Defect Recognition and Localization of Wind Turbine Blades
At present,the conventional methods for identifying defects in wind turbine blades in wind farms,such as traditional manual observation,sensor detection,and infrared thermal imaging technology,generally have the shortcomings of large workload,low recognition efficiency and accuracy,and high cost.Aiming at the shortcomings of conventional recognition methods and the difficulties of blade defect recognition,the authors propose an intelligent recognition method based on machine vision combined with deep learning.The accuracy of blade defect detection and recognition positioning can reach more than 90%,which is of great significance for improving the identification and positioning of defects such as cracks,lightning strikes and sand holes in wind turbine blades.
万方数据
维普
