Vegetation management by electricity networks is used to minimise contact between trees/other vegetation and our overhead power lines to minimise disruption to our customers and improve safety. The current solution used by many networks to scan overhead lines and nearby vegetation is remote sensing Light Detection and Ranging (LiDAR) and there are concerns that the cost of future scans could double. Networks are also seeking to reduce their carbon footprint and LiDAR scans are typically performed by aircraft that produce greenhouse gases and can cause noise pollution. Satelline aims to test satellite technology as an alternative to LiDAR and demonstrate whether the technology can accurately measure vegetation proximity to UK Power Networks’ overhead lines whilst exploring the costs and benefits of the technology.
Benefits
Base Cost
· Anticipated future LiDAR inspections every two years ~ £4m
· Annual tree cutting costs for ED2 ~ £20m
· CO2 emissions due to LiDAR activities every two years ~ 223 tonnes
Method Cost
· Implementation cost post project trial ~ £6m
· Annual running costs (Annual scans and interim partial scans for quality audit, full processing of data across HV and EHV network, support, hosting, maintenance and report customisation) ~ £1.5m
Benefits include:
· 15% optimisation savings to the tree cutting costs due to business transformation (efficiencies within vegetation management processes) and tree cutting prioritisation algorithm ~ £1.5m savings per year
· 4% annual reduction in the CIs and CMLs related to vegetation management ~ £111k and £208k respectively
· Societal benefit due to avoidance of 223 tonnes CO2 emissions every two years (satellites move with the earth’s natural gravity after being put in orbit and do not emit CO2) ~ £16k every two years
The expected project benefits if satellite technology is adopted during ED2 are £4.3m
A detailed cost benefit analysis will be performed during the project.
Learnings
Outcomes
The project met all of its key outcomes and deliverables. The final output provided clearances between the vegetation and conductors, criticality scores and a prioritised trim plan for all spans and feeders in the pilot area. This data was uploaded into a working satellite and AI-based SaaS platform for the pilot area, which was successfully demoed to operational teams. The solution included complete feeder- and span-level analytics like clearances and criticality score while additional data (e.g., last trim year) provides additional visibility as well.
Summary of the value drivers of a satellite and AI solution compared to a LiDAR-based approach
More frequent data updates (at least annual)
More cost-effective solution (compared to LiDAR) enables more regular data acquisition to create trim plans informed by regularly updated data
The high frequency also enables fast-growing species, which grow at faster rates than current assessment periods, to be captured in annual trim plans
Faster data collection & analytics
Automated workflows for data collection and subsequent AI-enabled analytics can draw up trim plans in c.2 weeks (compared to c.6 months for a LiDAR-based approach)
These trim plans are available significantly earlier than LiDAR analytics and tackle the highest priority areas faster
AI growth modelling
AI growth models analyse historical imagery to calculate vegetation growth rates
This is used to optimise trim dates for fast and slow-growing species to reduce outages while maintaining cost profile
On-demand data
Satellite imagery can be acquired on-demand to run spot checks in specific areas
For example, after storms, vegetation that has moved closer to the line or no longer needs trimming can be identified
Specific trim plans are then updated with the latest data
Imagery can also be used for automated post-trim audits
More frequent risk tree detection
Automated, annual detection of risk trees ensures more regular, proactive decision-making to tackle the highest priority trees
The high accuracy (98% radial clearance accuracy and 93% accuracy for encroachment classification, which will likely improve further with continued AI training) ensures that the Satelline solution can deliver significant value compared to the LiDAR approach when rolled out across the whole UK Power Networks network:
Operation and financial value
Optimised unplanned trimming costs by shifting unplanned trims to planned trims
Optimised trimming operations by deploying contractors surgically with the right equipment on the right spans that have highest risk/impact with 100% digital workflows
Reduced unplanned outages by 25%+ from more proactive trimming and in turn, help reduce Customer Interruptions (CIs) and Customer Minutes Lost (CMLs)
Advanced and automated risk tree detection to create a backlog of tasks to be done
Strategic value
E2E digitised workflows, for instant pre- and post- work inspections
Improved data-driven decision-making capabilities for employees
Strengthened customer value proposition with improved reliability for similar or reduced cost
Enhanced contractor partnership with better plan visibility and increased program control
Lessons Learnt
Lessons learned include:
The criticality of getting access to the right data (incl. operational data, cost data, customer data) early on, in order to precisely size the benefits of the new solution and compare qualitatively and quantitatively to the baseline.
Stakeholders were engaged regularly throughout the project for both progress checks and to gain input on the emerging work, particularly on the solution’s capabilities and the key deliverables (e.g., project report). For example, live demos of the solution were delivered; stakeholder feedback then influenced next steps and the final output of the solution. This pattern of stakeholder engagement for data collection and feedback worked well and should follow a similar pattern for a full-scale implementation.
The success enabled by full integration with the operational teams was highlighted at a small scale during this project and should be replicated for future operations-focused projects. For example, the field validation brought additional vegetation management opportunities (e.g., requirement to measure clearance differently for pole-mounted transformers and polymeric cables) to light by bringing the operations and working team together, as the assets were discussed live in-person.
