Inspection method for welds after heat treatment for large evaporator heating chambers

Abstract

The present application relates to the field of ultrasonic detection technology, and specifically to a method for inspecting welds after heat treatment in a large evaporator heating chamber, the method comprising: obtaining echo signals of probes at different angles on both sides of the weld; reordering the echo signals of probes at the same angle; performing curve fitting on the echo signals of all angle probes based on the weld width to obtain the projection curves of each probe on both sides of the weld; completing the data points of each probe based on the projection curve; comprehensively analyzing the distribution characteristics and numerical differences between the aligned echo signals and the aligned reverse order echo signals, and combining the data distribution characteristics between the corresponding projection curves to obtain a comprehensive pseudo-defect evaluation; reconstructing the projection curves of all probes without pseudo-defect waves, and combining with a neural network to obtain the inspection results of the weld after heat treatment. The present application aims to eliminate pseudo-defect waves, improve the quality of echo signals, and improve the accuracy of inspection.

Description

Technical Field

[0001] The present application relates to the technical field of ultrasonic testing, and in particular to a method for testing welds after heat treatment in a heating chamber of a large evaporator. Background Art

[0002] In the production process of large evaporator heating chambers, welding technology is needed to connect multiple metal parts that make up the heating chamber, such as the heating chamber wall composed of multiple metal plates. Heat treatment refers to the heating and cooling of the weld joint and its surrounding areas during or after welding to improve the properties of the metal and the welding quality. Heat treatment can improve the strength, toughness and fatigue resistance of the welding position, but it may also introduce crack defects. Crack defects may cause brittle failure of the weld during subsequent use, resulting in losses. Therefore, the internal quality of the weld needs to be tested after heat treatment.

[0003] In order to avoid secondary damage to the heating chamber structure during testing, non-destructive testing methods are required when testing welds after heat treatment, among which ultrasonic testing is one of the commonly used non-destructive testing methods. Ultrasonic testing has the advantages of high sensitivity, strong penetration, fast inspection speed, low cost, light equipment and harmlessness to the human body. Since ultrasonic testing is very sensitive, pseudo-defect waves may appear in the waveform, thus affecting the accuracy of subsequent crack defect detection results. Summary of the invention

[0004] In view of the above, it is necessary to provide a post-heat treatment inspection method for welds in large evaporator heating chambers to solve the above problems.

[0005] One embodiment of the present application provides a method for inspecting welds after heat treatment for a large evaporator heating chamber, the method comprising:

[0006] A probe with a preset angle is used to obtain echo signals on the corresponding detection path on both sides of the weld of the evaporator heating chamber;

[0007] The echo signals of each probe on one side of the weld are rearranged to obtain the reverse echo signals of each probe; the weld width is evenly divided, and according to the projection of the detection path corresponding to each probe on the weld width, the echo signals and reverse echo signals of each probe corresponding to each weld width interval are curve fitted to obtain the projection curves of each probe on both sides of the weld corresponding to each weld width interval; the data points in the echo signals and reverse echo signals of each probe are aligned according to the projection curves;

[0008] The evaluation of transverse pseudo-defects in each weld width interval is obtained by combining the distribution characteristics and numerical differences between the echo signals of all probes after alignment and the reverse echo signals after alignment corresponding to each weld width interval;

[0009] According to the fitting results of the projection curves of the echo signals of each probe corresponding to each weld width interval and the remaining probes on one side of the weld and the variation characteristics of the slope of each data point, combined with the transverse pseudo-defect evaluation, a comprehensive pseudo-defect evaluation of the echo signals of each probe corresponding to each weld width interval on one side of the weld is obtained;

[0010] Based on the comprehensive pseudo-defect evaluation, it is judged whether there is a pseudo-defect wave in the projection curve of each probe; the projection curves of all probes without pseudo-defect waves are reconstructed using a signal decomposition algorithm to obtain an optimized echo curve, and combined with a neural network, the inspection result after weld heat treatment is obtained.

[0011] The step of acquiring the reverse-sequence echo signals of each probe includes: arranging the echo signals of each probe on one side of the weld in reverse order according to the detection path to obtain the reverse-sequence echo signals of each probe.

[0012] The horizontal coordinate of the projection curve is the projection length of the ultrasonic detection path of each probe on the weld width, and the vertical coordinate is the amplitude of the echo signal.

[0013] The step of aligning the data points in the echo signals of each probe and the reverse echo signals according to the projection curve is specifically as follows:

[0014] Based on the number of echo signals of the probe corresponding to the longest angle of the detection path, the echo signals obtained by the remaining probes and the reverse echo signals are supplemented in combination with the projection curve.

