An Automatic Control Method and System for Directional High-Frequency Guided Waves in Steel Structure Weld Inspection

Abstract

This application discloses a directional high-frequency guided wave automatic control method and system for steel structure weld inspection, relating to the field of steel structure weld inspection technology. This method controls the inspection device to move along a travel track, and combines an infrared rangefinder and distance correction algorithm to ensure the inspection device moves along a predetermined route on the test piece without deviating from the route, thus guaranteeing the accuracy of the inspection signal. Simultaneously, an encoder is used for assisted positioning, recording the correspondence between probe displacement and signal in real time, providing parameters for subsequent data processing. The method also employs a bidirectional ultrasonic guided wave probe to simultaneously inspect welds on both sides, improving inspection efficiency. Furthermore, it can display the waveform detected by the probe in real time, facilitating rapid determination of the defect location. Finally, an intelligent recognition network is used to quickly identify whether the steel structure contains defects, the type of defect, and the location of the defect within the steel structure.

Description

Technical Field

[0001] This application relates to the field of steel structure weld inspection technology, specifically to a directional high-frequency guided wave automatic control method and system for steel structure weld inspection. Background Technology

[0002] Steel structures are widely used in modern architecture, bridges, and industrial facilities. The quality of their welds directly affects the safety and reliability of the structure, making weld inspection a crucial step in ensuring steel structure quality. During long-term service, cracks are highly susceptible to develop at the welds of steel structures. These cracks can be caused by various factors, including structural fatigue, material aging, design defects, or improper construction. If not detected and repaired promptly, these cracks may gradually propagate, eventually leading to structural failure and even serious safety accidents. To ensure the safety of steel structures, regular inspections are necessary to identify and repair potential safety hazards in a timely manner, preventing catastrophic damage caused by crack propagation. Furthermore, initial manufacturing defects can lead to reduced fatigue resistance or the migration of failure modes, necessitating inspection of steel components before they leave the factory to ensure structural quality and eliminate the impact of initial defects on the service life of the steel structure. In addition to weld defects, the surface of steel structures is also prone to corrosion damage due to environmental factors, requiring comprehensive inspection through efficient testing methods.

[0003] Traditional visual inspection methods for checking cracks in steel structure welds are inefficient. Due to environmental limitations, visibility is restricted in areas difficult to see, such as inside the structure, leading to missed defects. Furthermore, internal defects present in the steel before it leaves the factory cannot be detected promptly through simple visual inspection alone. Traditional inspection methods consume significant manpower and resources.

[0004] Existing technologies have also been proposed to use detection devices combined with intelligent algorithms to detect cracks. However, existing detection devices are prone to deviating from the preset detection route during the detection process, have low mobility, and the accuracy of the detection signal cannot be guaranteed. Summary of the Invention

[0005] To address the technical problems existing in current weld inspection technologies for steel structures, this application proposes a directional high-frequency guided wave automatic control method and system for weld inspection of steel structures.

[0006] This application is achieved through the following technical solution:

[0007] An automatic control method for directional high-frequency guided waves in steel structure weld inspection includes:

[0008] Start the bidirectional ultrasonic guided wave probe and infrared rangefinder, and control the bidirectional ultrasonic guided wave probe and infrared rangefinder to move on the running track. The weld seams on both sides of the probe are detected by the bidirectional ultrasonic guided wave probe.

[0009] The infrared rangefinder monitors the relative distance between the main structure of the instrument and the test piece in real time, and determines whether the relative distance is equal to the standard distance value. If it is, the subsequent steps are continued; otherwise, the correction amount is calculated using a distance correction algorithm and the relative distance is adjusted by controlling the three-axis displacement rail beam with a command.

[0010] The travel distance of the bidirectional ultrasonic guided wave probe is recorded by an coded pedometer, and the travel distance and the waveform signal collected by the bidirectional ultrasonic guided wave probe are imported into the ultrasonic guided wave instrument.

[0011] Signals are acquired from the ultrasonic guided wave instrument, processed, and then displayed as waveforms and imaged for defects.

[0012] The processed signal is input into the intelligent recognition model for multidimensional detection of damage information, and the damage information is summarized and output.

[0013] The traveling track is fixed on the foot support, which is attached to the test piece. The bidirectional ultrasonic guided wave probe and the infrared rangefinder are both mounted on the traveling track via a three-axis displacement beam.

[0014] In some implementations, it also includes:

[0015] When the waveform display is abnormal, the image acquisition device is activated to detect the image of the area where the bidirectional ultrasonic guided wave probe is located. The abnormality is analyzed based on the acquired image to determine the location and cause of the abnormality.

