In this time of nearly everything getting “smarter,” the potential to modernize home energy use, storage, and generation is within reach. However, current “smart” grid AMI (advanced metering infrastructure) technology, and its higher frequency data about residential power consumption, has fallen short of industry expectations because it was designed with the system, not the people, in mind.
Data about residential power consumption has fallen short of industry expectations because it was designed with the system, not the people, in mind.
VEIC has been exploring what can happen when we go beyond just whole-home metered energy consumption. What we call advanced home energy monitoring—or adaptive high-resolution metering—can shift this paradigm by offering customer-centered, data-rich solutions that arm people with more convenient and actionable ways to reduce waste and control their energy costs. The benefits are significant; however, this new technology also comes with challenges. Utilities and policy makers will need to navigate an increasingly complex landscape of new interventions, methods of measurement and verification, and resource management issues to keep pace with the customer and market dynamics of the future.
Obstacles aside, the more we align the interests of energy providers and consumers, resource decision-making will become more economically and environmentally efficient. VEIC believes advanced home energy monitoring offers a compelling pathway for integrating these interests, ultimately improving the individual and collective larger-scale energy benefits over the long term.
“Smart” meters just got much, much smarter
While many service areas have leveled up to the so-called “Smart Grid,” the industry is still learning how best to take advantage of these technologies. Despite being able to quickly and inexpensively record energy consumption, AMI (“smart”) meters report data as single measurement (energy consumption at the meter) that is relayed back to utilities’ databases on a daily or, at best, hourly basis. If a utility is deploying some form of analytics, each billing cycle the resident may receive comparative monthly breakdowns and general percentages reflecting high- and low-use times of day. This monthly, aggregated feedback is still a far cry from the responsive and accessible device-level detail that can truly help individuals identify and act on their energy opportunities.
New adaptive high-resolution metering products, on the other hand, promise a future that decodes all household energy usage consumption and presents and manages it in a more intuitive and intelligent way. Today, this kind of product is installed directly into the electrical panel, and local computational power (a.k.a. “edge computing”) disentangles the profiles of each appliance individually, millions of times per second, and sends that back to a central server (“the cloud”) that compares signals from the edge to find patterns and identify new devices. These local and remote insights are communicated through a well-designed smart phone app that fits into people’s daily lives. It keeps tabs on things in the background—like trends, goals, and problems—and upon request can display a real-time or historic view of energy stats broken down by device and other useful metrics, such as “Always On” usage that might otherwise go unnoticed.
So, how does it work? In general, these products use machine learning principles to read the electronic fingerprints of appliances and products to identify patterns of use in much the same way that voice recognition technology can distinguish one spoken word from another. Residents can monitor the performance of their refrigerator, water heater, dehumidifier, or solar array, etc., track down sources of inefficiency, and make informed decisions about their carbon footprint. Customizable activity alerts—Garage door just opened! Iron’s still on!—add a peace-of-mind element when away from home.
In the short term, the real payoff of this high-touch, human-centered interface is increased awareness, which can lead to positive behavior change over time. Consider the progress we’re seeing in fitness trackers. With the Fitbit – originally, there was added value to providing end-users daily step counts as a metric for health. As the technology advanced, Fitbit (and now others, like the Apple Watch) were able to integrate a heart sensor and motion tracker to identify exercise, sleep, and cardiovascular health patterns. One of Apple’s recent innovations is to use these data and more to detect if I person falls and to call for help when needed. Similarly with home energy technology, smart meters are able to add value through a more detailed stream of daily data, and when new sensors and technologies are added they gain the potential to provide other value such as fault detection, savings goal-setting and feedback, and an overall sense of security thanks to the peace-of-mind alerts. Device-specific energy breakdowns, and their downstream insights, offer glimpses into how energy is being used and suggest how that usage can be improved.
Beyond the benefits of residential engagement, such vastly more precise data signals have the potential to enable a whole ecosystem of convenient and trusted services like advanced controls, which automatically act on the user’s behalf so that energy decisions can be made efficiently and with confidence.
Pipe dream or proof?
In the fall of 2017, Efficiency Vermont launched an ambitious pilot program to explore the energy savings potential of intelligent and adaptive high-resolution metering products for the residential market. Augmented by behavioral science best practices, the pilot was conducted in partnership with Sense Labs, creator of the Sense home energy monitor. Along the way our teams gained valuable insights about this emerging market and how it can be accelerated to better serve customers and the energy system. We learned that through more precise energy accounting we could enable systems that served more accurate insights and personalized feedback, creating a higher-quality, high-return relationship between people and their energy use.
Through the pilot, my colleagues and I found that always-on devices—sometimes described as “vampire loads” from things such as computer monitors, entertainment consoles, and chargers—are a clear target for energy reductions. Shifting just 10 percent of the worst performing houses to match the median performance level would result in an overall energy savings of almost 8 percent of all of the electricity used by all of the pilot participants. Further, helping to reduce the median always-on usage by 10 percent would result in an additional total savings of 2.7 percent.
Here, the savings potential for three categories of home electricity usage is shown. The chart plots the percent of total energy for all houses that would be saved if the highest “X %” of houses used the average amount for each category. This shows the relatively small, and similar usage across homes for incandescent lighting, and refrigerators. In contrast, the savings potential for Always-On is much much higher.
While we focused on three well-known categories—refrigerators, incandescent lighting, and always-on—in the future we plan to track a broader range of efficiency gains based on a broader array of devices as we continue this work in Vermont and with other partners around the country. The biggest target category is HVAC performance, which requires modeling not only electrical use, but information about location, weather, house size, and equipment type, something that gets even better when we add another advanced technology – connected thermostats.
Currently, adaptive high-resolution metering costs more than just running analyses on existing AMI data, but it costs less than circuit-level monitoring and is more flexible and resilient to change. We hope to see more advances in this area as algorithms and models improve alongside reduced technology costs. In today’s hyper-connected world, it stands to reason that a tool combining convenience with the latest developments in smartphone metering integration could drastically impact home energy efficiency.
The medium strengthens the messenger
The increasing prevalence of smart phones and smarter things ensures that energy data will become more available, more meaningful, and more affordable for more people in the years to come, marking an improvement in our overall energy health. The challenge will be for energy stakeholders to keep up with advances in information technology, not to mention meet the high expectations of the services they deliver. It will take time, and perhaps a whole new operating system of energy engagement and economics. We should consider this not an obstacle, but an opportunity to test, engineer, and design for this new class of emerging technology.
In a world where it’s increasingly easy to measure, we should be moving to re-imagine our approach to management and success. In the future, high-resolution metering may be built into the core electrical infrastructure. Robust survey panels of homes might be established for various demographics, including underserved low-income and multi-family residents. This kind of representative sampling could then be used to track and understand larger market effects that drive policies, products, and people.
Although even the most aggressive forecasts indicate we’re a decade or two away from when most homes are smart and connected, this experience with the latest behind-the-meter products and services can help us start to conceptualize the grid systems of the future: where every citizen has access to nearly perfect energy data, and where decreasing the carbon footprint of a home will be achievable, and even the norm. We can and should help shape the evolution of this technology to not only match the needs of people, but to inspire them to be smarter consumers.
Read the full paper in our Resource Library