Smart grid communications smackdown: the cellular companies’ “combine-and-conquer” strategy
Oct 19, 2010
AT&T, Cellular South, Rogers Wireless, Sprint, T-Mobile, Verizon, CTIA and Qualcomm gathered at a press conference in Washington, D.C. this week to advance a shared vision of cellular networks as the smart grid communications backbone. Coverage, pricing, disaster recovery, bandwidth and security were among the topics discussed.
The event was orchestrated by grid communications company SmartSynch. The Jackson, Mississippi-based firm is nominally network agnostic. For instance, its GridRouter product claims to enable communications to any device over any network.
But with this announcement, SmartSynch made it clear it is betting heavily on cellular in its bid to distinguish itself in a sector crowded with me-too competitors.
.
Below I’ve listed some of the claims made by the cellular advocates. At the end, I’ll mention an issue they are not talking about… and it’s something that may make all the other claims moot.
Proponents of the public path make these assertions:
·Lower cost. Cellular companies have dramatically reduced their charges in the past 12 months, with rate plans dropping to pennies per meter per month. In addition, SmartSynch has lowered the cost of installing IP-addressable cellular communications inside meters.
·Better coverage. Don’t confuse voice coverage with meter coverage. Meter antennas are almost four times as powerful as those on cell phones. As a result, public networks can give almost universal coverage in most areas (according to proponents).
·Automatic upgrades. When cellular companies upgrade to 4G, smart grid networks get an automatic upgrade without investing in new equipment. By contrast, private networks inevitably require costly technology upgrades, according to fans of the cellular approach.
·Lower capital and maintenance costs. Using a public network does away with the need to install and maintain your own equipment.
·Faster installation. Since the cellular networks are already up and running, there's no delay to install and test equipment.
·More headroom. There is ample cellular bandwidth for utilities that want to expand past smart meters into distribution automation and grid applications. Some mesh networks, on the other hand, require upgrades.
·Prioritization is possible. When utilities use commercial carriers, their data is sent on a dedicated channel. It is not interspersed with text messages and phone calls. In addition, the 4G networks now being installed have superior quality of service (QoS) capabilities that -- in theory -- make it easier for utilities to make sure mission-critical data jumps to the head of the line.
Champions of private, RF mesh networks make their own claims:
·No, we have better coverage. With private networks, utilities can use a combination of wireless technologies to achieve coverage in urban, suburban and rural environments. By contrast, cellular networks typically have underserved or even unserved areas.
·Better availability. Advocates claim private networks can be engineered to higher levels of reliability than public networks (something the cellular companies dispute vigorously).
·Better latency. Applications such as distribution automation and wide area monitoring require very low latency. (Smart metering is much less demanding.) In general, private networks have superior latency.
·Prioritization. If a utility owns and operates its own private network, it can ensure that mission-critical traffic is properly prioritized.
What you rarely see discussed is the issue that may be the biggest single reason utilities prefer private networks. Their traditional business model is based around installing equipment, then getting a regulated return. Utilities that choose a public network forgo this opportunity. That explains why most of the early cellular pilots have been with municipals and co-ops that are not subject to the same regulatory cost-recovery mechanism.
I am convinced that the public network option has made enormous progress in the last 12 months – so much progress that any utility thinking about its smart grid communications optionsshould, at the very least, give it a hard look.
What I don't know is whether the cost and technical improvements will be enough to overcome the rate base issue. What’s your take? Use the nearby QuickPoll to cast your ballot.
.
UPDATE: After this article first came out, several vendors and utility professionals emailed to say that the rate base issue isn’t that big a deal. The bulk of the cost is in the meters and the comms boards inside. Those costs can go into the rate base whether you use private or public networks.
Most users of cellular telephones may not realize that cellular providers usually only provide 4-8 hours of backup power for their sites. If the power has been off for days (for whatever reason), most of the cellular networks will be on their face (even those sites that have generators may not still have fuel).
If the power utility wants to use those cellular links for control, they won't be there. What then?
If this is only going to be used to deal with individual consumers, it might be OK (depends on what we are trying to accomplish), but using it for higher-level command / control / status seems penny-wise and pound-foolish to me.
The power industry should operate its own network, and if it requires anything more than the narrow bandwidths provided by the existing low-frequency (below 540 kHz) networks already in use, then that network ought to be fiber-based to prevent RF-based snooping / intrusion.
Before we start thinking about what kind of network we need, though, isn't it time to have a serious discussion about WHAT kind of traffic model we envision, and where we envision that, and WHY??? Do we presume we need gigabits / second just because the IT department thinks it's cool?
We should follow that, perhaps, with a real discussion of what kind of network protocol we need. Why is IP the correct approach, considering it's the tool of choice for hackers? WHY???
After we iron out those questions, THEN we may be ready to start talking about the network we need.
Dave Maples - 10/20/2010 - 07:07
What versus How
Dave Maples makes am extremely cogent point in emphasizing the "What" of system design rather than "How." Security, data mass and reliability are just part of a system design. It's very easy to get tangled up in how to create a system without a clear understanding of what the system is expected to do.