[0016] The image acquisition device is mounted on the running track via a three-axis displacement beam.

[0017] In some implementations, it also includes:

[0018] The tilt angle is monitored by the built-in level of the bidirectional ultrasonic guided wave probe. If the tilt angle exceeds the threshold, an alarm is triggered, indicating that the detection effect at this position is poor, and the information of this position is recorded.

[0019] In some implementations, the following are included:

[0020] Three mutually perpendicular infrared rangefinders were used to measure the relative distances between the main structure of the instrument and the test piece in three directions.

[0021] The relative distances in the three directions are compared with their corresponding standard distance values. If any of the relative distances in the three directions monitored in real time deviates from its corresponding standard distance value, the difference between the relative distance value and the standard value is calculated to obtain the distance correction value.

[0022] Based on the distance correction value, a control command is generated to control the three-axis displacement beam to move in the corresponding direction, so that the relative distance between the infrared rangefinder and the test piece is equal to the standard distance value.

[0023] In some implementations, the training process of the intelligent recognition model includes:

[0024] Establish a guided wave data signal library, which contains a large number of ultrasonic guided wave signals in steel structures under different defect types and different defect degrees of the same defect;

[0025] The ultrasonic guided wave signals in the guided wave data signal library are preprocessed;

[0026] Time-frequency analysis is performed on the preprocessed signal to extract feature data, which is then labeled to construct a training dataset.

[0027] The intelligent recognition model is obtained by training the neural network using the training dataset.

[0028] In some embodiments, the bidirectional ultrasonic waveguide probe is two ultrasonic waveguide probes symmetrically arranged, and the ultrasonic wave emission directions of the two ultrasonic waveguide probes correspond to the welds on both sides of the bidirectional ultrasonic waveguide probe.

[0029] The two ultrasonic guided wave probes work synchronously to inspect the welds on both sides respectively;

[0030] The two sides of the bidirectional ultrasonic guided wave probe are designed with a ramp shape.

[0031] In some embodiments, a limiting block is provided on each side of the bidirectional ultrasonic guided wave probe to restrict the movement of the bidirectional ultrasonic guided wave probe during the moving detection process;

[0032] The outer sides of the two limiting blocks are pointed to remove debris from the surface of the steel structure.

[0033] On the other hand, this application also proposes a directional high-frequency guided wave automatic control system for steel structure weld inspection, comprising:

[0034] The drive module is used to start the bidirectional ultrasonic guided wave probe and the infrared rangefinder, and to control the bidirectional ultrasonic guided wave probe and the infrared rangefinder to move on the running track.

[0035] The detection module uses a pedometer to record the travel distance of the bidirectional ultrasonic waveguide probe, uses the bidirectional waveguide probe to detect the welds on both sides of the probe, and uses an infrared rangefinder to monitor the relative distance between the main structure of the instrument and the test piece in real time. It determines whether the relative distance is equal to the standard distance value. If it is, the travel distance and the acquired waveform signal are imported into the ultrasonic waveguide instrument; otherwise, the correction module is activated.

[0036] The correction module calculates the correction amount using a distance correction algorithm and transmits it to the drive module to generate control commands to control the three-axis displacement rail beam to adjust the relative distance.

[0037] The signal acquisition and analysis module acquires signals from the ultrasonic guided wave instrument, processes them, and displays waveforms and performs defect imaging.

[0038] In addition, the intelligent recognition module uses an intelligent recognition model to perform multi-dimensional detection of damage information and outputs the summarized damage information;

[0039] The traveling track is fixed on the foot support, which is attached to the test piece. The bidirectional ultrasonic guided wave probe and the infrared rangefinder are both mounted on the traveling track via a three-axis displacement beam.

[0040] In some implementations, it also includes:

[0041] The anomaly diagnosis module is used to activate the image acquisition device when the waveform display is abnormal, to detect the image of the area where the bidirectional ultrasonic guided wave probe is located through the image acquisition device, and to perform anomaly analysis based on the acquired image to determine the location and cause of the anomaly.

[0042] The image acquisition device is mounted on the running track via a three-axis displacement beam.

[0043] In some embodiments, the signal acquisition and analysis module includes:

[0044] The processing unit is used to process the acquired signals;

[0045] The display unit is used for waveform display and defect imaging;

[0046] In addition, the alarm unit monitors the tilt angle using the level built into the bidirectional ultrasonic guided wave probe. If the tilt angle exceeds the threshold, an alarm is triggered to indicate that the detection effect at this position is poor, and the information at this position is recorded.

