Liberty Pursuing Federal Grant Funds for System Hardening, Machine Learning Tools
The Empire District Electric Company d/b/a Liberty is working to secure federal grant funding for an innovative project that would add resilience to the most distant parts of its system and develop a unique “crowd-sourced” way of assessing the state of its power line equipment using Machine Learning. The company is applying for the U.S. Department of Energy’s (DOE) support through the Grid Innovation Partnerships (GRIP) grant program, enabled by the 2021 Infrastructure Investment and Jobs Act (IIJA).
“Project MVF” – which stands for “Most Vulnerable Feeders” – was conceived by the company’s planners to focus on the customers who live in the more distant and less populous areas of its service territory. By approaching its distribution grid “outside in,” the project seeks to focus its system-hardening efforts on distant and sparsely populated areas across all four states that Liberty serves.
To ensure that planned renewal and reinforcement work targets the areas most in need of upgrades, Liberty will decide the sequencing of work using a novel approach for asset data collection and analysis. The company will work to collect visual geotagged images of its system, including those sourced from its customers who capture them with regular cell phones for an appropriate fee. The visual data would then be used to train a machine learning algorithm that would help quickly develop and prioritize near- and longer-term system renewal and restoration needs – including during widespread storms.
If awarded the DOE funding, Liberty also plans to hire local high-school and college students along with recent graduates to help process the visual data and acquire some skills and familiarity with data science. At this juncture, the company has already passed the first significant DOE milestone where its concept paper was among 49% of proposals from across all 50 states to advance to the final evaluation stage. If the grant is awarded, as much as half of the project’s cost may be funded by the federal government, improving reliability and service efficiency at half of the rate impact.
Frequently Asked Questions
As conceived, the project would last five years, with the first year being dedicated to asset data collection and machine learning analysis to identify the sequencing of system hardening work that would take place over the next four years.
The company will perform the system renewal work regardless. While the company is also pursuing a variety of data science projects, the scale of the machine learning initiative proposed in Project MVF can realistically be pursued only with the help of outside funding.
No. The machine learning algorithm would help the professionals inspecting the company’s facilities – not replace the need for their skills and experience. The machine learning tools would be used to speed up the process of identifying the areas that the algorithm would predict are the most in need of repair or refurbishment form across hundreds of thousands of locations. The company’s trained staff would still make the recommendations on the appropriate course of action in each case – making their work more impactful.
As with the rest of its capital programs, Liberty would make applications to the appropriate state regulatory commissions for recovery of the remaining costs, subject to their findings of prudence through a detailed public review of the company’s costs and service delivery outcomes.
If approved, the project will create additional jobs in construction, warehousing and supporting industries including transportation, hospitality, and others. It will also create a unique opportunity for additional income for students and other residents through a variety of entry-level data science jobs.
As proposed, there will be multiple work sites throughout the company’s service territory. Watch this space for further project updates.
Project Most Vulnerable Feeders (MVF): In Brief
- Project MVF proposes to look at the distribution system “outside in” and focus power-grid upgrades and hardening work on the most distant portions of its system.
- The company will also work with data scientists and its own customers (including local students) to capture and process visual data to build new machine learning tools to help with damage or degradation assessment.
- The project proposal is being submitted to the U.S. Department of Energy’s infrastructure grant program. If selected, it will contribute up to half of the project’s cost.
- The project would create local jobs and training – as well as educational opportunities for students and recent graduates – in the areas of data science and power utilities operations.
We want to hear from our community members. Click below to send us questions, suggestions or invitations to explore partnerships.