As for cell system reliability one has to remember the Gulf coast cell companies that had propane powered back up generators. As soon as Katrina flooded their areas the propane tanks floated away and the cell system sputtered to a stop.
Don MacConnel - 10/20/2010 - 07:41
Automatic upgrades - really?
I would like to hear more about the "automatic upgrades". Are the cellular companies really saying that if they change technologies, making all their subscribers obsolete as they have in the past, they will send someone to each meter to upgrade the radio? For free? Is it only from 3G to 4G?
Also, are the cellular companies prepared to commit publicly to guaranteed reliability figures instead of saying, "we will negotiate privately with individual utilities to provide the reliability you want" as they have in the past?
Grant Gilchrist - 10/20/2010 - 09:19
Proof in the Pudding
SmartSynch evangelizing the virtues of public networks is fine, even if it is ridiculously self-serving. The biggest issue to-date has been the lack of contractual commitments by the public network providers. They want to use SmartGrid as a new growth opportunity for themselves, but they won't commit to improving the operating characteristics of their network in order to give utilities what they need to make a responsible decision. Coverage, latency, capacity and DR all need to be packaged in service level guarantees that give the utilities the protections they need to trust the operation of their key functions over public networks. Even during storm events. And the SLAs need to have penalties for non-performance, just like the utilities get punished by regulators for poor performance. Thus far, the carriers in our area will only talk vaguely about SLAs. Until they are ready to make commitments its simply a marketing conversation. Public networks can be a great strategy for utilities in theory. Public network providers need to move the discussion from theory to reality by offering credible SLAs that utilities can base operations upon. Until then, its just talk.
U. Til Exec - 10/20/2010 - 10:01
Cellular versus Private Networks
In my experience at any given point in time a point company can provide a "better" technical solution that does not rely on existing capability and/or existing standards. However when the existing technology is backed by a huge base and is in a competitive environment, the technology will continue to improve at a fast rate. If a specific implementation of the existing capability stumbles, there will be others to take its place.
However in a few years hence the point solution is out of date and no longer a "better" technology. Support wanes and each year the technology becomes less effective compared to its competitors. The point company is unable to compete with larger companies funded by larger pools of users and the "better" technologies becomes less effective. The user is trapped in obsolete technology.
Today Smart Grid can be implemented with "better" private mesh networks but is it a "smart" decision. In the long term the point solution user is left with three choices -- 1. Invest in updating the technology, 2. Replace the technology with new capability based on established technology and standards, 3. Limp along on out of date solutions.
Cellular technology is improving at a breakneck speed and the innovation is being financed by the Cellular company and its subscribers as a whole. Private networks may offer advantages today but where will the resources come from to keep this technology competitive with cellular solutions in the future.
There are any number of high technology examples where specialized point solutions were briefly popular (operating systems, private work stations, specialized database managers, etc.) where users have accepted short term "better" solutions only to regret their decisions as mainline technologies improved and short term "better" solutions were abandoned.
Utilities are not in the communications business. They should leave that communications sector to the people whose main function is communications.
Chuck Drinnan - 10/20/2010 - 10:23
Utility Networks
Good article and comments by Maples and Drinnan, but I think we must subdivide the vague term "smart grid networks" for better clarity. Mission-critical SCADA networks are likely to remain utility-controlled for the private network advocate reasons listed above. Residential AMI networks are likely to increasingly choose cellular for the public network advocate reasons listed above. Holistic network planning must make tradeoffs between those reasons to design optimal systems, but usually are made in fragmented fashion. And to add complexity, potential instrumentation with thousands of PMUs and other sensors have yet different performance requirements.
Bill Blessing - 10/20/2010 - 11:30
Mission critical application ?
I just read the referenced text from Tropos and SmartSync. In my opinion, SmartSync is more convincing, at least, for distribution area networks. I think communication on the bulk and the distribution network should be treated separatly.
The arguments from Tropos are valuable for transmission network, or the "bulk", where the automation and the "wide-area situational awareness" is critical. But what are the so-called "mission critical application" on a distribution feeders ? Distribution automation, so far, for remote network reconfiguration takes seconds and are not critical, it helps reducing outage time. Volt&Var are turned off when the feeders are in an abnormal condition (that happen a lot of time). Even with more intelligence, I think there will be still fuses blowing and protection tripping as "mission critical application" when the grid will become "dump" again.
Yes, we can acheived the so-called "microgrids" or even fast reconfiguration in the mid-term, but still, communication performance is not the most important issue. It's rather the availability and the flexibility of local load, local storage local production management that is problematic. Also, we should be sure wires and poles are still up for planned islanding after a strom, if it's down, what a public or a private network would do ?
It is obvious to say that both the utility and the telecom industry are self-serving. From a technical point of view, the telecom industry is the horse to bet on to make the grid smarter. However, I agree that major contractual commitments are required from public carriers.
David Beauvais - 10/20/2010 - 12:37
Emergency network VS M2M communications
Well, I didn't planned to spend the afternnon reading, but the Southern California filling mentionned in the Tropos paper is worth reading.