[0047] This application proposes a directional high-frequency guided wave automatic control method for steel structure weld inspection. By controlling the detection device to move along a travel track and combining an infrared rangefinder and a distance correction algorithm, the method ensures that the detection device can move on the test piece along a predetermined route without deviating from the route, thus guaranteeing the accuracy of the detection signal. Simultaneously, an encoder is used for assisted positioning, recording the correspondence between probe displacement and signal in real time, providing parameters for subsequent data processing. The method also employs a bidirectional ultrasonic guided wave probe to simultaneously inspect welds on both sides, improving inspection efficiency. Furthermore, it can display the waveform detected by the probe in real time, facilitating rapid determination of the defect location. Finally, an intelligent recognition network is used to quickly identify whether the steel structure contains defects, the type of defect, and the location of the defect within the steel structure.

[0048] Correspondingly, the directional high-frequency guided wave automatic control system for steel structure weld inspection proposed in this application also has the same technical effects as described above. Attached Figure Description

[0049] The accompanying drawings, which are included to provide a further understanding of the embodiments of this application and form part of this application, do not constitute a limitation on the embodiments of this application. In the drawings:

[0050] Figure 1 This is a schematic diagram of the automatic control method proposed in the embodiments of this application;

[0051] Figure 2 This is a schematic diagram of the CNN-LSTM hybrid model architecture used in the embodiments of this application;

[0052] Figure 3 Waveforms collected for embodiments of this application;

[0053] Figure 4 This is a schematic diagram of the detection device used in the embodiments of this application during operation;

[0054] Figure 5 This is a schematic diagram of the bidirectional ultrasonic guided wave probe structure according to an embodiment of this application;

[0055] Figure 6 This is a block diagram illustrating the principle of the automatic control system proposed in an embodiment of this application;

[0056] Figure reference numerals and corresponding component names:

[0057] 201-Test piece, 202-Weld, 203-Bidirectional ultrasonic guided wave probe, 204-Limit block, 205-Foot bracket, 206-Fixed end foot, 207-Traveling track, 208-Screw fixed end, 209-Three-axis displacement rail beam, 210-Infrared rangefinder, 211-Image acquisition device, 212-Encoded pedometer, 213-Stepper motor, 214-Coupling, 215-Ball screw, 216-Motor bracket, 217-Screw support end, 218-Support end foot, 500-Automatic control system, 501-Drive module, 502-Detection module, 503-Correction module, 504-Signal acquisition and analysis module, 505-Intelligent recognition module, 506-Abnormal diagnosis module. Detailed Implementation

[0058] In the following, the terms “comprising” or “may include” as used in the various embodiments of this application indicate the presence of a function, operation, or element of the invention and do not limit the addition of one or more functions, operations, or elements. Furthermore, as used in the various embodiments of this application, the terms “comprising,” “having,” and their cognates are intended only to indicate a specific feature, number, step, operation, element, component, or combination of the foregoing and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or adding one or more combinations of the foregoing.

[0059] In various embodiments of this application, the expression "or" or "at least one of A and / or B" includes any combination or all combinations of the words listed simultaneously. For example, the expression "A or B" or "at least one of A and / or B" may include A, may include B, or may include both A and B.

[0060] The terms used in the various embodiments of this application (such as "first," "second," etc.) may modify various constituent elements in the various embodiments, but do not limit the corresponding constituent elements. For example, the above terms do not limit the order and / or importance of the elements. The above terms are only used for the purpose of distinguishing one element from other elements. For example, a first user device and a second user device refer to different user devices, although both are user devices. For example, without departing from the scope of the various embodiments of this application, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element.

[0061] It should be noted that if a description is made of "connecting" one component to another, then the first component can be directly connected to the second component, and a third component can be "connected" between the first and second components. Conversely, when a component is "directly connected" to another component, it can be understood that there is no third component between the first and second components.

[0062] The terminology used in the various embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the various embodiments of this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. The terms (such as those defined in a generally used dictionary) are to be interpreted as having the same meaning as in the context of the relevant technical field and are not to be interpreted as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0063] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.

[0064] Example:

[0065] To address the problems of cumbersome inspection process, low inspection efficiency, inability to effectively detect internal defects, and low mobility of inspection equipment in existing steel structure weld inspection technologies, this embodiment proposes a directional high-frequency guided wave automatic control method for steel structure weld inspection.

[0066] like Figure 1 As shown, the automatic control method proposed in this embodiment includes:

[0067] Step 110: Start the bidirectional ultrasonic guided wave probe and infrared rangefinder, and control the bidirectional ultrasonic guided wave probe and infrared rangefinder to move on the running track. Use the bidirectional ultrasonic guided wave probe to detect the weld seams on both sides of the probe respectively.

