Redefining Vegetation Management with AI
2nd Jan, 20
Utility Vegetation Management (UVM) is a field that has been crying out for technological disruption. Doing away with the dependence on manual techniques and practicing an approach of prioritized UVM is the need of the hour, and Artificial Intelligence (AI) can be the biggest change-maker in this segment.
UVM includes a wide range of operations to maintain vegetation that threatens to encroach on the right of way of distributed assets. It involves hazard tree identification and removal, tree trimming, pruning, removing bushes by using saws and mowers, using herbicides and tree growth regulators, line clearance to keep power lines safe and clean, weed control, and more. Billions of dollars are being spent on VM every year worldwide. But wildfires and power outages caused by vegetation coming in contact with power lines continue to be a major problem across the utility industry.
A series of wildfires in the state of California in 2017 and 2018 brought up the shortcomings of the conventional system of vegetation management, leading to the loss of billions of dollars in wildfire-related liabilities.
Here’s a list of the common problems of traditional UVM that is facing the industry today:
• Poor efficiency: Traditional methods lack accuracy and efficiency, which increases the risks of fire and outage along feeders with high growth vegetation. It is hard to detect and remove hazard trees using traditional methods and cyclic trimming, as it relies greatly on unoptimized VM plans and less on predictive analytics.
• High costs: Charting out a fixed annual plan for VM and practicing it is an expensive affair. The cost of labor for monitoring powerlines and trying to calculate an optimal trimming or hazard removal plan without a comprehensive data set.
• Fixed annual cycle: The traditional VM approach usually relies on a fixed annual plan, which is often rigid and doesn’t take into account the possibility of accidents and unforeseen events.
• Service level contracts: Contracts at service level results in repeat work orders along the same feeders for various activities like trimming, herbicide, asset maintenance, and others.
The Way Ahead
Incorporating more scientific techniques in VM can be instrumental in dealing with these shortcomings. Cutting-edge technology like AI and satellite monitoring can play a significant role in transforming the current scenario for good. Technology can overcome the challenges mentioned above quite effectively and can lead to a dramatic reduction in cost and resource utilization.
Advanced data used in AI can increase accuracy in estimations, which in turn can lead to greater efficiency through machine learning. It can also be very effective in reducing risks as it uses detailed, actionable information through proven business case algorithms. This, in turn, can lead to cost-effectiveness through increased business efficiency.
Technologies like LiDAR are also doing a great deal today in terms of line planning and ongoing maintenance. With LiDAR, enterprises can now scan down over treetops and lines and take about 300,000 to 500,000 measurements per second with outstanding levels of accuracy. Vegetation managers can now couple data from LiDAR with biometric data for trees (that includes information on height, density, tree condition, type of tree, growth pattern, weather condition during data gathering and so on) to generate highly accurate estimations. When it comes to preventing wildfires, LiDAR can be extremely effective as it can also provide information on measurement and estimation of fire risks based on tree size, type, location and health and can also track insect and wildlife threats around lines, in rights-of-way and around structures.
AI is everywhere! It has the capability to drastically alter core industries in the next decade. Even though the technology is in a fledgling state, the opportunities that it has to offer are endless and the way it can transform the world is astounding. Utilizing AI for Utility Vegetation Management is a no-brainer, and more and more companies will adopt AI platforms to help boost their distributed asset management process to the next level.