If you are involved in Management or Customer Service and are responsible for communicating the value of smart meters to your utility customers, you don’t want to miss this online discussion - Communicating Smart Meter Value. more...
Join social media mavens Matthew Burks and Amanda Shewmake as they provide an insider's perspective on how HR, communications and marketing professionals in energy companies can harness the power of social media to be more effective and productive. more...
The convergence of power and information technologies in the smart grid has created opportunities for finer grained and broader controls of energy flows. These opportunities can improve electric service in multiple dimensions: lower cost, greater reliability, greater customer satisfaction, and more...
Significant cost over runs. Changing business requirements. A well thought out plan is essential. Attend this free webcast discussion to hear inside hear three experts in utility operations discuss what utilities need to evaluate when they are considering upgrading or more...
The smart grid is shifting the playing field for utilities. And when the game changes, it pays to be prepared. A nimble solutions partner can help you design the solutions that keep operations on track, even as new challenges come more...
Deliver a profitable, productive and commercially successful large scale CSP business in India. Building on the success of past events in USA, Europe & MENA, CSP Today brings to New Delhi the most relevant international experience for the concentrated solar more...
Two day conference that tackles the most important challenges. A blend of European knowledge from the companies who have been installing offshore wind turbines for the last decade alongside local state governing bodies and leading project developers. Permitting, securing long more...
Autovation 2010 is a not-to-miss educational forum that will attract utility executives from around the world looking for new ways to optimize their operations through automation technologies. more...
The North American convention provides a remarkable opportunity to play a part in guiding renewable energy policy for the 21st century. Attendees will create a resolution that, along with similar resolutions already drafted on four other continents, will help set more...
Hosted by the GridWise(R) Alliance and the U.S. Department of Energy, the GridWise Global Forum will convene thought leaders from the highest levels of government, business, NGOS, and academia from around the world to discuss the ultimate enabling potential of more...
Introduction to Natural Gas Trading & Hedging - This program provides a comprehensive understanding of the structures that underlie Natural Gas trading. Beyond Essentials: Option Applications in Energy - This course provides a solid practical and conceptual (non-quantitative) understanding of more...
Electric Business Understanding provides a comprehensive overview of the electric industry. Position yourself for career advancement by gaining a solid understanding of how the electric business works including key physical, market, and regulatory aspects and how market participants navigate this more...
Electric Market Dynamics offers participants an in-depth understanding of North American electric markets and how they function. Enhance your career by furthering your knowledge of market structures, pricing mechanisms, services offered in markets, and how various participants use the markets more...
Gas and Electric Business Understanding provides a comprehensive overview of the natural gas and electric industries. Position yourself for career success by gaining a solid understanding of how each business works, including key physical, market and regulatory aspects, as well more...
We know you have something to say!
There is an immediate need for articles on
the hot topics in the Power Industry!
EnergyPulse, like no other publication,
also provides a means for our readers to
immediately interact with experts like you.
How “smart” should meters be in order to create an effective “smart metering” system?
One side of the debate equates “smart metering” with “smart meters” -- that is, meters capable of data storage and manipulation. The other side reads “smart metering” as “smart applications for meter data.” They envision centralized intelligence manipulating meter data primarily within the “walls” of the utility’s IT system. It will surprise no one that proponents of “smart meters” tend to come from meter vendors, while proponents of “smart applications” -- this author included -- tend to come from applications vendors. Like ethical advocates everywhere, we are giving you the truth as our side sees it. Utilities, as the “jury,” must determine the answer.
To do that, it’s important to understand the way this debate has changed over time -- a debate altered radically through serious discussion of a relatively new application, meter data management (MDM). It’s also important to see how the idea of MDM is changing not just our view of metering data but also our view of how applications might work together across departmental lines in the future.
Where the Debate Began
It’s been little more than three years since the start of serious and widespread discussion about “smart metering.” Initial discussions focused largely on hardware and networks. What would the new meters look like? Should they be simply Time-of-Use or something more? How might they communicate with utilities? How should we time their installation?