[0068] Step 120: Using an infrared rangefinder, monitor the relative distance between the main structure of the instrument and the test piece in real time, and determine whether the relative distance is equal to the standard distance value. If it is, continue to the subsequent steps; otherwise, use the distance correction algorithm to calculate the correction amount and control the three-axis displacement rail beam to adjust the relative distance via command.

[0069] Step 130: Record the walking distance of the bidirectional ultrasonic guided wave probe using a pedometer, and import the walking distance and the waveform signal acquired by the bidirectional ultrasonic guided wave probe into the ultrasonic guided wave instrument for time alignment, preprocessing and format conversion.

[0070] Step 140: Acquire signals from the ultrasonic guided wave instrument, perform noise reduction and feature extraction, and display waveforms and perform defect imaging;

[0071] Step 150: Input the processed signal into the intelligent recognition model to perform multidimensional detection of damage information and output the summarized damage information;

[0072] The running track is fixed on the foot support, which is attached to the test piece. The bidirectional ultrasonic guided wave probe and infrared rangefinder are both mounted on the running track via a three-axis displacement rail beam.

[0073] It is understandable that steps 110 to 150 above are mainly for detecting defects in steel structure welds, but they can also be used to detect other defects in steel structures.

[0074] Optionally, the automatic control method proposed in this embodiment further includes:

[0075] When the waveform cannot be displayed normally, i.e., when the waveform is abnormal, the image acquisition device is activated. This device detects the image of the area where the bidirectional ultrasonic guided wave probe is located and performs anomaly analysis based on the acquired image. This analysis includes the location of the anomaly (the abnormal area is located by the joint encoding pedometer signal) and the cause of the anomaly. Feedback is then provided to facilitate subsequent maintenance and repair work. It is understandable that conventional image processing techniques can be used to analyze the location and cause of waveform anomalies, which will not be elaborated upon here.

[0076] Optionally, during the testing process, the received guided wave signals may not all be normal; sometimes, due to various on-site factors, the signal data may deviate from expectations. For example, during the climbing process, the probe will inevitably tilt, and a large angle of tilt will affect the propagation efficiency of ultrasonic waves between the probe and the test piece. Therefore, the automatic control method proposed in this embodiment further includes:

[0077] The probe's built-in level monitors the tilt angle, and a tilt angle threshold is set. During the detection process, when the probe tilts while traversing an obstacle, the built-in level also tilts, updating the tilt angle in real time to the signal acquisition and analysis module. When the tilt angle becomes too large, exceeding the tilt angle threshold, an alarm is triggered, indicating poor detection performance at that location, and the information at that location is recorded for further verification.

[0078] Optionally, in this embodiment, the infrared rangefinder is installed in three mutually perpendicular directions (the three coordinate axes in a coordinate system formed by the plane of the track and the direction perpendicular to the plane of the track). It emits infrared rays to monitor the current spatial position of the instrument in real time. The relative distance between the instrument and the test piece in the three directions is compared with the standard distance value. If any of the three relative distances deviates from the standard distance value, the difference between the relative distance value and the standard value is calculated to obtain a distance correction value. Based on the distance correction value, a command is sent to the three-axis displacement track beam to move in the corresponding direction, so that the relative distance between the infrared rangefinder and the test piece is equal to the preset standard distance value, thereby allowing the bidirectional ultrasonic guided wave probe to perform detection according to the preset route.

[0079] Optionally, in this embodiment, the travel distance of the bidirectional ultrasonic guided wave probe is recorded by an coded pedometer and imported into the ultrasonic guided wave instrument for time synchronization (time alignment) with the waveform signal to achieve accurate calibration of the defect location. The ultrasonic guided wave instrument performs analog-to-digital conversion and preprocessing (filtering, etc.) on the received waveform signal to generate standardized format data, providing a data foundation for subsequent noise reduction, time-frequency analysis, and defect imaging.

