Welding defect detection method and system for chemical metal pipeline

Abstract

The invention relates to the technical field of welding defect detection, in particular to a welding defect detection method and system for a chemical metal pipeline. The method comprises the following steps: acquiring a pipeline welding seam scanning image and an ultrasonic data value of the bimetal composite steel pipe at an initial stage; screening suspected crack regions according to shape distribution of different sub-regions in the pipeline welding seam scanning image at each moment in the initial stage; screening a crack trend time period according to the change degree of the ultrasonic data values at adjacent moments in the initial stage; according to the time interval between the crack trend time period and the current moment and the difference of the ultrasonic data values, state change necessary indexes are determined, and the ultrasonic probes to be adjusted are screened; according to the distance between each ultrasonic probe to be adjusted and the suspected crack area and the state change necessary indexes, the wheel rotating speed of the ultrasonic damage trolley is adjusted, and defect detection is conducted on the suspected crack area. The accuracy of the welding defect detection result of the bimetal composite steel pipe is improved.

Description

Technical Field

[0001] This invention relates to the field of welding defect detection technology, and specifically to a method and system for detecting welding defects in chemical metal pipelines. Background Technology

[0002] In chemical applications, the highly corrosive nature of the media transported in pipelines can cause significant damage to conventional pipelines, necessitating the use of specially structured pipes. Bimetallic composite pipes typically consist of two layers: an outer layer, usually made of carbon steel or low-alloy steel, and an inner layer, often made of stainless steel or nickel-based alloys. This enhances the pipe's corrosion and wear resistance, reducing the risk of media leakage or equipment damage during transport. However, during the welding process, bimetallic composite pipes are susceptible to hot cracking defects due to the combined effects of the material's metallurgical properties (such as the enrichment of low-melting-point eutectics) and welding stress. This is primarily caused by the thermal cycling effect on the heat-affected zone, which melts low-melting-point impurities, leading to cracking mainly in the coarse-grained region near the fusion line under welding stress. Therefore, welding defect detection of the bimetallic composite steel pipe is necessary.

[0003] To prevent secondary damage during the defect inspection of bimetallic composite steel pipes, ultrasonic testing is generally used. Existing methods involve installing ultrasonic probes on an ultrasonic testing carriage, which moves with the carriage to detect welding defects throughout the entire bimetallic composite steel pipe. However, in actual defect inspection, factors such as equipment and environment can affect the results, leading to discrepancies between the inspection results and the actual situation. Furthermore, it is impossible to focus on areas with suspected weld defects, resulting in low accuracy of the weld defect detection results for bimetallic composite steel pipes. Summary of the Invention

[0004] To address the issue of low accuracy in welding defect detection of bimetallic composite steel pipes using existing methods, this invention aims to provide a method and system for welding defect detection in chemical metal pipelines. The specific technical solution adopted is as follows: In a first aspect, the present invention provides a method for detecting welding defects in chemical metal pipelines, the method comprising the following steps: To acquire the pipe weld scanning image and ultrasonic data value of the bimetallic composite steel pipe in the initial stage of defect detection using an ultrasonic damage carriage; Based on the shape distribution of different sub-regions in the pipe weld scan images at each moment in the initial stage, suspected crack areas are screened; based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage, crack trend time periods are screened. Based on the time interval between the crack trend period and the current time, as well as the difference in ultrasonic data values, the necessary indicators for the state change of each ultrasonic probe are determined, and ultrasonic probes to be adjusted are selected; based on the distance between each ultrasonic probe to be adjusted and the suspected crack area and the necessary indicators for the state change, the sliding rate of each ultrasonic probe to be adjusted is obtained. The wheel speed of the ultrasonic damage trolley is adjusted based on the sliding rate to detect defects in suspected crack areas.

[0005] Preferably, the step of filtering suspected crack areas based on the shape distribution of different sub-regions in the pipe weld scan image at each moment in the initial stage includes: For any moment in the initial stage: extract the central region and two edge regions from the pipe weld scan image at that moment, and take both the central region and the edge regions as sub-regions; perform curve fitting on each sub-region, and take the average curvature of each sub-region as the first probability value of each sub-region; Based on the first probability value of each sub-region at consecutive moments in the initial stage, suspected crack regions are screened.

