石腾, 许波峰, 汪亚洲, 张金波, 赵振宙, 蔡新. 基于图像处理的风电叶片损伤识别定位系统[J]. 太阳能学报, 2024, 45(8): 565-571. DOI: 10.19912/j.0254-0096.tynxb.2023-0639
引用本文: 石腾, 许波峰, 汪亚洲, 张金波, 赵振宙, 蔡新. 基于图像处理的风电叶片损伤识别定位系统[J]. 太阳能学报, 2024, 45(8): 565-571. DOI: 10.19912/j.0254-0096.tynxb.2023-0639
Shi Teng, Xu Bofeng, Wang Yazhou, Zhang Jinbo, Zhao Zhenzhou, Cai Xin. DAMAGE IDENTIFICATION AND POSITIONING SYSTEM OF WIND TURBINE BLADES BASED ON IMAGE PROCESSING[J]. Acta Energiae Solaris Sinica, 2024, 45(8): 565-571. DOI: 10.19912/j.0254-0096.tynxb.2023-0639
Citation: Shi Teng, Xu Bofeng, Wang Yazhou, Zhang Jinbo, Zhao Zhenzhou, Cai Xin. DAMAGE IDENTIFICATION AND POSITIONING SYSTEM OF WIND TURBINE BLADES BASED ON IMAGE PROCESSING[J]. Acta Energiae Solaris Sinica, 2024, 45(8): 565-571. DOI: 10.19912/j.0254-0096.tynxb.2023-0639

基于图像处理的风电叶片损伤识别定位系统

DAMAGE IDENTIFICATION AND POSITIONING SYSTEM OF WIND TURBINE BLADES BASED ON IMAGE PROCESSING

  • 摘要: 设计一套基于图像处理的风电叶片损伤识别定位系统。首先,耦合图像滤波、分割和形态学处理等图像处理算法实现损伤区域的检测识别;然后,基于多边形拟合结果,结合质心定位算法和外接矩形的位置坐标实现叶片损伤的精确定位;最后,依据提取到的基础几何特征、形状因子和长短径之比等图像特征实现叶片损伤类型的准确判断;通过对比不同光照条件下的叶片损伤检测效果,验证了本系统具有一定的自适应能力。试验表明,本系统的平均检测准确率为90%,具备一定的可靠性和稳定性。

     

    Abstract: This paper presents a novel approach to the intelligent diagnosis of wind turbine blade damages, employing an image processing-based identification and positioning system. The system firstly utilises a combination of image processing algorithms, including image filtering, segmentation, and morphological processing, to detect and recognize damaged areas. The precise location of blade damage is determined through the integration of polygon fitting results, a centroid positioning algorithm, and the coordinates of the outer rectangle. Furthermore, the system accurately identifies the type of blade damage by extracting and analysing basic geometric features, shape factors, and the ratio of the long diameter to the short diameter. Comparative analysis of blade damage detection under varying lighting conditions confirms the system’s adaptive capabilities. Testing results shows that the average detection accuracy of this system is 90%, which has certain reliability and stability.

     

/

返回文章
返回