ABSTRACT
At present, wind turbine inspection mainly relies on manual inspection, there are low efficiency, high cost, high risk and other defects, there is an urgent need for intelligent inspection methods to improve efficiency. To this end, a system for wind turbine inspection is designed. The system consists of hybrid UAV equipment equipped with image recognition, precise positioning, intelligent obstacle avoidance, autonomous cruise and other modules. Using the U-Net surface defect detection method based on the parallel attention mechanism and the exposure correction of double illumination estimation for the defect recognition on the surface of wind turbine tower and the detection of wind turbine blade; based on the YOLOv4-tiny network to realise the obstacle recognition; the positioning system adopts the RTK carrier phase difference method to realise the target of centimetre level of positioning accuracy; the system has a better accuracy and efficiency of detection and achieves unmanned and intelligent cruising. The system has a good performance in terms of detection accuracy and efficiency, realises unmanned and intelligent inspection, and makes up for the drawbacks of the inspection of wind turbine units. Experiments show that the system can achieve the purpose of inspection of wind turbines, and its design ideas can also provide a reference for the inspection programme of other equipment.
- Zhao Kun. Welcoming new energy development with greater momentum [N]. China Electricity News, 2022, 01, 24 (02). DOI: 10.28061/n.cnki.ncdlb.2022.000107.Google ScholarCross Ref
- CHENG Qiang, LEI Lei. China's wind power development will continue to break the waves [N]. China Petrochemical News, 2023, 09, 04 (008). DOI: 10.28130/n.cnki.ncshb.2023.001440.Google ScholarCross Ref
- QIAN Cheng, YANG Zhen, SHI Dewei. Structural design and analysis of wall-climbing robot for wind turbine tower [J]. Mechanical Research and Application, 2022, 35 (06): 32-37. DOI: 10.16576/j.ISSN.1007-4414.2022.06.010.Google ScholarCross Ref
- Wu Beiping. Research on RTK Positioning Principle and Mathematical Model of GPS Network [D]. China University of Geosciences, 2003. 5.Google Scholar
- Chen, Yi-Ying. Research on the application of hybrid drones in wind farm development and operation and maintenance management [J]. Wind Energy, 2016, (08): 48-49.Google Scholar
- WEN Xinye, YANG Zhongwei, CHEN Chang. Research on the application of intelligent inspection of power transmission line by UAV [J]. China Equipment Engineering, 2021, (23): 31-32.Google Scholar
- DUCHOŇ F, BABINEC A, KAJAN M, Path planning with modified a star algorithm for a mobile robot [J]. Procedia Engineering, 2014, 96: 59-69.Google ScholarCross Ref
- KALA R, WARWICK K. Planning of multiple autonomous vehicles using RRT[C]// IEEE 10th International Conference on Cybernetic Intelligent Systems (CIS). Phuket:IEEE, 2011: 20-25.Google Scholar
- WEBB D J, BERG J. Kinodynamic RRT*: optimal motion planning for systems with linear differential constraints [J]. arXiv preprint arXiv:1205.5088, 2012.Google Scholar
Index Terms
- Design and application of unmanned aircraft inspection system for wind turbinesA UAV inspection system for wind turbines to improve inspection efficiency and safety
Recommendations
Human Supervised Multirotor UAV System Design for Inspection Applications
2018 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)Multirotor UAVs are widely used for aerial inspection applications where missions are accomplished either via manual or autonomous control. Human controlled UAVs require trained pilots which can be a barrier from using the technology for general ...
Comments