Meter vendors scored the first point in the debate -- possibly inadvertently. The shorthand name we’ve almost all adopted for networks of interval meters with two-way meter-to-utility communications is “smart metering.” That sounds a lot like something that focuses on meters. And why shouldn’t meters be “smart,” many argued. Distributed intelligence is clearly a trend. We have “smart cards,” “smart locks,” and scores of other everyday devices with embedded computing power. Intelligent energy and water meters seemed a natural evolution.
Meter vendors also scored a very strong second point. They looked at the way meter data was then handled at most utilities -- that is, within the billing system. Then they asked the question: how many electric utilities currently have billing applications that can handle the exponential data volume increases involved in smart metering? The answer was: probably fewer than 10 percent. Interval data is a huge burden. Manipulating merely interval data from a utility’s largest industrial customers can be such a burden on a billing system that many utilities handle industrial and residential billing in separate applications. What exactly would happen to billing applications when 100,000 residential meter reads per month suddenly turned into 7.2 million?
Applications vendors’ initial protestations about application strength—the thought-leadership equivalent of chest thumping -- failed to convince anyone, including the vendors themselves. Granted, billing systems capable of handling this data increase are readily available. The best of them can maintain throughput even with the interval data burden. We can even provide applications that can add to that new volume thousands of new online inquiries from customers tracking their daily electricity use via the Internet.
But what would be the cost? Would utilities see such billing systems as cost-effective? Of course, the same question might be asked about placing intelligence in the meter -- another costly proposition. Either way, it seemed, utilities were looking at significant financial outlays.
Enter: Meter Data Management
The answer to the dilemma arrived as soon as utilities started looking beyond billing.
As discussions of smart metering continued, more groups joined in. Network managers, strategic planners, field crew supervisors, and more saw that readily available interval data for all customers plus two-way communication between meter and utility could make possible projects previously relegated to wish lists. Field crew managers started calculating the savings available through remote meter turn-off and turn-on. Then they realized they could save even more if they could prevent a “truck roll” by first responding to a “lights out” call with a status check to the meter.
Network managers looked at how the analysis of block-by-block power flow data, available in near real time, might improve grid efficiency and supply portfolio management. How, they asked, might it help detect nested outages and speed restoration? Could interval reads better detect energy theft? Could two-way communications ease the cost of prepayment programs? How much might cash flow improve if customer service representatives could give disconnecting customers an up-to-the-minute meter read and the ability to pay with a credit card prior to leaving town?
Simultaneously, most utilities came to the conclusion that smart metering’s primary billing and conservation benefits -- demand-response programs, net metering, prepaid metering, and the like -- could take place no matter where in the system the “smartness” occurred. That was not the case, however, for some of the non-billing benefits, where central control of raw data would be a necessity.
Still, participants in the discussion could clearly see the impracticality of trying to force the billing system to service many other departments in real time. That was a classic non-starter. Software vendors’ response has been to offer a new application, meter data management, that pulls the raw meter data into a separate application. The separate MDM helps ensure that various departments receive what they need, when they need it, in appropriate forms. Their needs are not subordinated to billing, nor are they truncated by having only meter-processed data available.
Is Meter Data Management In Your Future?
The MDM concept is rapidly catching on. A new Chartwell study finds that 15 percent of utilities with fewer than half a million customers and 17 percent of those with more are already using MDM as metering data repositories. Even more startling for such a relatively new idea, 47 percent of the largest utilities are in the planning stages of MDM, and another 35 percent are considering the approach.
Movement toward MDM is considerably less marked at smaller utilities. The same study shows that 89 percent of utilities with fewer than 100,000 customers currently use their billing/customer care (CIS) systems as meter data repositories and that most (62 percent) are not even considering a change at this point.
Chartwell provides an explanation when it compares cooperatives in its study with large investor-owned utilities:
Cooperatives are smaller and therefore have fewer metered endpoints. They have smaller budgets and fewer back-office business requirements and needs.
A larger IOU with more than one million customers typically will have several departments staffed with specialized employees that, with the advent of advanced metering, now have a stake in the data. On the other hand, a cooperative with 17,000 members will have fewer departments staffed with employees who have extra-departmental duties and tasks.
Given the evident popularity of MDM among the largest utilities, one suspects that the 17 percent of them that Chartwell found not currently considering MDM will likely move to the “considering” category in the near future. But what of the thousands of other utilities where trends are not so clear? If your organization is among them, how can you best determine which path is best for you?
