航空学报 > 2025, Vol. 46 Issue (17): 131617-131617   doi: 10.7527/S1000-6893.2025.31617

基于双目散斑的3D-PSP技术及其在压气机叶栅中的应用

高丽敏(), 张峻, 欧阳波, 王磊   

  1. 西北工业大学 动力与能源学院,西安 710072
  • 收稿日期:2024-12-05 修回日期:2024-12-24 接受日期:2025-02-10 出版日期:2025-02-19 发布日期:2025-02-18
  • 通讯作者: 高丽敏 E-mail:gaolm@nwpu.edu.cn
  • 基金资助:
    国家自然科学基金(92152301);国家科技重大专项(J2019-Ⅱ-0016-0037)

Binocular speckle-based 3D-PSP technique and its applications in compressor cascade

Limin GAO(), Jun ZHANG, Bo OUYANG, Lei WANG   

  1. School of Power and Energy,Northwestern Polytechnical University,Xi’an 710072,China
  • Received:2024-12-05 Revised:2024-12-24 Accepted:2025-02-10 Online:2025-02-19 Published:2025-02-18
  • Contact: Limin GAO E-mail:gaolm@nwpu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(92152301);National Science and Technology Major Project of China (J2019-Ⅱ-0016-0037)

摘要:

三维压力重构对于压力敏感涂料(PSP)压力测量至关重要。提出了一种结合双目数字图像相关技术的三维压力重构方法,将PSP测量结果逐像素映射到重构的三维点云上,从而实现不依赖被测模型几何参数的三维压力重构。该方法通过在PSP涂层表面制备随机的散斑图案,并使用2台相机同步采集PSP图像。在PSP数据处理中融合互相关分析和双目立体视觉原理,实现被测表面三维压力的精确重构。将该方法成功应用于压气机叶栅表面压力测量,并通过与测压孔数据进行对比分析,PSP数据的相对偏差在1.2%以内。通过将重构得到的三维点云与CAD模型进行比较,进一步验证了测量结果的准确性,平均偏差为0.17 mm。

关键词: 压力敏感涂料, 数字图像相关, 双目立体视觉, 散斑, 三维压力重构

Abstract:

Reconstruction of 3D pressure is crucial for Pressure-Sensitive Paint (PSP) measurements. We propose a method for three-dimensional pressure reconstruction combining binocular digital image correlation technology. The PSP measurement results are mapped to the reconstructed three-dimensional point cloud pixel by pixel to realize the three-dimensional pressure reconstruction without relying on the geometric parameters of the measured model. The method involves creating random speckle patterns on the PSP coating surface and using two cameras to synchronously capture PSP images. By integrating cross-correlation analysis and the principles of binocular stereo vision in PSP data processing, we achieve precise reconstruction of the 3D pressure on the measured surface. This method has been successfully applied to the 3D pressure measurement on the surface of a compressor cascade, and a comparative analysis with pressure tap data has shown that the relative deviation of the PSP data is within 1.2%. Furthermore, the accuracy of the measurement results has been verified by comparing the reconstructed 3D point cloud with a CAD model, with an average deviation of 0.17 mm.

Key words: pressure-sensitive paint, digital image correlation, binocular stereo vision, speckle, three-dimensional pressure reconstruction

中图分类号: