Moving on from analog meters into AMI 1.0

Image 1. Analog meter

More than a decade ago, the US, Canada, Australia, and other countries decided that analog meters and manual readings had had enough time among us and there were a series of significant investments in deploying a new type of metering device that leveraged live communication and higher computing power: Advanced Metering Infrastructure (AMI). AMI enabled the deployment of basic technologies that established the first live data connection to the end customer. The objective of these first-of-their-kind AMI projects was to improve data access with the end goal of improving grid operations and customer billing for utilities and other energy providers. This enabled a massive incoming flow of data to the utility which, in turn, helped establish operational efficiencies by eliminating manual meter reads.

Even though this was clearly a move in the right direction to further digitize the grid, at that point,** utilities faced challenges** with the volume of data being sent and how to effectively take action based on it, and in many cases, data could not be leveraged to its full potential. As a result, many of the end-user or customer advantages promised by what it's now called AMI 1.0, did not occur or only partially materialized. Applications like real-time communication between the meter and the customer and related detailed usage reporting didn't come to reality.

Besides, some argue that this first round of deployments was too focused on energy service providers, and not on improving customer experience. This new gold mine of information that could be parsed and shared with targeted customers to improve efficient spending and inform distribution grid outage recovery are still largely untapped. In some cases, these tools and insights shared with the end customers to enable new types of offerings in the energy efficiency space, as an example, will simply not succeed as long as there's a lack of alignment in the underlining incentives of each stakeholder.

With the advancement of data analysis techniques and Moore's law still holding - where chips' processing power doubles every (approx.) 18 months - there's a clear need, or opportunity, to better manage and act on the data it's currently being mined but also enable finer and more localized grid capabilities when compared to older smart meters. The older ones, have measurements on an hourly basis and don't have the onboard capability to do fine-tuned edge computing needed to enable utilities to create, for example, grid-mapping projects, DER modeling, and fault analysis.

Current industry state

Figure 1. Current state of AMI adoption

Years have passed since the first AMI deployments. Over 30 states have adoption rates of over 50% with the national average at 65% or almost two every three meters. These high adoption values include highly populated states such as Texas, Florida, and California which are marked in green showing high coverage in both relative and absolute terms. Others are still lagging behind with relevant cases being New York and New Jersey for example marked in red, but nonetheless, these have been moving forward as well based on regulatory outcomes during last year.

Today, many utilities are evaluating and a few have already fully deployed different offerings to their customers based on AMI such as time-of-use rates, energy consumption data sharing, automated DR, and other offerings to improve the visibility of electricity costs and promote them to save it (kWh) or lower harmful demand peaks (kW) aligned with an ever-increasing view of providing customer-centric solutions. Nonetheless, some argue that utilities aren't capturing the full range of AMI capabilities as a report from the American Council for an Energy-Efficient Economy (ACEEE) shared with some capturing almost none.

While some new AMI rollouts are being approved, regulators are increasingly coming with new demands to optimize and fully tap on opportunities these provide and reach beyond the efforts of AMI 1.0 rollouts to empower customers with the data they generate. As a good example, PGE has worked to include residential and commercial customer portals to offer near-real-time data, data disaggregation for key end uses, behavioral tools like goal-setting, and connections to energy-efficiency programs. To do so, the utility created a specific new department in charge of disseminating the generated data throughout the utility for a wide set of purposes.

In some cases, local or national governments are making compulsory the easy sharing of consumption data including electricity and gas as well as in the case of some states in the US and Canada where the Green Button initiative is active and a number of utilities are part of. This effort is intended to provide a data-sharing standard for residential and business customers to have more choices on how to easily access their electricity or natural gas usage data and act upon it.

As new challenges are headed toward utilities such as higher DER penetration and the need to have visibility and actively manage them, AMI is particularly well set to provide needed insights on how to smartly manage these two-way communications. Aligned with this, a leading grid planning vendor has shared with Darcy that it's only a matter of time before EVSE deployments, for example, become relevant to distribution system planning processes. Power flow software providers are starting to take AMI data to model DER meter-level loads for some utilities and at that granularity. The impact of these will be harder and harder to ignore as times pass by and numbers increase.

AMI 2.0 and beyond

Image 2. Smart Meter

As we know, the future power grid is going to be more complex than in the past, with more DERs including PV systems, PV + storage ones, EV chargers, and smart electric appliances all enabling a two-way flow not only of electrons but of data as well. This will need solutions in the DER management space but also might need the help of AMI chips and collar devices that leverage artificial intelligence and machine learning that utilities will use to keep things running as smoothly as possible.

Giving utilities the flexibility to analyze and forecast hyperlocal issues like coincident capacity issues, which might have a low impact at a substation level but may be well enough to blow the local transformer, DER forecasting, VVO, and fault location among many others will be key in sustaining needed grid reliability. The new round of hardware and software solutions called AMI 2.0 could help in better managing different processes such as connecting new DERs and all-electric appliances to operating the grid-control systems that keep them in balance.

Besides the latest meter developments done by incumbents including Itron, Landis+Gyr, Sensus, and others, some interesting innovators are starting to play in the future of AMI too. As an example, Utilidata has teamed up with Nvidia and several utilities to develop and test in the field the next generation of grid-enhancing meters enabling next-gen computing at the very edge of the grid. The company’s patented machine learning software leverages real-time data from smart meters and other distribution grid devices to detect anomalies that are precursors to system failures and outages, optimize grid operations, and integrate exponentially more DERs onto the utility system.

The first iteration of their new technology is a device that can plug into digital smart meters that have already been deployed. AMI 1.0 doesn't have the computing power that these new chips have to make sub-minute decisions needed to optimize power flow between DERs and the grid at large. They also lack the programmable flexibility to adapt to emerging tasks, such as carrying out increasingly sophisticated analyses of the changing characteristics of a grid. As shared by some utilities, everyone could make use of more data to better manage system planning and interconnection studies, improve forecasting and do more scenario planning and for this, we need to step up AMI holistically.

As costs and deployment, volumes are always a concern, it's worth mentioning that AMI 2.0 doesn’t necessarily need to start from scratch with a “rip and replace” method. Many existing smart meters can be upgraded with new features through a software upgrade or small hardware add-ons. While utilities and municipalities have typically focused on physical, asset-based infrastructure, that focus is changing… and AMI 2.0 represents an opportunity to shift to these digital assets.

Darcy Coverage

In past DER events, we've covered related topics including Demand Response and Energy Efficiency performance-based programs event with solutions leveraging AMI data where Recurve and community choice aggregator MCE came to present. We've discussed similar AMI 2.0 applications but at the panel level during the smart panel and panel upgrade event with Span and Sunrun as well.

To continue our coverage of grid edge enhancing technologies, we are planning to have a forum later in Q3 about AMI 2.0 and would appreciate having your input! If you would like to share your needs and challenges in the space, discuss how we can support your research efforts, and even invite you to speak at the event, please feel free to leave your comments below or reach out directly to francisco@darcypartners.com.

Resources:

• Green Button
• Ontario Green Button
• GreenTech Media - US utilities not taking the most out of smart meters
• ACEEE - Leveraging advanced metering infrastructure to save energy
• Utilidata smart chip
• Utilidata advisory board
• Nvidia Jetson technology
• Zen and the art of rate design
• Woodmac AMI market projection