Abstract Wind turbine blades play a key role in the wind energy collection system. Deformation measurement for turbine blades is an important and efficient way to prevent blade damages. Here, a novel and low-cost measurement method of static deformation for wind turbine blades using home-made LiDAR was proposed, and then verified by measuring the flap-wise load bending of a wind turbine blade model. The horizontal and vertical angular resolution of the proposed system are 0.09° and 0.225°, meaning a high resolution of 500 × 200 distance points in a 45° × 45° field of view of each depth image. The visualization and processing of depth images was realized using Point Cloud Library. To fast segment the 3D wind turbine blade model from background data, we also proposed a segmentation approach based on Euclidean clustering and union-find algorithm. The experimental results show that this method can segment the blade model from acquired data effectively, reducing the time-cost by nearly 50%. Furthermore, the bending degrees and curvatures of the blade in different buckling loads can be calculated using processed data. The proposed method can realize low-cost, on-site, portable and visualized deformation measurement for wind turbine blades and would be a promising way to test blades and prevent blade damages.

中文翻译：

---

一种使用自制激光雷达和处理程序的风力涡轮机叶片静态变形测量和可视化新方法

摘要 风力发电机叶片在风能收集系统中起着关键作用。涡轮叶片的变形测量​​是防止叶片损坏的重要且有效的方法。在这里，提出了一种使用自制 LiDAR 的风力涡轮机叶片静态变形的新型低成本测量方法，然后通过测量风力涡轮机叶片模型的襟翼载荷弯曲进行验证。所提出系统的水平和垂直角分辨率分别为 0.09° 和 0.225°，这意味着在每个深度图像的 45° × 45° 视场中具有 500 × 200 个距离点的高分辨率。利用点云库实现深度图像的可视化和处理。为了从背景数据中快速分割 3D 风力涡轮机叶片模型，我们还提出了一种基于欧几里德聚类和联合查找算法的分割方法。实验结果表明，该方法可以有效地从获取的数据中分割出叶片模型，将时间成本降低近50%。此外，可以使用处理后的数据计算叶片在不同屈曲载荷下的弯曲度和曲率。该方法可以实现风力涡轮机叶片的低成本、现场、便携式和可视化变形测量，将成为测试叶片和防止叶片损坏的一种有前途的方法。