TMCP steel weld joint ultrasonic testing and debugging method
By integrating intelligent test blocks, AI judgment, and adaptive DAC generation, the problem of detection error caused by anisotropy in ultrasonic testing of TMCP steel welds is solved, achieving efficient and accurate defect quantification and automated detection, which is applicable to the detection of TMCP steel with multiple plate thicknesses and multiple directions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-12
- Publication Date
- 2026-05-01
AI Technical Summary
Traditional ultrasonic testing of TMCP steel welds suffers from anisotropy caused by grain orientation along the rolling direction, resulting in large deviations in beam angle calculation and high quantitative errors in defect location. Furthermore, existing solutions are inefficient and have cumbersome calibration processes, making it difficult to meet the accuracy and efficiency requirements of large-scale industrial testing.
The debugging method adopts an integrated intelligent test block, AI judgment, adaptive DAC generation and dynamic calibration. The intelligent test block automatically calibrates the sound velocity, AI assists in the determination of the rolling direction, the adaptive DAC generates a dynamic compensation factor, multiple probes work together to quantify, and temperature compensation is combined to achieve dynamic real-time calibration and accurate quantification of defects.
It has reduced the quantitative error of defects from 15% to below 5%, improved the detection efficiency by 25%-32%, shortened the calibration time by more than 65%, adapted to the detection of TMCP steel with multiple plate thicknesses and multiple directions, reduced human error, and improved the level of detection automation.
Smart Images
Figure CN121955183A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of steel weld inspection technology, and in particular to a TMCP steel weld ultrasonic testing and debugging method. Background Technology
[0002] TMCP steel is widely used in high-end equipment structures such as marine engineering due to its grain refinement and phase transformation strengthening properties. However, its rolling process causes the grains to be oriented along the rolling direction, exhibiting significant anisotropy in transverse wave sound velocity and attenuation.
[0003] Traditional ultrasonic testing uses ordinary IIW test blocks for calibration, which does not consider the differences in rolling / vertical direction. This leads to large deviations in the calculation of the sound beam angle, high quantitative errors in defect location, and a high risk of missed or misjudged defects. Existing solutions have shortcomings: manual determination of the rolling direction is inefficient, the calibration process is cumbersome, and there is a lack of dynamic compensation mechanisms, making it difficult to meet the accuracy and efficiency requirements of large-scale industrial testing.
[0004] To address the aforementioned pain points, this invention proposes a debugging method that integrates intelligent test blocks, AI judgment, adaptive DAC generation, and dynamic calibration. This method aims to solve the detection error problem caused by anisotropy, improve the level and efficiency of detection automation, and ensure the ultrasonic testing quality of high-end equipment structures. Summary of the Invention
[0005] The main technical problem solved by this invention is to provide a method for ultrasonic testing and debugging of TMCP steel welds, thereby solving one or more of the problems mentioned above in the prior art.
[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: a method for ultrasonic testing and debugging of TMCP steel welds, characterized by comprising the following steps: S1: Calibrate the transverse wave velocity and refraction angle of the ultrasonic testing equipment probe in TMCP steel; S2: Distance amplitude curves (DAC) are generated for the rolling direction and perpendicular to the rolling direction of TMCP steel to calibrate the detection sensitivity; S3: Before on-site inspection, confirm the rolling direction of the base material on both sides of the TMCP steel weld; S4: Select the corresponding calibration parameters based on the confirmed rolling direction to perform weld inspection.
[0007] In some implementations, step S2 employs adaptive DAC curve generation technology: after the instrument acquires the rolling direction determination result in real time, it automatically calls up the transverse hole wave height data of the corresponding direction test block, and combines it with the pre-stored anisotropic attenuation compensation factor (calculated by comparing the wave height difference of transverse holes of the same depth in the rolling / vertical direction, the formula is: compensation factor = 20lg(vertical direction wave height / rolling direction wave height)) to dynamically generate a DAC curve. The compensation factor is automatically updated by ±0.5dB every 10mm increase in plate thickness, achieving precise compensation for quantitative defects.
[0008] In some implementations, the TMCP steel test block used in step S1 is an integrated smart test block: the test block integrates the rolling direction area, the perpendicular rolling direction area and the 30mm / 50mm thickness stepped segment, and has built-in RFID tags to store parameters such as material sound velocity and attenuation coefficient. The instrument automatically completes the initial calibration by wirelessly reading the tag data. The test block surface is engraved with a QR code that links to the cloud database, which can update the calibration algorithm in real time.
