Ultrasonic-magnetic powder double flaw detection method based on wind power anchor plate weld

Abstract

This invention relates to the field of materials testing and analysis technology, specifically to a method for ultrasonic-magnetic particle dual flaw detection of wind turbine anchor plate welds. The method includes: acquiring the signal matrix for each layer; determining a fracture effectiveness index based on the signal matrix for each layer, assuming the weld depth at each location meets the standard; determining the degree of magnetic particle influence based on the angle between the direction of the effective fracture path at each layer and the main fracture direction of its corresponding region, combined with the fracture effectiveness index; determining whether to re-clean the corresponding layer based on the degree of magnetic particle influence; if necessary, performing magnetic particle cleaning on the corresponding layer and repeatedly executing the step of determining the degree of magnetic particle influence until it is less than or equal to a threshold value, and then obtaining the current quality inspection result of the wind turbine anchor plate weld. This invention, by determining the degree of magnetic particle influence at each layer, allows for targeted magnetic particle cleaning of different layers, effectively improving the accuracy of the quality inspection results.

Description

Technical Field

[0001] This invention relates to the field of materials testing and analysis technology, specifically to a method for quality inspection of wind turbine anchor plate welds using ultrasonic-magnetic particle dual flaw detection. Background Technology

[0002] Wind turbine anchor plates are key components in the foundation structure of wind turbine generators, primarily used to connect and secure the wind turbine tower to the foundation, ensuring the overall structural stability and load-bearing capacity. Since anchor plates are typically modular, they require welding to form a complete unit. To ensure the welding quality of the anchor plates, quality inspection of the weld seams is necessary. Currently, quality inspection of anchor plate weld seams often combines magnetic particle testing and ultrasonic testing. Magnetic particle testing involves spraying a suspension of magnetic powder onto the anchor plate surface; surface cracks accumulate due to the adsorption of the magnetic powder, and the location of the cracks is revealed through the aggregation of the magnetic powder. Ultrasonic testing utilizes the penetrating power of ultrasonic waves within the anchor plate; when encountering welding defects, the waves are emitted, and the internal defects are located by analyzing the echo signals.

[0003] Ultrasonic testing typically involves ultrasonic analysis at the crack location. However, residual magnetic particles can fill tiny surface cracks, causing additional reflections of ultrasonic waves at these residues. This results in echo signals containing false path lengths, leading to misinterpretations as internal defects. In other words, residues from magnetic particle testing interfere with the accuracy of ultrasonic testing. To overcome this, existing methods involve demagnetizing and cleaning the workpiece surface to reduce magnetic particle adhesion. However, in practice, complex crack structures prevent complete removal of magnetic particles, leaving residual particles to interfere with ultrasonic echoes. Consequently, the accuracy of wind turbine anchor plate weld quality inspection results based on residual magnetic particles is low. On the other hand, excessive cleaning of weld cracks due to residual magnetic particles exacerbates the severity of wind turbine anchor plate weld defects, resulting in more defective welds during the quality inspection process. Summary of the Invention

[0004] To address the technical problem of low accuracy in the quality inspection results of wind turbine anchor plate welds, the present invention aims to provide an ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds. The specific technical solution adopted is as follows:

[0005] One embodiment of the present invention provides a method for ultrasonic-magnetic particle dual flaw detection of weld seams in wind turbine anchor plates, the method comprising the following steps:

[0006] S1. After adaptively determining the number of arrays, the ultrasonic array is started to scan the wind turbine anchor plate after magnetic particle testing to obtain the acquisition curves of each position point on the wind turbine anchor plate; the number of arrays includes the number of longitudinal ultrasonic sensors and the number of transverse ultrasonic sensors; the horizontal axis of the acquisition curve is the frequency length and the vertical axis is the echo intensity, and one frequency length corresponds to one level.

[0007] S2, provided that the weld depth at each location point meets the standard, analyze the fracture uniformity of all initial paths within the weld crack area based on the signal matrix of each level, and determine the fracture effectiveness index of each weld crack area; the signal matrix is ​​composed of the echo intensity of each location point corresponding to the same level.

[0008] S3, determine the crack marking area; based on the angle between the direction of the effective fracture path of each level and its corresponding main fracture direction, and in conjunction with the fracture effectiveness index, determine the degree of magnetic particle influence of each level; the main fracture direction of the region is the sum vector direction corresponding to all initial paths within the crack marking area;

[0009] S4. Based on the degree of magnetic particle influence, determine whether to re-clean the corresponding layer. If necessary, perform magnetic particle cleaning on the corresponding layer, and then execute steps S1 to S3 to re-determine the degree of magnetic particle influence. Repeat this process until the degree of magnetic particle influence of all layers is less than or equal to the degree threshold, and obtain the current quality inspection result of the wind turbine anchor plate weld.

[0010] Furthermore, for any given location, determining whether the weld depth at that location meets the standard includes:

[0011] Obtain the disappearance time of the echo signal at the location point, and determine the distance corresponding to the disappearance time from the sound wave return time-distance correspondence table, which is used as the weld depth at the location point;

[0012] If the weld depth range is set, and the weld depth at the specified location point is not within the specified range, then the weld depth at the specified location point is determined to be substandard; otherwise, the weld depth at the specified location point is determined to be compliant.