[0080] Optionally, this embodiment utilizes a trained intelligent recognition model to identify the damage type and output the damage location, achieving multi-dimensional detection of damage information. The training process of the intelligent recognition model includes:

[0081] First, a guided wave data signal library is established, containing a large number of ultrasonic guided wave signals, including ultrasonic guided wave detection signals for steel structures of different shapes and thicknesses. Specifically, the steel structure can be inspected in a non-destructive state to obtain initial guided wave signals (i.e., guided wave signals in a non-destructive state). Then, the steel structure is subjected to operations such as loading, drilling, and corrosion to obtain ultrasonic guided wave signals in the steel structure under different defect types and different defect degrees of the same defect. These signals are then denoised to obtain more accurate guided wave signals. Further time-frequency analysis is performed to extract time-domain features, frequency-domain features, and time-frequency-domain features from the signals. Specifically, time-domain features such as peak amplitude and peak-to-peak time difference can be extracted from the denoised guided wave signals. Short-time Fourier transform can also be performed on the denoised guided wave signals to generate a time-frequency spectrum, extracting frequency-domain features such as frequency band energy distribution, instantaneous frequency, and frequency band energy proportion. At the same time, time-frequency-domain features such as waveform entropy can also be extracted. These features are used to train a neural network to obtain an intelligent recognition model. The neural network in this embodiment can adopt a CNN-LSTM hybrid model architecture, combining the local feature extraction of CNN with the temporal modeling advantages of LSTM to achieve high-precision classification. For example... Figure 2As shown, the hybrid model architecture includes an input data layer for inputting extracted feature data and guided wave timing signals; an LSTM timing processing layer (mainly including LSTM layers, Dropout layers, and their optional stacked structures) for capturing the temporal dynamic features of guided wave propagation; a CNN feature extraction layer (mainly including convolutional layers, pooling layers, and their repeated stacked structures) for extracting time-frequency domain features through convolution; a shared feature fusion layer (mainly including splicing or weighted and fully connected structures) for mapping features from different modalities to a shared feature space through multimodal fusion, parameter reuse, and task collaboration, serving as the feature basis for the classification task branch and the regression task branch; a classification task branch (mainly including fully connected layers and Softmax) for implementing defect type identification; and a regression task branch (mainly including fully connected layers and linear activation functions) for outputting the defect size corresponding to the identified defect type (e.g., the identified defect type is a crack, and the crack length is 3.2±0.3mm). The dataset, consisting of the original guided wave signal and its corresponding feature data, is proportionally divided into training and validation sets to ensure a balanced distribution of defects of each category across the subsets. The original guided wave signal or time-frequency feature vector is used as the time-series signal input, and parameters such as temperature and humidity are concatenated with the guided wave signal as additional input features. Batch data is input into the neural network, the output probability distribution is calculated, the predicted results are compared with the true labels, the cross-entropy loss is calculated, the gradient is calculated using the backpropagation algorithm, and the weights are updated using the Adam optimizer. The validation set accuracy is evaluated after each training round; if the accuracy does not improve for several consecutive rounds, training is terminated to prevent overfitting. Furthermore, different loss functions are needed for different recognition tasks. For defect category recognition, a classification loss function, such as cross-entropy loss, can be selected; for defect size recognition, a regression loss function, such as mean squared error, can be selected.

[0082] Alternatively, to address the issue of insufficient data volume, methods such as experimental data acquisition, finite element simulation data synthesis, and adversarial neural network generation of adversarial examples can be used to expand the data sample. When acquiring experimental data, different damage types and locations can be designed, the influence of other variables on the signal data can be considered, and multi-sensor arrays can be used to increase the data volume.

[0083] Optionally, the guided wave data signal library is also updated in real time, that is, during subsequent detection processes, the detected guided wave signals can be imported into the guided wave data signal library to expand the data volume.

[0084] Optionally, this embodiment displays the waveform, specifically a waveform amplitude versus displacement graph, thereby enabling rapid determination of the defect's location, such as... Figure 3 The waveform of the guided wave signal received by the ultrasonic guided wave probe at a certain location during the detection process is shown in the figure. Figure 3As can be seen, the horizontal axis represents the distance from a point in the detection direction to the ultrasonic guided wave probe, and the vertical axis represents the voltage amplitude measured by the probe. When a defect is encountered, the defect will cause the guided wave to generate an echo, which is received by the ultrasonic guided wave probe and generates a wave packet at the corresponding location. The ultrasonic guided wave instrument receives the raw guided wave signal from the bidirectional ultrasonic guided wave probe, performs analog-to-digital conversion and preprocessing, and then further processes it with noise reduction and feature extraction to obtain time-frequency features, which are input into the recognition model for defect identification. At the same time, the walking distance data of the coded pedometer can also be correlated with the time-frequency features to generate a space-signal matrix, thereby obtaining an image of the defect.