[0009] In some implementations, the determination of the rolling direction in step S3 adopts an AI-assisted machine learning model: when the probe scans along both sides of the weld, the instrument collects sound path, wave height and movement trajectory data, inputs them into the model trained based on the test data in document Table 1, and automatically outputs the rolling direction determination result with an accuracy of ≥95% and a determination time of ≤2 seconds, replacing manual reading determination.
[0010] In some implementations, the sound velocity calibration in step S1 adopts dynamic real-time calibration technology: during the detection process, the instrument identifies the plate thickness change by the probe movement trajectory and wave height change, and automatically calls the pre-stored corresponding plate thickness sound velocity value, without stopping the detection and recalibrating, thus improving the detection efficiency by ≥30%.
[0011] In some implementations, the defect quantification in step S4 uses a multi-probe collaborative weighted algorithm: three probes at 45°, 60°, and 70° are used to detect the same defect simultaneously. The instrument establishes a weighted model based on the wave height data of each probe and the attenuation differences of each angle in different directions recorded in the document (the weighting coefficient is set based on the sensitivity of the angle to anisotropy), and the defect size quantification error is ≤5%.
[0012] In some implementations, the secondary total internal reflection calibration in step S1 uses an intelligent peak recognition algorithm: the instrument automatically identifies the peaks of the primary / secondary total internal reflection, calculates the sound velocity and adjusts the time baseline. The entire process requires no manual operation, takes ≤3 seconds, and improves calibration accuracy by ≥20%.
[0013] In some implementations, the stepped sections of the test block are provided with a gradual thickness transition zone (thickness smoothly transitions from 30mm to 50mm), which can complete the sound velocity calibration of multiple plate thicknesses in one go, reduce the number of test block replacements, and is suitable for the inspection of workpieces with continuously varying thicknesses.
[0014] In some implementations, a temperature compensation factor is automatically incorporated during curve generation: the instrument has a built-in temperature sensor that adjusts the compensation factor according to the ambient temperature. The compensation factor is updated by ±0.3dB for every 5°C change in temperature, further improving the quantitative accuracy of defects.
[0015] The beneficial effects of this invention are: accuracy: the defect quantification error is reduced from the traditional 15%+ to below 5%, meeting the requirements of high-demand detection scenarios; Efficiency: Detection efficiency is improved by 25%-32%, and calibration time is reduced by more than 65%; Adaptability: Supports testing of TMCP steel with multiple plate thicknesses and multiple directions, eliminating the need for frequent test block replacements; Ease of use: AI-assisted judgment reduces reliance on experience, and automatic report generation reduces human error. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments 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, wherein: Figure 1 This is a flowchart of an ultrasonic testing and debugging method for TMCP steel welds according to the present invention.
[0017] Figure 2 This is a flowchart of a multi-probe coordinated detection method for ultrasonic testing and debugging of TMCP steel welds according to the present invention. Detailed Implementation
[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0019] like Figure 1 and Figure 2 As shown, the embodiments of the present invention include: Addressing the anisotropy problem in ultrasonic testing of TMCP steel welds, the present invention provides a complete intelligent debugging and testing solution, the components and steps of which are as follows: 1. Essential Constituent Elements Testing equipment: Digital ultrasonic flaw detector, such as the upgraded version of SIUI-CTS-2020, with built-in AI judgment module, adaptive DAC generation module, and temperature sensor; Probe set: 45° / 60° / 70° shear wave probe; Calibration block: An integrated intelligent block that integrates rolling direction / perpendicular rolling direction areas, 30mm / 50mm stepped sections, built-in RFID tags to store material sound velocity / attenuation coefficient, and surface engraved QR codes to link to cloud database; Auxiliary modules: AI machine learning model, dynamic sound velocity calibration algorithm, multi-probe weighted quantitative model.
[0020] 2. Core Implementation Steps Step 1: Smart test block initialization calibration. Place the probe on the surface of the smart test block. The flaw detector reads the pre-stored TMCP steel parameters in the test block through RFID and automatically completes the sound velocity / zero point calibration. Scan the QR code to update the cloud calibration algorithm.
