The Government of Ontario has recently mandated that electric and natural gas utilities provide consumers with easy access to energy usage data digitally. This initiative, known as Green Button, aims to foster a competitive market where innovative electrification and decarbonization solutions can more seamlessly reach interested consumers. Historically, the lack of digital access to billing and energy data from utilities has hindered the effective analysis of this complex dataset for homes and businesses.
The University of New Brunswick (UNB) is the first Canadian university to work with large volumes of utility billing and metering data from multiple utilities. This research seeks to determine whether this newly available data differs from historical data in terms of accuracy, completeness, and consistency with other available sources. A critical aspect of the investigation is to ensure that the quality of the data does not compromise Ontario’s efforts to decarbonize its economy, as unreliable data can obstruct clean energy solutions such as energy conservation and demand management planning, demand response, and virtual power plants.
Michael Murray, President of Mission:data Coalition—a non-profit advocate for energy consumers based in Washington state— and UNB researcher Dr. Francis Palma—an assistant professor in the faculty of computer science and the lead researcher on the project—emphasize the importance of consistency between monthly utility bills and Green Button energy data. Without this consistency, consumers cannot be confident in the accuracy of claims about the cost savings of moving off natural gas heating, for example. Data errors even pose a challenge for the utilities themselves. Robust analytical methods are necessary to assess these questions; otherwise, the burgeoning billion-dollar competitive market for energy-saving services will be at risk.
Utilities face challenges due to legacy back-office and customer engagement software. Multiple systems operating in parallel complicate the validation of energy data, especially when various vendors operate proprietary and siloed software. The lack of auditing and validation of the data infrastructure makes it difficult to establish an authoritative source of truth, particularly when meter readings from buildings are often archived after bills are created, making inconsistencies harder to trace.
To help address these challenges, Dr. Palma will employ machine learning techniques, which require significant data resources and time, to train models using (partially) labeled data. The project aims to minimize manual evaluations by adopting a mixed-method machine learning approach, specifically semi-supervised learning.
The findings from this research will have implications not only for other Canadian provinces but also globally. Many public agencies and businesses are required to report their energy usage and carbon emissions. The reliability of energy data is crucial. This data guides real-life economic decisions, such as multi-million-dollar building retrofits. Energy data underpins clean energy solutions that often provide small, incremental reductions in energy usage, which need to be better quantified for fair compensation and reporting.
Data inaccuracies can lead to financial discrepancies, affecting decisions and even power grid failures such as blackouts. As the power grid increasingly relies on distributed energy resources (DERs), quantified results are essential for the energy utility and market operator (such as Ontario’s IESO) to provide compensation.
Dr. Palma emphasizes the need for unbiased, modern analysis to establish the trustworthiness of these data sources. The research will quantitatively and qualitatively examine Ontario’s energy usage, billing, and account data to objectively benchmark performance and detect anomalies. In Ontario (and most areas of the world), energy data comes from various sources, including bills, legacy systems, outdated Electronic Business Transaction (EBT) infrastructures (e.g., the Ontario Energy Board’s EBT infrastructure), Green Button, email delivery, and multiple utility customer web and mobile App portals.
The Ontario Energy Board underscores that data output from utilities must ensure that the data delivered is the ‘best available.’ Given substantial investments in data sharing through Green Button implementations, it is vital to ascertain whether these data services meet high-quality standards and support the transition to zero-carbon energy.
In collaboration with Screaming Power and Mission:data Coalition, the research team has garnered support from subject matter experts who highlight the need for confidence in energy data. These partners have volunteered their data and insights for research purposes, aiming to educate and provide reliable outcomes.
Gary Michor, CEO of Screaming Power, highlights the long-standing impediment caused by the lack of digitized energy data in the energy industry. The research aims to explore how innovators can assist utilities, consumers, and grid operators through collaboration. It fosters new activities that support integration into other sectors, including the financial industry, for the adoption of conservation and greenhouse gas (GHG) reduction strategies.
From UNB’s perspective Dr. David MaGee, UNB Vice-President Research says, “This project exemplifies UNB’s commitment to tackling the large-scale challenges faced by our communities and our world. By having a better understanding and awareness of energy usage, Dr. Palma and his team will contribute broadly to our ability to address the critical impact of the energy sector on climate change.”
The Screaming Power and Mission:data Coalition team encourages Ontario utility ratepayers to share their building energy information via Green Button for university analysis. Screaming Power is developing a toolset to facilitate data sharing for research and engage consumers in societal improvements. The sharing of energy data with new innovators and consultants (third parties) is also encouraged, as working together as an open, interactive marketplace will significantly reduce costs and energy usage.
The findings from this four-month study will resonate in the U.S., where Green Button mandates touch over 36 million households. “This project will benefit consumers, utilities, policymakers, and Energy System Operators by providing an objective and unbiased benchmark for Green Button data accuracy and consistency,” says Michael Murray, President of Mission:data Coalition.
For interested parties in Ontario interested in donating their energy data for research, please reach out to Screaming Power through email at wanttoknow@screamingpower.com
About
Screaming Power
Screaming Power’s goal is to revolutionize strategic energy data management by providing a mobile big data cloud platform that connects directly with the energy user, the utility, the market (e.g., authorized third parties), and the Internet of Things (IoT) to provide effective education on energy conservation, cost savings and continuous energy improvement. Within the energy and sustainability marketplace, the Screaming Power team is subject matter experts and researchers focused on simplifying energy data management and driving change to transition to a zero-carbon economy.
Mission:data Coalition
A non-profit organization, Mission:data advocates for customer-friendly “energy data portability” policies to deliver benefits for consumers and enable an innovative, vibrant market for energy management services. Mission:data’s advocacy has led to Green Button availability for over 40 million consumers across North America.
University of New Brunswick
The University of New Brunswick (UNB) is Canada’s oldest English-language university, and New Brunswick’s only national comprehensive university. Founded in 1785, the multi-campus institution has a rich history as well as a dynamic focus on research, innovation, experiential learning and entrepreneurship. UNB welcomes more than 10,000 students from 100 countries to its campuses. With over 100 programs in 14 faculties, UNB helps students acquire the abilities needed to make an impact in their communities