[0006] Preferably, based on the first probability value of each sub-region at consecutive moments in the initial stage, suspected crack regions are screened, including: The sub-region with the first probability greater than a preset first threshold at each time is taken as the target region at each time. The area formed by the continuous target region on the bimetallic composite steel pipe in the initial stage is regarded as the suspected crack region.

[0007] Preferably, the step of filtering the crack trend time period based on the degree of change between adjacent ultrasonic data values ​​in the initial stage includes: The moment when the change in ultrasonic data value in the initial stage exceeds the preset change threshold is defined as the moment of crack trend; wherein, the change in ultrasonic data value at each moment in the initial stage is the absolute value of the difference between the ultrasonic data value at each moment in the initial stage and the previous moment. A continuous crack trend moment constitutes a crack trend time period.

[0008] Preferably, the step of determining the necessary indicators for the state change of each ultrasonic probe based on the time interval between the crack trend period and the current moment and the difference in ultrasonic data values ​​includes: The time interval between the current moment and the nearest crack trend time interval is denoted as the first time interval; For any ultrasonic probe, the difference between the ultrasonic data value at the current time and the last time in the crack trend time period closest to the current time is denoted as the first difference; the normalized result of the product of the first time interval and the first difference is determined as the necessary index for the state change of any ultrasonic probe.

[0009] Preferably, the ultrasound probes to be adjusted include: Ultrasonic probes whose state change indicators exceed the preset necessary thresholds are identified as ultrasonic probes to be adjusted.

[0010] Preferably, obtaining the sliding rate of each ultrasonic probe to be adjusted based on the distance between each probe to be adjusted and the suspected crack area and the necessary indicators of the state change includes: For any ultrasound probe to be adjusted: Calculate the normalized value of the product between the shortest distance between any of the ultrasonic probes to be adjusted and the suspected crack area and the necessary index for the state change of any of the ultrasonic probes to be adjusted. Determine whether any of the ultrasound probes to be adjusted is sliding towards the center region. If yes, calculate the first difference between the constant 1 and the normalized value, and determine the sliding speed of the ultrasound probe by multiplying the first difference with the current motion rate of the ultrasound probe to be adjusted. If no, calculate the first sum of the constant 1 and the normalized value, and determine the sliding speed of the ultrasound probe to be adjusted by multiplying the first sum with the current motion rate of the ultrasound probe to be adjusted.

[0011] Preferably, adjusting the wheel speed of the ultrasonic damage cart based on the sliding rate includes: The total number of ultrasonic probes installed on the ultrasonic damage cart is 2; If the sliding speed of two ultrasonic probes to be adjusted is not equal to the current motion speed, then the minimum sliding speed of the two ultrasonic probes to be adjusted is obtained, and the first normalized result of the difference between the current motion speed and the minimum value is calculated; the product of the current wheel speed of the ultrasonic damage cart and the first normalized result is rounded up and determined as the new wheel speed of the ultrasonic damage cart, and the wheel speed of the ultrasonic damage cart is adjusted. If the sliding speed of an ultrasonic probe to be adjusted is not equal to the current motion speed, then the sliding speed of the ultrasonic probe to be adjusted is taken as the new wheel speed of the ultrasonic damage cart, and the wheel speed of the ultrasonic damage cart is adjusted.

[0012] Secondly, the present invention provides a welding defect detection system for chemical metal pipelines, the system comprising: The data acquisition module is used to acquire the pipe weld scanning image and ultrasonic data value of the bimetallic composite steel pipe in the initial stage of the defect detection process of the bimetallic composite steel pipe using an ultrasonic damage carriage. The filtering module is used to filter suspected crack areas based on the shape distribution of different sub-regions in the pipe weld scan image at each moment in the initial stage; and to filter crack trend time periods based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage. The calculation module is used to determine the necessary indicators for the state change of each ultrasonic probe based on the time interval between the crack trend period and the current time and the difference in ultrasonic data values, and to screen the ultrasonic probes to be adjusted; based on the distance between each ultrasonic probe to be adjusted and the suspected crack area and the necessary indicators for the state change, the sliding rate of each ultrasonic probe to be adjusted is obtained. The defect detection module is used to adjust the wheel speed of the ultrasonic damage cart based on the sliding rate to detect defects in suspected crack areas.

[0013] Preferably, the data acquisition module includes an ultrasonic probe, an ultrasonic probe control component, and a line scanning component; and the wheel control component.

