Online detection system for laser welding defects based on ultrasonic technology

Abstract

The present application relates to the field of ultrasonic detection technology, and specifically to an online detection system for laser welding defects based on ultrasonic technology, the system comprising: a signal acquisition module, which uses ultrasonic TOFD technology to acquire signals and obtains the A-scan signal at the laser welding location, which contains the straight wave, defect wave, bottom wave and noise signal; a signal analysis module, including determining the signal-to-noise ratio of the A-scan signal; obtaining the optimal probe interval; determining the fluctuation difference of the straight wave in the A-scan signal; determining the porosity confidence; and a weld defect detection module, which detects welding defects based on the porosity confidence. The present application detects porosity defects near the surface of the weld and inside the weld, thereby improving the accuracy of welding defect detection.

Description

Technical Field

[0001] The present application relates to the field of ultrasonic detection technology, and in particular to an online detection system for laser welding defects based on ultrasonic technology. Background Art

[0002] Austenitic stainless steel is a material with good mechanical properties and strong corrosion resistance, and is widely used in many industries. However, the use conditions of stainless steel are usually relatively harsh, and the weld is the weakest area of ​​the complete stainless steel component. Due to the influence of the laser welding process and the harsh environment of high temperature, high pressure, corrosion, and radioactivity, the weld is prone to damage. Timely defect detection of stainless steel welds is an important guarantee for the safe operation of stainless steel equipment. Ultrasonic testing, as a non-destructive testing method, has become a common means of non-destructive testing compared with other methods, with its advantages of strong penetration, no pollution, convenience, and harmlessness to the human body. The principle of ultrasonic testing is to transmit ultrasonic waves to the inside of the material being tested through the transmitting probe, and use the receiving probe to receive the reflected echo reflected by the material. By observing and analyzing the delay and attenuation of the reflected echo, it is possible to determine whether there are defects inside the material being tested and to locate the location, size, and other information of the defects.

[0003] Ultrasonic TOFD (Time of Flight Diffraction Ultrasonic Testing) technology has the characteristics of fast detection efficiency and high detection accuracy, and is widely used in non-destructive testing scenarios of welds. However, when using TOFD technology to detect welds, if the porosity defect is very close to the weld surface, the defect wave of the porosity defect in the A-type scanning signal obtained by TOFD is superimposed on the straight wave of the weld surface, which is easy to cause missed detection and false detection, affecting the detection accuracy. Summary of the invention

[0004] In order to solve the above technical problems, an online detection system for laser welding defects based on ultrasonic technology is provided to solve the existing problems.

[0005] The solution to the technical problem of this application is to provide an online detection system for laser welding defects based on ultrasonic technology, the system comprising:

[0006] The signal acquisition module uses ultrasonic TOFD technology to collect signals and obtain the A-scan signal of the laser welding point, which includes the straight wave, defect wave, bottom wave and noise signal;

[0007] The signal analysis module analyzes the porosity defects on the weld surface and inside the weld through the signal amplitude in the A-scan signal, including:

[0008] (1) Determine the signal-to-noise ratio of the A-scan signal based on the difference between the amplitude of all signals in the A-scan signal and the amplitude of the noise signal; obtain the optimal probe interval based on the signal-to-noise ratio of the A-scan signal at different probe intervals;

[0009] (2) Determine the fluctuation difference of the direct wave in the A-scan signal based on the distribution of the amplitude of the direct wave in the A-scan signal each time the probe moves;

[0010] (3) determining the pore confidence according to the difference between the signal amplitudes of the A-scan signal of each probe movement and the comparison A-scan signal, and the fluctuation difference;

[0011] The weld defect detection module detects welding defects based on the porosity confidence.

[0012] Preferably, determining the signal-to-noise ratio of the A-scan signal comprises:

[0013] The maximum amplitudes of the defect wave signal and the noise signal are obtained respectively, which are recorded as the maximum defect amplitude and the maximum noise amplitude;

[0014] Calculate the mean of all signal amplitudes in the A-scan signal and the mean of all amplitudes of the noise signal respectively, and record them as the signal average amplitude and the noise average amplitude;

[0015] Recording the ratio of the maximum defect amplitude to the maximum noise amplitude as a first ratio;

[0016] Recording the ratio of the signal average amplitude to the noise average amplitude as a second ratio;

[0017] The first ratio and the second ratio are fused to determine the signal-to-noise ratio of the A-scan signal.