[0015] The step of obtaining the transverse pseudo defect evaluation of each weld width interval is as follows:

[0016] Based on the echo signals of each probe after alignment corresponding to each weld width interval and the difference between adjacent maximum value points in the aligned reverse echo signals, the distribution difference of each weld width interval in each probe is obtained;

[0017] The difference in the echo signal amplitudes of all data points in the projection curves of each probe on both sides of the weld corresponding to each weld width interval is merged, and combined with the distribution difference, a difference evaluation of the echo signals of each probe on both sides of the weld corresponding to each weld width interval is obtained;

[0018] The difference evaluation of the echo signals of all probes on both sides of the weld corresponding to each weld width interval is comprehensively evaluated to obtain the transverse pseudo-defect evaluation corresponding to each weld width interval.

[0019] The specific process of obtaining the distribution difference of each weld width interval in each probe is as follows:

[0020] The average level of the difference between all adjacent maximum points in the echo signals of the probes corresponding to each weld width interval after alignment is recorded as the first difference; the average level of the difference between all adjacent maximum points in the reverse echo signals of the probes corresponding to each weld width interval after alignment is recorded as the second difference; based on the change between the first difference and the second difference of each weld width interval in each probe, the distribution difference of each weld width interval in each probe is obtained.

[0021] The comprehensive pseudo-defect evaluation of the echo signals of each probe on one side of the weld corresponding to each weld width interval is specifically as follows:

[0022] According to the fitting result of the projection curve of the echo signals of any two probes on one side of the weld corresponding to each weld width interval, combined with the change characteristics of the slope of each data point in the projection curve corresponding to the echo signals of the any two probes, the relevant evaluation value of the echo signals of the any two probes on one side of the weld is obtained;

[0023] The inverse proportional mapping value of each probe corresponding to each weld width interval and the relevant evaluation values ​​of the remaining probes on one side of the weld is obtained, and combined with the transverse pseudo-defect evaluation corresponding to each weld width interval, the comprehensive pseudo-defect evaluation of each probe corresponding to each weld width interval on one side of the weld is obtained.

[0024] The process of obtaining the correlation evaluation value of the echo signals of any two probes on one side of the weld is specifically as follows:

[0025] The probes at least include: probe No. 1 and probe No. 2;

[0026] The relevant evaluation values ​​of probe No. 1 and probe No. 2 on the weld A side corresponding to the i-th weld width interval are recorded as , and its calculation formula is: ; In the formula, represents an exponential function with a natural constant as base, It represents the cumulative sum of all coefficients of the fitting polynomial of the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the cumulative sum of all coefficients of the fitting polynomial of the echo signal of probe No. 2 on the weld A side corresponding to the i-th weld width interval, It represents the goodness of fit of the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval, represents the goodness of fit of the echo signal of probe No. 2 on the weld A side corresponding to the i-th weld width interval, represents the number of data points in the echo signal corresponding to the i-th weld width interval, It represents the tangent slope of the rth data point of the projection curve of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the tangent slope of the rth data point of the projection curve of probe No. 2 on the weld A side corresponding to the i-th weld width interval;

[0027] The same calculation method as the correlation evaluation values ​​of probe No. 1 and probe No. 2 on the weld A side corresponding to the i-th weld width interval is used to obtain the correlation evaluation values ​​of any two probes on either side of the weld among all the probes corresponding to the i-th weld width interval.

[0028] The specific process of judging whether there is a pseudo defect wave in the projection curve of each probe is as follows:

[0029] When the comprehensive pseudo defect evaluation is greater than or equal to the preset pseudo defect threshold, it is determined that the corresponding projection curve has a pseudo defect wave and the projection curve is processed; otherwise, it is determined that the corresponding projection curve does not have a pseudo defect wave.

[0030] The processing of the projection curve is specifically as follows:

[0031] The projection curve with pseudo defect waves is replaced with the projection curve without pseudo defect waves corresponding to the same weld width interval. The first priority is to replace it with the projection curve of the same probe on the other side of the weld. The second priority is to replace it with the projection curve of the probe with the smallest angle difference from the currently used probe on the same side of the weld. The third priority is to replace it with the projection curve of the probe with the smallest angle difference from the currently used probe on the other side of the weld.

