Welding quality inspection method and platform for LNG marine pipe fittings

Abstract

The present application provides a welding quality inspection method and platform for LNG marine pipe fittings, which relates to the field of intelligent inspection technology, including: determining the welding damage intensity based on the spatial analysis of pipe fitting defect characteristics, calling the load-bearing performance evaluation channel to perform performance impact analysis, and outputting a first potential performance impact coefficient; using the standard pipe fitting welding image as a benchmark, performing a similarity traversal comparison on the pipe fitting surface image, and determining the second potential performance impact coefficient based on the deviation area ratio; determining the pipe fitting damage intensity based on the first and second potential performance impact coefficients, and if the damage intensity is greater than the intensity threshold, marking the pipe fitting for inspection. This application can solve the problem in existing methods that the actual damage condition of pipe fittings cannot be accurately judged due to the inability to conduct targeted detection of pipe fitting welding defects in combination with application scenarios. It can efficiently and accurately identify defects on the surface and inside of pipe fittings, significantly improving the accuracy and reliability of pipe fitting damage assessment.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection technology, and in particular to a welding quality detection method and platform for LNG marine pipe fittings. Background Art

[0002] During the manufacturing and maintenance of pipelines and fittings, welding quality is a key factor in ensuring the reliability and safety of pipe fittings. The welding quality of pipe fittings, especially those used in high-pressure, high-temperature and corrosive environments, such as LNG marine pipe fittings, is directly related to the load-bearing capacity and service life of the pipeline. Welding defects such as cracks, pores, slag inclusions, and lack of fusion often occur in weld joints and their heat-affected zones. If these defects are not discovered and repaired in a timely manner, they may lead to failure of the pipe fittings and even accidents.

[0003] At present, there is a technical problem in the existing pipe fitting welding quality inspection method that it is unable to conduct targeted inspection of pipe fitting welding defects in combination with the pipe fitting application scenario, resulting in the inability to accurately judge the actual damage condition of the pipe fitting. Summary of the Invention

[0004] The purpose of this application is to provide a welding quality inspection method and platform for LNG ship pipe fittings, so as to solve the technical problem that the existing pipe fitting welding quality inspection method cannot accurately judge the actual damage condition of the pipe fittings due to the inability to conduct targeted detection of pipe fitting welding defects in combination with the pipe fitting application scenarios.

[0005] In view of the above problems, the present application provides a welding quality detection method and platform for LNG marine pipe fittings.

[0006] In the first aspect, the present application provides a welding quality inspection method for LNG ship pipe fittings, which is implemented through a welding quality inspection platform for LNG ship pipe fittings, including: performing radiographic scanning and surface image acquisition on a target Y-type tee pipe fitting, performing internal defect identification and surface defect identification based on the pipe fitting radiographic image and the pipe fitting surface image, respectively, to obtain an internal defect feature distribution and a surface defect feature distribution; performing spatial fitting and maximum value retention processing on the internal defect feature distribution and the surface defect feature distribution, and modeling to generate a pipe fitting defect feature space; determining the welding damage intensity based on the pipe fitting defect feature space analysis, and calling a bearing performance evaluation channel to perform a performance impact analysis based on the welding damage intensity, and outputting a first potential performance impact coefficient; taking the standard pipe fitting welding image of the target Y-type tee pipe fitting as a reference, performing a similarity traversal comparison on the pipe fitting surface image, and determining a second potential performance impact coefficient based on the deviation area ratio; evaluating and determining the pipe fitting damage intensity based on the first potential performance impact coefficient and the second potential performance impact coefficient, if the pipe fitting damage intensity is greater than the risk intensity threshold, marking the target Y-type tee pipe fitting for inspection.

[0007] In the second aspect, the present application also provides a welding quality inspection platform for LNG ship pipe fittings, which is used to execute a welding quality inspection method for LNG ship pipe fittings as described in the first aspect, including: a defect feature distribution acquisition module, which is used to perform radiographic scanning and surface image acquisition on the target Y-type three-way pipe fitting, and perform internal defect identification and surface defect identification according to the pipe fitting radiographic image and the pipe fitting surface image, respectively, to obtain the internal defect feature distribution and the surface defect feature distribution; a defect feature space generation module, which is used to perform spatial fitting and maximum value retention processing on the internal defect feature distribution and the surface defect feature distribution, and model and generate the pipe fitting defect feature space; a first performance influence coefficient acquisition module, It is used to determine the welding damage intensity based on the spatial analysis of the pipe defect characteristics, and call the bearing performance evaluation channel to perform performance impact analysis according to the welding damage intensity, and output a first potential performance impact coefficient; the second performance impact coefficient acquisition module is used to perform a similarity traversal comparison on the pipe surface image based on the standard pipe welding image of the target Y-type tee pipe fitting as a benchmark, and determine the second potential performance impact coefficient according to the deviation area ratio; the pipe damage intensity determination module is used to determine the pipe damage intensity based on the first potential performance impact coefficient and the second potential performance impact coefficient evaluation. If the pipe damage intensity is greater than the risk intensity threshold, the target Y-type tee pipe fitting is marked for inspection.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The target Y-shaped tee pipe fitting is subjected to radiographic scanning and surface image acquisition. Internal and surface defect identification are performed based on the pipe fitting radiographic image and surface image, respectively, to obtain internal and surface defect feature distributions. Spatial fitting and maximum value retention are then performed on the internal and surface defect feature distributions to model and generate a pipe fitting defect feature space. The weld damage intensity is then determined based on the pipe fitting defect feature space analysis. A load-bearing performance evaluation channel is then invoked based on the weld damage intensity to perform a performance impact analysis and output a first potential performance impact coefficient. Furthermore, a similarity traversal comparison is performed on the pipe fitting surface image using a standard pipe fitting weld image as a reference, and a second potential performance impact coefficient is determined based on the deviation area ratio. Finally, the pipe fitting damage intensity is assessed based on the first and second potential performance impact coefficients. If the pipe fitting damage intensity exceeds a risk intensity threshold, the target Y-shaped tee pipe fitting is marked for inspection. This method can efficiently and accurately identify defects on the surface and inside of pipe fittings, significantly improving the accuracy and reliability of pipe fitting damage assessment, thereby enabling early detection of potential risks and ensuring the safety and reliability of pipe fitting use.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0012] Figure 1 A schematic flow chart of a welding quality inspection method for LNG marine pipe fittings is provided for this application;