[0018] Preferably, the method for obtaining the optimal probe interval is: calculating the signal-to-noise ratios of all A-scan signals obtained under different probe intervals, and taking the probe interval distance corresponding to all the maximum values ​​of the signal-to-noise ratios as the optimal probe interval.

[0019] Preferably, determining the fluctuation difference of the through wave in the A-scan signal includes:

[0020] Move the probe at the optimal probe interval along the length direction of the weld to obtain multiple moving A-scan signals;

[0021] Clustering the maximum amplitude of the through wave in the A-scan signal of each movement to obtain multiple clusters;

[0022] According to the maximum amplitude of each direct wave and the membership status between each cluster, the membership index of each direct wave is determined;

[0023] The product of the mean of the membership index of all direct waves in each cluster and the degree of dispersion is used as the discrete change index of each cluster;

[0024] The discrete change index and the number of all clusters are combined to determine the fluctuation change index of each cluster;

[0025] The fluctuation variation indexes of all clusters are fused to determine the fluctuation difference of the through wave in the A-scan signal.

[0026] Preferably, the determining of the membership index of each direct wave includes:

[0027] The sum of the membership between each direct wave and all clusters is recorded as the first membership value;

[0028] Count the number of direct waves with a membership degree greater than a preset threshold to all clusters, and record it as the membership number of each direct wave;

[0029] The ratio between the number of all clusters and the membership number is recorded as the second membership value;

[0030] The first membership value and the second membership value are merged to determine the membership index of each direct wave.

[0031] Preferably, the determining of the pore confidence comprises:

[0032] According to the difference in propagation time between the defect wave signal and the through wave signal, the welding defect factor of the A-scan signal in each movement is calculated;

[0033] Obtain the peak value of the signal amplitude in the A-scan signal of each movement; count the number of peaks of the A-scan signal in each movement;

[0034] Select the A-scan signal of any movement during all the movements and record it as the comparison A-scan signal;

[0035] Determine the signal difference coefficient of each movement according to the difference in signal amplitude, number of peaks, and signal change trend between the A-scan signal of each movement and the comparison A-scan signal;

[0036] According to the discrete degree of all signal difference coefficients, the first confidence level of the pore defect is determined;

[0037] The first confidence level is combined with the fluctuation difference of the through wave in the A-scan signal to determine the pore confidence level.

[0038] Preferably, the calculation of the welding defect factor of the A-scan signal in each movement includes:

[0039] The ultrasonic diffraction time difference method is used to calculate the propagation time of the defect wave signal and the direct wave signal;

[0040] Obtain the maximum amplitude between the defect wave signal and the bottom wave signal, and record it as the maximum amplitude of the signal;

[0041] If the propagation time of the defect wave is equal to the propagation time of the through wave, the welding defect factor is assigned a value of 0; otherwise, the welding defect factor is the maximum amplitude of the signal.

[0042] Preferably, the signal difference coefficient of each movement includes:

[0043] Determine the overall signal difference of each movement according to the difference between the range of the signal amplitude of the A-scan signal of each movement and the comparison A-scan signal, and the difference of the welding defect factor;

[0044] The difference between the number of peaks of the A-scan signal of each movement and the comparison A-scan signal is recorded as the local signal difference of each movement;

[0045] Analyzing the difference between the A-scan signal of each movement and all signal amplitudes of the comparison A-scan signal to determine the difference in signal trend of each movement;

[0046] The overall signal difference, the local signal difference and the signal trend change difference are integrated to determine the signal difference coefficient for each movement.

[0047] Preferably, determining the overall signal difference for each movement comprises:

[0048] Calculate the range of the signal amplitude of the A-scan signal in each movement;

[0049] Determine the difference between the range of the A-scan signal of each movement and the comparison A-scan signal as the range difference of each movement;

[0050] Determine the difference between the welding defect factor of the A-scan signal of each movement and the comparison A-scan signal as the defect impact difference of each movement;

[0051] The defect impact difference is fused with the range difference to determine the overall signal difference for each move.

[0052] Preferably, the detection of welding defects includes: normalizing the porosity confidence, when the normalized result is less than a preset defect threshold, there is no porosity defect at the weld, otherwise, there is a porosity defect at the weld.