[0013] Furthermore, the step of analyzing the fracture uniformity of all initial paths within the weld crack region based on the signal matrix of each level, and determining the fracture effectiveness index for each weld crack region, includes:

[0014] For any signal matrix, determine the echo intensity histogram of the signal matrix, analyze the echo intensity distribution characteristics corresponding to the weld crack based on the echo intensity histogram, and obtain each weld crack region corresponding to the signal matrix.

[0015] For any weld crack region, the shortest path from the highest sound intensity point in the weld crack region to each edge point of the weld crack region is taken as the initial path.

[0016] Based on the echo intensity at each location point on each initial path, the severity of the fracture in each initial path is determined, and then the effective fracture path and the ineffective fracture path are distinguished by the severity of the fracture.

[0017] The fracture effectiveness index of the weld crack region is determined based on the maximum proportion of consecutively adjacent invalid fracture paths and the proportion of fracture severity of all the effective fracture paths.

[0018] Further, obtaining the weld crack region includes:

[0019] Determine the highest peak of the echo intensity histogram, mark the position point corresponding to the echo intensity of the highest peak as 0 in the signal matrix, and mark the position points corresponding to the echo intensity other than the echo intensity corresponding to the highest peak as 1, to obtain the marking matrix;

[0020] The marker matrix is ​​subjected to morphological erosion processing, and then the locations marked as 1 in the morphologically eroded marker matrix are subjected to connected component analysis to obtain the weld crack region.

[0021] Furthermore, determining the severity of the breakage of each initial path based on the echo intensity at each location point on each initial path includes:

[0022] For any initial path, the echo intensity at each location point on the initial path is processed by first-order difference to obtain each difference value;

[0023] Calculate the difference between each difference value and the mean value corresponding to all difference values, and use it as the difference offset; standardize the difference offset to obtain the severity of the initial path breakage.

[0024] Further, obtaining the effective fracture path and the invalid fracture path includes:

[0025] For any weld crack region, the fracture severity of all initial paths in the weld crack region is arranged in descending order to obtain a fracture severity sequence;

[0026] Calculate the difference between two adjacent fracture severity in the fracture severity sequence, and obtain the numerical range between the larger fracture severity corresponding to the two fracture severity with the maximum difference and the maximum fracture severity.

[0027] Initial paths within the specified numerical range are identified as valid break paths, while initial paths outside the specified numerical range are identified as invalid break paths.

[0028] Furthermore, based on the maximum proportion of consecutively adjacent invalid fracture paths and the proportion of fracture severity of all said valid fracture paths, fracture effectiveness indicators for the weld crack region are determined, including:

[0029] The maximum number of consecutively adjacent invalid fracture paths is counted, and the ratio of this maximum number to the total number of all initial paths within the weld crack area is used as the first fracture effectiveness factor.

[0030] A first cumulative value of the fracture severity of the effective fracture path within the weld crack region is determined, and the ratio of the first cumulative value to the second cumulative value of the fracture severity of all initial paths is used as a second fracture effectiveness factor.

[0031] By combining the first fracture effectiveness factor and the second fracture effectiveness factor, the fracture effectiveness index of the weld crack region is determined.

[0032] Further, determining the crack marking area includes:

[0033] An initial path is obtained from the weld crack region of the adjacent level to form an initial path pair, and the number of overlapping points of each initial path pair is determined; where, the adjacent level represents the length of two frequencies with adjacent numerical values;

[0034] For each adjacent level, obtain the maximum number of overlapping location points, and use the initial path of the maximum number of overlapping location points to the weld crack area to be the crack marking area of ​​the corresponding adjacent level.

[0035] Furthermore, determining the degree of magnetic particle influence at each level based on the angle between the direction of the effective fracture path at each level and the corresponding main fracture direction in the region, combined with the fracture effectiveness index, includes:

[0036] The main direction interval is determined by the maximum value of the angle corresponding to each level, an angle histogram is generated based on all the angles, and the number of paths located in the main direction interval is counted.

[0037] Based on the proportion of the number of paths and the fracture effectiveness index, the degree of magnetic particle influence at each level is determined.

[0038] Furthermore, the step of determining whether to re-clean the corresponding layer based on the degree of influence of the magnetic powder includes:

[0039] A threshold is set. For any level, if the degree of magnetic powder influence in that level is greater than the threshold, it is determined that the level needs to be cleaned again; otherwise, it is determined that the level does not need to be cleaned again.

[0040] The present invention has the following beneficial effects:

[0041] This invention provides a dual ultrasonic-magnetic particle inspection method for wind turbine anchor plate welds. The method adaptively determines the number of arrays, ensuring complete coverage of the entire anchor plate and facilitating the acquisition of echo intensity (acquisition curves) at different frequencies and lengths at each location point, thus improving the robustness of the ultrasonic array scanning. Based on the acquisition curves, the method first determines whether the weld depth at each location point meets the standard, saving inspection time and improving efficiency. If the weld depth does not meet the requirements, it can be directly identified as a defective product. For welds at locations where the depth meets the standard, the method combines the crack characteristics affected and unaffected by magnetic particle to determine the fracture effectiveness index for each weld crack region. Different fracture effectiveness indices exist for different weld crack regions at different levels, facilitating subsequent analysis of the degree of magnetic particle influence at different levels. By determining the degree of magnetic particle influence at each level, targeted magnetic particle cleaning can be performed at different levels while minimizing crack severity, improving not only the efficiency and effectiveness of magnetic particle cleaning but also the accuracy of wind turbine anchor plate weld quality inspection results. Attached Figure Description

[0042] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the steps of an ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds according to an embodiment of the present invention.