[0013] Figure 2 This is a schematic diagram of a process for determining welding damage intensity in a welding quality inspection method for LNG marine pipe fittings in this application;

[0014] Figure 3 This is a structural schematic diagram of a welding quality inspection platform for LNG marine pipe fittings.

[0015] Description of reference numerals:

[0016] Defect feature distribution acquisition module 11, defect feature space generation module 12, first performance impact coefficient acquisition module 13, second performance impact coefficient acquisition module 14, pipe damage strength determination module 15. DETAILED DESCRIPTION

[0017] This application provides a welding quality inspection method and platform for LNG marine pipe fittings, addressing the technical issue with existing pipe welding quality inspection methods, which cannot accurately determine the true damage condition of pipe fittings due to the inability to conduct targeted inspections of pipe welding defects in conjunction with pipe application scenarios. This method can efficiently and accurately identify surface and internal defects in pipe fittings, significantly improving the accuracy and reliability of pipe damage assessments, thereby enabling early detection of potential risks and ensuring the safety and reliability of pipe fittings.

[0018] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0019] For example 1, please refer to the attached Figure 1 The present application provides a welding quality inspection method for LNG marine pipe fittings, which is applied to a welding quality inspection platform for LNG marine pipe fittings and specifically includes the following steps:

[0020] S100: performing radiographic scanning and surface image acquisition on a target Y-shaped tee pipe fitting, performing internal defect recognition and surface defect recognition based on the pipe fitting radiographic image and the pipe fitting surface image, respectively, to obtain internal defect feature distribution and surface defect feature distribution.

[0021] Furthermore, step S100 of the present application further includes:

[0022] S110: Based on a multi-scale convolutional neural network, an internal defect recognition model and a surface defect recognition model are constructed for the purpose of identifying predetermined defect features, wherein the predetermined defect features include defect type and size information, and the defect types include at least cracks, pores, and slag inclusions; S120: Internal defects are identified on the pipe X-ray image using the internal defect recognition model, and the internal defect feature distribution is output; S130: Surface defects are identified on the pipe surface image using the surface defect recognition model, and the surface defect feature distribution is output.

[0023] Specifically, Y-type tee pipe fittings play an important role in LNG marine pipe fittings, especially in the liquefied natural gas (LNG) transportation, storage and distribution system. LNG marine pipe fittings are pipeline system components designed specifically for LNG ships. They have special requirements for low temperature resistance, corrosion resistance, high strength and safety. As part of this system, Y-type tee pipe fittings are responsible for diverting or converging liquefied natural gas in the pipeline system. The quality of their welding directly affects the operational safety of the entire pipeline system.

[0024] First, the target Y-type tee pipe fitting is subjected to radiographic scanning and surface image acquisition. Radiographic scanning is a commonly used method in nondestructive testing, primarily used to detect potential defects within the pipe fitting. For Y-type tee pipe fittings, radiographic scanning can effectively reveal internal defects in welded joints, such as pores, slag inclusions, lack of fusion, and cracks. Based on the material and thickness of the pipe fitting, appropriate radiographic equipment, typically an X-ray machine, is selected. The energy of the radiation needs to be selected based on the thickness and material of the pipe fitting to ensure that it can penetrate and form a clear image. A radiographic image of the pipe fitting is obtained. Surface image acquisition is used to inspect the surface defects of the pipe fitting, such as cracks, weld beading, and pores. Surface defects can have a direct impact on the sealing and load-bearing capacity of the pipe fitting, so surface inspection is also crucial. Using a high-definition industrial camera, the welded joints and pipe surface of the target Y-type tee pipe fitting are photographed from multiple angles to acquire surface images of the pipe fitting.

[0025] The multi-scale convolutional neural network uses convolution operations to extract features at different scales, thereby effectively capturing defects of different sizes. By using multiple layers of convolution kernels and pooling layers, the network can gradually extract features from simple to complex, effectively distinguishing and locating different types of defects. Next, based on the multi-scale convolutional neural network, with the purpose of identifying predetermined defect features, an internal defect recognition model and a surface defect recognition model are constructed. Among them, internal defect recognition focuses on defects inside pipe fittings. The model needs to identify internal defects such as cracks and pores. It includes an input layer, a multi-scale convolutional layer, a pooling layer, a fully connected layer, and an output layer. The input data of the input layer is the X-ray image of the pipe fitting, and the output data is the internal defect type and defect size information. Surface defect recognition focuses on possible defects on the surface of the pipe fitting, such as cracks, pores, weld nodules, etc. It also includes an input layer, a multi-scale convolutional layer, a pooling layer, a fully connected layer, and an output layer. The input data of the input layer is the surface image of the pipe fitting, and the output data of the output layer is the surface defect type and defect size information.