[0053] This application has at least the following beneficial effects:

[0054] The present application determines the signal-to-noise ratio of the A-scan signal by analyzing the difference between all signal amplitudes and the noise signal amplitude in the A-scan signal; determines the optimal probe interval for the signal-to-noise ratio of the A-scan signal at different probe intervals, which has the beneficial effect of avoiding the influence of the growth direction of the columnar crystal on the ultrasonic defect detection to the greatest extent and improving the detection accuracy; by moving the probe and determining the fluctuation difference of the straight wave according to the distribution of the amplitude of the straight wave in the A-scan signal during each movement process, which has the beneficial effect of taking into account the situation when the porosity defect appears near the surface of the weld, the porosity defect near the surface is detected; secondly, according to the difference between the signal amplitudes of the A-scan signal of each probe movement, the signal difference coefficient of each movement is calculated to determine the porosity confidence, which has the beneficial effect of analyzing the situation when the porosity defect appears inside the weld, detecting the porosity defect inside the weld, and solving the problem that only the defects inside the weld are analyzed during TOFD detection, but the porosity defects near the surface are ignored, thereby improving the accuracy of welding defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The following is a further detailed description of the laser welding defect online detection system based on ultrasonic technology of the present application in conjunction with the accompanying drawings.

[0056] Figure 1 A block diagram of an online detection system for laser welding defects based on ultrasonic technology provided by an embodiment of the present application;

[0057] Figure 2 A schematic diagram of an A-scan signal provided by an embodiment of the present application;

[0058] Figure 3 A schematic diagram of the position of an ultrasonic probe provided in one embodiment of the present application;

[0059] Figure 4 A flowchart of the steps of a signal analysis module provided in one embodiment of the present application;

[0060] Figure 5 A flowchart of a method for obtaining the fluctuation difference of a through wave in an A-scan signal provided by an embodiment of the present application;

[0061] Figure 6 A flowchart of the steps of a method for obtaining pore confidence provided in one embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of this application more clear, the following is a further detailed description of the online detection system for laser welding defects based on ultrasonic technology proposed in this application in conjunction with the accompanying drawings and implementation examples. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0063] Unless defined otherwise, 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 application belongs.

[0064] See also Figure 1 , which shows a block diagram of an online laser welding defect detection system based on ultrasonic technology provided by an embodiment of the present application. The system includes: a signal acquisition module, a signal analysis module, and a welding defect detection module.

[0065] The signal acquisition module uses ultrasonic TOFD technology to acquire A-scan signals at laser welding locations, including through waves, defect waves, bottom waves, and noise signals.

[0066] The broadband square wave pulse generator is used to transmit and receive pulse energy. The transmitting probe and the receiving probe are placed on both sides of the weld area (the area constructed by the maximum width of the weld) of the welded object to be detected. That is, the weld area is located in the center of the two probes. The initial horizontal interval between the two probes is 2 ( The maximum width of the weld is taken), and the ultrasonic signal is collected, stored and processed to obtain the A-scan signal during the TOFD detection process, including the straight-through wave, defect wave and bottom wave. Other signals except the straight-through wave, defect wave and bottom wave are recorded as noise signals.

[0067] Further, the A scan signal schematic diagram provided in the embodiment of the present application is as follows Figure 2 shown.

[0068] Among them, the through wave refers to the ultrasonic wave emitted from the transmitting probe and reaching the receiving probe along the surface of the welding object without passing through any defects or materials; the defect wave means that when the ultrasonic wave encounters defects or other irregular shapes in the material, the wave will diffract at the edge of the defect; the bottom wave refers to the signal that the ultrasonic wave propagates to the bottom of the material and is reflected back from the bottom.

[0069] Preferably, as an embodiment of the present application, the welding object is austenitic stainless steel material; the transmitting probe and the receiving probe are ultrasonic TOFD detection probes with a center frequency of 10MHz and a chip diameter of 3mm, the front edge of the probe is 8mm, and the probe wedge is 60°; during the acquisition of ultrasonic signals, the sampling frequency is 60MHz; secondly, the schematic diagram of the probe position is as shown in Figure 3 As shown, Z is the welding object, HF is the welding area, FS and JS are the transmitting and receiving probes respectively. is the initial separation distance between the two probes.

[0070] At this point, the A-scan signal of the laser welding point is obtained, including the through wave, defect wave, bottom wave and noise signal.

[0071] The signal analysis module analyzes the porosity defects on the weld surface and inside the weld through the signal amplitude in the A-scan signal.

[0072] The flowchart of the steps of the signal analysis module provided by the embodiment of the present application is as Figure 4 shown.

[0073] Step 1: Determine the signal-to-noise ratio of the A-scan signal according to the difference between all signal amplitudes and the noise signal amplitude in the A-scan signal; determine the optimal probe interval for the signal-to-noise ratio of the A-scan signal at different probe intervals.