Your first information stop should probably be your program and regulatory strategists. Exactly how committed is your jurisdiction and your utility to programs that mandate universal interval billing? If you’re detecting a “wait and see” attitude with respect to demand-response programs, if you hear words like “pilot programs,” if debates focus on the “most cost-effective path to fill environmental commitments,” you are unlikely to need to make a decision immediately about meter data management. Until you are certain that a flood of new data is about to drown current applications, a meter data management RFP is probably premature. A good billing system and some outsourced services can likely get you through almost any pilot program. You don’t need to spend $25 million on new hardware and software, as one utility recently did, to service the few hundred residential and small business customers who ultimately committed to advanced metering programs.
The second information stop should be your billing department. Is it failing to provide the flexibility and data handling your large industrial customers increasingly demand? You might be able to cost-justify an MDM based solely on your ability to satisfy those customers, keep them in the community, and retain a larger portion of their long-term business.
Alternatively, you may uncover serious discussion of a CIS replacement. If so, you will probably want to ensure that the product ultimately selected can work effectively either with or without meter data management.
A third information source is your long-term IT plan. Today, many utility IT departments are already floundering under the burden of maintaining integrations among an ever-growing set of point solutions. IT is unlikely to welcome the news that they might soon have another – MDM -- to cope with. They may, however, be amenable to an MDM from a vendor that handles a major share of the integration burden in ways transparent to the customer.
Integration is not, of course, the only IT issue. It is not too soon to re-examine long-term plans for data warehousing, security, and spatial data in light of the varying capabilities of MDM systems now emerging on the market.
The Structure of the Future
A growing number of analysts believe that utility software will evolve in the direction of offloading data-heavy requirements into applications that can service higher-level functions. Multi-application reporting and business intelligence systems are increasingly available. Schedulers seem an obvious next step. In this context, so does MDM.
That doesn’t mean that advocates of software embedded in the meter are necessarily wrong. Distributed processing is clearly preferable for some applications. Central control is clearly mandatory for others.
That said, it is also true that a great deal rests on your decision about where to locate “system smarts.” That decision will influence the way business processes work across applications for many years into the future. The more information you can obtain, the more of your departments you can involve, the better that decision is likely to be.
For information on purchasing reprints of this article, contact Tim Tobeck ttobeck@energycentral.com. Copyright 2010 CyberTech, Inc.
Regarding your [a great deal rests on your decision about where to locate “system smarts.” ] question, it seems obvious to me that people involved in this debate are likely simply confusing terminology. There would seem to me to be no way to operate an intelligent metering system to full benefit without storing the interval data centrally in near-real-time. I doubt there is much realistic debate on that question. The question I see as unsettled is "where should the decision-making intelligence (demand / market response action decisions for each site,etc) reside in a new smart metering system?" I've seen some propose that also should be located centrally (at the utility HQ) which to me is a serious design error which would, in the long term as the yet-un-thought-of applications of these systems evolve, result in strangulation of the system by the (necessary) imposition of all the standard access / security / liability constraints on a customer's individual parameter settings.
The intelligence belongs out at the meter, but the data storage belongs at the central location.
Eric Christenson 10.30.07
Customers have the best understanding of their own usage assumptions and actual patterns. Many customers would be curious to see more than the vanilla analysis that often goes with MDM customer analysis engines.
I hope that utilities understand that raw interval data should be available to all individual customers.
Frank Hyde 10.30.07
This article illuminates some of the conflicting forces in the market on how the architecture of the new data systems should be designed. Guerry makes an interesting point that meter vendors want to see the meter evolve into a data server.
However, the need to perform validation, editing and estimation (VEE) operations on the data before it is distributed is a sufficient reason alone to have the data collected and resident at a central server, namely the MDM.
Furthermore, the ongoing reduction in data communications and storage costs mitigates the arguments about the high cost for central storage. Terabyte scale disk drives are inexpensive and are being installed in homes to be used for media servers. Full length videos involving gigabyte quantities of data are routinely distributed over the internet.
