Research On Test Method For Wing Deformation Based On 3D-DSCM

Since the airplane wing deformation directly affects the performance of the aircraft and personnel safety, it's important to survey the wing deformation law. The traditional methods of wing deformation test always have many defects such as measurement points less. On the contrary, a new measure method called three dimensional digital speckle correlation method(3D-DSCM) which is developed in recent years has advantages like high precision, real-time performance, non-contract measurement and is little affected by environment. Using this method in wing deformation measurement can overcome the shortcomings of traditional methods. So in this paper, the method of the airplane wing deformation test based on 3D-DSCM is going to be researched.At first, the calculation principle of the two-dimensional digital speckle correlation method is analyzed and the depth information calculation formula of the binocular stereo vision is deduced. The key technologies of the 3D digital speckle correlation method are proposed to research, including the camera calibration method, the essence of the matching and 3D reconstruction.Then, the wing deformation test system is designed and the testing system layout is built. In this layout, the core part of the deformation measurement is researched, consists ofestablishing the model of the wing deformation, and selecting the ZNSSD correlation function which can be anti-interference effectively. And then the deformation initial value's importance for the sub-pixel search is pointed out and the NR method and IC-GN method are compared by theoretical and simulation.Next, a new kind of estimation method to estimation the initial value of the deformation parameters is put forward. This method improved the estimation accuracy of the initial value of deformation parameters using the matching accuracy of marking point which was a high reflection coefficient circle landmark pasted on the speckle image. When matching the marking point, we improved a SIFT algorithm which will limit the extreme value point in a remarkable edge area and greatly improve the effectiveness of the landmark detection in this paper according to the research of the image features, and verified the algorithm using experiment.At last, a measurement system is built. The first experiment validated the estimation method of the initial value of deformation parameters to show that the method was able to provide precise subpixel iterative initial value and can meet the real-time requirements of industry. The second experiment implemented the deformation measurement scheme by making the wing model which well validated the effectiveness of the scheme.