[0032] This application has at least the following beneficial effects:

[0033] The present application first uses a probe with a preset angle to obtain echo signals on the corresponding detection path on both sides of the weld in the evaporator heating chamber, which helps to obtain echo signals from multiple directions, so that the quality of the weld can be evaluated more comprehensively and the missed detection caused by angle problems can be reduced; the echo signals of each probe on one side of the weld are rearranged to obtain the reverse echo signals of each probe, which helps to correspond the echo signals of each probe on both sides of the weld based on the position of the same weld width; the weld width is evenly divided, and the echo signals are curve-fitted according to the projection of the detection path corresponding to each probe on the weld width to obtain the projection curve of each probe on both sides of the weld corresponding to each weld width interval, which helps to unify the different numbers of echo signals generated on the inspection paths corresponding to probes at different angles, and then analyze the shape characteristics of the echo signals corresponding to each section of the weld width; the echo signals of each probe and the data points in the reverse echo signals are aligned according to the projection curve, which helps to compare the echo signals obtained by probes at different angles later. Analysis is performed to obtain more accurate results; the distribution characteristics and numerical differences between the echo signals of all aligned probes corresponding to each weld width interval and the aligned reverse echo signals are integrated to obtain a transverse pseudo-defect evaluation for each weld width interval, and its beneficial effect is to compare the projection curves on different sides of the same angle, and to preliminarily determine whether a pseudo-defect wave exists by transverse comparison; according to the fitting results of the projection curves of the echo signals of each probe corresponding to each weld width interval and the remaining probes on one side of the weld and the change characteristics of the slope of each data point, combined with the transverse pseudo-defect evaluation, a comprehensive pseudo-defect evaluation of the echo signals of each probe corresponding to each weld width interval on one side of the weld is obtained, and its beneficial effect is to further combine and compare the longitudinal comparison of the projection curves on the same side at different angles, so that the subsequent judgment of the pseudo-defect wave is more accurate, and the projection curve without pseudo-defect waves is reconstructed to obtain an optimized echo curve, which is helpful to improve the quality of echo signals and the accuracy of the detection results after weld heat treatment in the heating chamber of a large evaporator. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flow chart of a method for inspecting welds after heat treatment for a large evaporator heating chamber provided for this application;

[0035] Figure 2 Schematic diagram of the detection path of the probes at different angles provided in this application on both sides of the weld;

[0036] Figure 3 A flowchart for obtaining the lateral pseudo defect evaluation provided for this application. DETAILED DESCRIPTION

[0037] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0039] It should also be noted that the terms "first" and "second" in this application and its drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The method disclosed in the embodiments of the present application or the method shown in the flow chart includes one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged with each other, and some steps can also be deleted.

[0040] 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.

[0041] This application first proposes a method for inspecting welds after heat treatment in a large evaporator heating chamber, which is applied in the field of ultrasonic testing technology. Figure 1 , the method comprises the following steps:

[0042] S1: Use a probe with a preset angle to obtain echo signals on the corresponding detection path on both sides of the weld in the evaporator heating chamber.

[0043] This application uses ultrasonic flaw detection equipment to detect welds. Specifically, this embodiment uses three types of shear wave oblique probes with a single crystal of 45°, 60° and 70° with a chip size of 20mm×22mm, sets the ultrasonic probe frequency to 2.5MHz, and performs detection on both sides of the weld to be detected. Since the width of the weld is determined in this embodiment, all echo signals of all probes on the vertical plane from one side of the weld to the other side of the weld are obtained. Due to the different angles of the probes, the time length corresponding to the echo signal is also different.

[0044] The left and right sides of the weld area are marked as side A and side B, and the 45° probe, 60° probe and 70° probe are marked as probes 1, 2 and 3 respectively. The detection path diagram of probes at different angles on both sides of the weld is shown in the figure below. Figure 2 As shown; Figure 2In the figure, A1, A2, and A3 represent the detection paths of probes No. 1, 2, and 3 on the A side of the weld respectively; B1, B2, and B3 represent the detection paths of probes No. 1, 2, and 3 on the B side of the weld respectively.

[0045] Each angle of the probe has a corresponding echo signal on both sides of the weld, where each element in the echo signal represents the echo signal amplitude at each echo time. In this embodiment, the sampling frequency of the echo signal amplitude is 5 MHz; the implementer can set it according to the actual situation. The echo time can be used to determine the depth and position of the defect, and the echo signal amplitude can show the changes in the sound wave when it propagates inside the material, thereby reflecting the information of the defect characteristics.

[0046] S2: Rearrange the echo signals of each probe on one side of the weld to obtain the reverse echo signals of each probe; divide the weld width equally, and perform curve fitting on the echo signals and reverse echo signals of each probe corresponding to each weld width interval according to the projection of the detection path corresponding to each probe on the weld width, to obtain the projection curves of each probe on both sides of the weld corresponding to each weld width interval; align the data points in the echo signals and reverse echo signals of each probe according to the projection curve.

