Wind Turbine Drone Inspection Using Visual Feedback Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wind power generator inspection systems using GPS-based drones face challenges in accurately determining the location and inspecting the state of wind power generators, particularly when they are stopped at an angle or bent, and are prone to GPS drift due to ionospheric errors, leading to potential collisions and incorrect inspections.
Innovation Solution
An inspection system and method that utilizes a drone to capture image information and sensor detection data, which is then analyzed by an inspection server using deep learning algorithms to determine the location and state of the wind power generator, allowing for real-time control of the drone's operations through a mobile device to ensure accurate inspection and flexible response to abnormal states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If GPS-based route is used for drone inspection, then the drone can fly a determined route, but the drone may collide with the wind power generator when it is stopped at a certain angle or bent
Solution Approach 1:
The system continuously captures images of the wind power generator and uses deep learning algorithms to determine its real-time location and state. This feedback loop allows the drone to adjust its flight route dynamically, avoiding collisions when the generator is stopped at certain angles or bent, while maintaining ease of operation through automated control.
Solution Approach 2:
The inspection system transitions from a static GPS-based predetermined route to a dynamic route that adapts in real-time based on the wind power generator's actual position and state. The drone's flight path is continuously adjusted according to feedback from image analysis, enabling reliable collision avoidance while maintaining operational simplicity.
2Extent of automation
If GPS-based route is used for drone inspection, then the drone can fly automatically, but GPS drift caused by ionospheric error may cause the drone to fly to a location other than intended
Solution Approach 1:
The system introduces an intermediary deep learning-based image recognition system between the GPS navigation and the drone's actual position. This intermediary corrects GPS drift by comparing the recognized position of the wind power generator from images with the GPS coordinates, thereby maintaining high location precision while preserving autonomous flight capability.
Solution Approach 2:
The system replaces reliance on purely mechanical/GPS-based navigation with an optical recognition system using cameras and deep learning algorithms. This substitution corrects GPS drift caused by ionospheric errors by visually identifying the wind power generator's location, thereby maintaining measurement precision while keeping the drone fully autonomous.
3Measurement precision
If sensor and computer are mounted on drone to solve GPS issues, then location determination accuracy improves, but it becomes difficult to use a small drone
Solution Approach 1:
The system extracts the complex computer processing unit from the drone and relocates it to a ground-based server. Only lightweight sensors and image capture devices remain on the drone, maintaining location determination accuracy through cloud-based deep learning processing while preserving the ability to use small, lightweight drones for inspection.
Solution Approach 2:
The system introduces a ground-based server as an intermediary between the drone's sensors and the deep learning processing. This allows the drone to carry only minimal sensing equipment while achieving high location determination accuracy through remote computational power, thereby avoiding the weight penalty of on-board computers.
Data Source
AI summary
An inspection system for a wind power generator includes a drone that transmits image information obtained by capturing images of a wind power generator and the surroundings of the wind power generator, and sensor detection information for detecting the wind power generator and the surroundings of the wind power generator; an inspection server that receives, from the drone, the image information and the sensor detection information as a transmission; and a mobile device that receives the image information and the sensor detection information, and controls operation of the drone by transmitting at least one instruction to the drone. On the basis of at least one piece of information from among the image information and the sensor detection information, the inspection server may identify a location of the wind power generator and inspect a current state of the wind power generator.