[0074] Through the analysis of austenitic stainless steel materials, it is found that the grains inside austenitic stainless steel materials will form coarse columnar crystal structures when heated during laser welding. The columnar crystals have the same growth direction in a certain area of the weld, while the growth directions of the columnar crystals in different areas are different, so that the austenitic stainless steel weld shows anisotropic characteristics as a whole. As a result, when detecting the stainless steel weld by ultrasonic waves, its material structure will cause adverse effects such as rapid attenuation of the ultrasonic beam, large noise influence, and distortion of the sound wave propagation direction, seriously affecting the accuracy of ultrasonic TOFD detection. Therefore, it is necessary to analyze the influence degree of the weld characteristics of austenitic stainless steel on the propagation of ultrasonic waves.

[0075] Furthermore, if there are no porosity defects in the welding area, the receiving probe will only receive two pulse signals, namely the direct wave and the bottom wave. If there are porosity defects, the receiving probe will also receive the pulse signal of the defect wave of the defect; in order to locate each characteristic wave in the A-scan signal of ultrasonic waves, such as the direct wave, the defect wave, and the bottom wave, a welding defect factor is constructed to preliminarily judge whether there are defects at the stainless steel weld after laser welding. Specifically:

[0076] According to the acoustic delay a in the probe wedge, the longitudinal wave sound velocity c, the defect depth h, and the thickness H of the stainless steel to be detected, where the defect depth is the initial depth obtained by ultrasonic TOFD detection at the initial interval 2 between the two probes. The calculation method of the ultrasonic detection depth is a well-known technology and will not be elaborated here;

[0077] Adopt the ultrasonic time-of-flight diffraction method to calculate the propagation time of the diffraction waves at both ends of the defect when the defect depth is d , where the ultrasonic time-of-flight diffraction method is a well-known technology and will not be elaborated here.

[0078] It should be noted that when 0 < d < H, is the propagation time of the defect wave at the defect end with a depth of ; represents the propagation time of the direct wave when represents the propagation time of the bottom wave.

[0079] The ultrasonic diffraction time difference method is used to calculate the propagation time of the defect wave signal and the direct wave signal;

[0080] Obtain the maximum amplitude between the defect wave signal and the bottom wave signal, and record it as the maximum amplitude of the signal;

[0081] If the propagation time of the defect wave is equal to the propagation time of the straight wave, the welding defect factor is assigned a value of 0; if the propagation time of the defect wave is not equal to the propagation time of the straight wave, the welding defect factor is the maximum amplitude of the signal;

[0082] It should be noted that if the propagation time of the defect wave is equal to the propagation time of the straight wave, there is no porosity defect in the weld. If the propagation time of the defect wave is not equal to the propagation time of the straight wave, it means that there is a defect in the weld. Therefore, a larger weight is given to analyze the impact of the porosity defect.

[0083] Furthermore, in order to analyze the influence of the growth direction of columnar crystals in stainless steel welds on the current probe spacing, the signal-to-noise ratio between the defect wave signal and the noise signal is constructed, specifically:

[0084] The maximum amplitudes of the defect wave signal and the noise signal are obtained respectively, which are recorded as the maximum defect amplitude and the maximum noise amplitude;

[0085] Calculate the mean of all signal amplitudes and the mean of the noise signal amplitude in the A-scan signal, and record them as the signal average amplitude and the noise average amplitude;

[0086] Recording the ratio of the maximum defect amplitude to the maximum noise amplitude as a first ratio;

[0087] Recording the ratio of the signal average amplitude to the noise average amplitude as a second ratio;

[0088] The first ratio and the second ratio are combined to determine the signal-to-noise ratio of the A-scan signal;

[0089] It is understandable that the fusion relationship can be specifically expressed as a multiplication relationship, an addition relationship, and the idempotence of an exponential function, and the present application does not impose any special limitation on this.

[0090] Preferably, as an embodiment of the present application, the product of the first ratio and the second ratio is used as the signal-to-noise ratio of the A-scan signal.

[0091] It should be noted that if the value of the signal-to-noise ratio is larger, it means that the attenuation degree of the columnar crystal on the incident direction of the sound beam at the current interval is smaller and the interference degree of the scattering noise is smaller, and the detection effect of TOFD is better at this time. If the value of the signal-to-noise ratio is smaller, it means that the columnar crystal has a greater impact on the incident direction of the sound beam at the current interval, and it is necessary to adjust the interval between the probes to achieve more accurate laser welding defect detection.