[0021] Step 2: The rolling direction AI determination probe scans vertically along both sides of the weld to collect sound path / wave height / trajectory data, inputs it into the AI model, and outputs the determination result within 2 seconds.
[0022] Step 3: Adaptive DAC curve generation. Based on the determined direction, call the corresponding test block transverse hole data and calculate the attenuation compensation factor. The calculation formula is: compensation factor = 20lg(vertical wave height / rolling wave height). Combined with temperature sensor data, dynamically generate the DAC curve.
[0023] Step 4: During dynamic real-time calibration testing, the changes in plate thickness are identified, and the corresponding pre-stored sound velocity for the plate thickness is automatically called up without stopping the machine for calibration.
[0024] Step 5: Multi-probe collaborative quantitative analysis. 45° / 60° / 70° probes simultaneously detect defects. The flaw detector calculates the defect size according to the weights (45°:0.4, 60°:0.3, 70°:0.3), and the quantitative error is ≤5%.
[0025] The following are three examples and comparative examples for verification: Example 1: Adaptive DAC and Temperature Compensation Objective: To address the influence of anisotropy and temperature on the quantitative analysis of defects in TMCP steel. 1. Equipment Preparation Flaw detector: SIUI-CTS-2020 upgraded version, with built-in temperature sensor, adaptive DAC module, and AI judgment model; Probe: 70° shear wave probe Test block: T=30mm integrated intelligent test block; rolling direction + vertical direction area, built-in RFID tag to store parameters: sound velocity in the rolling direction 3100m / s, vertical direction 3300m / s; surface engraved with QR code to link to cloud calibration algorithm; Auxiliary tool: Temperature recorder, accuracy ±0.5℃.
[0026] 2. Calibration Procedure a. Smart Test Block Initialization The probe is placed in the rolling direction area of the test block, and the flaw detector reads the pre-stored parameters through RFID, automatically completing the sound velocity (3100m / s) and zero-point calibration (lead edge 5mm); the cloud algorithm is updated by scanning the QR code.
[0027] b. AI determination of rolling direction The probe scans along both sides of the weld in a direction perpendicular to the weld (path: from 10mm to 10mm to the right of the weld, speed 5mm / s, 20 data points collected). Instrument data acquisition: sound path (21mm for a 15mm aperture), wave height (Ø3-6), trajectory coordinates; The AI model outputs the following judgment result: rolling direction (time taken: 1.8s, accuracy: 97%).
[0028] c. Adaptive DAC curve generation Call the 15mm vertical hole wave height data of the test block; Calculate the attenuation compensation factor: The temperature sensor reads an ambient temperature of 25℃, and adjusts it according to the preset rule (compensation factor +0.28dB for every 5℃ increase in temperature) as follows: The instrument automatically generates a DAC curve that includes a compensation factor and sets the alarm threshold to DAC + 6dB.
[0029] 3. Testing Operation The probe scans the weld seam in a sawtooth pattern (speed ≤100mm / s, overlap rate ≥10%). When a defect wave (wave height Ø3-7, sound path 25mm) is detected, the instrument locks the defect location.
[0030] 4. Data Processing Defect quantitative dimensions: (The angle of refraction is 72°, calculated from the calibrated speed of sound) Application of compensation factor adjustment: Final size = 7.7mm × (1-0.042) = 7.4mm Output report: Defect location (5mm to the left of the weld center), size (7.4mm), nature (point defect).
[0031] Example 2: Smart test block and dynamic calibration Objective: To solve the calibration efficiency problem of continuously varying thickness workpieces. 1. Equipment Preparation Flaw detector: SIUI-CTS-2020 upgraded version, with built-in dynamic sound velocity module and gradual thickness recognition algorithm; Probe: 60° shear wave probe; Test block: Integrated intelligent test block, including a 30mm to 50mm gradual transition zone, with built-in RFID storage of 3100m / s sound velocity at 30mm and 3080m / s sound velocity at 50mm; Auxiliary tool: calipers.
[0032] 2. Calibration Procedure a. Smart Test Block Initialization The probe is placed in a 30mm section of the test block, and the parameters are read by RFID. The sound velocity and zero point are automatically calibrated to 3100m / s. Pre-stored dynamic sound velocity library: Import document table 1 T=50mm rolling direction sound velocity 3080m / s.
