CN-121702335-B - Flatness measurement adjusting method based on suspension parabolic curve

CN121702335BCN 121702335 BCN121702335 BCN 121702335BCN-121702335-B

Abstract

The invention provides a flatness measurement adjusting method based on a suspension parabola, and relates to the technical field of data processing. The pixel adjustment module acquires the actual projection track of the steel wire through image acquisition, perspective correction and distortion correction, homography mapping and central line extraction. The parabolic adjusting module calculates a difference value between a vertical coordinate of the theoretical parabolic calculation and an actual track, calculates the difference value through a cooperative control unit, and outputs a cooperative control instruction to electromagnetic force actuators distributed along the steel wire, so that the shape of the steel wire is actively adjusted through distributed vertical electromagnetic force. The two modules work cooperatively according to a closed loop with a fixed period until the deviation between the actual track of the steel wire and the theoretical parabola is smaller than a set threshold value, so as to form a stable and high-precision dynamic measurement reference. The active establishment and the dynamic maintenance of the suspension cable reference are realized, the precision and the reliability of flatness measurement are improved, and an efficient and intelligent solution is provided for detecting the flatness of the large-size component.

Inventors

ZHENG CHONG
HUANG WENBO
Li Xiameng
WANG GANG

Assignees

成都成亚航空科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260213

Claims (7)

1. A flatness measurement adjusting method based on suspension parabola is realized by adopting a pixel adjusting module and a parabola adjusting module, and is characterized in that the pixel adjusting module at least collects images comprising steel wires, perspective correction and distortion correction are carried out on the collected images, steel wire images in the images are mapped onto a virtual correction plane parallel to a theoretical plane of the steel wires through homography transformation, and a linear mapping relation between pixel coordinates and actual physical coordinates of the images is established in a coordinate system of the correction plane, so that the corresponding actual physical dimension of each pixel on a measurement plane is ensured to be constant; The parabolic adjusting module comprises a cooperative control unit and a plurality of electromagnetic force actuators which are uniformly distributed along the axial direction of the measured steel wire, extracts the actual projection track of the steel wire, generates a theoretical parabolic curve corresponding to the actual projection track of the steel wire according to the length and the track of the steel wire, and calculates the difference value of the vertical coordinate of the actual projection track of the steel wire and the theoretical parabolic curve at the position corresponding to the horizontal coordinate; The pixel adjusting module and the parabola adjusting module work cooperatively according to a fixed period and sequentially execute the closed loop flow of image acquisition, perspective correction, track extraction, difference value calculation and cooperative control output until the parabola adjusting module judges that the vertical coordinate deviation of the actual projection track of the steel wire and the theoretical parabola datum line at each detection point is smaller than a set threshold value, and the system executes subsequent flatness detection and calculation.
2. The adjustment method for flatness measurement based on suspension parabola according to claim 1, wherein the pixel adjustment module adopts a calibration method based on a three-dimensional target, the three-dimensional target comprises a plurality of mark points with known three-dimensional coordinates, the internal and external parameters of a camera and the distortion coefficient of a lens are obtained simultaneously through single shooting, and a homography transformation matrix from the pixel coordinates of an image to the physical coordinates of a theoretical plane of a steel wire is directly calculated and generated.
3. The method of claim 1, wherein the pixel adjustment module further comprises an embedded real-time dynamic calibration network consisting of a plurality of fixedly mounted intelligent reference cells, each cell comprising a uniquely encoded micro-scale pattern and attitude sensor for real-time monitoring and compensating for reference plane distortions due to environmental vibrations, temperature changes or mounting surface deformations.
4. The method for adjusting flatness measurement based on suspended parabola according to claim 1, wherein perspective correction and distortion correction adopt a layered progressive structure: the first layer is dynamic perspective correction based on feature tracking and pose estimation, and the camera pose is estimated in real time and a homography transformation matrix is dynamically updated through tracking stable auxiliary visual features in a visual field; And the second layer is on-line estimation and compensation of distortion residual based on the geometric characteristics of the steel wire image, a residual distortion field model is built by fitting the deviation of the central line of the steel wire, and fine adjustment and compensation of pixel level are carried out.
5. The method for adjusting flatness measurement based on suspended parabola according to claim 1, wherein the method for establishing a linear mapping relation between image pixel coordinates and actual physical coordinates and adopting hierarchical weighted fusion and dynamic self-calibration comprises the following steps: dividing the corrected image into a plurality of subareas based on hierarchical mapping coefficient calculation weighted by the regional confidence coefficient, fitting a local affine transformation model in each area, and determining the confidence coefficient weight according to fitting residual errors; The mapping coefficient based on closed loop feedback is dynamically self-calibrated, and the scale error is monitored and the mapping coefficient is dynamically adjusted by identifying a micro-scale reference scale fixedly arranged in an image so as to compensate the scale change caused by thermal drift of a camera.
6. The adjustment method for flatness measurement based on suspension parabola according to claim 1, characterized in that the precise center line of the steel wire in the corrected image is extracted, and a fusion extraction method based on multi-scale linear enhancement and gray profile iterative modeling is adopted: The first stage performs robust initial positioning based on multi-scale Hessian matrix linear response to generate a pixel-level center point set; And carrying out sub-pixel refinement on the second level based on gray level profile physical model fitting, fitting a gray level profile through a Gaussian model, and calculating sub-pixel level offset of the center point.
7. The adjustment method for flatness measurement based on suspended cable parabola according to claim 1 is characterized in that the parabola adjustment module further comprises a comparison and difference sequence generation link of an actual track and a theoretical parabola reference, and a space-time registration-elastic matching-vectorization difference flow is adopted, and comprises the following steps of D1, unifying a time stamp and a coordinate system, transforming an actual track point to be consistent with a control coordinate system, D2, adopting an elastic matching algorithm based on a dynamic time regularity concept, matching theoretical points for each actual point, calculating a sub-pixel level longitudinal coordinate difference value, and D3, integrating differences according to the position of an actuator, forming an error vector sequence, constructing an error vector field and carrying out space analysis.

