Grants and Contributions:

Title:
Anomaly Detection for Advanced Metering Infrastructure
Agreement Number:
EGP
Agreement Value:
$25,000.00
Agreement Date:
Feb 7, 2018 -
Organization:
Natural Sciences and Engineering Research Council of Canada
Location:
Ontario, CA
Reference Number:
GC-2017-Q4-00704
Agreement Type:
Grant
Report Type:
Grants and Contributions
Additional Information:

Grant or Award spanning more than one fiscal year (2017-2018 to 2018-2019).

Recipient's Legal Name:
Grolinger, Katarina (The University of Western Ontario)
Program:
Engage Grants for universities
Program Purpose:

Utilismart provides meter data management and advanced metering infrastructure software solutions for utilities, municipalities, industrial, commercial, and residential consumers. Presently, Utilismart serves over 100 utility consumers, primarily small and midsize public power utilities. Data managed by Utilismart is used for energy billing, for monetary settlements between utility and electricity system operators (such as IESO in Ontario), and other data services. Although Utilismart already follows best practices in data management and conforms to data management regulations, the company is continuously working on improving its processes and is always seeking opportunities to provide new data-related services.x000D
This project aims to evaluate and improve Utilismarts existing data processes and flows and apply machine learning to detect deviations from normal and expected behaviours. Machine learning, specifically anomaly detection, will be used to detect potentially faulty data (due to transmission, meter installation, data processing, or other errors) enabling staff to focus on checking identified data segments. Moreover, anomaly detection will identify unexpected energy consumption patterns such as high weekend energy consumption of an office building closed over weekend. Consumers then can be notified to further investigate possible causes. Utilismart will benefit through additional guaranties of data quality and improved internal processes. Moreover, anomaly detection will enable the company to provide new energy-related services and open new market opportunities in energy conservation and management.