[0092] Furthermore, by adjusting the interval between the probes to find the optimal probe interval, the influence of the columnar crystal growth direction on ultrasonic defect detection is avoided, thereby improving the detection accuracy. Specifically:

[0093] Starting from the initial position of the probe, the horizontal spacing distance between the two probes is increased or decreased in sequence according to the adjustment interval of d, and the signal-to-noise ratio of the A-scan signal collected at each probe interval is calculated, where the maximum spacing distance between the two probes is b times the maximum width of the weld, and the maximum spacing distance between the two probes does not exceed the width of the stainless steel;

[0094] According to the signal-to-noise ratio of the A-scan signal under different probe intervals, the probe interval distance corresponding to the maximum signal-to-noise ratio of all A-scan signals is taken as the optimal probe interval;

[0095] Preferably, as an embodiment of the present application, the horizontal spacing distance between the two probes is adjusted according to an adjustment distance of 1 mm, wherein the maximum spacing distance between the two probes is taken as 8 times the maximum width of the weld. As other implementation methods, the implementer can set it according to the actual situation.

[0096] It should be noted that when adjusting the probe interval, only the horizontal interval is adjusted, the corresponding angle of the probe remains unchanged, and the two probes are still on the same horizontal line.

[0097] At this point, the optimal probe spacing is obtained.

[0098] Step 2: Determine the fluctuation difference of the direct wave in the A-scan signal according to the distribution of the amplitude of the direct wave in the A-scan signal each time the probe moves.

[0099] Because laser welding is automated by industrial machines, which are stable, even if there are pores inside the weld after welding, the depth of the pore defects is relatively consistent, which is reflected in the A-scan signal as a mixture of the straight-through wave and the defect wave. Although the waveform and amplitude of the straight-through wave change, the degree of amplitude change is relatively consistent.

[0100] Furthermore, the probe is moved along the length direction of the weld to detect the porosity defects near the surface of the stainless steel, specifically:

[0101] For the probe at the optimal probe interval, move the probe along the length direction of the weld, perform TOFD-A scanning on the weld, and obtain multiple moving A-scan signals;

[0102] Preferably, as an embodiment of the present application, the moving step length when the probe is moved is 1 mm. As other implementation modes, the implementer can set it according to the actual situation.

[0103] Furthermore, according to the characteristic that the depth of pore defects near the surface is relatively consistent, the amplitudes of all direct waves are clustered to analyze the membership index of each direct wave, which is as follows:

[0104] Clustering the maximum amplitude of the through wave in the A-scan signal of each movement to obtain multiple clusters;

[0105] Preferably, as an embodiment of the present application, a fuzzy mean clustering algorithm is used for clustering, and the number of clusters is obtained by the elbow rule. The fuzzy mean clustering algorithm and the elbow rule are well-known technologies and will not be described in detail.

[0106] It is understandable that this embodiment uses the fuzzy mean clustering algorithm for fuzzy clustering. As other implementation methods, implementers can use other methods in the prior art, such as K-means clustering, hierarchical clustering algorithm, etc. This embodiment does not impose any special restrictions on this.

[0107] It should be noted that the membership of each element can be directly obtained through the fuzzy mean clustering algorithm. When other clustering methods are used for clustering, the membership of each element can be measured by calculating the distance between each element and the cluster center of all clusters.

[0108] The sum of the membership between each direct wave and all clusters is recorded as the first membership value;

[0109] Count the number of direct waves with a membership degree greater than a preset threshold to all clusters, and record it as the membership number of each direct wave;

[0110] The ratio between the number of all clusters and the membership number is recorded as the second membership value;

[0111] The first membership value and the second membership value are merged to determine the membership index of each direct wave;

[0112] Preferably, as an embodiment of the present application, the preset threshold value is 0.5; the product of the first membership value and the second membership value is used as the membership index of each direct wave.

[0113] It should be noted that if there are no defects in the weld after laser welding, no matter how the probe moves along the length of the weld, because the propagation path of the ultrasonic wave is consistent, the arrival time and corresponding amplitude of the direct wave and the bottom wave are relatively consistent, then the number of clusters should be 1; and the degree of membership of each direct wave to each cluster is high; secondly, if the degree of membership of each direct wave to a cluster is greater than the preset threshold value of 0.5, it indicates that the direct wave has a membership relationship with the cluster.

