by Jeff St. John
January 07, 2021

Whose fault is it that the first generation of U.S. smart meters hasn’t lived up to its potential? And how can the next generation avoid repeating the same mistakes?

To be clear, the first question makes a big assumption — that the first decade or so of advanced metering infrastructure (AMI) rollouts, which have led to more than 100 million smart meters deployed across North America, have failed to deliver the functionalities and values they originally promised. 

But that assumption is backed up by a decade’s experience of unfulfilled expectations, particularly from the customer perspective. Promises of ubiquitous real-time meter-to-customer communication links and home area networking have petered out into mere thousands of customers connected via outdated wireless technologies. Even using historical meter data to engage and inform customers has fallen short of industry expectations for many utilities, according to multiple studies over the past decade. 

Opportunities to put AMI to use in utility-facing applications have often required significant additional technology implementations and workarounds to achieve. That’s been the case for using smart meters to detect power outages, to inform conservation voltage reduction implementations, and to design and implement time-of-use pricing or optimize energy efficiency or demand response offerings.

“The track record on AMI is a mixed bag,” Alicia Barton, CEO of FirstLight Power and former CEO of the New York State Energy Research and Development Agency (NYSERDA), said in an interview. “I think there are a lot of places in the cleantech space where we’ve lived that experience — a lot of hype, and then the realization that it hasn’t quite lived up to the expectations.” 

It’s a bit too easy to blame this state of affairs entirely on utilities, however. To be sure, some have overpromised smart meter capabilities that happen to lie outside their primary business case of automated billing and other “cash register” functions. But regulators that approve or deny AMI business cases — and the regulatory structures that organize that activity — also have a role to play. 

Simply put, every smart meter implementation plan must balance the costs of deploying millions of meters with the benefits they can deliver, Barton said. And many of the future use cases for smart meters can’t be determined as part of this cost-benefit calculation unless they’ve been proven in previous implementations — something that’s more or less impossible to do for use cases that have yet to emerge. 

That’s a big problem, because many states with aggressive decarbonization goals, like New York, may well rely on advanced smart meter capabilities to measure the value of distributed energy resources (DERs) such as rooftop solar and behind-the-meter batteries, integrate more and electric vehicles and demand-side flexibility into their grids, and modernize the exchange of data between utilities and their energy-enabled customers. 

With “most utility ratepayer investments, you have a rate case, you outline cost and benefits, and those get judged retrospectively,” Barton said. “There are important reasons for that, [including] to make sure that money’s being invested wisely.” 

At the same time, she said, “We’re just starting to scratch the surface of the grid edge computing capabilities” needed from next-generation meters to integrate renewables and enable electrification at large scale. 

New York, for example, has “extremely aggressive decarbonization goals built into law [that] will require the state to deploy massive amounts of DERs,” as well as targets to electrify buildings and vehicles, she said. “We have only 20 years to get from here to there.” 

Smart meters “have a lifespan that will take up a good portion of that time horizon,” Barton noted. Utilities now in the midst of AMI deployments in New York, and those awaiting regulator approval for new or replacement smart meters, “can’t afford another cycle of missed opportunities.” 

Nor can the industry risk seeing AMI deployment plans rejected by regulators “because the cost-benefit [calculation] can be marginal across the traditional use cases,” she said. Over the past few years, regulators in MassachusettsVirginia, Kentucky and New Mexico have blocked multimillion-unit smart meter deployments, citing concerns of cost-effectiveness and a lack of clear metrics on how they’ll benefit customers. 

2020 report prepared by E9 Insights and Plugged in Strategies for the U.S. Department of Energy highlighted the challenges facing utility regulators in assessing the costs and benefits of AMI investments with “identifiable costs but uncertain future benefits.” These proceedings are complicated by the fact that they envision large-scale changes to the traditional role of the utility, which is framed as evolving from a supplier of electricity to an orchestrator of customer-owned DERs. 