[0026] Further, based on the pipe inspection log, with the target Y-type tee pipe fitting as a constraint, a sample pipe fitting radiographic image set and a sample internal defect feature set (internal defect type and defect size information) are collected as an internal defect training data set; the internal defect training data set is used to supervise the internal defect recognition model. First, the preprocessed radiographic image is used as input, the defect type and size are used as labels, and the input data is forward propagated through the network to generate a predicted output; then the defect type and size output by the network are compared with the actual label, and the loss value is calculated through the loss function; then, based on the calculated loss, the back propagation algorithm is used to update the weights in the network, and an optimization algorithm (such as Adam or SGD) is used to minimize the loss function, thereby optimizing the performance of the network; iterative training is repeated until the loss function converges to obtain a trained internal defect recognition model; on the other hand, a sample pipe fitting surface image set and a sample surface defect feature set (surface defect type and defect size information) are collected as a surface defect training data set, and the surface defect recognition model is supervised and trained using the surface defect training data set until convergence to obtain a trained surface defect recognition model. By building internal defect recognition models and surface defect recognition models based on multi-scale convolutional neural networks, welding defects in pipe fittings can be efficiently and accurately identified and classified, significantly improving the accuracy and speed of detection.

[0027] Then, the pipe fitting radiographic image is input into the internal defect recognition model for internal defect recognition, and the internal defect feature distribution (including the location, type, size information of each internal defect, etc.) is output; the pipe fitting surface image is input into the surface defect recognition model for surface defect recognition, and the surface defect feature distribution (including the location, type, size, etc. of each surface defect) is output.

[0028] S200: performing spatial fitting and maximum value retention processing on the internal defect feature distribution and the surface defect feature distribution to generate a pipe defect feature space by modeling.

[0029] Specifically, the internal defect feature distribution and the surface defect feature distribution are spatially fitted and maximum value retention processed. The purpose of spatial fitting is to model the defect feature distribution in the image space and create a unified spatial model to describe the distribution of defects inside and outside the pipe. Through this process, the defects can be spatially located and the severity and impact range of the defects can be determined. First, the spatial coordinates of all defects (such as pixel coordinates in the image) are converted into a physical coordinate system, and the coordinate range is normalized to ensure the consistency of spatial data. Then, based on the defect position, size information and defect type, the defect distribution is spatially fitted using interpolation methods (such as bilinear interpolation, spline interpolation, etc.). Through spatial fitting, a smooth defect distribution model can be generated to show the relative impact area of ​​each defect. The purpose of maximum value retention processing is to retain the most serious and significant defects in pipe fittings. These defects usually have a significant impact on the safety and performance of pipe fittings. This processing step selects the spatial characteristics of defects and retains only those critical defects that may cause performance degradation. First, the maximum value areas in the defect distribution map after spatial fitting are identified through algorithms (such as local maximum value detection, non-maximum value suppression, etc.). These areas usually correspond to the locations of the most serious defects in the pipe fitting. The maximum value can indicate the location and intensity of the defect, that is, the location where the defect has the greatest impact on the overall structure; then, the maximum value area is screened and only the most serious defects (such as crack tips, pore areas, etc.) are retained. By setting the threshold, minor defects with less impact on performance are removed; by highlighting the maximum value area, the visualization effect of the defect's impact on overall performance is enhanced, providing a basis for subsequent performance evaluation and decision-making.

[0030] Then, combining the results of spatial fitting and maximum value processing, a three-dimensional or two-dimensional defect feature space is created. This space reflects the defect distribution, defect intensity, and defect impact range of the pipe fitting, resulting in the pipe fitting defect feature space. This spatial model effectively shows the distribution, type, and impact of various defects in the pipe fitting, providing strong support for subsequent pipe fitting health assessment, maintenance decision-making, and performance prediction.

[0031] S300: Determine welding damage intensity based on the pipe defect feature space analysis, call a load-bearing performance evaluation channel to perform performance impact analysis according to the welding damage intensity, and output a first potential performance impact coefficient.

[0032] Further, if Figure 2 As shown, step S300 of this application also includes:

[0033] S310: Based on the pipe fitting defect feature space, perform transverse defect feature analysis and longitudinal defect feature analysis respectively to determine the transverse defect area ratio, transverse defect distribution discreteness and longitudinal defect depth; S320: Obtain the material properties and working environment characteristics of the target Y-type tee pipe fitting, perform corrosion intensity simulation analysis based on the material properties and working environment characteristics, and determine the corrosion intensity coefficient; S330: Determine the welding damage intensity based on the transverse defect area ratio, transverse defect distribution discreteness, longitudinal defect depth and corrosion intensity coefficient evaluation.

[0034] Specifically, based on the pipe defect feature space, transverse defect feature analysis and longitudinal defect feature analysis are performed respectively. The transverse defect feature analysis focuses on the distribution of defects in the transverse direction of the pipe (usually the cross-sectional direction of the pipe), including the area ratio and distribution dispersion of the defects. The transverse defect area ratio refers to the ratio of the area occupied by the defect to the entire cross-sectional area in the cross-sectional area of ​​the pipe. The transverse defect distribution dispersion is used to assess whether the defects are evenly distributed across the cross-sectional area of ​​the pipe. The smaller the distribution dispersion, the more concentrated the defects are in a certain part of the pipe. For example, the standard deviation or variance is used to measure the degree of distribution dispersion of transverse defects across the cross-sectional area of ​​the pipe to obtain the transverse defect area ratio and transverse defect distribution dispersion. The longitudinal defect feature analysis focuses on the distribution of defects in the longitudinal direction of the pipe (along the pipeline axis), especially the depth of the defects (the vertical depth from the pipe surface to the defect). The longitudinal defect depth refers to the vertical distance from the pipe surface to the defect location. The average defect depth is calculated by averaging the defect depths of multiple internal defects, and the average defect depth is set as the longitudinal defect depth.