[0047] Due to the high sensitivity of ultrasonic detection, it is easy to be affected by stains on the weld surface to produce pseudo defect waves. Pseudo defect waves may cause errors in subsequent analysis, so it is necessary to analyze the existence of pseudo defect waves first. Oil and other stains in the weld area may cause irregular distortion of sound waves during propagation, resulting in pseudo defect waves. When the probe detects the weld area at different positions, the performance of the pseudo defect waves generated by irregular scattering will be quite different, while the echo generated by the real defect will not change much when the original waveform is detected at different positions. Therefore, by comparing the differences between the echo signals of the same probe on different sides of the weld, the existence of pseudo defect waves is analyzed and eliminated.

[0048] The echo signals obtained by the probes at the same angle on both sides of the weld are processed. First, the echo signals of the probe on the B side of the weld are arranged in reverse order according to the detection path to obtain the reverse echo signals of the probe on the B side of the weld. All echo signals of the probe on the A side of the weld remain unchanged, so that the amplitudes of the echo signals on both sides A and B correspond one to one based on the weld width.

[0049] It should be noted that since the same weld area is being tested, the internal thickness of the weld area remains unchanged, so the ultrasonic wave of the same probe propagates the same distance inside the weld under the same ultrasonic parameters, that is, the detection path is the same, which means the echo time is the same. Since the measurements are made from the A side and the B side of the weld area respectively, after the echo signal of the probe on the B side of the weld is arranged in reverse order, the position of each element in the reverse echo signal corresponds to the detection position of each element of the echo signal of the probe on the A side of the weld at the same weld width. This means that whether it is the signal on the A side or the B side, it can represent that the ultrasonic wave reaches the lateral position with the same distance from the vertical plane of the A side of the weld.

[0050] Furthermore, the echo signals of all probes on the weld A side and the reverse echo signals on the weld B side are aligned based on the weld width. The echo signals of all probes on the weld A side are processed, including the echo signals of all probes on the weld B side after reverse arrangement, because the weld width is consistent, and the projection length of the ultrasonic detection path of probes at different angles on the weld A side at the corresponding angle is equal to the weld width.

[0051] Based on this, the weld width is evenly divided, and the echo signals of each probe on the weld A side and the reverse echo signals on the weld B side corresponding to each weld width interval are obtained, and curve fitting is performed to obtain the projection curves of each probe on both sides of the weld corresponding to each weld width interval. This embodiment uses a polynomial regression algorithm of a third-order polynomial for fitting, and calculates the goodness of fit of the fitting result of each echo signal segment. The greater the goodness of fit, the better the fitting result can reflect the change of ultrasonic signal inside the weld. The polynomial regression algorithm and the calculation of goodness of fit are both well-known technologies. The horizontal axis of the fitting curve is the projection length of the echo signal detection path on the weld width, and the vertical axis is the echo signal amplitude.

[0052] At different angles, the corresponding ultrasonic detection path lengths of each weld width interval are different, and thus the density of echo signals of different probes corresponding to each weld width interval is different. Therefore, the number of echo signals of the probe corresponding to the longest detection path angle is used as the benchmark, and the echo signals obtained by the remaining probes and the reverse echo signals are supplemented in combination with the projection curve. For example, the echo signals of probes 1, 2, and 3 on the weld A side are recorded as , , ,because The echo signal has a longer path, so If the number of signals in Based on the number of signals in Obtained from the corresponding projection curve , The echo signal amplitude corresponding to the projection length of the echo signal detection path on the weld width is used to complete the alignment of data points in the echo signals of different probes.

[0053] S3: The distribution characteristics and numerical differences between the echo signals of all aligned probes and the aligned reverse echo signals corresponding to each weld width interval are combined to obtain the transverse pseudo-defect evaluation of each weld width interval.

[0054] After aligning the data points in the echo signals of different probes, the echo signal lengths of all probes are consistent. First, the echo signals of the same probe on both sides of the weld are analyzed to obtain the lateral pseudo-defect evaluation of each probe corresponding to each weld width interval.

[0055] The average level of the difference between all adjacent maximum points in the echo signals of the probes corresponding to each weld width interval after alignment is recorded as the first difference; the average level of the difference between all adjacent maximum points in the reverse echo signals of the probes corresponding to each weld width interval after alignment is recorded as the second difference; based on the change between the first difference and the second difference of each weld width interval in each probe, the distribution difference of each weld width interval in each probe is obtained.

[0056] The differences in the echo signal amplitudes of all data points in the projection curves of each probe on both sides of the weld corresponding to each weld width interval are fused, and combined with the distribution difference to obtain a difference evaluation of the echo signals of each probe on both sides of the weld corresponding to each weld width interval.