[0044] Figure 2 This is a flowchart illustrating the implementation of step S2 in an embodiment of the present invention. Detailed Implementation

[0045] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the technical solution proposed according to the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0047] The application scenarios targeted by this invention can be:

[0048] Demagnetizing the workpiece before cleaning its surface can reduce the adhesion of magnetic powder to some extent. However, in practice, the contact surfaces of surface cracks are complex. Ultrasonic testing of the workpiece quality, performed without a clear understanding of the cleaning effect, may still be affected by interference from the magnetic powder interface, leading to inaccuracies in the workpiece quality inspection results. Here, the workpiece can be a wind turbine anchor plate that requires welding.

[0049] To improve the accuracy of weld quality inspection, the surface of the wind turbine anchor plate weld is first treated, and magnetic powder is spread for inspection, resulting in a wind turbine anchor plate after magnetic powder inspection. Next, based on steps S1 to S3, the degree of magnetic powder influence of the wind turbine anchor plate after magnetic powder inspection is quantitatively analyzed. Finally, it is determined whether the degree of magnetic powder influence determined at each level meets the preset requirements. If a certain level does not meet the preset requirements, the level is cleaned in a targeted manner until the degree of magnetic powder influence at each level meets the preset requirements, and then the weld quality inspection result of the wind turbine anchor plate in the current cleaning state is obtained.

[0050] Among them, the surface treatment of the wind turbine anchor plate weld refers to grinding the weld position until the surface roughness meets the operation inspection standard; the magnetic particle inspection specifically involves using an atomizing sprayer to spray magnetic suspension liquid, and controlling the concentration and flow rate of the magnetic suspension liquid to spray the surface of the weld position.

[0051] Specifically, one embodiment of the present invention provides a method for ultrasonic-magnetic particle dual flaw detection of wind turbine anchor plate welds, such as... Figure 1 As shown, it includes the following steps:

[0052] S1. After adaptively determining the number of arrays, the ultrasonic array is started to scan the wind turbine anchor plate after magnetic particle testing to obtain the acquisition curves of each position point on the wind turbine anchor plate.

[0053] Here, the number of arrays includes the number of longitudinal ultrasonic sensors and the number of transverse ultrasonic sensors. The horizontal axis of the acquisition curve represents the frequency length, and the vertical axis represents the echo intensity.

[0054] As an exemplary implementation, step S1 described above can be achieved through the following steps:

[0055] The first step is to determine the number of arrays to use when scanning with an ultrasound array.

[0056] To align the ultrasonic array with the surface dimensions of the wind turbine anchor plate workpiece, the number of arrays is adaptively determined during ultrasonic testing. This helps align each position point on the wind turbine anchor plate with each ultrasonic unit in the ultrasonic array, facilitating subsequent prior assessment of the crack environment at the corresponding position of the ultrasonic array, such as obtaining the weld crack area.

[0057] Specifically, the length of the weld location and the detection range of a single sensing unit are obtained, and then the ratio of the two lengths (the larger length to the smaller length) is calculated. This ratio is then rounded up, and the rounded result is used as the number of transverse ultrasonic sensors. Similarly, the number of longitudinal ultrasonic sensors is obtained.

[0058] The length of the weld can be the length of the workpiece or the length of the fluorescent imaging image of the wind turbine anchor plate, and the width of the weld can be the width of the workpiece or the width of the fluorescent imaging image.

[0059] The second step involves using the number of arrays as the operating parameter of the ultrasonic array, and then starting the ultrasonic array to scan the wind turbine anchor plate after magnetic particle testing to obtain the acquisition curves of each location point on the wind turbine anchor plate.

[0060] Here, the number of transverse sensing units in the ultrasonic array is the number of transverse ultrasonic sensors, and the number of longitudinal sensing units is the number of longitudinal ultrasonic sensors.

[0061] In this embodiment, the ultrasonic array simultaneously transmits sound waves to collect echo intensity at different locations. The sound wave locations differ under different pulse results, which is due to the influence of internal voids and fractures in the weld, as well as the multi-layered nature of the welding cracks. The echo intensity at the same location may differ for different frequency lengths, therefore it is necessary to obtain the acquisition curves for each location on the wind turbine anchor plate. One frequency length corresponds to one layer of the wind turbine anchor plate.

[0062] Thus, this embodiment has obtained the acquisition curves of each location point on the wind turbine anchor plate, that is, obtained the echo intensity of each location point at different frequency lengths.

[0063] S2, under the premise that the weld depth at each location point meets the standard, analyze the fracture uniformity of all initial paths in the weld crack area according to the signal matrix of each level, and determine the fracture effectiveness index of each weld crack area.

[0064] Here, the signal matrix is ​​composed of the echo intensity of each location point corresponding to the same level. That is, the echo intensity of each location point in the coverage area of ​​each ultrasonic sensor under a single level is placed at the row and column positions corresponding to the sensing unit in the ultrasonic array, which can form a signal matrix of a single level.

[0065] The fracture effectiveness index can be obtained by analyzing the uniformity of echo intensity decay and the concentration of path direction of the effective fracture path in the weld crack area. The larger the fracture effectiveness index, the more obvious the crack characteristics that are not affected by magnetic powder, and the less the crack in the weld crack area is affected by magnetic powder.

