CN-121976925-A - Wind power blade three-dimensional track dynamic monitoring method integrating images and laser point clouds

CN121976925ACN 121976925 ACN121976925 ACN 121976925ACN-121976925-A

Abstract

The invention belongs to the technical field of blade track monitoring, and in particular relates to a wind power blade three-dimensional track dynamic monitoring method integrating images and laser point clouds, which comprises the steps of fixing a camera and a scanner through a rigid connecting device, and using the camera and the scanner as a wind power blade three-dimensional track dynamic monitoring system; the method comprises the steps of calibrating a camera, calibrating the relative position and the gesture between the camera and a scanner, acquiring image data and laser point cloud data in the normal use process of the wind driven generator, performing motion compensation correction on the laser point cloud data, calculating three-dimensional coordinates corresponding to monitoring points, acquiring a time sequence monitoring point image side coordinate sequence, and calculating a three-dimensional coordinate sequence corresponding to the time sequence monitoring point image side coordinate sequence, namely a three-dimensional motion track of the monitoring points on the wind driven generator blade in the operation period. The method for monitoring the three-dimensional track of the wind power blade by fusing the image and the laser point cloud technology has the advantages of no auxiliary mark, no complex calibration, flexible deployment, high precision and capability of efficiently supporting the safety evaluation and maintenance decision of the blade.

Inventors

LI CAILIN
WANG ZHIYONG
GUO BAOYUN
LU XIAOFENG
LI XINGXING
LI ZHUTIE
ZHANG LEIAN

Assignees

山东理工大学

Dates

Publication Date: 20260505
Application Date: 20260407

Claims (9)