Lastly, with hourly based pricing (AKA RTP), bill calculations requires interval data. Modern customer information systems will permit customers to access their energy costs in quasi real-time from central data servers. This information must be validated through the VEE processes.
Guerry has it right, it is far more efficient for the meter data consuming applications and customer information systems to be designed around a centralized enterprise scale telecommunications and servers.
Amatsia Kashti 10.30.07
To consider the benefits or smartness of a system using only known variables such as the current beneficiaries, can be equated to Bill Gates famous statement that 64kb of memory should suffice for the home computer.
The range of uses and application of MDM and the new modes of interaction with the consumer are likely to open a new stream of communication application.
This article seem to concentrate on the (important) facet of utilities benefits only.
Anantharamiah Raja Rao 10.31.07
If you had been designing the system from scratch you would have thought that the meter is only a tool to provide the information of consumption - kwh, kvarh, etc all of course time tagged etc. On could also say that the reason the legacy meter gave the reading was not because the utility wanted the customer to know the reading but because his meter reader would go there some time and read the meter. So if one can ignore 'legacy' one would consider that the 'meter' should be only a dumb device to give data to the central office of the utility. The smartness should be in the central office. Now the question arises about what information the utility wants to give to the customer in today's context of demand response etc- and for this probably the best answer is a separate 'customer information device - CID' which is tailored to give this information - the information may vary from utility to utility and from time to time. This would not affect the basic network which takes data from the 'meter' and sends it to the utility. This of course would hurt the meter manufacturers but if the network which sends the data from the 'meter' to the utility and back to the 'CID' could be a wireless system such as 'Zigbee', the whole thing would be very economical. As far as the usage of the data collected from the 'meter' is concerned it seems that the sky is the limit to what can be done - and it seems that newer and newer usages will be found - in other words putting smartness in the utility would make the system future proof to a large extent.
Len Gould 11.1.07
I disagree with Mr. Rao's philosophy, that the central system is where all the smarts should reside. It seems to me quite ridiculously complex and costly to propose that a single central system of software should be made to be aware of every issue involved in e.g. whether to cycle for peak control, every one of a million or more customer's air conditioners. How would it acquire up-to-date on-site temperatures? How would it know if a customer has moved since yesterday? How would it know if someone happened to be home on a particular afternoon and didn't want their unit cycled? etc. etc. etc.
Now multiply that complexity by the correct number of appliances on each premisis which should really be controlled in a good load management system.
Now add the control of small local on-site CHP units (whispergen, GE's SOFC, Honda's unit, solar PV and Thermal, all the others about to be introduced).
Now do a proper job of handling the interaction with grid-wise PHEV's
It just goes on. Coupled with the fact that small distributed, relatively simple controllers are a LOT cheaper per control point than large centralized remote systems with their communications requirements, redundancy, reliability, high-skilled maintainance crews, etc. Large central systems are NOT the place to put the controls. To me that seem so obvious as to not even need saying.
Anantharamiah Raja Rao 11.1.07
I think Mr Gould has missed the point - we were talking of where the smartness of the meter should be - whether in the meter or the central office of the utility. The central office of the utility need not be 'one' per utility - in fact the sizes of utlities and the number of customers vary so much. The central office should be so chosen as to make the further processing of data from the meters practical and useful - and as I had said probably no one really knows what uses will be thought of in the future. So should the smartness be in the meter????
Alan Dickman 11.1.07
Great article. I tend to work with the over-500K meter utilities. The article provides good perspective on the smaller utilities.
The debate over centralized vs. decentralized is going to go on for awhile. It really depends on the perspective that you at the problem from. Among other observations, Advanced Metering view is fantastically different from a Intelligrid view. It also depends on where you are in the lifecycle of meter, network, and grid technologies.
My company (IBM) takes a 360 view, called Intelligent Utility Network. In that view, network latency alone suggests utilities must push some grid automation (smarts) into the substation. However, many elements of Advanced metering and grid automation are candidates for centralization. But add to this some of the flavors of analytics needed to support grid automation, outage detection, condition-based maintenance, and you conclude that so much data is centralized in a data center but distributed across multiple applications. What you really need is a very flexible architectures (enterprise, information, integration, etc.) that accommodates centralized and de-centralized functions
And no company ever seems to be a mirror image of another!