[0066] As an exemplary implementation, step S2 described above can be achieved through... Figure 2 Steps S21 to S25 shown are implemented as follows:

[0067] S21, for each location point, determine whether the weld depth at the location point meets the standard.

[0068] Specifically, the disappearance time of the echo signal at the location point is obtained, and the distance corresponding to the disappearance time is determined from the sound wave return time-distance correspondence table, which is used as the weld depth of the location point; a weld depth range is set, and if the weld depth of the location point is not within the weld depth range, it is determined that the weld depth of the location point does not meet the standard, otherwise it is determined that the weld depth of the location point meets the standard.

[0069] In this embodiment, for any location point, the time when the echo signal disappears is first determined, i.e., the disappearance time. Then, the distance corresponding to the disappearance time is determined from the prior acoustic wave return time-distance correspondence table, and the distance is used as the weld depth La of the current location point. The preset weld depth L' and the allowable weld depth offset tolerance ΔL' are input into the control system. If the weld depth La satisfies L'-ΔL'>La>L'+ΔL', that is, the weld depth of the location point is within the weld depth range (L'-ΔL', L'+ΔL'), then the weld depth of the location point is determined to be up to standard. Otherwise, the weld depth of the location point is determined to be down to standard. If the weld depth of any location point is down to standard, the wind turbine anchor plate that has been welded can be directly identified as a defective weld workpiece.

[0070] It should be noted that the echo signal at the location point disappears at the moment of disappearance. The echo situation at the location point can characterize whether the ultrasonic wave has reached the bottom surface. When the echo signal disappears, it means that the bottom surface has been reached. Therefore, the weld depth at the location point can be obtained by the moment of disappearance of the echo signal at the location point.

[0071] S22. If the weld depth at each location point meets the standard, then for any signal matrix, determine the echo intensity histogram of the signal matrix, analyze the echo intensity distribution characteristics corresponding to the weld crack based on the echo intensity histogram, and obtain the weld crack region corresponding to each signal matrix.

[0072] Here, the weld crack area can be determined by analyzing the echo intensity distribution of all locations at the same level. The more locations corresponding to a certain echo intensity, the less likely those locations are to belong to the weld crack area, and the greater the likelihood that they belong to the normal area of ​​the wind turbine anchor plate. To facilitate subsequent analysis of the magnetic particle cleaning of the weld, i.e., to determine the degree of magnetic particle influence, it is necessary to screen out the weld crack area. The worse the magnetic particle cleaning of the weld, the greater the degree of magnetic particle influence, and the two show a positive correlation.

[0073] Provided that the weld depth at all locations meets the standard, determine the echo intensity histogram for each signal matrix. The horizontal axis of the echo intensity histogram can be the echo intensities arranged in descending order of value, and the vertical axis can be the number of locations for each echo intensity.

[0074] Based on the echo intensity histogram analysis, the echo intensity distribution characteristics corresponding to the weld crack are obtained, and the signal matrix for each weld crack region is obtained, including:

[0075] Specifically, the highest peak of the echo intensity histogram is determined, and the position point corresponding to the echo intensity of the highest peak is marked as 0 in the signal matrix, while the position points corresponding to the echo intensity other than the echo intensity corresponding to the highest peak are marked as 1, thus obtaining a marking matrix. Morphological etching is performed on the marking matrix, and then connected component analysis is performed on the position points marked as 1 in the morphologically etched marking matrix to obtain the weld crack region.

[0076] In this embodiment, the peak values ​​in the histogram correspond to densely distributed areas of echo intensity. The highest peak typically represents the background or dominant signal, while the second highest peak may be a real weak signal, such as a defect echo or target reflection. By excluding the highest peak in the echo intensity histogram, the most significant background peaks can be filtered out, retaining other potential effective signal peaks, i.e., the remaining peaks are used as peaks to be labeled. During ultrasonic testing, the highest peak may be an echo from a uniform part of the material, while the peaks to be labeled correspond to information about cracks or pores.

[0077] Therefore, in the signal matrix, the position point corresponding to the peak to be marked is marked as 1, and the remaining position points are marked as 0; the marked signal matrix is ​​subjected to morphological erosion to remove the slight adhesion between abnormal position points and realize the segmentation of the complete crack position; the position points marked as 1 in the morphologically eroded marking matrix are subjected to connected component analysis to obtain the weld crack region.

[0078] It should be noted that the morphological corrosion process is existing technology and is not within the scope of protection of this invention; therefore, it will not be described in detail here. Of course, there are cases where multiple weld crack regions correspond to a certain level, and there are also cases where there are zero weld crack regions.

[0079] S23, For any weld crack region, take the shortest path from the highest sound intensity point in the weld crack region to each edge point of the weld crack region as the initial path.

[0080] It should be noted that uniform weld seams generally do not have echoes. Once an echo appears, it is either generated by the bottom surface of the workpiece or by a significant crack.

[0081] In this embodiment, the point of highest sound intensity is the location of the highest echo intensity, which represents the point of strongest reflection. This point typically corresponds to the crack initiation point or the location of the most severe damage. The reflection intensity of the crack usually gradually decreases from the point of highest sound intensity towards the edge, exhibiting a radial distribution. Therefore, the shortest path from the point of highest sound intensity to each edge point of the weld crack region is used as the initial path. The initial path can characterize the crack distribution within the weld crack region.