[0114] Furthermore, based on the membership index of each straight wave, the fluctuation difference of the straight wave in the A-scan signal during all movements is determined to reflect the possibility of the existence of porosity defects near the weld surface, specifically:

[0115] The product of the mean of the membership index of all direct waves in each cluster and the degree of dispersion is used as the discrete change index of each cluster;

[0116] Preferably, as an embodiment of the present application, the degree of dispersion can be measured by variance and standard deviation, and for other implementations, the implementer can adopt other methods, such as coefficient of variation, information entropy, etc., and this embodiment does not impose any special restrictions on this.

[0117] The discrete change index and the number of all clusters are combined to determine the fluctuation change index of each cluster;

[0118] The fluctuation change indexes of all clusters are merged to determine the fluctuation difference of the through wave in the A-scan signal;

[0119] Preferably, as an embodiment of the present application, the product of the discrete change index and the number of all clusters is used as the fluctuation change index of each cluster; secondly, the sum of the fluctuation change indexes of all clusters is used as the fluctuation difference of the through wave in the A-scan signal.

[0120] It should be noted that if there is no porosity defect, the amplitude of the straight wave is consistent, the number of clusters is 1, and the membership index of each straight wave in each cluster is relatively consistent, the mean is also small, and the fluctuation difference is small; if there is a porosity defect, the number of clusters is greater than or equal to 2, and the mean of the membership index of all straight waves in each cluster is large, then the fluctuation difference is large. Therefore, the greater the fluctuation difference, the greater the possibility of the presence of porosity defects near the surface of austenitic stainless steel.

[0121] Further, the step flow chart of the method for obtaining the fluctuation difference of the through wave in the A-scan signal provided in the embodiment of the present application is as follows: Figure 5 shown.

[0122] At this point, the fluctuation difference of the through wave in the A-scan signal is obtained.

[0123] Step 3: Determine the pore confidence according to the difference between the signal amplitudes of the A-scan signal of each probe movement and the A-scan signal of the comparison A-scan signal, and the fluctuation difference.

[0124] Since porosity defects are holes caused by the failure of gas to be discharged in time during high-temperature welding, although the depth of the pores is relatively consistent, the pore size and pore spacing are not consistent, and it is possible that near-surface porosity defects and internal defects of the weld are scanned at the same time. Therefore, when the probe moves, the corresponding amplitude of the defect wave in the signal obtained by TOFD-A scanning will fluctuate to a certain extent.

[0125] Furthermore, by comparing the differences between the A-scan signals scanned each time the probe moves, the signal difference coefficient is determined to reflect the possibility of the presence of porosity defects inside the weld, specifically:

[0126] Obtain the peak value of the signal amplitude in the A-scan signal of each movement;

[0127] Preferably, as an embodiment of the present application, the AMPD algorithm is used to obtain the peak. As other implementation methods, the implementer can use other methods, such as the derivative method, the local extreme value method, etc., and this embodiment does not impose any special restrictions on this.

[0128] Calculate the welding defect factor of the A-scan signal in each movement; count the number of peaks of the A-scan signal in each movement;

[0129] Calculate the range between the signal amplitudes of the A-scan signal in each movement;

[0130] Select the A-scan signal of any movement during all the movements and record it as the comparison A-scan signal;

[0131] Preferably, as an embodiment of the present application, the A-scan signal of the first movement in all movement processes is selected as the comparison A-scan signal. As other implementation methods, the implementer can determine the comparison A-scan signal by himself, and this embodiment does not impose any special restrictions on this.

[0132] Determine the difference of the welding defect factor between the A-scan signal and the A-scan signal of each movement as the defect influence difference of each movement;

[0133] Preferably, as an embodiment of the present application, the absolute value of the difference between the A-scan signal of each movement and the welding defect factor of the comparison A-scan signal is used as the defect impact difference of each movement.

[0134] Determine the range difference of each movement according to the difference between the A-scan signal of each movement and the range of the comparison A-scan signal;

[0135] Merging the defect impact difference with the range difference to determine the overall signal difference for each movement;

[0136] Preferably, as an embodiment of the present application, the absolute value of the difference between the A-scan signal of each movement and the range of the comparison A-scan signal is taken as the defect impact difference of each movement; secondly, the sum of the defect impact difference and the range difference is taken as the overall signal difference of each movement.

[0137] The difference between the number of peaks of the A-scan signal of each movement and the comparison A-scan signal is recorded as the local signal difference of each movement;

[0138] Preferably, as an embodiment of the present application, the absolute value of the difference between the A-scan signal of each movement and the number of peaks of the comparison A-scan signal is used as the local signal difference of each movement.

