Does giving customers the ability to opt-out of AMI programs create new problems?
During the long, drawn-out case of PG&E’s smart metering project, the president of the California Public Utilities Commission (CPUC), Michael Peevey, made an announcement in which he directed the president of PG&E to “bring to this commission a proposal or a series of proposals that will allow customers with an aversion to wireless devices the option of being metered without the use of wireless technology.”1 Peevey added: “In other words, I am directing PG&E to prepare a proposal for our consideration that will allow some form of opt-out for customers who object to these devices—at reasonable cost, to be paid by the customers who choose to opt-out.”2
Although the recommendation to have customers opt-out might "calm some of the emotion around this issue,”3 as Peevey stated, and earn back some level of customers’ trust for the short term, implementing this recommendation would potentially harm the smart grid movement by settling on a lower version of an AMI system. This version would change accepted AMI and data collection principles, and delay deployment, and it could set a poor model for future smart grid projects in the residential sector.
Negative public perception toward PG&E has arisen since the 2009 class action lawsuit, which in itself followed the company’s bankruptcy in the midst of the California energy crisis of 2001-2001. Thus, it’s not entirely shocking that a large number of customers may want to distance themselves from the noise and the fear of the unknowns associated with PG&E’s smart grid deployment by choosing this option. Their personal decisions, however, would have societal impacts that affect many more people.
Assuming that even a small number of customers chose to opt-out, the situation could potentially erode the value of the economic and performance benefits that were projected for full-scale deployment. A mixed environment of smart and traditional meters inevitably would create a hodge-podge system that may ultimately require a new set of tools and processes for management and maintenance. Such a constantly changing environment would require, at a minimum, a customized meter reading practice—one that would need to know who has a meter that must be manually read today and who in the route would have one in the next cycle. This environment ultimately could result in disparities among meter data management, data analysis, and the billing process, which likely could defeat significant meter reading values behind the smart metering system in the first place and impact plans for using this information for more effective smart-grid operations in the future.
As the utilities migrate from their older types of metering methods toward AMI, they could gradually limit the meter reading routes in each area or city, and will eventually seek to eliminate them when the entire system in the area is mature enough for remote reading meters. This approach could be compromised if there are adopters of the opt-out option. Also, with smart meters, data is not limited to billing only; it contains outage information, tamper indication, power quality, and more, which a supporting communication system collects more than once a month. There are many internal uses for frequently transmitted information that is captured by a data acquisition system or meter data management system, such as monitoring trends and power quality, and plotting demand graphs for small and large geographical zones, which provide significant value to help optimize responses to identified issues like outages, alarms/events, power quality abnormalities, and over/under voltages.
For a utility to have a mixed environment of both old and smart meters, the management would now need to optimize their focus of a “meter-for-billing-only” philosophy and practice, and possibly forego the value-added services of frequently transmitted data that goes beyond the revenue cycle needs. If data for a large number of customers remains limited to consumption data (kilowatt-hours), data analysis, and trend analysis for any other parameter such as power quality, then it becomes less meaningful or limited to those areas where a significant number of volunteers participate in these programs. Likewise, monitoring energy consumption and demand for an entire territory or large zones in any time period smaller than a month (e.g., weekly or daily) would become almost impossible. This could push the utility back to make projections from sampling and surveying, which could result in inaccuracies and ineffectiveness in the data that major visions of smart grid initiatives such as demand response programs are dependent on.
Since the business case and economics of AMI are based on a saturation model that includes uniformity of coverage, rate designs, and time of use programs, such offerings have their foundation in uniform services. Customers’ option to opt-out from this large-scale project will make the business case analysis of the project even more difficult for the utility.
Furthermore, utilities would need to track customers that still use old meters and customers that have the new ones. This new subclass of customers would lead to stock and inventory management database issues. Also, the call centers would have to be provisioned to distinguish these subclasses and offer vastly different treatment regimes. This would become more complicated as customers move and churn, since the new owner of a house or property may choose a different metering option than the previous owner had. These issues would not be impossible to resolve, but they would be time consuming and costly for any utility regardless of its size.
In regard to communication networks, design and installation of an appropriate system to communicate with a large number of metering devices would require knowledge of their quantities and positions—information that the utility will no longer possess. Local area networks (LAN) are usually designed to transmit metering data from the point of metering to the backhaul infrastructure. Wide area networks (WAN) are designed to carry data to the meter data management systems. The best solution for a utility with a mixture of remote-reading and manual-reading meters would be to design and implement an inclusive and scalable communication infrastructure that accommodates the maximum and minimum service coverage. The utility most likely would have to opt for the maximum condition, even if today there were less need for that, to avoid risks of major upgrades in the future due to more manual-reading meters that may later join the remote-reading system. Such a large infrastructure would function far less efficient, since it would communicate with only parts of the metering system at any given time.