[0014] The present invention has at least the following beneficial effects: 1. This invention first combines the characteristics of the pipe weld of bimetallic composite steel pipe, and selects suspected crack areas based on the shape distribution of different sub-regions in the pipe weld scanning image of bimetallic composite steel pipe at each moment in the initial stage of the welding defect detection process. It then selects crack trend time periods based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage, and selects ultrasonic probes to be adjusted based on the time interval between the crack trend time periods and the current moment and the difference in ultrasonic data values. Finally, it adjusts the wheel speed of the ultrasonic damage cart based on the distance between the ultrasonic probe to be adjusted and the suspected crack area, so as to achieve the purpose of focusing on the detection of suspected crack areas and improving the accuracy of the weld defect detection results of bimetallic composite steel pipe. 2. The welding defect detection system for chemical metal pipelines provided by this invention is equipped with an automated ultrasonic flaw detection trolley. The trolley automatically and intelligently slides on the pipeline to automatically identify possible cracks and defects on the weld surface. While ensuring the accuracy of welding defect detection results, it greatly improves the defect detection efficiency of bimetallic composite steel pipes. Attached Figure Description

[0015] 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.

[0016] Figure 1 A flowchart illustrating a welding defect detection method for chemical metal pipelines provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a sub-region division result provided in an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, describes a welding defect detection method and system for chemical metal pipelines based on the present invention.

[0018] 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.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for a welding defect detection method and system for chemical metal pipelines provided by the present invention.

[0020] An embodiment of a welding defect detection system for chemical metal pipelines: The welding defect detection system for chemical metal pipelines mainly includes a data acquisition module, a screening module, a calculation module, and a defect detection module.

[0021] The data acquisition module is used to acquire the pipe weld scanning image and ultrasonic data value of the bimetallic composite steel pipe in the initial stage of the defect detection process of the bimetallic composite steel pipe using an ultrasonic damage carriage. The filtering module is used to filter suspected crack areas based on the shape distribution of different sub-regions in the pipe weld scan image at each moment in the initial stage; and to filter crack trend time periods based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage. The calculation module is used to determine the necessary indicators for the state change of each ultrasonic probe based on the time interval between the crack trend period and the current time and the difference in ultrasonic data values, and to screen the ultrasonic probes to be adjusted; based on the distance between each ultrasonic probe to be adjusted and the suspected crack area and the necessary indicators for the state change, the sliding rate of each ultrasonic probe to be adjusted is obtained. The defect detection module is used to adjust the wheel speed of the ultrasonic damage cart based on the sliding rate to detect defects in suspected crack areas.

[0022] The screening module of the welding defect detection system for chemical metal pipelines analyzes and processes the collected pipeline weld scan images and ultrasonic data to screen out suspected crack areas and crack trend time periods, and transmits the screening results to the calculation module of the bimetallic composite steel pipe welding defect system.

[0023] The calculation module of the bimetallic composite steel pipe welding defect system uses the screening results from the combined screening module to determine the ultrasonic probes to be adjusted and the sliding rate of each ultrasonic probe to be adjusted. The calculation module then transmits the calculation results to the defect detection module of the bimetallic composite steel pipe welding defect system.

[0024] After the ultrasonic flaw detection carriage stops, the pipe weld scan images at each moment are stitched together from the cloud database to obtain a complete pipe weld scan image containing defect markers, thus completing the welding defect detection of the bimetallic composite steel pipe. The defect markers include depth indicators corresponding to the ultrasonic data.

[0025] An embodiment of a welding defect detection method for chemical metal pipelines: This embodiment proposes a method for detecting welding defects in chemical metal pipelines, such as... Figure 1 As shown, a welding defect detection method for chemical metal pipelines according to this embodiment includes the following steps: Step S1: Obtain the pipe weld scanning image and ultrasonic data value of the bimetallic composite steel pipe in the initial stage of defect detection using an ultrasonic damage carriage.