[0082] The reasons for choosing the shortest path are: the shortest path from the point of highest sound intensity to the edge is most likely to approximate the actual direction of crack propagation, which can avoid introducing irrelevant tortuous paths to interfere with the analysis; the shortest path algorithm can be calculated efficiently, avoiding the computational burden caused by complex path fitting; the stress concentration zone near the crack tip is usually symmetrically distributed, and the direction of the shortest path is often consistent with the direction of the maximum principal stress, which conforms to the crack propagation criterion in fracture mechanics.

[0083] S24. Based on the echo intensity at each location point on each initial path, determine the severity of the fracture in each initial path, and then distinguish between effective fracture paths and ineffective fracture paths based on the severity of the fracture.

[0084] The first step is to determine the severity of the breakage of each initial path based on the echo intensity at each location point on each initial path.

[0085] First, it should be noted that, in the absence of magnetic powder, the more obvious the deviation between echo intensities on any initial path, the more uneven the fracture surface of that initial path, and the greater the severity of the fracture. On the other hand, magnetic powder accumulates and adheres to the fracture surface, easily forming a plane, thereby destroying the fracture characteristics of the crack. The more uniform the fracture surface, the less severe the fracture.

[0086] Here, the severity of fracture refers to the abruptness of the decay of echo intensity. The greater the severity of fracture, the more significant the abrupt change in echo intensity of the initial path, which may correspond to local non-uniform defects such as cracks and voids. The smaller the severity of fracture, the more gradual the change in echo intensity, and the better the material continuity.

[0087] Specifically, for any initial path, the echo intensity at each location point on the initial path is processed by first-order difference to obtain each difference value; the difference between each difference value and the mean value corresponding to all difference values ​​is calculated as the difference offset; the difference offset is standardized to obtain the severity of the breakage of the initial path.

[0088] In this embodiment, each difference value is taken on the current initial path, and the y-th difference value is denoted as ΔQ. y The mean of all differences is denoted as The standard deviation of all differences is denoted as σ. ΔQ ; Calculate the y-th difference value ΔQ y Mean of all differences The absolute value of the difference between the values ​​is calculated. The smaller the absolute value of the difference, the closer the difference values ​​are to the average difference value, the more significant the uniformity of the current intensity decay, and the less severe the fracture of the current initial path. After obtaining the absolute value of the difference corresponding to the y-th difference value, in order to facilitate the assessment of the attenuation of the current initial path, the absolute value of each difference value is standardized using the standard deviation of all difference values ​​to obtain the standardized values. In order to amplify the analysis of fracture details, the standardized values ​​are cubed, and then the average of all cubed values ​​is calculated to obtain the fracture severity of the current path.

[0089] The implementation process of performing first-order difference processing on the current initial path is existing technology and is not within the scope of protection of this invention, so it will not be described in detail here.

[0090] As an example, the formula for calculating the severity of the break in the current initial path can be:

[0091] In the formula, J represents the severity of the break in the current initial path, W represents the number of difference values ​​corresponding to the current initial path, and ΔQ y σ represents the y-th difference value corresponding to the current initial path. ΔQ This represents the standard deviation of all difference values ​​corresponding to the current initial path. This represents the average of all difference values ​​corresponding to the current initial path.

[0092] The second step is to distinguish between effective and ineffective fracture paths based on the severity of the fracture.

[0093] Distinguishing between initial paths at the same level helps quantify the dominant fracture direction in the current weld crack region based on the overall fracture severity distribution. Here, an effective fracture path refers to an initial path with significant fracture severity that dominates the fracture direction; an ineffective fracture path may be less severe, possibly representing noise or minor damage. Furthermore, differentiating between effective and ineffective fracture paths facilitates the subsequent determination of fracture effectiveness indicators for the weld crack region.

[0094] Specifically, for any weld crack region, the fracture severity of all initial paths in the weld crack region is arranged in descending order to obtain a fracture severity sequence; the difference between two adjacent fracture severity in the fracture severity sequence is calculated, and the numerical range between the larger fracture severity corresponding to the maximum difference and the maximum fracture severity is obtained; the initial paths within the numerical range are determined as valid fracture paths, and the initial paths outside the numerical range are determined as invalid fracture paths.

[0095] S25. Based on the maximum proportion of consecutively adjacent invalid fracture paths and the proportion of fracture severity of all valid fracture paths, determine the fracture effectiveness index of the weld crack region.

[0096] Here, the fracture effectiveness index refers to the degree to which the cracks within the weld crack region are actual cracks, that is, the degree to which they conform to the actual crack direction characteristics and uniformity characteristics. The more concentrated the effective path direction within the weld crack region and the greater the fracture severity, the higher the fracture effectiveness index of the weld crack region.

[0097] Specifically, the indicators for determining the fracture effectiveness of the weld crack region include:

[0098] The first step is to count the maximum number of consecutively adjacent invalid fracture paths, and then use the ratio of the maximum number to the total number of all initial paths within the weld crack area as the first fracture effectiveness factor.

[0099] In this embodiment, the first fracture effectiveness factor can characterize the proportion of invalid fracture paths in the weld crack area, i.e., the invalid ratio. The larger the invalid ratio, the fewer the effective fracture paths, the more concentrated the direction of the effective fracture paths, and the stronger the fracture effectiveness.

[0100] The second step is to determine the first cumulative value of the fracture severity of the effective fracture path within the weld crack area, and use the ratio of the first cumulative value to the second cumulative value of the fracture severity of all initial paths as the second fracture effectiveness factor.

