but its so wrong thing”… Oh and internet service interruptions compared to electricity outages…I wouldn’t take that comparison too far. For one thing, for just about anyone with a cable or DSL modem a power outage constitutes an internet outage (by definition). So of course if there’s just one more cable outage than electric in a year, then its more cable outages…

        Anyways, those fundamentals that you talk about needing to be re-engineered…well that’s only really the case if you’re stuck with the paradigm of synchronous or induction-based interconnections. If you’re a proponent of microgrids – parallel, non-synchronous microgrids that just treat utility grid power as one among many sources of power and that look to the utility like nothing more than pure load reduction (to be a bit more specific)— then the approach to interconnection is going to have to be inverter based. That’s the only way large scale deployment of DG microgrids is going to happen.

        The next step will be to get those microgrids to either talk to each other or talk to the utility substation — a much easier task than you make it out to be. Remember, the speed of communications along an ethernet cable or a WLAN are talking place at the kHz level and above…a few orders of magnitude faster than the 60Hz phase frequency for utility power. So no its not that hard to get the digital controls to respond in ‘real time’. I’m pretty sure that GE has a nice little black box that does this for the microgrid bus already.
        Technically its not that difficult…but then we’re back again to those slow moving utilities…moving a few orders of magnitude slower than the 60Hz power they send through their wires.