[0085] Optionally, in this embodiment, the support columns are attached to the test piece (steel structure), and the travel track is fixed to the support columns on both sides. The testing device is mounted on the travel track via a three-axis displacement beam. The testing device can move freely on the travel track via the three-axis displacement beam, thereby enabling the testing device to move on the test piece along a predetermined route. Specifically, as shown... Figure 4 and Figure 5 As shown, the traveling track 207 is mounted on a base support consisting of feet (fixed-end feet 206 and supporting-end feet 218) and feet brackets 205. A permanent magnet is installed at the bottom of this base support to attract the traveling track onto the test piece 201. The detection device (consisting of a bidirectional ultrasonic guided wave probe 203, an infrared rangefinder 210, and an image acquisition device 211, etc.) can move freely on the traveling track 207 via a three-axis displacement beam 209 to detect weld defects 202 in the test piece 201. Optionally, this embodiment uses a stepper motor combined with a ball screw to drive the three-axis displacement beam 209, thereby realizing the movement of the detection device, as detailed below. Figure 4 As shown, the ball screw 215 is mounted on the screw fixed end 208 and the screw support end 217. The ball screw 215 on the screw fixed end 208 side is connected to the stepper motor 213 through the coupling 214. The coded pedometer 212 is mounted on the stepper motor 213. The stepper motor 213 is mounted on the motor bracket 216. The motor bracket 216 and the screw fixed end 208 are mounted on the fixed end foot post 206. The screw support end 217 is mounted on the support end foot post 218. Both the fixed end foot post 206 and the support end foot post 218 are mounted on the foot post bracket 205.

[0086] It is understood that the triaxial displacement rail beam 209 is a device capable of displacement adjustment in three axial directions. It includes adjustment beams in three axial directions: one extending along the running track 207 (Y direction), one perpendicular to the running track 207 (Z direction), and one perpendicular to the YZ plane (X direction). The bidirectional ultrasonic guided wave probe 203, infrared rangefinder 210, and image acquisition device 211 (i.e., the detection device) are all mounted on the adjustment beam in the X direction. The infrared rangefinder 210 and image acquisition device 211 are located on either side of the bidirectional ultrasonic guided wave probe 203. The three infrared rangefinders 210 emit infrared rays in the three directions to achieve distance detection in each direction. The image acquisition device 211 faces the area where the bidirectional ultrasonic guided wave probe 203 is located to acquire images of the bidirectional ultrasonic guided wave probe 203 for fault location and fault cause analysis, etc.

[0087] Optionally, such as Figure 4 and Figure 5 As shown, the bidirectional ultrasonic guided wave probe 203 used in this embodiment consists of two symmetrically arranged ultrasonic guided wave probes, with the ultrasonic wave emission direction facing the weld (when detecting weld defects). It can be understood that for detecting other defects, the ultrasonic wave emission direction only needs to correspond to the test piece. During detection, the two ultrasonic guided wave probes work simultaneously, detecting the weld on both sides of the probe, thus improving detection efficiency. To prevent the probe from tipping over during movement, a limiting block 204 is set on each side of the bidirectional ultrasonic guided wave probe (i.e., two limiting blocks 204 are set), which restricts the probe's movement during the detection process. Simultaneously, the outer sides of the two limiting blocks 204 are pointed, which can remove debris from the steel structure surface, ensuring that the probe can always perform normal detection work. For the probe, this embodiment sets the two sides of the bidirectional ultrasonic guided wave probe into a ramp shape, allowing it to easily overcome any debris not removed from the surface of the test piece without tripping.

[0088] Optionally, the probe holder is equipped with a movable connection, allowing for the replacement of different types of ultrasonic guided wave probes to meet the inspection needs of different plate thicknesses and different types of welds.

[0089] This embodiment also proposes an example of a directional high-frequency guided wave automatic control system for steel structure weld inspection, such as... Figure 6 As shown, the automatic control system 500 includes:

[0090] The drive module 501 is used to start the bidirectional ultrasonic guided wave probe and the infrared rangefinder, and to control the bidirectional ultrasonic guided wave probe and the infrared rangefinder to move on the running track.

[0091] The detection module 502 uses a pedometer to record the distance traveled by the bidirectional ultrasonic guided wave probe, uses the bidirectional ultrasonic guided wave probe to detect the welds on both sides of the probe, and uses an infrared rangefinder to monitor the relative distance between the main structure of the instrument and the test piece in real time. It determines whether the relative distance is equal to the distance standard value. If it is, the collected waveform signal is imported into the ultrasonic guided wave instrument; otherwise, the correction module 503 is driven to work.

[0092] The correction module 503 calculates the correction amount using a distance correction algorithm and transmits it to the drive module 501, which controls the three-axis displacement rail beam to adjust the relative distance via commands.

[0093] The signal acquisition and analysis module 504 acquires signals from the ultrasonic guided wave instrument, processes them, and displays the waveforms and performs defect imaging.