[0035] Obtain the material properties and operating environment characteristics of the target Y-shaped tee pipe fitting. Common materials for Y-shaped tee pipe fittings include stainless steel, carbon steel, and alloy steel. These materials have significant variations in corrosion resistance. Therefore, material properties, including corrosion rate, are crucial factors in determining the corrosion intensity of pipe fittings. The corrosion intensity of pipe fittings in actual operating environments is affected by a variety of environmental factors, including temperature, humidity, and liquid composition (e.g., water, acid, and salt). The mean temperature, mean humidity, and mean component concentration can be used as operating environment characteristics. A corrosion intensity simulation analysis is then performed based on these material properties and operating environment characteristics. For example, a linear corrosion model is constructed for simulation analysis. The corrosion intensity coefficient is calculated based on the relationship between the corrosion rate, the environment, and the material. This coefficient measures the impact of the corrosive environment on the pipe fitting. A larger corrosion intensity coefficient indicates more severe corrosion.

[0036] Finally, a comprehensive assessment of the welding damage intensity is performed based on the transverse defect area ratio, transverse defect distribution dispersion, longitudinal defect depth, and corrosion intensity coefficient. The welding damage intensity is positively correlated with the transverse defect area ratio, longitudinal defect depth, and corrosion intensity coefficient, and negatively correlated with the transverse defect distribution dispersion. First, the transverse defect area ratio, transverse defect distribution dispersion, longitudinal defect depth, and corrosion intensity coefficient are dimensionlessly processed to ensure that features of different scales and units have the same influence. Dimensionless processing typically employs normalization or standardization methods to ensure that all features are on the same scale, avoiding disproportionate influence of certain features on the results due to their large numerical ranges. Then, a weighted allocation is performed based on the influence of the transverse defect area ratio, transverse defect distribution dispersion, longitudinal defect depth, and corrosion intensity coefficient on the welding damage intensity. The greater the influence, the greater the corresponding weight. Then, a weighted calculation is performed based on the influence of the transverse defect area ratio, transverse defect distribution dispersion, longitudinal defect depth, and corrosion intensity coefficient on the welding damage intensity to obtain the welding damage intensity. The welding damage intensity is used to assess the overall health of the pipe fitting. A higher damage intensity indicates a greater potential risk during use of the pipe fitting.

[0037] Furthermore, step S300 of this application also includes:

[0038] S340: Pre-training a bearer performance evaluation channel, wherein the bearer performance evaluation channel includes P basis evaluators.

[0039] Furthermore, step S340 of this application also includes:

[0040] S341: Based on the historical maintenance log of the target Y-type tee pipe fitting, a three-dimensional feature distribution set of sample pipe fitting defects is collected, and the proportion of pipe fitting load-bearing failures within a predetermined time window of the three-dimensional feature distribution of different sample pipe fitting defects is statistically analyzed, and the sample pipe fitting load-bearing failure proportion is obtained and set as the sample performance influence coefficient to obtain the sample performance influence coefficient set; S342: Integrate the three-dimensional feature distribution set of sample pipe fitting defects and the sample performance influence coefficient set to obtain a sample training set, and divide it into P equal parts to obtain P training arrays; S343: Use the P training arrays to train the generator and discriminator of the generative adversarial network until P basis evaluators are obtained after convergence, and integrate to obtain the load-bearing performance evaluation channel.

[0041] Specifically, based on the historical maintenance logs of the target Y-shaped tee pipe fittings, defect information for the sample pipe fittings is extracted from the maintenance logs. These defects include internal and surface defects. The three-dimensional features of the defects may include cracks, pores, slag inclusions, and other types of defects, as well as their location, size, and depth in three-dimensional space. Then, based on the collected defect data, a three-dimensional feature distribution set of sample pipe fitting defects is constructed. The defect feature of each pipe fitting sample is represented as a data point in three-dimensional space. Furthermore, the proportion of pipe fitting load-bearing failures within a predetermined time window (e.g., within a week) for each sample pipe fitting's three-dimensional feature distribution is statistically analyzed. Load-bearing failure refers to the loss of load-bearing capacity of a pipe fitting due to defects or damage during operation, typically manifested as crack propagation, pore enlargement, or reduced strength due to slag inclusions. The number of failures occurring within the predetermined time window (e.g., within a week) is counted, and the proportion of sample pipe fitting load-bearing failures is calculated as the sample performance impact coefficient. This coefficient reflects the potential impact of defects on the load-bearing capacity of each pipe fitting during its service life, resulting in a sample performance impact coefficient set.

[0042] The three-dimensional feature distribution set of sample pipe defects and the sample performance influence coefficient set are further integrated, that is, the three-dimensional feature distribution of sample pipe defects and the corresponding sample performance influence coefficients are used as a set of training data to obtain a sample training set; then the sample training set is divided into P equal parts, where P is an integer greater than 10. The value of P can be set according to actual needs, such as being set to 20 to obtain P training arrays. Then, a first training array is randomly selected from the P training arrays without replacement, and supervised training is performed on the generative adversarial network. The generative adversarial network is a neural network framework consisting of a generator and a discriminator. The generator generates fake samples, and the discriminator provides feedback to the generator by distinguishing between real and fake samples. Through this adversarial training, the generator can learn how to generate realistic samples. In this case, the generator will generate load-bearing performance data for the pipe fittings based on the defect characteristics of the sample pipe fittings (such as three-dimensional defect distribution, performance impact coefficient, etc.), and the discriminator will evaluate the load-bearing performance of the samples, that is, the load-bearing capacity of the pipe fittings under specific defect conditions. The generator's task is to generate load-bearing performance data for the pipe fittings based on the three-dimensional characteristics of the pipe fitting defects. This data should reflect the impact of the defects on the pipe fittings' load-bearing capacity. The discriminator receives the real load-bearing performance data and the fake performance data generated by the generator and outputs a probability value representing the real and fake data. The discriminator's goal is to distinguish between real data and generated data as accurately as possible and provide feedback to the generator. The generative adversarial network is trained using the first training array. In each round of training, the generator attempts to generate increasingly realistic load-bearing performance data. The discriminator is trained based on the generated data and real data to improve its ability to distinguish between generated data and real data. The training continues until the network converges, the generator is able to generate sufficiently realistic load-bearing performance data, and the discriminator's ability is also enhanced. The trained first base evaluator is then output.