[0101] In this embodiment, the second fracture effectiveness factor can characterize the fracture uniformity of the effective fracture path in the weld crack region. The more obvious the fracture uniformity, the stronger the fracture effectiveness.

[0102] The third step is to combine the first fracture effectiveness factor and the second fracture effectiveness factor to determine the fracture effectiveness index of the weld crack area.

[0103] In this embodiment, both the first fracture effectiveness factor and the second fracture effectiveness factor are positively correlated with the fracture effectiveness index. The larger the first fracture effectiveness factor and the second fracture effectiveness factor are, the larger the fracture effectiveness index is.

[0104] As an example, the formula for calculating the fracture effectiveness index of the weld crack region can be:

[0105] In the formula, D represents the fracture effectiveness index of the weld crack region, and n Y' n represents the maximum number of consecutively adjacent invalid fracture paths within the weld crack region. Y Sum(J) represents the total number of initial paths within the weld crack region. Y” J represents the first cumulative value indicating the severity of fracture along the effective fracture path within the weld crack region. all,Y The second cumulative value represents the severity of fracture across all initial paths within the weld crack region.

[0106] In the formula for calculating the fracture effectiveness index, n Y and J all,Y Not only can it analyze the proportion, but it can also... and By limiting the range to between 0 and 1, the influence of dimensions is eliminated when multiplying different types of data.

[0107] By referring to the process for determining the fracture effectiveness index of any weld crack region mentioned above, the fracture effectiveness index of each weld crack region can be obtained.

[0108] Thus, this embodiment has obtained the fracture effectiveness index for each weld crack region.

[0109] S3, determine the crack marking area; based on the angle between the direction of the effective fracture path of each layer and the corresponding main fracture direction of the region, and in combination with the fracture effectiveness index, determine the degree of magnetic particle influence of each layer.

[0110] Here, the crack marking area is the weld crack area with the greatest impact on the longitudinal extension of weld cracks in adjacent layers, and the direction corresponding to the crack marking area is the main fracture direction of the region. The degree of magnetic particle influence refers to the impact of residual magnetic particles in the magnetic particle inspection process on the quality of ultrasonically inspected welds. The greater the degree of magnetic particle influence, the less reliable the quality inspection results for wind turbine anchor plate welds are, that is, the greater the possibility of deviation in the accuracy of the quality inspection results.

[0111] As an exemplary implementation, determining the crack marking area includes:

[0112] The first step is to obtain an initial path from the weld crack region of the adjacent layer to form an initial path pair, and determine the number of overlapping points of each initial path pair; where adjacent layers represent the lengths of two frequencies with adjacent numerical values.

[0113] In this embodiment, the number of overlapping points in the initial path pair refers to the number of points with the same position on the two initial paths. The larger the number of overlapping points, the greater the impact of the crack in the initial path of the upper level extending longitudinally to the lower level in adjacent levels.

[0114] The second step is to obtain the maximum number of overlapping position points for each adjacent level, and use the initial path of the maximum number of overlapping position points to the weld crack area as the crack marking area of ​​the corresponding adjacent level.

[0115] In this embodiment, the maximum number of overlapping position points not only needs to be the position point with the most overlap among all initial path pairs corresponding to adjacent levels, but also needs to be greater than the preset number of overlapping position points. That is, when the maximum number of overlapping position points is less than the preset number of overlapping position points, it can be directly determined that there is no crack marking area in the corresponding adjacent level.

[0116] The preset number of overlapping points can be taken as an empirical value of 3, and the implementer can set it according to the specific situation without making a specific limitation. By setting the preset number of overlapping points, unnecessary calculations can be reduced to a certain extent, ensuring the accuracy of the main fracture direction of the region determined based on the crack marking area.

[0117] As an exemplary implementation, based on the angle between the direction of the effective fracture path at each level and the corresponding main fracture direction in the region, and in conjunction with the fracture effectiveness index, the degree of magnetic particle influence at each level is determined, including:

[0118] Here, the direction of the effective fracture path refers to the direction of the vector formed by each point on the effective fracture path, and the direction of the main fracture in the region is the direction of the sum vector corresponding to all initial paths within the crack marking area. By analyzing the angle between the direction of the effective fracture path and the direction of the main fracture in the region, the distribution characteristics of the direction of the effective fracture path can be statistically analyzed. The more concentrated the direction of the effective fracture path at a certain level is in the main direction interval, that is, the more effective fracture paths there are in the main direction interval, and the stronger the fracture effectiveness, the more significant the crack characteristics of the corresponding level are, and the less the crack is affected by magnetic powder.

[0119] The first step is to determine the main direction interval by the maximum value of the angle corresponding to each level, generate an angle histogram based on all angles, and count the number of paths located in the main direction interval.

[0120] In this embodiment, the main direction interval can be (θ) max -5°, θ max +5°), determine each angle in the main direction interval in the angle histogram, and take the sum of the number of effective fracture paths corresponding to each angle as the number of paths in the main direction interval corresponding to the corresponding level.

[0121] It should be noted that the maximum angle can represent the case of the greatest deflection, which is the main direction of crack propagation. Therefore, the main direction range can be determined based on the maximum angle. The allowable offset angle of the main direction range can be set to 5°. The implementer can also set it according to the specific actual situation. No specific limit is made here.

[0122] The second step is to determine the degree of magnetic particle influence at each level based on the proportion of the number of paths and the fracture effectiveness index.