“The promise and potential of AMI’s many associated use cases coupled with its significant cost are driving not only increasing expectations for more details and specifics but also an expectation for new types of information,” the report states. “This may include questions about AMI’s role in the company’s vision, how AMI will be used to achieve policy and/or legislative objectives, what future investments will be needed, and — most importantly — what it means for customers. Meeting these expectations so that commissions and other parties have the reassurance they need to approve an application is no small task.” 

Regulatory models to encompass smart meters’ full potential

That’s why Barton and other AMI enthusiasts are excited about new regulatory approaches to consider the potential benefits of next-generation smart meters, or “AMI 2.0” in industry parlance. 

Take the new Benefits Implementation Plan that is guiding National Grid’s AMI rollout in New York. Under order from the New York Public Service Commission, it requires the utility to specify and prioritize the quantified and unquantified benefits it will pursue with its smart meter network, including “grid edge computing capabilities” and “value-added access to useful data for customers and distributed energy resource providers,” as well as to set clear milestones for meeting those goals.  

This focus on unquantified benefits is a step beyond the benefit-cost analysis approach common to traditional utility ratemaking; it's more akin to the way that NYSERDA and other agencies support as-yet-untested technologies, Barton noted. 

The idea is to get stakeholders to agree on “benefits we collectively believe, and that the regulators can understand, are at least likely to occur — and we’re going to allow the utilities to deploy that, and measure and adjust if needed, going forward,” she said. “We’re going to have to innovate in real time; we’re going to have to iterate in real time.” 

While some of New York’s utilities like National Grid and Avangrid are still in their early stages of AMI deployments, others like Con Edison are already in the final phases. But the state as a whole is nearly a decade behind California and Texas in rolling out smart meters — in part “because some of the concerns around those earlier experiences,” she said. 

Grid voltage control technology provider Utilidata has worked with National Grid and other utilities to tap smart meter data to inform its operations, Jess Melanson, president and chief operating officer, said in an interview last month. It’s also teamed up with AMI vendors Itron and Landis+Gyr to embed its software in their latest generation of smart meters. 

“A computer and sensor at the very end of the line where all the customer electrification and DERs are happening [are] incredibly valuable,” he said. “We’ve talked to utilities that have said, 'We’d love to engage in a conversation about future benefits with our regulator and future stakeholders' — but a [benefit-cost analysis] is not the place to do it.” 

The Benefits Implementation Plan model, by creating a venue for setting a list of benefits to achieve along with milestones to achieve them, “creates a parallel path that says you have to plan for this stuff too — and a roadmap we can all refer to as the grid evolves so that we can realize these future benefits.” 

Other East Coast states are following a similar path. New Jersey’s major utilities are a bit behind New York in their AMI plans. Late last year, the state’s Board of Public Utilities launched a public AMI work session to elicit stakeholders’ views on balancing costs and functionalities of smart meters in meeting the state’s clean energy goals. 

Connecticut’s Public Utilities Regulatory Authority (PURA) has taken a step ahead of utility AMI proposals with a proceeding to seek stakeholder ideas on how smart meters can support the state’s broader clean energy goals. The idea is to spur discussions between industry participants, consumer advocates, environmental groups and other interested parties about the value of various novel and untested AMI uses cases beyond core metering functions, he said. 

As part of that PURA process, “we’ve had very specific conversations about how meter-based applications can improve energy efficiency results from VVO [volt/VAR optimization], about how grid edge visibility can improve preparedness for DERs,” and how smart meters can enable EV adoption, Melanson said. 

Barton noted that this is a vital step for utilities to make sure they’re reaching out beyond their metering departments to ensure their AMI plans will meet the needs of other parts of their operations.

“It’s not obvious that data in and of itself leads to really smart uses of that data. It does take a fair amount of innovation and collaboration around trying to derive the insights.” 

Melanson pointed out how the “first wave” of AMI-enabled conservation voltage reduction and VVO applications yielded “a whole range of cautionary tales” on this front. “The meters didn’t actually capture voltage information and send it back, or they needed bandwidth they didn’t have.”