1. The wind power blade three-dimensional track dynamic monitoring method integrating the images and the laser point cloud is characterized by comprising the following steps of: S1, fixing a camera and a scanner through a rigid connection device, and using the camera and the scanner as a dynamic monitoring system of a three-dimensional track of a wind power blade, wherein the scanner is a three-dimensional laser scanner; S2, calibrating the camera by adopting a plane calibration plate to obtain accurate internal azimuth elements of the camera; S3, calibrating the relative position and the posture between the camera and the scanner; s4, monitoring the three-dimensional track of the wind power blade by using the calibrated dynamic monitoring system in the normal use process of the wind power generator, and obtaining image data and laser point cloud data; S5, carrying out motion compensation correction on the laser point cloud data obtained through monitoring frame by frame; s6, determining adjacent three-dimensional points corresponding to each monitoring point in the laser point cloud data, and calculating three-dimensional coordinates corresponding to the monitoring points, wherein the monitoring points are feature points to be tracked selected from wind power blades; S7, in an initial frame of the image data, aiming at the selected monitoring point, continuously tracking the monitoring point in each subsequent frame of image to obtain a time sequence monitoring point image side coordinate sequence; s8, calculating a three-dimensional coordinate sequence corresponding to the time sequence monitoring point image space coordinate sequence, wherein the three-dimensional coordinate sequence completely describes continuous position change of the monitoring point in space, namely a three-dimensional motion track of the monitoring point on the wind power blade in the operation period.
2. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing the image and the laser point cloud according to claim 1, wherein in the step S1, an aluminum alloy section bar spliced rigid connecting device is adopted to fix an industrial camera and a scanner, a 6061 aluminum alloy square tube is taken as a framework, an adaptive integrated bracket is spliced according to the overall dimension and an installation interface of the camera and the scanner, a threaded hole site matched with a camera holder interface and a mounting Kong Jingzhun at the bottom of the scanner is reserved on the integrated bracket, and the integrated bracket is fastened and installed through a stainless steel bolt; The middle part of the integral bracket is provided with a thick aluminum alloy rib plate to realize triangular reinforcement, a hollowed-out channel is reserved at the corresponding position of a camera lens and a scanning port of a scanner to provide an acquisition visual field, and the bottom of the integral bracket is provided with an adjustable positioning clamping seat to be in seamless butt joint with an external monitoring bracket/cradle head so as to avoid relative displacement and shaking of the camera and the scanner in the data acquisition process.
3. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing the images and the laser point clouds according to claim 1, wherein in the step S3, at least four cylinders are adopted as calibration reference objects, the relative positions and the postures between the camera and the scanner are calibrated, firstly, the cylinders are arranged and data are synchronously collected, then, the two-dimensional center coordinates of the circular end face of the cylinders are extracted based on the images, the three-dimensional center coordinates of the circular end face of the cylinders are extracted based on the laser point clouds, and finally, the relative positions and the postures of the camera relative to the scanner are calculated.
4. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing images and laser point clouds according to claim 3, wherein in the step S3, the specific steps of calibrating the relative position and the posture between the camera and the scanner are as follows: S3.1, arranging cylinders and synchronously acquiring data, namely, enabling the round end faces of the cylinders to face towards a camera and a scanner, and enabling the cylinders to be uniformly distributed in an imaging picture of the camera; S3.2, extracting two-dimensional circle center coordinates of the circular end faces of the cylinders based on the images, namely performing edge detection on the acquired images, extracting contour edges of the circular end faces of the cylinders, fitting the extracted contour edges by adopting an ellipse fitting method, taking the center coordinates of the fitted ellipse as the two-dimensional circle center coordinates of the end faces under an image coordinate system, and recording as I is the serial number of the cylinder, , 、 Respectively representing X-axis and Y-axis coordinate values of the circle center of the circular end face of the ith cylinder under an image coordinate system, wherein n is the number of cylinders; S3.3, extracting three-dimensional center coordinates of the circular end face of the cylinder based on the laser point cloud, namely extracting the circular end face of the cylinder in the laser point cloud, extracting the contour of the extracted end face to obtain contour points of the circular end face, performing circle fitting on the contour points again to obtain the center coordinates of the circular end face, wherein the center coordinates are the three-dimensional center coordinates of the circular end face of the cylinder under a scanner coordinate system , 、、 Respectively representing X-axis, Y-axis and Z-axis coordinate values of the circle center of the circular end face of the ith cylinder under a scanner coordinate system; s3.4, calculating the relative position and posture of the camera relative to the scanner based on the corresponding center coordinates in the image and the laser point cloud And Combining collineation conditional equation in photogrammetry, using spatial back-convergence algorithm to solve for relative position between camera and scanner And an attitude rotation matrix , 、、 Respectively representing the X-axis, Y-axis and Z-axis coordinate values of the projection center of the camera under the coordinate system of the scanner, 、、 Respectively representing the course angle, pitch angle and roll angle of the camera relative to the scanner.
5. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing images and laser point clouds according to claim 1, wherein in the step S4, a camera and a scanner are triggered simultaneously when each frame of data is acquired in the monitoring process, so as to obtain the image frames and the laser point clouds data which are strictly aligned in time.
6. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing images and laser point clouds according to claim 4, wherein in the step S5, the process of motion compensation correction is as follows: s5.1, taking the center of a rotating main shaft of the wind power blade as a coordinate origin Taking the rotation plane of the wind power blade as A plane, the direction perpendicular to the plane is In the axial direction, any wind power blade direction is Axial direction based on Axial direction and direction of the shaft Axis direction determination The axial direction establishes a rectangular space coordinate system of the wind driven generator body Converting all scanning point coordinates into the coordinate system, wherein the scanning points are each independent three-dimensional laser sampling point contained in single-frame laser point cloud data; S5.2 at In the plane of Establishing a polar coordinate system for an origin, and 、 Conversion of coordinates from rectangular to polar The lower part of the upper part is provided with a lower part, The coordinates remain unchanged, wherein, Is the polar diameter, which represents To the scanning point at The linear distance of the projection point in the plane; Is the polar angle, representing the slave Around the positive axis Counterclockwise rotating to an included angle between the original point and the connecting line of the projection point of the scanning point; S5.3, calculating the time interval between adjacent scanning points according to the scanning frequency, the field angle and the angle resolution of the scanner The wind power blade operates at an angular velocity of Is combined with And (3) with Calculating the angle correction under polar coordinates For a pair of Performing reverse compensation to obtain corrected polar coordinates ; S5.4, restoring the corrected polar coordinates to Rectangular plane coordinates and the original The coordinates are reconstructed into three-dimensional rectangular coordinates, the new three-dimensional rectangular coordinates are converted back to a scanner coordinate system, and the three-dimensional coordinates after motion compensation are obtained , 、、 Respectively representing the X-axis, Y-axis and Z-axis coordinate values of the jth scanning point in the kth frame laser point cloud data under the scanner coordinate system.
7. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing images and laser point clouds according to claim 6, wherein in the step S6, adjacent three-dimensional points of each monitoring point are searched in single-frame laser point cloud data, and the process is as follows: s6.1, according to the camera imaging model, the motion compensated three-dimensional coordinates are obtained Projecting to the image plane of the corresponding frame to obtain Corresponding two-dimensional projection coordinates, noted as , 、 Respectively representing the X-axis coordinate value and the Y-axis coordinate value of the jth scanning point in the kth frame laser point cloud data under an image coordinate system; S6.2, aiming at a certain monitoring point on a kth frame of image, searching P points which are closest to the Euclidean distance of the monitoring point and are not collinear in a laser scanning point projection coordinate set corresponding to the frame as projection points, wherein P is more than or equal to 10, and the kth frame of image and the kth frame of laser point cloud data are time-space alignment data pairs which are synchronously triggered and collected; S6.3, based on the obtained P projection points, obtaining the corresponding three-dimensional laser point coordinates under the scanner coordinate system, and marking as , 、、 And respectively representing three-dimensional laser point coordinate components corresponding to the p-th projection point in the k-th frame image under a scanner coordinate system.
8. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing images and laser point clouds according to claim 7, wherein in the step S6, the three-dimensional coordinates corresponding to the calculation monitoring points are specifically: s6.4, fitting by using P projection points and adopting a quadric equation: (1); Wherein A, B, C, D, E, F, G, H, I, J is a coefficient to be solved, each coefficient is obtained through least square method solving, and a geometric model of the k frame monitoring point adjacent to the surface of the blade is obtained; S6.5, forming a space ray by the image point of the monitoring point in the image and the projection center of the camera, and calculating the intersection point of the space ray and the geometric model of the blade surface at the monitoring point, namely, the three-dimensional coordinate of the monitoring point.
9. The method for dynamically monitoring the three-dimensional track of the wind power blade by fusing images and laser point clouds according to claim 8, wherein in the step S6.5, the equation of the space light is as follows: (2); (3); In the formula, 、 The coordinate values of the monitoring point of the kth frame on the X axis and the Y axis of the imaging plane of the camera are respectively; 、、 the coordinate values of the monitoring points of the kth frame are respectively X-axis, Y-axis and Z-axis under the rectangular coordinate system of the space of the wind driven generator body; 、 focal lengths in the X-axis and Y-axis directions of the camera are respectively; 、 coordinate values of an X-axis and a Y-axis of a projection point of a camera optical axis on a camera imaging plane are respectively; 9 orthogonal rotation matrix elements formed by the directional cosine of the camera attitude angle; establishing a fitting equation: (4); Solving simultaneous equations consisting of the formulas (2), (3) and (4) to obtain a kth frame monitoring point Corresponding three-dimensional coordinates 。