[0123] In this embodiment, for any level, the more effective fracture paths located in the main direction interval of the level, the more concentrated the effective fracture directions are, and the less the level is affected by magnetic particles; the larger the fracture effectiveness index of all weld crack regions in the level, the smaller the directional influence of the ineffective fracture paths in the level, the stronger the fracture uniformity, and the less the level is affected by magnetic particles.

[0124] As an example, the formula for calculating the degree of magnetic particle influence at each level can be:

[0125] In the formula, C represents the degree of magnetic particle influence at each level, n' represents the number of effective fracture paths located in the main directional interval at each level, and n ” This represents the total number of valid fracture paths in each level. This represents the average fracture effectiveness index of all weld crack regions at each level. exp represents an exponential function with the natural constant e as the base. exp(-) is used to normalize the data to achieve negative correlation.

[0126] In the formula for calculating the degree of magnetic particle influence, each layer has its corresponding degree of magnetic particle influence. and All showed a negative correlation with the degree of influence of magnetic particles. and The larger the value, the smaller the influence of the magnetic powder at the corresponding level.

[0127] Thus, this embodiment has obtained the degree of magnetic particle influence at each level.

[0128] S4. Determine whether to perform magnetic particle cleaning again based on the degree of magnetic particle influence. If necessary, perform magnetic particle cleaning on the corresponding level, and then execute steps S1 to S3 to redetermine the degree of magnetic particle influence. Repeat this process until the degree of magnetic particle influence on all levels is less than or equal to the degree threshold, and obtain the quality inspection result of the current wind turbine anchor plate weld.

[0129] The determination of whether to repeat the magnetic particle cleaning operation based on the degree of magnetic particle impact includes:

[0130] A threshold is set. For any level, if the magnetic powder influence of that level is greater than the threshold, it is determined that the level needs to be cleaned again; otherwise, it is determined that the level does not need to be cleaned again.

[0131] In this embodiment, the degree threshold can be set to 0.3. The implementer can set the degree threshold according to the specific actual situation. If the implementer has high requirements for the magnetic powder cleaning effect, the degree threshold can be set to a smaller value, and vice versa.

[0132] When the degree of magnetic particle influence at each level is less than or equal to the degree threshold, it indicates that the magnetic particle cleaning effect at each level is good. Based on the premise of good magnetic particle cleaning effect, ultrasonic testing of weld quality can effectively improve the accuracy and reliability of the current quality inspection results of wind turbine anchor plate welds.

[0133] Regarding obtaining the current quality inspection results of wind turbine anchor plate welds, including:

[0134] First, the crack images (320-400nm band) obtained after magnetic particle inspection are overlaid with the crack areas of each weld after eliminating the influence of magnetic particle. Then, the multi-channel fusion function of software, such as Avizo software, is used to achieve cross-spectral alignment. Next, the abnormal interference areas are highlighted by adjusting the transparency of the fluorescent layer, thus obtaining the quality inspection results. The process of using ultrasound-magnetic particle to achieve quality inspection is existing technology and is not within the scope of protection of this invention, so it will not be described in detail here.

[0135] In summary, this invention provides an ultrasonic-magnetic particle dual flaw detection quality inspection method for wind turbine anchor plate welds. The adaptive determination of the array number ensures complete coverage of the entire wind turbine anchor plate by the ultrasonic array, facilitating the acquisition of echo intensity (i.e., acquisition curves) at different frequency lengths at each location point, thus improving the robustness of the ultrasonic array scanning. Based on the acquisition curves, the weld depth at each location point is first determined to meet the standard, improving quality inspection efficiency. If the weld depth does not meet the requirements, it can be directly identified as a defective product. For weld depths at various locations that meet the standard, the fracture effectiveness index for each weld crack region is determined by combining the crack characteristics affected and unaffected by magnetic particle. Different fracture effectiveness indices are corresponding to different weld crack regions at different levels, facilitating subsequent analysis of the degree of magnetic particle influence at different levels. By determining the degree of magnetic particle influence at each level, targeted magnetic particle cleaning can be performed at different levels while minimizing crack severity, effectively improving the accuracy of wind turbine anchor plate weld quality inspection results.

[0136] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for quality inspection of wind turbine anchor plate welds using ultrasonic-magnetic particle dual flaw detection, characterized in that, Includes the following steps: S1. After adaptively determining the number of arrays, the ultrasonic array is started to scan the wind turbine anchor plate after magnetic particle testing to obtain the acquisition curves of each position point on the wind turbine anchor plate. The number of arrays includes the number of longitudinal ultrasonic sensors and the number of transverse ultrasonic sensors. The horizontal axis of the acquisition curve is the frequency length, and the vertical axis is the echo intensity. One frequency length corresponds to one level. S2, provided that the weld depth at each location point meets the standard, analyze the fracture uniformity of all initial paths within the weld crack area based on the signal matrix of each level, and determine the fracture effectiveness index of each weld crack area; the signal matrix is ​​composed of the echo intensity of each location point corresponding to the same level. S3, determine the crack marking area; based on the angle between the direction of the effective fracture path of each level and its corresponding main fracture direction, and in conjunction with the fracture effectiveness index, determine the degree of magnetic particle influence of each level; the main fracture direction of the region is the sum vector direction corresponding to all initial paths within the crack marking area; S4. Based on the degree of magnetic particle influence, determine whether to re-clean the corresponding layer. If necessary, perform magnetic particle cleaning on the corresponding layer, and then execute steps S1 to S3 to re-determine the degree of magnetic particle influence. Repeat this process until the degree of magnetic particle influence of all layers is less than or equal to the degree threshold, and obtain the current quality inspection result of the wind turbine anchor plate weld. The step of analyzing the fracture uniformity of all initial paths within the weld crack region based on the signal matrix of each level, and determining the fracture effectiveness index of each weld crack region, includes: For any signal matrix, determine the echo intensity histogram of the signal matrix, analyze the echo intensity distribution characteristics corresponding to the weld crack based on the echo intensity histogram, and obtain each weld crack region corresponding to the signal matrix. For any weld crack region, the shortest path from the highest sound intensity point in the weld crack region to each edge point of the weld crack region is taken as the initial path. Based on the echo intensity at each location point on each initial path, the severity of the fracture in each initial path is determined, and then the effective fracture path and the ineffective fracture path are distinguished by the severity of the fracture. The fracture effectiveness index of the weld crack region is determined based on the maximum proportion of consecutively adjacent invalid fracture paths and the proportion of fracture severity of all the effective fracture paths.