[0139] Analyzing the difference between the A-scan signal of each movement and all signal amplitudes of the comparison A-scan signal to determine the difference in signal trend of each movement;

[0140] Preferably, as an embodiment of the present application, the DTW distance between the A-scan signal of each movement and all signal amplitudes of the comparison A-scan signal is taken as the signal trend difference of each movement; as other implementation methods, the implementer can adopt other existing calculation methods, such as cosine similarity, etc., and this embodiment does not impose any special restrictions on this.

[0141] The overall signal difference, the local signal difference and the signal trend change difference are integrated to determine the signal difference coefficient for each movement;

[0142] Preferably, as an embodiment of the present application, the product of the overall signal difference, the local signal difference and the signal trend change difference is used to determine the signal difference coefficient for each movement.

[0143] It should be noted that if the fluctuation trend of the A-scan signal of each movement is relatively consistent with the fluctuation trend of the comparison A-scan signal, indicating that no defect is scanned, the overall signal difference, local signal difference and signal trend change difference are small, and the signal difference coefficient is smaller; if a pore defect is scanned, the defect wave caused by the pore defect fluctuates greatly, and the corresponding amplitude is greater than the through wave and the bottom wave, then the overall signal difference, local signal difference and signal trend change difference will all be large, and the signal difference coefficient will also be larger; therefore, the larger the signal difference coefficient, the greater the possibility of scanning a pore defect.

[0144] Furthermore, the porosity confidence is determined according to the signal difference coefficient of each movement to reflect the overall defect level inside the weld of austenitic stainless steel, specifically:

[0145] According to the discrete degree of all signal difference coefficients, the first confidence level of the pore defect is determined;

[0146] The first confidence level is combined with the fluctuation difference of the through wave in the A-scan signal to determine the pore confidence level;

[0147] Preferably, as an embodiment of the present application, the first confidence level of the pore defect is the standard deviation of the signal difference coefficients of all scans; the product of the first confidence level and the fluctuation difference of the through wave in the A-scan signal is used as the pore confidence level.

[0148] Further, the step flow chart of the method for obtaining the pore confidence provided in the embodiment of the present application is as follows: Figure 6 shown.

[0149] It should be noted that the larger the first confidence level is, the larger the porosity confidence level is, which means the greater the possibility of porosity defects inside the weld.

[0150] At this point, the pore confidence is obtained.

[0151] The weld defect detection module detects welding defects based on the porosity confidence.

[0152] Based on the porosity confidence level, the stainless steel welds are evaluated for defects, specifically:

[0153] The porosity confidence is normalized. If the normalized result is less than the preset defect threshold, it is determined that there is no porosity defect at the weld. If the normalized result is greater than or equal to the preset defect threshold, it is determined that there is a porosity defect at the weld, thereby realizing the detection of laser welding defects.

[0154] Preferably, as an embodiment of the present application, the preset defect threshold value is 0.5.

[0155] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0156] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0157] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the present application. It should be pointed out that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the content of the technical solution of the present application, shall fall within the protection scope of the technical solution of the present application.