Description

Wind power blade three-dimensional track dynamic monitoring method integrating images and laser point clouds Technical Field The invention belongs to the technical field of blade track monitoring, and particularly relates to a wind power blade three-dimensional track dynamic monitoring method integrating images and laser point clouds, which is suitable for three-dimensional track dynamic monitoring of in-service wind power blades. Background The three-dimensional running track data of the in-service wind power blade is an important basis for analyzing the dynamic stress condition of the blade, evaluating the structural health state and early warning potential fatigue damage, directly relates to long-term safe and stable running of the wind power blade, and is also a core data support for a wind power system to develop structural safety evaluation, life prediction and maintenance decision. At present, three-dimensional track measurement of in-service wind power blades in the industry mainly adopts a stereoscopic vision method, a laser scanning technology and other means, and although basic track measurement can be realized, a high-efficiency monitoring scheme adapting to the actual running scene of the wind power blades is not formed yet. The traditional measuring method has obvious technical bottlenecks and application defects in practical application, and is difficult to meet the high-precision, long-term and interference-free monitoring requirements of the wind power blade, a stereoscopic vision system is not only required to complete a complex external field calibration process, high in calibration difficulty and complicated in later maintenance, but also is greatly influenced by environmental light, weather conditions such as cloudy, sunny, rainy and snowy, and the like, the measuring precision is also influenced by the weather conditions, and the stability is insufficient in an outdoor complex wind field environment, and the laser scanning method is required to additionally install auxiliary devices such as a reflecting target ball or a mark point on the surface of the blade, so that the labor and material costs of field installation and daily maintenance are greatly increased, the original structural design of the blade is damaged, the pneumatic performance of the blade is interfered, and even the structural integrity of the blade is influenced, and the actual running requirements of the wind power blade are contradicted. Disclosure of Invention According to the defects in the prior art, the invention aims to provide the wind power blade three-dimensional track dynamic monitoring method for fusing the image and the laser point cloud, wherein the wind power blade three-dimensional track is monitored by the fused image and the laser point cloud technology, no auxiliary mark exists, complex calibration is avoided, deployment is flexible, accuracy is high, and blade safety assessment and maintenance decision can be efficiently supported. In order to achieve the above purpose, the invention provides a wind power blade three-dimensional track dynamic monitoring method for fusing images and laser point clouds, which comprises the following steps: S1, fixing a camera and a scanner through a rigid connection device, and using the camera and the scanner as a dynamic monitoring system of a three-dimensional track of a wind power blade, wherein the scanner is a three-dimensional laser scanner; S2, calibrating the camera by adopting a plane calibration plate to obtain accurate internal azimuth elements of the camera; S3, calibrating the relative position and the posture between the camera and the scanner; s4, monitoring the three-dimensional track of the wind power blade by using the calibrated dynamic monitoring system in the normal use process of the wind power generator, and obtaining image data and laser point cloud data; S5, carrying out motion compensation correction on the laser point cloud data obtained through monitoring frame by frame; s6, determining adjacent three-dimensional points corresponding to each monitoring point in the laser point cloud data, and calculating three-dimensional coordinates corresponding to the monitoring points, wherein the monitoring points are feature points to be tracked selected from wind power blades; S7, in an initial frame of the image data, aiming at the selected monitoring point, continuously tracking the monitoring point in each subsequent frame of image to obtain a time sequence monitoring point image side coordinate sequence; s8, calculating a three-dimensional coordinate sequence corresponding to the time sequence monitoring point image space coordinate sequence, wherein the three-dimensional coordinate sequence completely describes continuous position change of the monitoring point in space, namely a three-dimensional motion track of the monitoring point on the wind power blade in the operation period. In the S1, an aluminum alloy section splicing type rigid connec