2. The ultrasonic-magnetic particle dual flaw detection quality inspection method for wind turbine anchor plate welds according to claim 1, characterized in that, For any given location, determine whether the weld depth at that location meets the standard, including: Obtain the disappearance time of the echo signal at the location point, and determine the distance corresponding to the disappearance time from the sound wave return time-distance correspondence table, which is used as the weld depth at the location point; If the weld depth range is set, and the weld depth at the specified location point is not within the specified range, then the weld depth at the specified location point is determined to be substandard; otherwise, the weld depth at the specified location point is determined to be compliant.

3. The ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds according to claim 2, characterized in that, Obtaining the weld crack region includes: Determine the highest peak of the echo intensity histogram, mark the position point corresponding to the echo intensity of the highest peak as 0 in the signal matrix, and mark the position points corresponding to the echo intensity other than the echo intensity corresponding to the highest peak as 1, to obtain the marking matrix; The marker matrix is ​​subjected to morphological erosion processing, and then the locations marked as 1 in the morphologically eroded marker matrix are subjected to connected component analysis to obtain the weld crack region.

4. The ultrasonic-magnetic particle dual flaw detection quality inspection method for wind turbine anchor plate welds according to claim 2, characterized in that, The determination of the severity of fracture in each initial path based on the echo intensity at each location point on each initial path includes: For any initial path, the echo intensity at each location point on the initial path is processed by first-order difference to obtain each difference value; Calculate the difference between each difference value and the mean value corresponding to all difference values, and use it as the difference offset; standardize the difference offset to obtain the severity of the initial path breakage.

5. The ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds according to claim 2, characterized in that, Obtaining the valid fracture path and the invalid fracture path includes: For any weld crack region, the fracture severity of all initial paths in the weld crack region is arranged in descending order to obtain a fracture severity sequence; Calculate the difference between two adjacent fracture severity in the fracture severity sequence, and obtain the numerical range between the larger fracture severity corresponding to the maximum difference and the maximum fracture severity. Initial paths within the specified numerical range are identified as valid break paths, while initial paths outside the specified numerical range are identified as invalid break paths.

6. The ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds according to claim 2, characterized in that, Based on the maximum percentage of consecutively adjacent invalid fracture paths and the percentage of fracture severity across all valid fracture paths, fracture effectiveness indicators for the weld crack region are determined, including: The maximum number of consecutively adjacent invalid fracture paths is counted, and the ratio of this maximum number to the total number of all initial paths within the weld crack area is used as the first fracture effectiveness factor. A first cumulative value of the fracture severity of the effective fracture path within the weld crack region is determined, and the ratio of the first cumulative value to the second cumulative value of the fracture severity of all initial paths is used as a second fracture effectiveness factor. By combining the first fracture effectiveness factor and the second fracture effectiveness factor, the fracture effectiveness index of the weld crack region is determined.

7. The ultrasonic-magnetic particle dual flaw detection quality inspection method for wind turbine anchor plate welds according to claim 2, characterized in that, The determination of the crack marking area includes: An initial path is obtained from the weld crack region of the adjacent level to form an initial path pair, and the number of overlapping points of each initial path pair is determined; where, the adjacent level represents the length of two frequencies with adjacent numerical values; For each adjacent level, obtain the maximum number of overlapping location points, and use the initial path of the maximum number of overlapping location points to the weld crack area to be the crack marking area of ​​the corresponding adjacent level.

8. The ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds according to claim 1, characterized in that, The determination of the magnetic particle influence degree at each level, based on the angle between the direction of the effective fracture path at each level and the corresponding main fracture direction in the region, combined with the fracture effectiveness index, includes: The main direction interval is determined by the maximum value of the angle corresponding to each level, an angle histogram is generated based on all the angles, and the number of paths located in the main direction interval is counted. Based on the proportion of the number of paths and the fracture effectiveness index, the degree of magnetic particle influence at each level is determined.

9. The ultrasonic-magnetic particle dual flaw detection method for wind turbine anchor plate welds according to claim 1, characterized in that, The step of determining whether to re-clean the corresponding layer based on the degree of magnetic powder influence includes: A threshold is set. For any level, if the degree of magnetic powder influence in that level is greater than the threshold, it is determined that the level needs to be cleaned again; otherwise, it is determined that the level does not need to be cleaned again.

Images