The current generation of smart meters come with more robust and flexible computing capacity to host various applications like high-fidelity voltage monitoring, he noted. “That has huge benefits for hosting capacity, system planning [and] DER interconnection.” 

But utilities that haven’t yet laid out plans to make use of those capabilities might be constrained by a strict cost-benefit approach from investing in the technology to enable them. “You may not be able to fully execute a DER integration solution on day one — and you may not need to, depending on your service territory. But don’t buy a meter without getting grid edge visibility in order to execute on those use cases.” 

Planning for an AMI-enabled grid edge 

Metering and grid technology vendor Landis+Gyr was among the industry participants submitting to Connecticut’s PURA docket, laying out the scope of “first-wave” and “second-wave” smart meter use cases. 

“The transformation of the utility industry is coming,” John Romero, Landis+Gyr’s vice president of industry solutions, said in an interview. “We’re going to have solar and batteries in people’s homes, and that’s going to change the way the grid edge interacts.” 

The rise of electric vehicles and the potential for vehicle-to-grid applications mean that “in the next five to six years, you’re going to be able to plug your car into your home, and it will be a battery.” 

“We have to be stepping back and thinking about what technology we can apply today that’s available for the next 10, 15, 20 years [and] that will allow the utility industry to adapt to where this technology is heading.” 

Like its fellow AMI vendors, Landis+Gyr is also embedding much more computing power and flexibility into its meters, which can enable a number of applications that can create more value for utilities and customers alike. The company has already started to make use of this flexibility to manage customer-owned solar systems by translating solar inverter communications protocols to distribution grid protocols for utility Arizona Public Service’s Solar Partners Program, he said.

Last year’s move to embed startup Sense’s energy disaggregation technology into Landis+Gyr’s latest generation of meters will allow utilities to “pinpoint exactly what in a customer’s home is using the most energy,” he added. That can allow customers to get much better data about their energy use and allow utilities to fine-tune efficiency program offerings, rate designs and other personalized approaches to energy management, he said. 

Of course, consumer advocates have plenty of reason to be leery of how much of this value will translate to better service, lower costs or more choices for utility customers. Most of the first wave of AMI deployments placed few requirements on utilities to deliver on promised customer benefits. Even those that did, such as Illinois’ Energy Infrastructure Modernization Act of 2011, have come under fire for failing to hold utilities to account.  

That’s why the novel regulatory approaches being taken in states like New York, Connecticut and New Jersey need to give consumers and ratepayer advocates an equal say in the process, he said. It’s also why an AMI deployment plan “shouldn’t be pitching the technology — it should be focused on what benefits that technology can provide to customers. Can it provide more real-time energy use data? How does it give them better price signals? How does it give them more energy efficiency options?” 

“There are different perspectives on what values technology can bring. A lot of it starts with agreeing on the fundamentals,” he said. In this sense, a collaborative approach is superior to a traditional ratemaking process conducted via comment and reply filings, he noted. “That’s the value of having everyone in a room, instead of sending documents back and forth: It takes forever to get agreement.” 

Barton highlighted the challenge of the “inevitable silos these conversations take place in. That is something that is a feature of regulatory proceedings, almost of necessity. You have proceedings on distributed energy resources in one forum, you have rate cases in another forum, you have smart grid dockets that might be totally separate from the rate cases or DER dockets.” 

But, she added, “There’s a huge value proposition here, not just for core utility business functions but toward doing the things that DER providers want to do, which is to leverage this technology to make distributed generation interconnections faster, smarter, lower-cost, to make those assets better connected to the grid and able to provide services to the grid,” she said. 

Beyond that, “the grid we’ve been relying on will have to take a few huge leaps forward to support electric vehicles, electric buildings — electric everything,” she said. “We have to have a clear-eyed focus on the fact that we can’t approach this conversation the way we have in the past. […] The decisions we make now will lock us into certain paths. We can’t afford to not look ahead.”