GridEyeS

GridEyeS

Improved vegetation management and more resilient electric grids through satellites and artificial intelligence.

Improved vegetation management and more resilient electric grids through satellites and artificial intelligence.

Vegetation management is one of the largest operating expenses for many power distribution utilities. Current processes are highly manual and work intensive utilizing helicopters, drones and crews visually inspecting all sections of the utility line. Improper line maintenance leads to increased weather and vegetation induced outages, and a heightened risk for wildfire ignition.  

Through the GridEyeS project with HVL, eSmart, European Space Agency (ESA) and ENTSO-e, StormGeo is researching and developing a new satellite and drone technology approach to vegetation management that shows promise to significantly reduce inspection and vegetation management costs by well over $100 per mile of line per year while also reducing the risks of wildfire. GridEyeS uses AI to process low-cost government satellite images, commercial high-resolution satellite images, and drone imagery, so that each tree’s risk metadata such as distance from line, height, health, and growth rate, can be used to classify the risk levels for all of the vegetation along the grid.

The project aims to create an end to end electric grid monitoring platform to give electric utilities and transmission companies more accurate risk assessments, faster outage management and better restoration strategies.

 

GridEyeS uses information from multiple sources, making vegetation management more efficient, by identifying high risk regions so utility companies can prioritize drone and manual inspection activities.

From Manual to Automatic

Go beyond manual grid inspections based on a calendar rotation, and prioritize your maintenance based on a machine-learning algorithm that identifies high risk areas so that you can more efficiently allocate resources.

  • GridEyeS utilizes a new family of machine learning algorithms, the Multi-task Deep Recurrent Neural Network (MDRNN), to analyze data from multiple sources – satellite imagery, weather forecast, LiDAR, drones and field observations from experts and sensors – while continually improving algorithm performance.

Reduce Costs, Increase Efficiency

Through the streamlined end-to-end system, GridEyeS aims to maximize drone surveillance planning, resilience assessment and post-event restoration planning by reducing the amount of “hit or miss” inspections with the use of LiDAR and imagery from helicopters and drones.

  • Current line inspection methods are costly and time-consuming, by identifying high-risk areas, you can allocate more expensive resources when and where they are needed most.

Advanced Monitoring & Grid Maintenance

Improper line maintenance leads to outages and increased wildfire spark risks. The goal of GridEyeS is to break down the barriers of efficient grid maintenance by bringing together experts across disciplines and combining state of the art techniques such as dynamic weather predictions and load forecasting, artificial intelligence and satellite-based decision making.

  • In addition to meeting maintenance regulations, GridEyeS can improve contract bidding by providing greater insight into the work that needs to be completed, and allow remote inspections to ensure that the work was properly completed.
  • The solution can also be used to detect changes to the physical environment, like construction activity near power lines.

 

 Image vector.9