[0026] Start the ultrasonic flaw detection cart and set the ultrasonic data acquisition frequency to the same frequency as the pipeline scanning image acquisition frequency. In specific applications, the implementer sets the image and ultrasonic data acquisition frequencies according to the specific situation, and the preset speed is also set by the implementer according to the specific situation, which will not be elaborated further here. This embodiment acquires pipeline scanning images and ultrasonic data values ​​of the bimetallic composite steel pipe at each moment in the initial stage. It should be noted that the unit of the ultrasonic data values ​​acquired in this embodiment is millivolts. The initial stage is a set of all historical moments whose time interval between the current moment is less than or equal to the preset duration. Considering that when using an ultrasonic flaw detection carriage to inspect bimetallic composite steel pipes for defects, it is necessary to adjust the sliding speed of the ultrasonic probe in a timely manner based on the initial inspection results, so that the ultrasonic flaw detection carriage can focus on inspecting areas suspected of weld defects, setting the initial stage too long will result in excessive time required to detect actual crack areas, leading to low inspection efficiency. Setting the initial stage too short will result in insufficient data for reference inspection results, and the basis for setting the speed of the ultrasonic flaw detection carriage will not be sufficient, thus affecting the accuracy of subsequent defect inspection results of bimetallic composite steel pipes. The initial stage can be set to 10% to 20% of the total time required for the ultrasonic flaw detection carriage to perform a complete inspection of the bimetallic composite steel pipe. That is, the preset time is 10% to 20% of the total time required for the ultrasonic flaw detection carriage to perform a complete inspection of the bimetallic composite steel pipe. This time is neither too long nor too short, ensuring both overall inspection efficiency and sufficient data for reference inspection results. The total time required for the ultrasonic damage carriage to perform a complete inspection of a bimetallic composite steel pipe can be calculated by averaging the inspection times of multiple inspections of the bimetallic composite steel pipe in historical inspection processes. These historical inspections include defect-free bimetallic composite steel pipes and bimetallic composite steel pipes with varying degrees of crack defects. For example, if the average inspection time of multiple inspections of bimetallic composite steel pipes in historical inspection processes is 10 minutes, and the preset time is set to 10% of the total time required for the ultrasonic damage carriage to perform a complete inspection of the bimetallic composite steel pipe, then the preset time is 1 minute.

[0027] For each pipe scan image, the weld region is extracted. Each pipe scan image containing the weld region is recorded as a single pipe weld scan image. The extraction of the weld region can be achieved using a semantic segmentation network. Semantic segmentation networks and their training processes are existing technologies and will not be elaborated upon here. Thus far, this embodiment has acquired pipe weld scan images and ultrasonic data values ​​of the bimetallic composite steel pipe at each moment in the initial stage.

[0028] Step S2: Based on the shape distribution of different sub-regions in the pipe weld scan image at each moment in the initial stage, screen for suspected crack areas; and based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage, screen for crack trend time periods.

[0029] Compared with liquefaction cracking defects, crystallization cracking defects in bimetallic composite steel pipes are mainly distributed in the center of the weld, while liquefaction cracking defects are mainly distributed at the edge of the weld. At the same time, since both crystallization cracking and liquefaction cracking are surface cracks that exist on the surface of the steel pipe weld, if crystallization cracking and liquefaction cracking exist in the pipeline weld scanning image at a single moment, there will be relatively obvious depressions on the upper edge lines of the corresponding central area and the two side edge areas.

[0030] After acquiring the pipeline weld scan images, each pipeline weld scan image is divided into sub-regions. For each pipeline weld scan image, the least squares method is used to fit a straight line to the weld region to obtain the corresponding fitted line segment. The two line segments formed by the pixels in the pipeline weld scan image that are parallel to the fitted line segment and are at a preset first distance from the fitted line segment are used as the boundary lines of the central region. The closed region formed by the two boundary lines of the central region and the edge lines of the pipeline weld scan image that intersect with it is the central region. The two regions of the pipeline weld scan image other than the central region are used as the two edge regions. Both the central region and the edge regions are used as sub-regions, that is, each pipeline weld scan image is divided into multiple sub-regions. As one implementation method, the preset first distance can be determined as follows: Analyze the scanned images of pipe welds of bimetallic composite steel pipes with confirmed crack defects of varying degrees from historically acquired images, statistically analyze the average width of the cracks in these images, and use half of the average width as the preset first distance. That is, twice the preset distance is approximately equal to the crack width. When a crack defect occurs, the central area is roughly the location of the crack defect. It should be noted that the specifications and materials of these bimetallic composite steel pipes with existing crack defects are the same as those of the bimetallic composite steel pipe to be inspected in this embodiment. This method of determining the preset first distance based on the statistical analysis results of crack areas in the scanned images of pipe welds of bimetallic composite steel pipes with confirmed crack defects of varying degrees from historically acquired images allows the central area divided based on the first distance to better adapt to different degrees of crack defects, improving the accuracy of the crack area division range. As another implementation method, the preset first distance can also be set to an empirical value of 50 pixels. In specific applications, the implementer can choose the setting method of the preset first distance according to the specific situation. Figure 2 As shown in the figure, a is the fitted line segment, a1 and a2 are the boundary lines of the central region, A represents the central region, and B and C represent the edge regions.

