Norway’s largest distribution system operator, Elvia, partners with StormGeo for an R&D project to develop a machine learning model predicting the risk of power outages.
More than two million people in the southeastern part of Norway rely on Elvia for a steady flow of electricity to their homes and businesses. To keep the lights on, Elvia operates and monitors close to 65,000 kilometers of power grid lines, a distance equivalent to one and half times the earth’s circumference.
Continuously working to improve its operations, Elvia recently partnered with StormGeo for a two-year R&D project to enhance its power outage prediction capabilities and reduce unplanned downtime due to severe weather.
Together with Elvia, StormGeo aims to develop a new model that leverages machine learning algorithms to deliver data-driven predictions on the likelihood of power outages. These predictions will consider geographical, historical, and material factors, which, in combination with weather conditions, influence outage risks.
“Today, we rely on a combination of weather forecasts, experience, and intuition to estimate the correct level of power outage preparedness – a method that has proven not sufficiently reliable, unverifiable, and prone to human biases,” says Stein Orrem, Department Manager Operations at Elvia. “By leveraging StormGeo’s unparalleled experience with data science, vegetation management, and weather intelligence, we hope to develop a state-of-the-art model to quantify the outage risks in selected areas.”
By leveraging StormGeo’s unparalleled experience with data science, vegetation management, and weather intelligence, we hope to develop a state-of-the-art model to quantify the outage risks in selected areas.Stein Orrem
Department Manager Operations at Elvia
The R&D project marks a continuation of a long-standing and successful partnership with Elvia and StormGeo. Over the last ten years, Elvia has leveraged StormGeo’s traffic light scheme to define critical criteria for severe weather preparedness. If successful, the new model will complement this scheme and help Elvia utilize resources accurately in the face of a potential power outage.
“StormGeo’s extensive experience working with electric utilities worldwide makes us uniquely positioned to assist Elvia with intelligent power outage prediction, providing insights and best practices from other companies grappling with similar challenges,” says Jostein Mælan, CCO at StormGeo. “We see this as an exciting opportunity to demonstrate how digitalization can reduce costs and help secure a reliable electricity supply.”
The R&D project will be completed in early 2024. If the project results are satisfactory, StormGeo aims to leverage them to help other utilities proactively and accurately plan for weather-induced power outages.