Len Gould 11.6.07
Alan Dickman points out a genuine problem, hinted at by others. "no company ever seems to be a mirror image of another! " The fact that utilities have learned nothing yet from the ridiculous experience of late last century of custom-designing from scratch every one of perhaps 200 nuclear power stations. One is beginning (finally) to hear some talk of solidly standardizing the next generation of reactors.... (but).... one also has to wonder where incentives actually lie for regulated utilities, e.g. the higher they can get costs accepted by regulator boards, the higher their profits are.
Jim Beyer 11.6.07
Why could someone develop an open standard for meters? Then anyone can build them, as long as they are compliant with the standard? I guess that would be too simple....
Jose Antonio Vanderhorst-Silverio 11.13.07
An effective smart metering system should develop under competition of business models for several market segments of the power industry. Innovations should be the jury.
Let the Innovations Locate the Smarts By José Antonio Vanderhorst-Silverio, Ph.D.
Guerry Waters, Vice President - Industry Strategy, Oracle SPL, and all comments so far, except perhaps that of Amatsia Kashti, have taken for granted the existence of the utility now and forever. Mr Waters even go as far as saying “Utilities, as the “jury,” must determine the answer.”
I understand that all stakeholders and the general public will be better served if the answer is determined by competition in the market, and not by the utilities under debates.
In the article Disintegrating the Grid and Retail Worlds, I wrote “Instead of trying to integrate the grid and the retail sides of the utilities, CIOs should take the results of an essential system analysis that supports the EWPC market architecture and design breakthrough paradigm shift of the power industry.”
Such market competition can be developed, in several market segments, among retail side of utilities by expanding their actions from state level to federal level in the U.S. and from country level to EU level in Europe.
Jim Beyer asks an important question: “Why could someone develop an open standard for meters?” One answer is found in the article The Sixth Disruptive Technology, is as follows:
The interface standards mentioned in the article [The Critical Role of Advanced etering Infrastructure in a World Demanding More Energy] should enable the separation of transportation and retail, which no longer will be regulated with price controls, as retail will be the subject of competition. As firmware downloads may differentiate 2GRs [Second Generation Retailer - 2GR] business plans, I am happy to recall the business case of a very low cost worldwide meter that I envisioned in my article a Dominican strategy, which was published in the May-June 2006 issue of the IEEE Power&Energy Magazine, just like they are doing for the US$100.00 laptop computer and the US$40.00 or so cellphone. Such low cost meter could be very promising for power service in the Bottom of the Pyramid.
Jose Antonio Vanderhorst-Silverio 11.13.07
An affordable model for retailers’ enterprise solutions business model innovations, based on Free Open Source Solutions, is reported by Bill Opalka at Knowledge 2007. The model suggests how the electric industry can become an intelligent enterprise. SAP American, IBM and Oracle Utilities shared their incumbent visions, with IBM's seeming to be closer to the EWPC.
In brief: At first glance, an upstart airline and a century-old electric utility might not seem to have much in common. Not so says the Chief Information Officer of Virgin America Airlines, who kicked off Knowledge 2007 this morning. Bill Maguire explained how IT for an airline barely 10 weeks in the sky, created by larger-than-life British entrepreneur and adventurer Richard Branson, and a utility striving to become an Intelligent Enterprise aren't all that different.
While the airlines deregulated kind of successfully, the electric industry’s barriers to innovation need to be taken down first. It is only after the introduction of the EWPC market architecture and design breakthrough paradigm that the conclusions of Mr. Branson would make sense.
And the key driver in Virgin's technology is reliance on Free Open Source Solutions. FOSS provides much of the company's software and e-commerce solutions. It's very low-risk to the business, cuts costs and helps reduce head count, all major concerns of the CIO, Maguire says. Virgin, with relationships with outside partners in customer service and other functions, has an IT staff of 17, where a similar enterprise might be expected to have 100 people. Another real-world cost benefit -- its e-commerce site cost $700,000, when a similar operation might be expected to run $3.5 million to $5 million.
FOSS is the “start up” key element for the airline business model innovation that is trying to compete with enterprise solutions of the Three Giants on the Stage (Bill’s second post).