The CPUC, which is focused on making California a national leader in modernizing the electric grid4, may now be setting a precedence that could become a wrong model for energy utilities and their customers around the nation, by implementing smart grid systems that sacrifice many of the benefits of new technology.
Some of the main objectives raised by customers in regard to smart meters have been: fire safety, wireless signal power (or electromagnetic field radiation), and metering accuracy.
There is no difference between traditional and smart meters in terms of the fire hazard associated with electric meters. In the rare occasion that a meter is not installed correctly/tightly into the socket, a situation known as “hot socket” could develop, which can cause the temperature at the connection point to rise to a potentially hazardous level. Utility employees throughout the country usually go through special training to prevent and deal with hot socket incidents. This situation does not depend on the type of meter. With smart meters, the utility receives alarms when a major parameter changes (e.g., loss of a phase), so it is more likely to be notified and react to such a condition and prevent consequences such as fire.
Research indicates that the power range of wireless signals from smart meters is much less severe than customers’ WiFi networks in their homes or places of work or recreation (including Starbucks and McDonalds), which are likely located near cell towers or often accessed by customers that use their wireless or cell phones to make phone calls.
It is crucial for utilities to educate customers about the myths and facts of communication with smart meters. Not all smart meters communicate through wireless meshes. A considerable portion of such communication takes place through industrial wires such as fiber optics, phone lines, serial communication wires, and power line carriers. Meters that do communicate wirelessly transmit signals in power ranges much lower than the limits set by the U.S. Federal Communications Commission (FCC). The Maine Center for Disease Control & Prevention (an office of the Maine Department of Health and Human Services) conducted a health study on smart meters, reported to the Maine Public Utilities Commission. As part of this study, a comparison was made between the non-ionizing radiation of smart meters and other wireless devices used in today’s common households. The findings suggest that the smart meter’s average and maximum radiation power are not greater than those of a regular wireless router or cell phone. (Refer to the table under "Downloads" in the right-hand column to review these findings.)
Also, a recent research study by the Electric Power Research Institute (EPRI) indicates that even radiation generated by a rack of 10 smart meters has much less power than the limits set by the FCC.5
In regard to customers’ concerns about inaccurate billing, a significant number of evaluation tests on smart meters by various entities under various conditions have now proven that smart meters are more accurate than traditional ones in many ways, from the class accuracy under standard condition (maximum ±0.2% or ±0.5% for smart meters vs. ±2% error for most traditional meters), to operating with consistent accuracy under unstable conditions such as temperature, voltage, load, and frequency.
Peevey had also previously said, “We spent $1.5 million with a firm in Houston who looked into the accuracy of the smart meters because that was the first claim that people made. We found them to be 99.9995 percent accurate, much more accurate than the current meters that can be off by 2 percent. Frankly, the problem has been that old meters, when they ran slow, nobody complained.”6
The CPUC should rely on the outcomes of scientific and economic investigations into smart meters and educate customers as much as possible about the benefits and nature of a smart grid and energy management. In addition, the CPUC should support the utilities to achieve their visions, remain active in the irreversible path of the industry, and try to solve the problems as they appear, instead of setting a precedent, which would offer a solution that makes the problem even more complex.
In its response to the CPUC’s order, which was filed on March 24, 2011, PG&E has offered to allow customers the option to have their meters’ radios turned-off. To do this, the customer would incur “a reasonable up-front charge and an ongoing monthly charge—either in the form of a fixed monthly fee or a volumetric per-kWh rate adder applied to the customer’s energy usage.”7 In this proposal, PG&E presented a table that includes various fixed and monthly options for customers, with figures that are subject to change based on the actual number of customers that choose to opt-out. Early reports indicate that the customers are not pleased with this proposal.8
References:
1. “Peevey orders PG&E to let customers opt-out of AMI,” SmartGridToday.com, March 11, 2011.
2. Ibid.
3. Anna McCarthy, “SmartMeter opt-outs proposed,” Mill Valley Herald, March 16, 2011.
4. California Public Utilities Commission, “Fact Sheet: California Leads the Nation in Modernizing its Electric Grid,” July 2010, 1.
5. Electric Power Research Institute, “Radio-Frequency Exposure Levels from Smart Meters: A Case Study of One Model,” Southern California Gas Company website, February 2011.
6. “CPUC President Bullish on Smart Meters,” SmartMeters.com, March 14, 2011.
7. Application of Pacific Gas and Electric Company (U 39 M) for Approval of Modifications to its SmartMeterTM Program A. 11-03-014, before the Public Utilities Commission of the State of California, March 24, 2011.
8. Jesse Burst, “PG&E files game-changing smart meter opt out plan. But will it work?” SmartGridNews.com, March 25, 2011.
Share your thoughts and insights on this article. Join the discussion at KEMA’s Utility of the Future blog.