[0043] Continue to use other training data to supervise the training of the generator and discriminator of the generative adversarial network to obtain P basis evaluators, and obtain the load-bearing performance evaluation channel according to the combination of the P basis evaluators.

[0044] S350: Obtain the maximum welding damage intensity within a predetermined historical time zone, and determine the number of calls to the base evaluator of the load-bearing performance evaluation channel based on the welding damage intensity and the maximum welding damage intensity evaluation, which is set to Q; S360: Randomly activate Q selected base evaluators among the P base evaluators, use the Q selected base evaluators to perform performance impact analysis on the pipe defect feature space, output Q performance impact coefficients, and obtain the first potential performance impact coefficient after average calculation.

[0045] Specifically, the maximum weld damage intensity within a predetermined historical time zone (e.g., the most recent month) is obtained. The ratio of the weld damage intensity to the maximum weld damage intensity is multiplied by P and rounded to an integer to obtain the number of base evaluator calls, set to Q. For example, assuming the weld damage intensity is 0.2, the maximum weld damage intensity is 0.5, and P is 20, then Q is 8. Q selected base evaluators are then randomly activated from the P base evaluators. The pipe defect feature space is input into these Q selected base evaluators for performance impact analysis, outputting Q performance impact coefficients. These Q performance impact coefficients are then averaged to obtain a first potential performance impact coefficient.

[0046] By selecting an appropriate number of base evaluators for load-bearing performance analysis according to the welding damage intensity, the required number of base evaluators can be flexibly and dynamically adjusted according to the actual state of the pipe fitting, so that more base evaluators can be used for more accurate load-bearing performance evaluation when the damage is more serious. When the damage is lighter, the calculation amount is reduced, the evaluation efficiency is improved, and the rational utilization of computing resources is improved.

[0047] S400: Using the standard pipe welding image of the target Y-shaped tee pipe fitting as a reference, performing similarity traversal comparison on the pipe fitting surface image, and determining a second potential performance impact coefficient according to the deviation area ratio.

[0048] Furthermore, step S400 of this application also includes:

[0049] S410: performing welding area cropping on the standard pipe welding image and the pipe surface image respectively to obtain a standard welding area image and a pipe welding area image; S420: performing grayscale processing on the standard welding area image and the pipe welding area image to obtain a standard welding area grayscale image and a pipe welding area grayscale image; S430: randomly selecting a first grayscale area in the pipe welding area grayscale image according to a predetermined size, wherein the predetermined size is a 5 by 5 pixel area; S440: comparing the first grayscale area with the standard welding area grayscale image The method further comprises the following steps: performing traversal similarity comparison on the area of ​​the predetermined size in the image, and setting the maximum similarity as the first area similarity of the first grayscale area; S450: continuing to perform random area selection and traversal similarity comparison in the grayscale image of the pipe welding area, and obtaining multiple grayscale areas and multiple area similarities; S460: setting the grayscale area whose area similarity is less than the similarity threshold as the height difference area, and calculating the area ratio of the height difference area to the area of ​​the grayscale image of the pipe welding area to obtain the deviation area ratio, and determining the second potential performance impact coefficient according to the deviation area ratio.

[0050] Specifically, a standard pipe welding image of a target Y-shaped tee pipe fitting (typically generated from process standards or empirical data) is obtained. The weld area is then cropped from the standard pipe welding image and the pipe surface image, respectively. The weld area typically refers to a specific portion of the pipe weld, such as the weld seam. An edge detection algorithm (such as Canny edge detection) is used to determine the location of the weld area. The image is cropped based on the detected edge information to obtain an image containing only the weld area, resulting in a standard weld area image and a pipe welding area image. The standard weld area image and the pipe welding area image are further grayscale processed, converting the color image to grayscale using an RGB to grayscale formula. The grayscale conversion is then performed to obtain a standard weld area grayscale image and a pipe welding area grayscale image. A predetermined size (5 by 5 pixel area) is then obtained. Specifically, a 5 by 5 pixel sub-region is selected from the pipe welding area grayscale image, and a first grayscale region of the predetermined size is randomly selected from the pipe welding area grayscale image.

[0051] Next, a traversal similarity comparison is performed between the first grayscale region and the regions of predetermined size within the standard weld region grayscale image. Specifically, a similarity comparison is performed between the first grayscale region (a 5x5 region randomly selected from the pipe weld region grayscale image) and each region of predetermined size within the standard weld region grayscale image. The similarity between the first grayscale region and each traversed region is evaluated using, for example, structural similarity (SSIM). The entire standard weld region grayscale image is traversed, the similarity between the first grayscale region and each 5x5 region is calculated, and the maximum similarity is selected as the first region similarity. Multiple 5x5 pixel regions are then randomly selected from the pipe weld region grayscale image, with regions that do not overlap with previous regions being selected to ensure diversity in the traversal process. The traversal similarity comparison step described in S440 is repeated for each newly selected grayscale region, performing a similarity comparison between each randomly selected region and a region in the standard weld region grayscale image, calculating the similarity, and obtaining multiple grayscale regions and multiple region similarities.

[0052] A similarity threshold is set, typically based on experimental or empirical values. This threshold is used to determine the similarity between the grayscale region and the standard weld region. Grayscale regions with a region similarity less than the similarity threshold are then defined as high-difference regions. If the region similarity is less than the similarity threshold, the region is considered a high-difference region, indicating a significant difference from the standard weld region, possibly due to a welding defect, error, or non-compliant region. The ratio of the high-difference region area (the sum of the areas of multiple high-difference regions) to the grayscale image of the pipe weld region is further calculated to obtain a deviation area ratio. Finally, a second potential performance impact coefficient is determined based on the deviation area ratio. The deviation area ratio and the second potential performance impact coefficient are positively correlated, meaning that a larger deviation area ratio corresponds to a larger second potential performance impact coefficient, indicating a greater impact on the pipe's load-bearing performance.