[0031] Normally, crystallization cracks and liquefaction cracks are distributed as cracks of a certain length and do not exist as single points. Therefore, if these two types of cracks exist on the weld of a steel pipe, the line scanner will continuously detect the abnormality in the edge area at certain times during scanning.

[0032] For any moment in the initial stage: curve fitting is performed on each sub-region in the pipeline weld scan image at that moment, and the average value of the curvature corresponding to each sub-region is taken as the first probability value of each sub-region. Each sub-region has a corresponding first probability value.

[0033] The sub-regions with a first probability greater than a preset first threshold at each time moment are designated as target regions at each time moment. The regions formed by consecutive target regions on the bimetallic composite steel pipe in the initial stage are designated as suspected crack regions. As one implementation method, the preset first threshold can be determined by analyzing multiple bimetallic composite steel pipes with confirmed crack defects of varying degrees from historical inspection results. Specifically, a set of pipe weld scan images of these confirmed crack-defect bimetallic composite steel pipes is collected, the first probability value of the crack region in these pipe weld scan images is calculated, and then the distribution characteristics of the first probability value are statistically analyzed. Based on these distribution characteristics, the preset first threshold is set as the percentile value that can effectively screen out crack regions, such as the 70th percentile. This method of determining the preset first threshold based on the statistical analysis results of pipe weld scan images of bimetallic composite steel pipes with confirmed crack defects in history allows the preset first threshold to better adapt to crack defects of different degrees, improving the accuracy of suspected crack region screening results. As another implementation method, the preset first threshold can also be directly set to an empirical value of 0.7. In specific applications, the implementer can choose the setting method of the preset first threshold according to the specific situation.

[0034] Thus, using the above method, this embodiment has identified suspected crack areas.

[0035] When inspecting steel pipes, the ultrasonic probe moves left and right along both sides of the weld seam, emitting ultrasonic waves inward at a certain angle to detect cracks in the weld seam. When the ultrasonic probe does not detect any suspected cracks, its movement speed remains constant to ensure detection efficiency. However, once the ultrasonic probe begins to detect a suspected crack area, the computing chip in the corresponding ultrasonic probe control component starts transmitting the suspected crack area data from the weld seam scanning module. The image data and the ultrasound data acquired by the ultrasound probe are intelligently analyzed and calculated to control the ultrasound probe in real time.

[0036] When using an ultrasonic probe to inspect steel pipes, the ultrasonic data collected after reflection will fluctuate within a certain range due to the different material distribution in different parts of the steel pipe, resulting in many fine burrs in the waveform. When contacting cracks, the intensity of ultrasonic wave reflection increases, causing the local waveform to be greatly amplified and forming obvious high peaks.

[0037] Based on the above characteristics, for any moment in the initial stage other than the first moment: the absolute value of the difference between the ultrasonic data value at that moment and the previous moment is taken as the change value of the ultrasonic data value at that moment. Using this method, the change value of the ultrasonic data value at each moment in the initial stage other than the first moment can be obtained. The moment when the change value of the ultrasonic data value in the initial stage is greater than the preset change threshold is determined as the crack trend moment. Using this method, multiple crack trend moments can be screened out. As one implementation method, the preset change threshold can be determined in the following way: analyze multiple bimetallic composite steel pipes that have been confirmed to have crack defects in historical detection, use an ultrasonic flaw detection carriage to detect these bimetallic composite steel pipes, use the method provided in this embodiment to determine the change value of the ultrasonic data value at each moment, count the change value of the ultrasonic data value when the ultrasonic carriage detects cracked areas and non-cracked areas, and set the preset change threshold at the critical level that can effectively distinguish between cracked areas and non-cracked areas. As another implementation method, the preset change threshold can also be set to an empirical value of 100 millivolts. In specific applications, the implementer can set it according to the specific situation.