[0033] b. Rolling direction determination The probe scanned a 30mm segment, and the AI determined it to be in the vertical direction (takes 2.0s).
[0034] 3. Testing Operation The probe moves from the 30mm segment to the 50mm segment in a linear scanning motion (speed ≤120mm / s); When the instrument scans to the gradual transition zone (thickness from 30mm to 50mm), it recognizes the change in wave height (from Ø3-5 to Ø3-4) and the extension of sound path (from 20mm to 35mm), and automatically triggers dynamic calibration. Call the pre-stored 50mm sound velocity 3080m / s, and continue to detect defects in the 50mm segment (wave height Ø3-6, sound path 40mm) without stopping the machine.
[0035] 4. Data Processing Defect size for a 30mm segment: 20×cos(68°)≈7.5mm; Defect size for the 50mm segment: 40×cos(66°)≈16.2mm; Compared with the actual dimensions (7.8mm and 16.9mm), the errors are 3.8% and 4.1% respectively (average 4.8%, corresponding to the quantitative error in the comparison table). Output report: Defect distribution in continuously thickened welds (1 defect in the 30mm section and 2 defects in the 50mm section).
[0036] Example 3: AI Judgment + Multi-Probe Collaboration Objective: Improve the accuracy of defect quantification 1. Equipment Preparation Flaw detector: SIUI-CTS-2020 upgraded version, with built-in multi-probe weighted module and AI model; Probe set: 45°, 60°, 70° shear wave probes; Test block: T=50mm rolling direction test block, with built-in RFID storage for sound velocities at various angles: 45°→3150m / s, 60°→3120m / s, 70°→3100m / s; Auxiliary tool: probe holder.
[0037] 2. Calibration Procedure a. Smart Test Block Initialization Three probes are placed on the test block, and the parameters are read by RFID to automatically calibrate the sound velocity of each probe. Set weighting coefficients: Based on the attenuation differences at each angle in Table 1 of the document, the weights are 45°: 0.4, 60°: 0.3, and 70°: 0.3.
[0038] b. AI determination of rolling direction Three probes simultaneously scan both sides of the weld, collecting data from each probe, and AI determines it to be the rolling direction (time taken 1.9s).
[0039] 3. Testing Operation The bracket holds three probes to inspect welds using a zigzag scanning motion (speed ≤80mm / s). When a defect is detected, the three probes record the wave heights as follows: 45° (Ø3-8), 60° (Ø3-10), and 70° (Ø3-12).
[0040] 4. Data Processing Weighted wave height calculation: Defect quantitative dimension: 15mm according to the DAC curve; Compared to the actual size of 15.6mm, the error is 3.8%; Output report: Defect location, size, and nature.
[0041] Comparative example: Traditional detection methods Core technology: Calibration of ordinary IIW test blocks + manual judgment steps: Calibrate the velocity of sound (default 3200 m / s) and zero point using a standard IIW test block; The rolling direction is determined by manually observing the markings on the edge of the plate (takes 10.3 seconds). DAC curves were fabricated using a standard RB-3 test block (without compensation). Defects were detected using a single probe, and dimensions were calculated manually; result: quantitative error 16.5%.
[0042] Performance Comparison Table index Example 1 Example 2 Example 3 Comparative Example Defect quantification error (%) 4.2 4.8 3.9 16.5 Rolling direction determination time (s) 2.1 2.0 1.8 10.3 Detection efficiency improvement rate (%) 25 32 28 0 Calibration time (s) 5.3 4.5 5.0 18.2 Accuracy rate (%) 96.2 95.8 97.5 82.0 The principle behind this solution is as follows: Directional compensation: Obtain direction-specific parameters by using rolling / vertical direction test blocks, and adjust the DAC curve accordingly; Intelligent automation: AI makes judgments to replace manual labor, and dynamic calibration adapts to changes in plate thickness, reducing human intervention; Multi-dimensional collaboration: Weighted quantification by multiple probes combined with temperature compensation further improves accuracy.
[0043] The advantages of this technical solution are: Accuracy: The defect quantification error has been reduced from over 15% to below 5%, meeting the requirements of high-demand testing scenarios; Efficiency: Detection efficiency is improved by 25%-32%, and calibration time is reduced by more than 65%; Adaptability: Supports testing of TMCP steel with multiple plate thicknesses and multiple directions, eliminating the need for frequent test block replacements; Ease of use: AI-assisted judgment reduces reliance on experience, and automatic report generation reduces human error.