[0094] In addition, the intelligent recognition module 505 uses an intelligent recognition model to perform multi-dimensional detection of damage information and outputs the summarized damage information.

[0095] The running track is fixed on the foot support, which is attached to the test piece. The bidirectional ultrasonic guided wave probe and infrared rangefinder are both mounted on the running track via a three-axis displacement rail beam.

[0096] Optionally, the automatic control system 500 proposed in this embodiment further includes:

[0097] The anomaly diagnosis module 506 is used to activate the image acquisition device when the waveform cannot be displayed normally, i.e. when the waveform is abnormal. The image acquisition device detects the image of the area where the bidirectional ultrasonic guided wave probe is located, and performs anomaly analysis based on the acquired image, including the abnormal location (abnormal area of ​​the joint coding pedometer positioning signal) and the cause of the anomaly, and provides feedback to facilitate subsequent maintenance work.

[0098] Optionally, the signal acquisition and analysis module 504 in this embodiment includes:

[0099] The processing unit is used to process the acquired signal; specifically, wavelet threshold denoising can be used to eliminate environmental noise, retain the defect reflection wave component, perform short-time Fourier transform on the denoised signal to generate a time spectrum, and extract features such as peak amplitude, peak-to-peak time difference, frequency band energy ratio and waveform entropy.

[0100] The display unit is used for waveform display and defect imaging. Specifically, it displays the changes in signal amplitude of each channel in real time with displacement as the horizontal axis, automatically marks areas where the amplitude exceeds the threshold, and assists in manual verification. It can also associate the position data of the coded pedometer with time-frequency features to generate a space-signal matrix, thereby obtaining an image of the defect.

[0101] Additionally, the alarm unit monitors the tilt angle using a built-in level on the probe. If the tilt angle exceeds a threshold, an alarm is triggered and the location information is recorded.

[0102] The distance correction algorithm, intelligent recognition model, and detection device used in the automatic control system proposed in this embodiment are the same as those in the above-mentioned automatic control method, and will not be repeated here.

[0103] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0104] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0106] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0108] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for automatic control of directional high-frequency guided waves for weld inspection of steel structures, characterized in that, include: Start the bidirectional ultrasonic guided wave probe and infrared rangefinder, and control the bidirectional ultrasonic guided wave probe and infrared rangefinder to move on the running track. The weld seams on both sides of the probe are detected by the bidirectional ultrasonic guided wave probe. The relative distance between the main structure of the instrument and the test piece is monitored in real time using the infrared rangefinder. The relative distance is then determined to be equal to the standard distance value. If it is, subsequent steps are continued; otherwise, a correction amount is calculated using a distance correction algorithm, and the relative distance is adjusted by controlling the three-axis displacement beam via commands. This process includes: using three mutually perpendicular infrared rangefinders to measure the relative distance between the main structure of the instrument and the test piece in three directions; comparing the relative distances in the three directions with their corresponding standard distance values; if any of the real-time monitored relative distances in the three directions deviates from its corresponding standard distance value, the difference between the relative distance value and the standard value is calculated to obtain a distance correction value; and generating control commands based on this distance correction value to control the three-axis displacement beam to move in the corresponding direction, so that the relative distance between the infrared rangefinder and the test piece equals the standard distance value. The travel distance of the bidirectional ultrasonic guided wave probe is recorded by an coded pedometer, and the travel distance and the waveform signal acquired by the bidirectional ultrasonic guided wave probe are imported into the ultrasonic guided wave instrument. Signals are acquired from the ultrasonic guided wave instrument, processed, and then displayed as waveforms and imaged for defects. The processed signal is input into the intelligent recognition model for multi-dimensional detection of damage information, and the damage information is summarized and output. The training process of the intelligent recognition model includes: establishing a guided wave data signal library, which contains a large number of ultrasonic guided wave signals in steel structures with different defect types and different defect degrees; preprocessing the ultrasonic guided wave signals in the guided wave data signal library; performing time-frequency analysis on the preprocessed signals, extracting feature data from the signals, labeling them, and constructing a training dataset; and using the training dataset to train the neural network to obtain the intelligent recognition model. The traveling track is fixed on the foot support, which is attached to the test piece. The bidirectional ultrasonic guided wave probe and the infrared rangefinder are both mounted on the traveling track via a three-axis displacement beam.

2. The method for automatic control of directional high-frequency guided waves for steel structure weld inspection according to claim 1, characterized in that, Also includes: When the waveform display is abnormal, the image acquisition device is activated to detect the image of the area where the bidirectional ultrasonic guided wave probe is located. The abnormality is analyzed based on the acquired image to determine the location and cause of the abnormality. The image acquisition device is mounted on the running track via a three-axis displacement beam.