[0053] By performing a forward similarity comparison between the pipe surface image and the standard pipe welding image, the second potential performance impact coefficient is obtained, which can be used to evaluate the load-bearing performance of the pipe from another perspective, thereby further improving the accuracy and reliability of the pipe damage strength analysis.

[0054] S500: Determine the pipe damage intensity based on the first potential performance impact coefficient and the second potential performance impact coefficient. If the pipe damage intensity is greater than a risk intensity threshold, mark the target Y-type tee pipe for inspection.

[0055] Furthermore, step S500 of this application also includes:

[0056] S510: Configure a first influence weight and a second influence weight, wherein the first influence weight is 0.85 and the second influence weight is 0.15; S520: Based on the first influence weight and the second influence weight, perform weighted calculation on the first potential performance influence coefficient and the second potential performance influence coefficient to obtain the pipe damage intensity.

[0057] Specifically, first, a first impact weight and a second impact weight are configured, where the first impact weight is 0.85 and the second impact weight is 0.15. Then, based on the first and second impact weights, a weighted calculation is performed on the first and second potential performance impact coefficients, and the weighted calculation result is set as the pipe damage intensity. By comprehensively considering the first and second potential performance impact coefficients through weighted calculation, the pipe damage intensity can be comprehensively assessed from multiple perspectives, effectively improving the accuracy and reliability of pipe damage assessment.

[0058] Obtain a risk intensity threshold. The risk intensity threshold is a preset standard value used to distinguish whether a pipe fitting is within an acceptable damage range. It is usually set based on the working environment, design standards, and safety requirements of the pipe fitting. The damage intensity of the pipe fitting is judged based on the risk intensity threshold. If the damage intensity of the pipe fitting is greater than the risk intensity threshold, it indicates that the degree of damage to the pipe fitting has exceeded the predetermined safety standard and may affect the bearing capacity or stability of the pipe fitting. The target Y-type tee pipe fitting is then marked for maintenance. If the damage intensity of the pipe fitting is less than or equal to the risk intensity threshold, it indicates that the damage degree of the pipe fitting is acceptable and no maintenance is required.

[0059] In summary, the welding quality inspection method for LNG marine pipe fittings provided in this application has the following technical effects:

[0060] The target Y-shaped tee pipe fitting is subjected to radiographic scanning and surface image acquisition. Internal and surface defect identification are performed based on the pipe fitting radiographic image and surface image, respectively, to obtain internal and surface defect feature distributions. Spatial fitting and maximum value retention are then performed on the internal and surface defect feature distributions to model and generate a pipe fitting defect feature space. The weld damage intensity is then determined based on the pipe fitting defect feature space analysis. A load-bearing performance evaluation channel is then invoked based on the weld damage intensity to perform a performance impact analysis and output a first potential performance impact coefficient. Furthermore, a similarity traversal comparison is performed on the pipe fitting surface image using a standard pipe fitting weld image as a reference, and a second potential performance impact coefficient is determined based on the deviation area ratio. Finally, the pipe fitting damage intensity is assessed based on the first and second potential performance impact coefficients. If the pipe fitting damage intensity exceeds a risk intensity threshold, the target Y-shaped tee pipe fitting is marked for inspection. This method can efficiently and accurately identify defects on the surface and inside of pipe fittings, significantly improving the accuracy and reliability of pipe fitting damage assessment, thereby enabling early detection of potential risks and ensuring the safety and reliability of pipe fitting use.

[0061] In the second embodiment, based on the welding quality inspection method of a LNG ship pipe fitting in the above embodiment, the present application also provides a welding quality inspection platform for LNG ship pipe fittings, please refer to the attached Figure 3 ,include:

[0062] A defect feature distribution acquisition module 11 is configured to perform radiographic scanning and surface image acquisition on a target Y-type tee pipe fitting, perform internal defect identification and surface defect identification based on the pipe fitting radiographic image and pipe fitting surface image, respectively, to obtain an internal defect feature distribution and a surface defect feature distribution. A defect feature space generation module 12 is configured to perform spatial fitting and maximum value retention processing on the internal defect feature distribution and the surface defect feature distribution to model and generate a pipe fitting defect feature space. A first performance impact coefficient acquisition module 13 is configured to determine the welding damage intensity based on the pipe fitting defect feature space analysis, call a load-bearing performance evaluation channel based on the welding damage intensity to perform performance impact analysis, and output a first potential performance impact coefficient. A second performance impact coefficient acquisition module 14 is configured to perform a similarity traversal comparison on the pipe fitting surface image based on a standard pipe fitting welding image of the target Y-type tee pipe fitting, and determine a second potential performance impact coefficient based on the deviation area ratio. A pipe fitting damage intensity determination module 15 is configured to evaluate and determine the pipe fitting damage intensity based on the first potential performance impact coefficient and the second potential performance impact coefficient. If the pipe fitting damage intensity is greater than a risk intensity threshold, the target Y-type tee pipe fitting is marked for inspection and repair.

[0063] Furthermore, the welding quality inspection platform for LNG marine pipe fittings is also used to: construct an internal defect recognition model and a surface defect recognition model based on a multi-scale convolutional neural network for the purpose of identifying predetermined defect features, wherein the predetermined defect features include defect type and size information, and the defect types include at least cracks, pores and slag inclusions; use the internal defect recognition model to perform internal defect recognition on the pipe fitting radiographic image and output the internal defect feature distribution; use the surface defect recognition model to perform surface defect recognition on the pipe fitting surface image and output the surface defect feature distribution.