[0044] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.
Claims
1. A method for ultrasonic testing and debugging of TMCP steel welds, characterized in that: Includes the following steps: S1: Calibrate the transverse wave velocity and refraction angle of the ultrasonic testing equipment probe in TMCP steel; S2: Distance amplitude curves were generated for the rolling direction and perpendicular to the rolling direction of TMCP steel to calibrate the detection sensitivity; S3: Before on-site inspection, confirm the rolling direction of the base material on both sides of the TMCP steel weld; S4: Select the corresponding calibration parameters based on the confirmed rolling direction to perform weld inspection.
2. The ultrasonic testing and debugging method for TMCP steel welds according to claim 1, characterized in that: In step S2, adaptive DAC curve generation technology is used: After the instrument acquires the rolling direction determination result in real time, it automatically calls the transverse hole wave height data of the corresponding direction test block, and combines it with the pre-stored anisotropic attenuation compensation factor. The compensation factor is calculated by comparing the wave height difference of transverse holes of the same depth in the rolling / vertical direction. The formula is: compensation factor = 20lg(vertical direction wave height / rolling direction wave height). The DAC curve is dynamically generated. The compensation factor is automatically updated by ±0.5dB every 10mm increase in plate thickness, so as to achieve precise compensation for quantitative defects.
3. The ultrasonic testing and debugging method for TMCP steel welds according to claim 1, characterized in that: The TMCP steel test block used in step S1 is an integrated smart test block: the test block integrates the rolling direction area, the perpendicular rolling direction area and the 30mm / 50mm thickness stepped segment, and has built-in RFID tags to store parameters such as material sound velocity and attenuation coefficient. The instrument automatically completes the initial calibration by wirelessly reading the tag data. The test block surface is engraved with a QR code that links to the cloud database, which can update the calibration algorithm in real time.
4. The ultrasonic testing and debugging method for TMCP steel welds according to claim 1, characterized in that: In step S3, the determination of the rolling direction adopts an AI-assisted machine learning model: when the probe scans along both sides of the weld, the instrument collects sound path, wave height and movement trajectory data, inputs them into the model trained based on the test data in document Table 1, and automatically outputs the rolling direction determination result with an accuracy of ≥95% and a determination time of ≤2 seconds, replacing manual reading determination.
5. The ultrasonic testing and debugging method for TMCP steel welds according to claim 1, characterized in that: The sound velocity calibration in step S1 adopts dynamic real-time calibration technology: During the detection process, the instrument identifies the plate thickness change by the probe movement trajectory and wave height change, and automatically calls the pre-stored corresponding plate thickness sound velocity value, without stopping the detection and recalibrating, thus improving the detection efficiency by ≥30%.
6. The ultrasonic testing and debugging method for TMCP steel welds according to claim 1, characterized in that: In step S4, the defect quantification adopts a multi-probe collaborative weighted algorithm: three probes at 45°, 60° and 70° are used to detect the same defect simultaneously. The instrument establishes a weighted model based on the wave height data of each probe and the attenuation difference of each angle in different directions recorded in the document. The defect size quantification error is ≤5%.
7. The ultrasonic testing and debugging method for TMCP steel welds according to claim 1, characterized in that: The secondary total internal reflection calibration in step S1 uses an intelligent peak recognition algorithm: the instrument automatically identifies the peaks of the primary / secondary total internal reflection, calculates the sound velocity and adjusts the time baseline. The entire process requires no manual operation, takes ≤3 seconds, and improves calibration accuracy by ≥20%.
8. The ultrasonic testing and debugging method for TMCP steel welds according to claim 3, characterized in that: The stepped sections of the test block are designed with a gradual thickness transition zone, which can complete the sound velocity calibration of multiple plate thicknesses in one go, reduce the number of test block replacements, and is suitable for the inspection of workpieces with continuously varying thicknesses.
9. The ultrasonic testing and debugging method for TMCP steel welds according to claim 2, characterized in that: Temperature compensation factor is automatically incorporated during curve generation: The instrument has a built-in temperature sensor that adjusts the compensation factor according to the ambient temperature. The compensation factor is updated by ±0.3dB for every 5°C change in temperature, further improving the quantitative accuracy of defects.