LiDAR-Guided Drones Provide High-Resolution Wind Turbine Inspections
- Mar 11, 2019 7:53 pm GMT
This item is part of the Special Issue - 2019-06 - US Wind Power, click here for more
Automated drones using LiDAR technology and high-resolution cameras capture consistent, high quality data for wind turbine inspections. These inspections are safer and more efficient, while providing high-quality data, so wind asset owners can conduct preventative inspections that ultimately save costs and optimize energy production.
How Does it Work?
A customized LiDAR sensor precisely guides the drone along a 4-path inspection, eliminating drift due to variable wind conditions and controlling drone velocity to minimize motion blur. The viewing angle on the four sides of the blade is perpendicular, and 100% of the blade is inspected, without the need for pitching the blades.
Inspection data can be captured with a 20 mega pixel camera equipped on DJI M210, which provides high-resolution images for routine O&M inspections; or a 47 mega pixel camera equipped on DJI M600 for ultra-high resolution warranty or insurance claim adjustment inspections.
A specialized flight plan is built based on a pre-developed pseudo 3D model of the wind turbine to be inspected. Once loaded with the 3D flight plan, the drone is ready to perform fully automated, and precisely repeatable, turbine blade inspections.
The LiDAR system is used not only to navigate close to the blade but also to obtain spatial information about the blade. After a mission is completed, the data import process associates each image with the corresponding spatial information from the LiDAR system. By merging the spatial meta data with the images, and adding automated blade edge detection and image brightness correction, it becomes possible to determine the position of a defect (within 10 cm) on the blade and its dimensions (within 5mm).
Once data is uploaded, data processing and analysis begins.
In the case of Measure, a 3-step process designed to report accurate and actionable information is used. Step 1 is Automated Classification based on pattern recognition techniques that automatically identify images with defects, measures and locates each defect, and then preliminarily classify those defects into 5 category levels. During Step 2, Manual Review, Measure engineers review classified images and correct any false positives and/or false negatives, feeding those corrections back into the image processing algorithm to improve future data processing accuracy. In the final step, Blade Specialist Review, Measure’s team of former LM Wind and GE blade repair engineers provide a QA/QC check on the inspection findings and conduct an engineering review, adding expert commentary on any severe defects (Class 3-5).
Upon completion of data processing and analysis, Measure delivers an easy-to-use, actionable report. PDF reports provide a clear punch-list of defects, including high-resolution imagery and the location, size, type, and severity of each defect. An online web portal allows for secure storage of inspection data and makes it easier for the asset manager to access detailed inspection data, as well as to look through the history of inspection results to perform portfolio analytics.
No discussions yet. Start a discussion below.
Get Published - Build a Following
The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.
If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.