[0038] In this embodiment, after selecting multiple crack trend moments, the time period consisting of consecutive crack trend moments is recorded as the crack trend time period, that is, one or more crack trend time periods are obtained.

[0039] Step S3: Based on the time interval between the crack trend period and the current time and the difference in ultrasonic data values, determine the necessary indicators for the state change of each ultrasonic probe and screen the ultrasonic probes to be adjusted; based on the distance between each ultrasonic probe to be adjusted and the suspected crack area and the necessary indicators for the state change, obtain the sliding rate of each ultrasonic probe to be adjusted.

[0040] In this embodiment, the crack trend time period is determined in step S2. Next, based on the time interval between the crack trend time period and the current time and the difference in ultrasonic data values, the necessity of changing the sliding state of each ultrasonic probe is analyzed, ultrasonic probes to be adjusted are selected, and the sliding rate of each ultrasonic probe to be adjusted is determined.

[0041] Specifically, the time interval between the current moment and the crack trend time interval closest to the current moment is denoted as the first time interval. For any ultrasonic probe, the absolute value of the difference between the ultrasonic data value at the current moment and the ultrasonic data value at the last moment of the crack trend time interval closest to the current moment is denoted as the first difference. The normalized result of the product of the first time interval and the first difference is determined as the necessary index for the state change of the ultrasonic probe. There are many methods for normalizing the data; the maximum-minimum normalization method or other normalization methods can be used to ensure that the normalization result is within the range of [0, 1]. Using this method, the necessary index for the state change of each ultrasonic probe can be obtained. Ultrasonic probes with a necessary index for state change greater than a preset necessary threshold are identified as ultrasonic probes to be adjusted.

[0042] As one implementation method, the preset necessary threshold can be determined in the following way: using the ultrasonic flaw detection cart in this embodiment to detect multiple bimetallic composite steel pipes with different degrees of crack defects confirmed in historical test results, statistically analyzing the distribution characteristics of the necessary state change indicators when the ultrasonic probe on the ultrasonic flaw detection cart does not detect crack defects and when it detects crack defects, the preset necessary threshold is set at a critical level that can distinguish between the detected location as a normal area (area without cracks) and a cracked area, for example, it can be 0.6; as another implementation method, the preset necessary threshold can also be directly set to the empirical value of 0.6. In specific applications, the implementer can set it according to the specific situation.

[0043] The horizontal position of the ultrasonic probe on the pipe is not the same at different times, so the sliding distance required to move different ultrasonic probes to the suspected crack area is also different at different times.

[0044] For any ultrasonic probe to be adjusted: First, obtain the shortest distance between the probe and the suspected crack area. Then, calculate the normalized value of the product of the shortest distance and the necessary state change index of the probe. There are many methods for normalizing the data; the maximum-minimum normalization method or other normalization methods can be used, so that the normalized value of the product of the shortest distance and the necessary state change index is [0, 1]. Determine whether the ultrasonic probe is sliding towards the center region. If yes, calculate the difference between the constant 1 and the normalized value, and record this difference as the first difference. The product of the first difference and the current motion rate of the ultrasonic probe is determined as the sliding rate of the probe. If no, calculate the sum of the constant 1 and the normalized value, and record this sum as the first sum. The product of the first sum and the current motion rate of the ultrasonic probe is determined as the sliding rate of the probe. Using the above method, calculate the sliding rate of each ultrasonic probe to be adjusted.

[0045] Thus, this embodiment has obtained the sliding rate of each ultrasonic probe to be adjusted.

[0046] Step S4: Adjust the wheel speed of the ultrasonic damage trolley based on the sliding rate to perform defect detection on the suspected crack area.

[0047] As the ultrasonic flaw detection carriage moves forward continuously, ideally, when the ultrasonic probe begins to detect the cracked area of ​​the pipe weld, the carriage needs to slow down to allow the ultrasonic probe to perform a more detailed ultrasonic scan of the weld in order to make the ultrasonic probe detection more accurate.

