eSmart Systems R&D initiatives: AI-based risk model for vegetation near power lines
Many utilities face major challenges and costs related to vegetation near power lines. In a research and development project supported by the Research Council of Norway, eSmart Systems has partnered up with utilities to develop an operational software solution for the assessment of vegetation and safety near power line spans.
Vegetation insight to prevent tree falls
The name of the project is "AI-based risk model for vegetation near power lines". Its members are five utilities, the Norwegian Institute of Bioeconomy Research (NIBIO), and eSmart Systems. As a result of this project, the utilities are flying drones to collect data on growth conditions and vegetation.
“We are finishing up the first, out of two rounds of data collection, and have collected thousands of images of power lines and the surrounding forest”, says Irina Pirozhinskaia, Electrical Engineer at Ringerikskraft.
With the data collected, the project will try to identify which trees have fallen from one year to another. Furthermore, it will identify which properties of the trees and their place of growth are predictive of tree falls.
“We hope to help develop a system that can detect tree falls before they happen”, says August Johan Evensen, Forest Engineer at Gudbrandsdal Energi Nett.
Using AI to analyze data
eSmart Systems will use AI to analyze the data and identify the important properties of vegetation. After that, NIBIO uses the research to determine risk.
“Our main input in this project is to develop a mathematical risk model for tree falls on power lines. In addition, we are providing basic inputs to the field and drone measurement protocol for the recording of important tree and forest property variables”, said Svein Solberg, Senior Researcher at NIBIO.
By combining the probability of treefall with the consequences of it occurring at span level on the grid, the utilities can plan and prioritize the vegetation management more efficiently than with current practices while also reducing the number of outages.
Innovation to be at the forefront
“The advantage of this project is that we can sit together and look at 3D images, get an overview of the area, and see which trees can pose a danger to the power lines”, says August Johan Evensen.
Evensen is positive about participating in research projects that can improve operations and help prevent power outages. The same goes for Irina Pirozhinskaia at Ringerikskraft who believes R&D projects are of great value.
“We get the opportunity to work with R&D, increase the competence of the company, and work towards a more profitable operation,” said Pirozhinskaia.
At eSmart Systems, Research & Innovation (R&I) is a cornerstone of the company.
“Our software is breaking new ground in the utility industry and we need to be at the forefront of understanding the market and all technology trends and possibilities. By cooperating with forward-leaning customer partners and reputable research institutes, eSmart Systems strives to gain the insight needed to be at the forefront of our industry,” said Erik Åsberg, CTO at eSmart Systems.
- Project title: AI-based risk model for vegetation near power lines
- Project goal: To develop an operational software solution for assessment of vegetation safety near power line spans, that utilities can use for prioritization and planning, based on local data on growth conditions and vegetation.
- Project Manager: eSmart Systems
- Partners: NIBIO (Norwegian Institute of Bioeconomy Research), Norgesnett AS, Valdres Energi Nett AS, Dalane Energi Nett AS, Ringerikskraft Nett AS, Gudbrandsdal Energi Nett AS
- Project period: 04-2020 to 06-2022
- Budget: 11 mill NOK
- Funding granted: 4.4 mill NOK