Description

Flatness measurement adjusting method based on suspension parabolic curve Technical Field The invention relates to the technical field of data processing, in particular to a flatness measurement adjusting method based on a suspension parabola. Background Flatness is one of the key geometric tolerances for evaluating the surface quality of mechanical components, and in the fields of aerospace, ship manufacturing, large-scale machine tools, wind power equipment and the like, the flatness accuracy of large-size components (such as wing panels, ship body sections and machine tool tables) directly influences the assembly quality, structural stability and service performance of the large-size components. Therefore, the development of the large-size flatness measurement technology with high precision, high efficiency and strong adaptability has important engineering significance. Currently, flatness measurement methods can be mainly divided into two main categories, contact type and non-contact type. The contact method (such as dial indicator and three-coordinate measuring machine) obtains data by physical contact of the measuring head and the measured surface, although the precision is higher, the inherent limitations of easy deformation of the thin-wall part, low measuring efficiency, harsh requirements on site environment and the like of the measuring force exist, and the requirements of quick and in-place detection of large-size components are difficult to meet. Non-contact methods (such as a laser tracker, a laser interferometer and digital photogrammetry) avoid the influence of contact force, but the problems of expensive equipment, complex system, weak environment anti-interference capability, requirement of multi-station moving and splicing and the like are faced in large-scale measurement, and the popularity and the field applicability are limited. Under the background, the suspension cable method is regarded as a classical large-size straightness and flatness indirect measurement method and is paid attention again. The basic principle is that a stable, continuous and predictable catenary or parabola is formed by utilizing the natural sagging of the thin steel wires with the two tensioned ends under the action of dead weight, and the catenary or parabola is used as a natural datum line for space measurement. And calculating the flatness error by measuring the deviation of each point of the measured surface relative to the datum line. The method has the outstanding advantages of simple equipment, low cost, infinite theoretical measurement range, no need of complex environmental control and the like. However, the conventional suspension cable method faces a fundamental technical problem in practical application, namely, how to accurately obtain and maintain the spatial position of a theoretical parabola under the practical working condition, namely, the problem of 'accurate establishment and dynamic maintenance of a parabola reference'. The concrete steps are as follows: the original shape uncertainty is that the theoretical shape (parabola) of the steel wire drooping under the dead weight is uniquely determined by parameters such as the material, the diameter, the span, the tension at two ends and the like. However, in practical erection, the initial static form of the steel wire often has obvious deviation from a theoretical parabola due to the influence of various factors such as height errors of supporting points at two ends, installation eccentricity, uneven weight distribution of the steel wire, inaccurate initial tension control and the like. If not corrected, the deviation is directly transmitted as a systematic error to the final flatness measurement. Environmental disturbance sensitivity-perturbation of air flow, environmental temperature change, mechanical vibration, etc., which are difficult to completely avoid at the measuring site, can lead to continuous low-frequency swing or slow morphological drift of the suspension cable. In the traditional method, a static 'disposable' datum line cannot be kept stable in a dynamic environment, so that the measuring datum itself becomes an error source, the measuring precision is severely restricted, and the method is particularly unfavorable for long-time or online measurement. Reference calibration and tracing are difficult, namely, in order to convert the relative displacement measured by an image or a sensor into absolute space coordinates, high-precision calibration must be performed on a parabolic reference. The traditional method relies on placing a precise calibration object or a standard ruler with known length in a measurement visual field, but the method is often complicated in operation (accurate coplanar placement is needed), high in invasiveness (measurement space is occupied), and the calibration result is only in an initial state, so that errors caused by deformation of the standard ruler or drift of an imaging system in the measurement pro