[0048] If the number of ultrasonic probes to be adjusted is 2, and the sliding speed of both probes is not equal to the current motion speed, then the minimum sliding speed of the two probes is obtained. The absolute value of the difference between the current motion speed and this minimum value is calculated. This absolute value reflects the difference between the two. The normalized result of this absolute value is recorded as the first normalization result, which takes the value [0, 1]. The product of the current wheel speed of the ultrasonic damage cart and the first normalization result is rounded up and determined as the new wheel speed of the ultrasonic damage cart. The wheel speed of the ultrasonic damage cart is then adjusted. If the number of ultrasonic probes to be adjusted is 1, and the sliding speed of this probe is not equal to the current motion speed, then the sliding speed of this probe is used as the new wheel speed of the ultrasonic damage cart. For all other cases, the wheel speed of the ultrasonic damage cart is not adjusted, ensuring that the speed at the next moment is consistent with the speed at the current moment, and the detection continues. Once the ultrasonic damage carriage has completed its inspection of the entire bimetallic composite steel pipe, it stops. The pipe weld scan images from each moment are then stitched together from the cloud database to obtain a complete scan image of the pipe weld containing defect markers, thus completing the welding defect detection of the bimetallic composite steel pipe. The defect markers include the depth values ​​corresponding to the ultrasonic data.

[0049] Thus, the method provided in this embodiment has been used to detect welding defects in bimetallic composite steel pipes.

[0050] This embodiment first combines the characteristics of the weld seam of the bimetallic composite steel pipe with the shape distribution of different sub-regions in the weld seam scanning image of the bimetallic composite steel pipe at each moment in the initial stage of the welding defect detection process to screen out suspected crack areas. Then, based on the degree of change between the ultrasonic data values ​​at adjacent moments in the initial stage, it screens the crack trend time period. Combining the time interval between the crack trend time period and the current moment and the difference in ultrasonic data values, it screens the ultrasonic probe to be adjusted. Finally, it adjusts the wheel speed of the ultrasonic damage cart based on the distance between the ultrasonic probe to be adjusted and the suspected crack area, so as to achieve the purpose of focusing on the detection of suspected crack areas and improving the accuracy of the weld seam defect detection results of bimetallic composite steel pipe.

[0051] The welding defect detection system for chemical metal pipelines provided in this embodiment is equipped with an automated ultrasonic flaw detection trolley. The trolley automatically and intelligently slides on the pipeline to automatically identify cracks and defects that may exist on the weld surface. While ensuring the accuracy of welding defect detection results, it greatly improves the defect detection efficiency of bimetallic composite steel pipes.

[0052] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting welding defects in chemical metal pipelines, characterized in that, The method includes the following steps: To acquire the pipe weld scanning image and ultrasonic data value of the bimetallic composite steel pipe in the initial stage of defect detection using an ultrasonic damage carriage; Based on the shape distribution of different sub-regions in the pipe weld scan images at each moment in the initial stage, suspected crack areas are screened; based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage, crack trend time periods are screened. Based on the time interval between the crack trend period and the current time, as well as the difference in ultrasonic data values, the necessary indicators for the state change of each ultrasonic probe are determined, and ultrasonic probes to be adjusted are selected; based on the distance between each ultrasonic probe to be adjusted and the suspected crack area and the necessary indicators for the state change, the sliding rate of each ultrasonic probe to be adjusted is obtained. The wheel speed of the ultrasonic damage trolley is adjusted based on the sliding rate to detect defects in suspected crack areas.

2. The method for detecting welding defects in chemical metal pipelines according to claim 1, characterized in that, The step of filtering suspected crack areas based on the shape distribution of different sub-regions in the pipe weld scan images at each moment in the initial stage includes: For any moment in the initial stage: extract the central region and two edge regions from the pipe weld scan image at that moment, and take both the central region and the edge regions as sub-regions; perform curve fitting on each sub-region, and take the average curvature of each sub-region as the first probability value of each sub-region; Based on the first probability value of each sub-region at consecutive moments in the initial stage, suspected crack regions are screened.

3. The method for detecting welding defects in chemical metal pipelines according to claim 2, characterized in that, Based on the first probability value of each sub-region at consecutive moments in the initial stage, suspected crack regions are screened, including: The sub-region with the first probability greater than a preset first threshold at each time is taken as the target region at each time. The area formed by the continuous target region on the bimetallic composite steel pipe in the initial stage is regarded as the suspected crack region.

4. The method for detecting welding defects in chemical metal pipelines according to claim 1, characterized in that, The step of filtering the crack trend time period based on the degree of change between adjacent ultrasonic data values ​​in the initial stage includes: The moment when the change in ultrasonic data value in the initial stage exceeds the preset change threshold is defined as the moment of crack trend; wherein, the change in ultrasonic data value at each moment in the initial stage is the absolute value of the difference between the ultrasonic data value at each moment in the initial stage and the previous moment. A continuous crack trend moment constitutes a crack trend time period.