It says that "SouthernLINC Wireless Restore the Service More Quickly Than Commercial Wireless Carriers".
In this text, there is no reference to M2M communications, what I think is reuired for smart grid to operate. It stress how such a network was usefull for voice communication during emergency field operations. This is a different issue.
Finally, this case tell use more on the importance of such a network for a society, than why it should be paid by rate payers instead of tax payers.
David Beauvais - 10/20/2010 - 13:40
The Voices of Painful Experiences
The more things appear to change, the more they really stay the same. I've been an electric utility engineer for more than 40 years. In Year One, I discovered that our relay techs had rigged up a clock to measure the cumulative outage time on a dedicated MST&T circuit involved in a transfer-trip scheme which provided primary protection for a rather expensive transformer. It wasn't unusual to walk into that control house and see an elapsed outage time of several hours in one week, and that was after several years of raising %&## with the telco.
In Year 40, I heard fellow engineers from a Southern Co. subsidiary and from Duke Energy discuss inadequate responses to problems by various cell phone companies involved in their distribution SCADA communications.
Lesson - DO YOUR OWN COMMUNICATIONS!
David L Gilmer - 10/20/2010 - 20:51
Where's the 3rd option?
The fact that both the poll options above and the article ignore utility owned cellular networks is bizarre. We know Nth American utilities are lobbying for the spectrum required for this option and in Australia it is a clear choice of some utilities.
Yochai Glick - 10/21/2010 - 06:25
U Til to Build their own Vehicles & Roads??
A Vehicle and Road are used to get a person and/or equipment from one place to another - a huge need for a Utility.
A Communications Network is used to get information from one place to another - a huge need for a Utility. Should both be built and maintained by the Utility??
I used to think that the Regulated Return mentioned above was why Utilities have been in the "Build my own" mode but looking at the huge total cost of ownership (manitenance, upgrades, human capitol) that may not get a return would lesson the rational behind this business model.
I think it is truely a technology evolution issue.
We are typically talking about Wireless networks here not land line networks. Utilities have no problem using telco landline networks. The evolution of Wireless networks has happend over a more narrow time frame and 10 years ago a Utility had to build their own to get what they needed. Today that is not the case. Public Wireless networks have evolved to the point where they can provide what the utility needs. However as the U Til Exec says above there has to be proof and SLAs. The past is the past. There was a time and a place for a Utility Owned network. Those times are gone or are going away quickly.
I hope I do not have to pay for a 3G/4G Utlity owned network when there are others deployed or being deployed that could be used.
U Til Customer - 10/21/2010 - 14:39
Not so Simple
We have been studying the communication problem of the smartgrid for quite a while, the Public vs Private problem has risen and we haven't yet been able to find a solution.
The thing is that cellular networks has a really poor reliability, we have found that instead of the "instant upgrade" you get the "the stuff you bought last year is outdated and unreliable and you have to buy a new better one now to solve your problems". Our ARM communication fails at least once a week on GPRS and at least a couple of minutes a day on 3G.
My company is in Chile and we had a real proof of the 4 hour backup on cellular networks after the earthquake (Cellular network went off exactly 4 hours later, where it was still standing), moreover, most of the cellular towers are connected through microwaves and the earthquake affected the alignment of the antennas so in some cases we got electricity before cellular.
I know that for AMI infrastructure this is OK, and you can probably use public instead of private, and you will probably prefer private over public for network operations. But that's the problem, AMI and network control don't need much of a bandwidth, and if you build a private network with current technology just for control you'll get bandwidth to spare to add AMI, so why pay for 2 networks?.
If you go private I don't think the current better solution will be a problem later, you probably wont need to update the network because it will just do what's needed from it (Not like you'll be changing the meters 2 years from now), the problem will be with the equipment becoming obsolete and wont been able to find spares.
Time frames is a huge problem for smartgrids, communications and utility worlds live in totally different time frames, so should Utilities start thinking on smaller time frames when it comes to smartgrid? If so, shouldn't we be asking our meter providers for cheaper less durable meters in order to adjust them to communication technology timeframes?. Or should we be asking to our communications providers for a technology that will last for 10 or 20 years?
Sebastian Bernstein - 10/22/2010 - 13:07
DA Use of Cellular Networks
Cellular back-haul is a part of the distribution automation strategy of many of the utilities that I work with. I have seen many successful deployments of cellular devices on the distribution side. That said, there is always room for improvement. Network uptime SLAs by the public carriers would be a vote of confidence that would really catch utility executives' attentions. Qualcomm's GOBI software delineated radio should be incorporated into the cellular gateways to liberate utility companies from being held hostage to only one carrier network and give them the freedom of public network choice.
Three new demonstration projects caught our attention - a smart grid effort in Albuquerque's business district, a rapid recovery transformer study in Texas and a trial involving low voltage current sensor technologies in the UK. They also got us to thinking: At this stage in the smart grid build out, if you could design a demonstration project, what would it entail? That's our latest Tuesday Topic; click for the details.
Twitter
Facebook
Linkedin
RSS