Claims

1. The online detection system for laser welding defects based on ultrasonic technology is characterized by: The system comprises: The signal acquisition module uses ultrasonic TOFD technology to collect signals and obtain the A-scan signal of the laser welding point, which includes the straight wave, defect wave, bottom wave and noise signal; The signal analysis module analyzes the porosity defects on the weld surface and inside the weld through the signal amplitude in the A-scan signal, including: (1) Determine the signal-to-noise ratio of the A-scan signal based on the difference between the amplitude of all signals in the A-scan signal and the amplitude of the noise signal; obtain the optimal probe interval based on the signal-to-noise ratio of the A-scan signal at different probe intervals; (2) Determine the fluctuation difference of the direct wave in the A-scan signal based on the distribution of the amplitude of the direct wave in the A-scan signal each time the probe moves; (3) determining the pore confidence level according to the difference between the signal amplitudes of the A-scan signal of each probe movement and the comparison A-scan signal, as well as the fluctuation difference; A weld defect detection module, which detects welding defects based on the porosity confidence; Determining the signal-to-noise ratio of the A-scan signal includes: The maximum amplitudes of the defect wave signal and the noise signal are obtained respectively, which are recorded as the maximum defect amplitude and the maximum noise amplitude; Calculate the mean of all signal amplitudes in the A-scan signal and the mean of all amplitudes of the noise signal respectively, and record them as the signal average amplitude and the noise average amplitude; Recording the ratio of the maximum defect amplitude to the maximum noise amplitude as a first ratio; Recording the ratio of the signal average amplitude to the noise average amplitude as a second ratio; The first ratio and the second ratio are combined to determine the signal-to-noise ratio of the A-scan signal; The step of determining the fluctuation difference of the through wave in the A-scan signal includes: Move the probe at the optimal probe interval along the length direction of the weld to obtain multiple moving A-scan signals; Clustering the maximum amplitude of the through wave in the A-scan signal of each movement to obtain multiple clusters; According to the maximum amplitude of each direct wave and the membership status between each cluster, the membership index of each direct wave is determined; The product of the mean of the membership index of all direct waves in each cluster and the degree of dispersion is used as the discrete change index of each cluster; The discrete change index and the number of all clusters are combined to determine the fluctuation change index of each cluster; The fluctuation change indexes of all clusters are merged to determine the fluctuation difference of the through wave in the A-scan signal; The step of determining the pore confidence level comprises: According to the difference in propagation time between the defect wave signal and the through wave signal, the welding defect factor of the A-scan signal in each movement is calculated; Obtain the peak value of the signal amplitude in the A-scan signal of each movement; count the number of peaks of the A-scan signal in each movement; Select the A-scan signal of any movement during all the movements and record it as the comparison A-scan signal; Determine the signal difference coefficient of each movement according to the difference in signal amplitude, number of peaks, and signal change trend between the A-scan signal of each movement and the comparison A-scan signal; According to the discrete degree of all signal difference coefficients, the first confidence level of the pore defect is determined; The first confidence level is combined with the fluctuation difference of the through wave in the A-scan signal to determine the pore confidence level.

2. The laser welding defect online detection system based on ultrasonic technology as claimed in claim 1 is characterized in that: The method for obtaining the optimal probe interval is: calculating the signal-to-noise ratios of all A-scan signals obtained under different probe intervals, and taking the probe interval distances corresponding to the maximum values ​​of all the signal-to-noise ratios as the optimal probe interval.

3. The online detection system for laser welding defects based on ultrasonic technology as claimed in claim 1 is characterized in that: The determining of the membership index of each direct wave comprises: The sum of the membership between each direct wave and all clusters is recorded as the first membership value; Count the number of direct waves with a membership degree greater than a preset threshold to all clusters, and record it as the membership number of each direct wave; The ratio between the number of all clusters and the membership number is recorded as the second membership value; The first membership value and the second membership value are merged to determine the membership index of each direct wave.

4. The online detection system for laser welding defects based on ultrasonic technology as claimed in claim 1, characterized in that: The calculation of the welding defect factor of the A-scan signal in each movement includes: The ultrasonic diffraction time difference method is used to calculate the propagation time of the defect wave signal and the direct wave signal; Obtain the maximum amplitude between the defect wave signal and the bottom wave signal, and record it as the maximum amplitude of the signal; If the propagation time of the defect wave is equal to the propagation time of the through wave, the welding defect factor is assigned a value of 0; otherwise, the welding defect factor is the maximum amplitude of the signal.

5. The laser welding defect online detection system based on ultrasonic technology as claimed in claim 1 is characterized in that: The signal difference coefficient of each movement includes: Determine the overall signal difference of each movement according to the difference between the range of the signal amplitude of the A-scan signal of each movement and the comparison A-scan signal, and the difference of the welding defect factor; The difference between the number of peaks of the A-scan signal of each movement and the comparison A-scan signal is recorded as the local signal difference of each movement; Analyzing the difference between the A-scan signal of each movement and all signal amplitudes of the comparison A-scan signal to determine the difference in signal trend of each movement; The overall signal difference, the local signal difference and the signal trend difference are integrated to determine the signal difference coefficient for each movement.

6. The ultrasonic-based laser welding defect online detection system according to claim 5, characterized in that: Determining the overall signal difference for each movement includes: Calculate the range of the signal amplitude of the A-scan signal in each movement; Determine the difference between the range of the A-scan signal of each movement and the comparison A-scan signal as the range difference of each movement; Determine the difference between the welding defect factor of the A-scan signal of each movement and the comparison A-scan signal as the defect impact difference of each movement; The defect impact difference is fused with the range difference to determine the overall signal difference for each move.

7. The online detection system for laser welding defects based on ultrasonic technology as claimed in claim 1, characterized in that: The detection of welding defects includes: normalizing the porosity confidence, when the normalized result is less than a preset defect threshold, there is no porosity defect at the welding position, otherwise, there is a porosity defect at the welding position.

Images