In brief: It's not often that SAP American, IBM and Oracle Utilities share the same podium, but they were all present at K2007 to give their respective visions on Enterprise Solutions -- Perspectives from the Market Leaders during the first afternoon of the conference.
Jose Antonio Vanderhorst-Silverio 11.13.07
K2007 Retailers Enterprise Solutions . . . continued
In addition, IBM seems to be closer to a non-utility or customer oriented statement as follows:
Gerry Metzler, a partner in IBM Global Business Services, expects the electric utility market moving over the next 10 years from the "passive persistence" model currently in vogue, to a "participatory network" in which the customer is involved in the decision-making, becoming the trailblazer in determining energy consumption. One question that remains to be answered is if the changes will derive from operational concerns or customer decisions. But there will be massive capital expenditures, whcih the industry's perceived slow pace in decision-making, as opposed to say, telecom, may provide some benefit.
I claim that the shift from “passive persistence” to “participatory network” is nothing more than the paradigm shift from the slow pace vertically integrated utility to the competitive innovators that will develop The Sixth Disruptive Technology of EWPC. EWPC aims to both operational concerns and customer decisions.
Please read the two related articles on business model innovations enterprise solutions needs.
In brief: An effective smart metering system should develop under competition of business models for several market segments of the power industry. Innovations should be the jury.
In brief: Instead of trying to integrate the grid and the retail sides of the utilities, CIOs should take the results of an essential system analysis that supports the EWPC market architecture and design breakthrough paradigm shift of the power industry.
Len Gould 11.15.07
Jose: Would you please "get it"? Adding a set of theoretically "competitive" retailers to our existing de-regulation system does NOT make a free and open market. NO DE_REGULATION UNTIL THERE EXISTS AN OPEN AND COMPETITIVE FREE MARKET FOR ALL CUSTOMERS.
Jose Antonio Vanderhorst-Silverio 11.15.07
In EWPC there are 8 possible End-State (UK was developed on 4), only one of which is the generic market model paradigm: retail competition with active demand (UK had no active demand) and ultraquality transportation (UK has separate transmission and distribution and no ultraquality identified). That is the essence."
EWPC is NOT the UK Model By José Antonio Vanderhorst-Silverio, Ph.D.
[Len this what I wrote to] Adrian: thanks for your comments. I hope you get to see this very brief response that goes right to the essence. [EWPC is not a existing market.]
If you didn't have the time to read the "Synthesis Proposal Agreement of EWPC," you will see that EWPC it is NOT the UK model, but a new market architecture and design breakthrough paradigm that emerged in the last two years. This is the summary of my discovery as simple, but not simpler, as it can be said:
There are "8 possible End-State (UK was developed on 4), only one of which is the generic market model paradigm: retail competition with active demand (UK had no active demand) and ultraquality transportation (UK has separate transmission and distribution and no ultraquality identified). That is the essence."
Ultraquality transportation is the key requirement to develop A Futures Market under EWPC (hit link please), which the UK model lacks. That is why the UK model, as you say “has failed to deliver adequate investment in new generation…” It has also failed to deliver “demand side participation,” because it only considered the 4 possible End-States at the outset.
Please don’t confuse EWPC with the UK model. EWPC is an extension of Fred Schweppe et al Spot Pricing of Electricity. Under EWPC, system reliability is first and foremost. Instead of first generation (and incumbent) retailers, as the UK has, EWPC has Second Generation Retailer - 2GR (no incumbents) that participate in long run power system planning to integrate demand.
The important example of the Dominican Republic is not that of the vegetables, which we certainly have. I have tried hard, as Len can attest, to have Dominican like electric retailers. The example is that of customer differentiation, giving the opportunity to integrate demand and to practice true spot pricing of electricity. The opportunity is to integrate demand into power system planning, operation and control. It is about the opportunities available on demand side risk management.
In Free Market and Central Planning, Under R1E2, I wrote of the discovery that “To optimize the transportation system, it is required to consider total social (demand, transport, supply) welfare needs, and not just the optimization of transmission, distribution, or both, by themselves.” I used to say that electricity was not a commodity (I did in 1995 at an IEEE meeting). But once you have ultraquality transportation in the least cost controlled grid, electricity becomes the best commodity in an open market.