[0064] Furthermore, the welding quality inspection platform for LNG marine pipe fittings is also used to: perform transverse defect feature analysis and longitudinal defect feature analysis based on the pipe fitting defect feature space, and determine the transverse defect area ratio, transverse defect distribution discreteness and longitudinal defect depth; obtain the material properties and working environment characteristics of the target Y-type tee pipe fitting, perform corrosion intensity simulation analysis based on the material properties and working environment characteristics, and determine the corrosion intensity coefficient; and evaluate and determine the welding damage intensity based on the transverse defect area ratio, transverse defect distribution discreteness, longitudinal defect depth and corrosion intensity coefficient.

[0065] Furthermore, the welding quality inspection platform for LNG marine pipe fittings is also used to: pre-train a load-bearing performance evaluation channel, wherein the load-bearing performance evaluation channel includes P basis evaluators; obtain the maximum welding damage intensity within a predetermined historical time zone, and determine the number of calls of the basis evaluators of the load-bearing performance evaluation channel based on the welding damage intensity and the maximum welding damage intensity evaluation, which is set to Q; randomly activate Q selected basis evaluators among the P basis evaluators, use the Q selected basis evaluators to perform performance impact analysis on the pipe fitting defect feature space, output Q performance impact coefficients, and obtain the first potential performance impact coefficient after average calculation.

[0066] Furthermore, the welding quality inspection platform for LNG marine pipe fittings is also used to: collect a three-dimensional feature distribution set of sample pipe fitting defects based on the historical maintenance log of the target Y-type tee pipe fitting, and statistically analyze the proportion of pipe fitting load-bearing failures within a predetermined time window within the three-dimensional feature distribution of different sample pipe fitting defects, obtain the sample pipe fitting load-bearing failure proportion and set it as the sample performance influence coefficient to obtain the sample performance influence coefficient set; integrate the sample pipe fitting defect three-dimensional feature distribution set and the sample performance influence coefficient set to obtain a sample training set, and divide it into P equal parts to obtain P training arrays; use the P training arrays to train the generator and discriminator of the generative adversarial network until P basis evaluators are obtained after convergence, and integrate to obtain the load-bearing performance evaluation channel.

[0067] Furthermore, the welding quality inspection platform for LNG ship pipe fittings is also used to: perform welding area cropping on the standard pipe fitting welding image and the pipe fitting surface image respectively to obtain a standard welding area image and a pipe fitting welding area image; perform grayscale processing on the standard welding area image and the pipe fitting welding area image to obtain a standard welding area grayscale image and a pipe fitting welding area grayscale image; randomly select a first grayscale area in the pipe fitting welding area grayscale image according to a predetermined size, wherein the predetermined size is a 5 by 5 pixel area; perform traversal similarity comparison on the first grayscale area and the area of ​​the predetermined size in the standard welding area grayscale image, and set the maximum similarity as the first area similarity of the first grayscale area; continue to perform random area selection and traversal similarity comparison on the pipe fitting welding area grayscale image to obtain multiple grayscale areas and multiple area similarities; set the grayscale area whose area similarity is less than the similarity threshold as a height difference area, and calculate the area ratio of the height difference area to the area of ​​the pipe fitting welding area grayscale image to obtain the deviation area ratio, and determine the second potential performance impact coefficient based on the deviation area ratio.

[0068] Furthermore, the welding quality inspection platform for LNG ship pipe fittings is also used to: configure a first influence weight and a second influence weight, wherein the first influence weight is 0.85 and the second influence weight is 0.15; based on the first influence weight and the second influence weight, perform a weighted calculation on the first potential performance influence coefficient and the second potential performance influence coefficient to obtain the damage strength of the pipe fitting.

[0069] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The welding quality inspection method and specific examples of a LNG ship pipe fitting in the aforementioned embodiment 1 are also applicable to the welding quality inspection platform of a LNG ship pipe fitting in this embodiment. Through the aforementioned detailed description of the welding quality inspection method of a LNG ship pipe fitting, those skilled in the art can clearly understand the welding quality inspection platform of a LNG ship pipe fitting in this embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here. As for the platform disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For relevant details, please refer to the description of the method section.

[0070] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0071] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. The welding quality inspection method of LNG marine pipe fittings is characterized by: Methods include: Perform radiographic scanning and surface image acquisition on the target Y-shaped tee pipe fitting, perform internal defect recognition and surface defect recognition based on the pipe fitting radiographic image and pipe fitting surface image, and obtain the internal defect feature distribution and surface defect feature distribution; Performing spatial fitting and maximum value retention processing on the internal defect characteristic distribution and the surface defect characteristic distribution to generate a pipe defect characteristic space by modeling; Determining the welding damage intensity based on the pipe defect feature space analysis, and calling the load-bearing performance evaluation channel to perform performance impact analysis according to the welding damage intensity, and outputting a first potential performance impact coefficient; Taking the standard pipe welding image of the target Y-type tee pipe fitting as a reference, performing similarity traversal comparison on the pipe fitting surface image, and determining the second potential performance impact coefficient according to the deviation area ratio; Determine the pipe fitting damage intensity based on the first potential performance impact coefficient and the second potential performance impact coefficient, and if the pipe fitting damage intensity is greater than a risk intensity threshold, mark the target Y-type tee pipe fitting for inspection and maintenance; Determining the welding damage intensity based on the pipe defect feature space analysis includes: Based on the pipe defect feature space, performing transverse defect feature analysis and longitudinal defect feature analysis respectively to determine the transverse defect area ratio, transverse defect distribution dispersion and longitudinal defect depth; Obtaining material properties and working environment characteristics of the target Y-type tee pipe fitting, performing a corrosion intensity simulation analysis based on the material properties and working environment characteristics, and determining a corrosion intensity coefficient; Determining the welding damage intensity based on the transverse defect area ratio, transverse defect distribution dispersion, longitudinal defect depth and corrosion intensity coefficient; The load-bearing performance evaluation channel is called according to the welding damage intensity to perform a performance impact analysis, and a first potential performance impact coefficient is output, including: A pre-trained bearer performance evaluation channel, wherein the bearer performance evaluation channel includes P basis evaluators; Obtaining the maximum welding damage intensity within a predetermined historical time zone, and determining the number of calls of the base evaluator of the load-bearing performance evaluation channel according to the welding damage intensity and the maximum welding damage intensity, which is set as Q; Randomly activate Q selected basis evaluators among the P basis evaluators, use the Q selected basis evaluators to perform performance impact analysis on the pipe defect feature space, output Q performance impact coefficients, and obtain the first potential performance impact coefficient after average calculation.