5. The method for detecting welding defects in chemical metal pipelines according to claim 1, characterized in that, The necessary indicators for determining the state changes of each ultrasonic probe are determined based on the time interval between the crack trend period and the current moment, as well as the differences in ultrasonic data values. These indicators include: The time interval between the current moment and the nearest crack trend time interval is denoted as the first time interval; For any ultrasonic probe, the difference between the ultrasonic data value at the current time and the last time in the crack trend time period closest to the current time is denoted as the first difference; the normalized result of the product of the first time interval and the first difference is determined as the necessary index for the state change of any ultrasonic probe.

6. The method for detecting welding defects in chemical metal pipelines according to claim 5, characterized in that, Screening ultrasound probes to be adjusted, including: Ultrasonic probes whose state change indicators exceed the preset necessary thresholds are identified as ultrasonic probes to be adjusted.

7. The method for detecting welding defects in chemical metal pipelines according to claim 2, characterized in that, The step of obtaining the sliding rate of each ultrasonic probe to be adjusted based on the distance between each probe to be adjusted and the suspected crack area and the necessary indicators of state change includes: For any ultrasound probe to be adjusted: Calculate the normalized value of the product between the shortest distance between any of the ultrasonic probes to be adjusted and the suspected crack area and the necessary index for the state change of any of the ultrasonic probes to be adjusted. Determine whether any of the ultrasound probes to be adjusted is sliding towards the center region. If yes, calculate the first difference between the constant 1 and the normalized value, and determine the sliding speed of the ultrasound probe by multiplying the first difference with the current motion rate of the ultrasound probe to be adjusted. If no, calculate the first sum of the constant 1 and the normalized value, and determine the sliding speed of the ultrasound probe to be adjusted by multiplying the first sum with the current motion rate of the ultrasound probe to be adjusted.

8. The method for detecting welding defects in chemical metal pipelines according to claim 1, characterized in that, The adjustment of the wheel speed of the ultrasonic damage cart based on the sliding rate includes: The total number of ultrasonic probes installed on the ultrasonic damage cart is 2; If the sliding speed of two ultrasonic probes to be adjusted is not equal to the current motion speed, then the minimum sliding speed of the two ultrasonic probes to be adjusted is obtained, and the first normalized result of the difference between the current motion speed and the minimum value is calculated; the product of the current wheel speed of the ultrasonic damage cart and the first normalized result is rounded up and determined as the new wheel speed of the ultrasonic damage cart, and the wheel speed of the ultrasonic damage cart is adjusted. If the sliding speed of an ultrasonic probe to be adjusted is not equal to the current motion speed, then the sliding speed of the ultrasonic probe to be adjusted is taken as the new wheel speed of the ultrasonic damage cart, and the wheel speed of the ultrasonic damage cart is adjusted.

9. A welding defect detection system for chemical metal pipelines, said system implementing the method of claim 1, characterized in that, The system includes: The data acquisition module is used to acquire the pipe weld scanning image and ultrasonic data value of the bimetallic composite steel pipe in the initial stage of the defect detection process of the bimetallic composite steel pipe using an ultrasonic damage carriage. The filtering module is used to filter suspected crack areas based on the shape distribution of different sub-regions in the pipe weld scan image at each moment in the initial stage; and to filter crack trend time periods based on the degree of change between ultrasonic data values ​​at adjacent moments in the initial stage. The calculation module is used to determine the necessary indicators for the state change of each ultrasonic probe based on the time interval between the crack trend period and the current time and the difference in ultrasonic data values, and to screen the ultrasonic probes to be adjusted; based on the distance between each ultrasonic probe to be adjusted and the suspected crack area and the necessary indicators for the state change, the sliding rate of each ultrasonic probe to be adjusted is obtained. The defect detection module is used to adjust the wheel speed of the ultrasonic damage cart based on the sliding rate to detect defects in suspected crack areas.

10. The welding defect detection system for chemical metal pipelines according to claim 9, characterized in that, The data acquisition module includes an ultrasonic probe, an ultrasonic probe control component, and a line scanning component; the defect detection module includes a wheel control component.

Images