3. The method for automatic control of directional high-frequency guided waves for steel structure weld inspection according to claim 1, characterized in that, Also includes: The tilt angle is monitored by the built-in level of the bidirectional ultrasonic guided wave probe. If the tilt angle exceeds the threshold, an alarm is triggered, indicating that the detection effect at this position is poor, and the information of this position is recorded.

4. The automatic control method for directional high-frequency guided waves for steel structure weld inspection according to any one of claims 1-3, characterized in that, The bidirectional ultrasonic waveguide probe consists of two symmetrically arranged ultrasonic waveguide probes, with the ultrasonic wave emission directions of the two ultrasonic waveguide probes corresponding to the weld seams on both sides of the bidirectional ultrasonic waveguide probe. The two ultrasonic guided wave probes work synchronously to inspect the welds on both sides respectively; The two sides of the bidirectional ultrasonic guided wave probe are designed with a sloping shape.

5. The method for automatic control of directional high-frequency guided waves for steel structure weld inspection according to claim 4, characterized in that, A limiting block is provided on each side of the bidirectional ultrasonic guided wave probe to restrict the movement of the bidirectional ultrasonic guided wave probe during the moving detection process; The outer sides of the two limiting blocks are pointed to remove debris from the surface of the steel structure.

6. A directional high-frequency guided wave automatic control system for steel structure weld inspection, characterized in that, include: The drive module is used to start the bidirectional ultrasonic guided wave probe and the infrared rangefinder, and to control the bidirectional ultrasonic guided wave probe and the infrared rangefinder to move on the running track. The detection module uses a pedometer to record the travel distance of the bidirectional ultrasonic waveguide probe, uses the bidirectional waveguide probe to detect the welds on both sides of the probe, and uses an infrared rangefinder to monitor the relative distance between the main structure of the instrument and the test piece in real time. It determines whether the relative distance is equal to the standard distance value. If it is, the travel distance and the acquired waveform signal are imported into the ultrasonic waveguide instrument; otherwise, the correction module is activated. The correction module calculates the correction amount using a distance correction algorithm and transmits it to the drive module to generate control commands to control the three-axis displacement beam to adjust the relative distance. This includes: using three mutually perpendicular infrared rangefinders to measure the relative distance between the instrument's main structure and the test piece in three directions; comparing each relative distance in one of the three directions with its corresponding standard distance value; if any of the real-time monitored relative distances in the three directions deviates from its corresponding standard distance value, calculating the difference between the relative distance value and the standard value to obtain a distance correction value; and generating control commands based on this distance correction value to control the three-axis displacement beam to move in the corresponding direction, so that the relative distance between the infrared rangefinders and the test piece equals the standard distance value. The signal acquisition and analysis module acquires signals from the ultrasonic guided wave instrument, processes them, and displays waveforms and performs defect imaging. Furthermore, an intelligent recognition module is included, which utilizes an intelligent recognition model to perform multi-dimensional detection of damage information and outputs the summarized damage information. The training process of the intelligent recognition model includes: establishing a guided wave data signal library, which contains a large number of ultrasonic guided wave signals in steel structures under different defect types and different defect degrees; preprocessing the ultrasonic guided wave signals in the guided wave data signal library; performing time-frequency analysis on the preprocessed signals, extracting feature data from the signals, labeling them, and constructing a training dataset; and training the neural network using the training dataset to obtain the intelligent recognition model. The traveling track is fixed on the foot support, which is attached to the test piece. The bidirectional ultrasonic guided wave probe and the infrared rangefinder are both mounted on the traveling track via a three-axis displacement beam.

7. The directional high-frequency guided wave automatic control system for steel structure weld inspection according to claim 6, characterized in that, Also includes: The anomaly diagnosis module is used to activate the image acquisition device when the waveform display is abnormal, to detect the image of the area where the bidirectional ultrasonic guided wave probe is located through the image acquisition device, and to perform anomaly analysis based on the acquired image to determine the location and cause of the anomaly. The image acquisition device is mounted on the running track via a three-axis displacement beam.

8. The directional high-frequency guided wave automatic control system for steel structure weld inspection according to claim 6, characterized in that, The signal acquisition and analysis module includes: The processing unit is used to process the acquired signals; The display unit is used for waveform display and defect imaging; In addition, the alarm unit monitors the tilt angle using the level built into the bidirectional ultrasonic guided wave probe. If the tilt angle exceeds the threshold, an alarm is triggered to indicate that the detection effect at this position is poor, and the information at this position is recorded.

Images