2. The welding quality inspection method for LNG marine pipe fittings according to claim 1, characterized in that: Internal defects and surface defects are identified based on the pipe X-ray image and pipe surface image, respectively, to obtain the internal defect feature distribution and surface defect feature distribution, including: Based on a multi-scale convolutional neural network, an internal defect recognition model and a surface defect recognition model are constructed for the purpose of identifying predetermined defect features, wherein the predetermined defect features include defect type and size information, and the defect types include at least cracks, pores, and slag inclusions; Performing internal defect recognition on the pipe radiographic image using the internal defect recognition model, and outputting the internal defect feature distribution; Surface defect recognition is performed on the pipe surface image using the surface defect recognition model, and the surface defect feature distribution is output.

3. The welding quality inspection method for LNG marine pipe fittings according to claim 1, characterized in that: Pre-training load performance evaluation channel, including: Based on the historical maintenance logs of the target Y-shaped tee pipe fitting, a three-dimensional feature distribution set of sample pipe fitting defects is collected. The proportion of pipe fitting load-bearing failures within a predetermined time window due to the three-dimensional feature distribution of defects of different sample pipe fittings is statistically analyzed. The proportion of sample pipe fitting load-bearing failures is set as the sample performance impact coefficient, and a sample performance impact coefficient set is obtained. Integrating the three-dimensional characteristic distribution set of sample pipe defects and the sample performance influence coefficient set to obtain a sample training set, and dividing it into P equal parts to obtain P training arrays; The generator and discriminator of the generative adversarial network are trained using the P training arrays until convergence to obtain P basis evaluators, which are integrated to obtain the bearing performance evaluation channel.

4. The welding quality inspection method for LNG marine pipe fittings according to claim 1, characterized in that: Based on the standard pipe welding image of the target Y-type tee pipe fitting, a similarity traversal comparison is performed on the pipe fitting surface image, and the second potential performance impact coefficient is determined according to the deviation area ratio, including: Performing welding area cropping on the standard pipe fitting welding image and the pipe fitting surface image respectively to obtain a standard welding area image and a pipe fitting welding area image; Performing grayscale processing on the standard welding area image and the pipe welding area image to obtain a standard welding area grayscale image and a pipe welding area grayscale image; Randomly selecting a first grayscale area in the grayscale image of the pipe welding area according to a predetermined size, wherein the predetermined size is a 5 by 5 pixel area; Performing a traversal similarity comparison between the first grayscale area and an area of ​​a predetermined size in the standard welding area grayscale image, and setting the maximum similarity as the first area similarity of the first grayscale area; Continue to randomly select regions and perform traversal similarity comparison in the grayscale image of the pipe welding area to obtain multiple grayscale regions and multiple region similarities; The grayscale area with a regional similarity less than a similarity threshold is set as a high-difference area, and the area ratio of the high-difference area to the grayscale image of the pipe welding area is calculated to obtain the deviation area ratio, and the second potential performance impact coefficient is determined based on the deviation area ratio.

5. The welding quality inspection method for LNG marine pipe fittings according to claim 1, characterized in that: Evaluating and determining the damage intensity of the pipe fitting according to the first potential performance impact coefficient and the second potential performance impact coefficient includes: Configure a first influence weight and a second influence weight, wherein the first influence weight is 0.85 and the second influence weight is 0.15; Based on the first influence weight and the second influence weight, a weighted calculation is performed on the first potential performance influence coefficient and the second potential performance influence coefficient to obtain the pipe damage intensity.

6. LNG marine pipe welding quality inspection platform, characterized by: The steps for implementing the welding quality inspection method of LNG marine pipe fittings according to any one of claims 1 to 5 include: The defect feature distribution acquisition module is used to perform radiographic scanning and surface image acquisition on the target Y-type tee pipe fitting, perform internal defect recognition and surface defect recognition based on the pipe fitting radiographic image and pipe fitting surface image, and obtain the internal defect feature distribution and surface defect feature distribution; A defect feature space generation module is used to perform spatial fitting and maximum value retention processing on the internal defect feature distribution and the surface defect feature distribution, and to model and generate a pipe defect feature space; a first performance impact coefficient obtaining module, configured to determine the welding damage intensity based on the pipe defect feature space analysis, and call the load-bearing performance evaluation channel to perform performance impact analysis based on the welding damage intensity, and output a first potential performance impact coefficient; A second performance impact coefficient obtaining module is used to perform a similarity traversal comparison on the surface image of the target Y-shaped tee pipe fitting based on the standard pipe welding image of the target pipe fitting, and determine the second potential performance impact coefficient according to the deviation area ratio; The pipe fitting damage intensity determination module is used to evaluate and determine the pipe fitting damage intensity based on the first potential performance impact coefficient and the second potential performance impact coefficient. If the pipe fitting damage intensity is greater than the risk intensity threshold, the target Y-type tee pipe fitting is marked for inspection and maintenance.

Images