[0057] Specifically, since the analysis method of the echo signals corresponding to the probes at different angles for each weld width interval is consistent, this embodiment takes probe No. 1 as an example, and compares the echo signal of probe No. 1 on the weld A side with the reverse echo signal on the weld B side. The difference evaluation of the echo signals of probe No. 1 on both sides of the weld corresponding to the i-th weld width interval is The calculation formula is: ; In the formula, The average value of the absolute value of the difference between each two adjacent maximum value points in the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval is recorded as the first difference; The average value of the absolute value of the difference between each two adjacent maximum value points in the echo signal of probe No. 1 on the weld B side corresponding to the i-th weld width interval is recorded as the second difference; represents the number of data points in the echo signal corresponding to the i-th weld width interval, It represents the echo signal amplitude of the rth data point of the projection curve of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the echo signal amplitude of the rth data point of the projection curve of probe No. 1 on the weld A side corresponding to the i-th weld width interval.

[0058] It should be noted that the greater the difference between the first difference and the second difference in the echo signal of each probe and the inverse echo signal, the greater the difference in the number of maximum points and the distribution difference of the maximum points on the projection curve, and the greater the difference in the projection curve of the probe on both sides of the weld, and the greater the possibility of the occurrence of pseudo defect waves; when the difference in the amplitude of the corresponding signal points of the projection curve of the probe on both sides of the weld is greater, the greater the difference in the waveform of the projection curve of the probe on both sides of the weld, which also means that the possibility of defect waves is greater, that is, the greater the difference evaluation.

[0059] Furthermore, since the pseudo-defect wave caused by the same stain will increase the difference between the echo signals of all probes on both sides of the weld corresponding to each weld width interval, the difference evaluation of the echo signals of all probes on both sides of the weld corresponding to each weld width interval is comprehensively evaluated to obtain the lateral pseudo-defect evaluation corresponding to each weld width interval, specifically, the normalized result of the sum of the difference evaluations of the echo signals of all probes on both sides of the weld corresponding to each weld width interval. This embodiment adopts a linear normalization method.

[0060] Among them, the acquisition flow chart of horizontal pseudo defect evaluation is as follows: Figure 3 shown.

[0061] It should be noted that the greater the difference in the projection curves of the probe on both sides of the weld, that is, the greater the difference evaluation, the greater the transverse pseudo-defect evaluation corresponding to each weld width interval.

[0062] S4: Based on the fitting results of the projection curves of the echo signals of each probe corresponding to each weld width interval and the remaining probes on one side of the weld and the change characteristics of the slope of each data point, combined with the lateral pseudo-defect evaluation, a comprehensive pseudo-defect evaluation of the echo signals of each probe corresponding to each weld width interval on one side of the weld is obtained.

[0063] By comparing the echo signals of the same probe on both sides of the weld horizontally, we preliminarily judged the possibility of pseudo-defect waves appearing in each weld width interval, that is, the possibility of pseudo-defect waves appearing on each ultrasonic detection path. However, we cannot determine which side of the weld has the pseudo-defect wave in the echo signal corresponding to the high horizontal pseudo-defect evaluation. Further analysis is needed to determine and eliminate the pseudo-defect waves. Therefore, this application longitudinally compares the differences in echo signals of different probes on the same side of the weld.

[0064] Since the polynomial fitting results of the projection curves of the echo signals of the probes on both sides of the weld corresponding to each weld width interval can reflect the shape characteristics of the projection curves, that is, the changing trend of the data; based on this, according to the fitting results of the projection curves of the echo signals of any two probes on one side of the weld corresponding to each weld width interval, combined with the changing characteristics of the slopes of the data points in the projection curves corresponding to the echo signals of the any two probes, the relevant evaluation values ​​of the echo signals of the any two probes on one side of the weld are obtained.

[0065] Specifically, the relevant evaluation values ​​of probe No. 1 and probe No. 2 corresponding to the i-th weld width interval on the weld A side are recorded as , and its calculation formula is: ; In the formula, represents an exponential function with a natural constant as base, It represents the cumulative sum of all coefficients of the fitting polynomial of the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the cumulative sum of all coefficients of the fitting polynomial of the echo signal of probe No. 2 on the weld A side corresponding to the i-th weld width interval, It represents the goodness of fit of the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval, represents the goodness of fit of the echo signal of probe No. 2 on the weld A side corresponding to the i-th weld width interval, represents the number of data points in the echo signal corresponding to the i-th weld width interval, It represents the tangent slope of the rth data point of the projection curve of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the tangent slope of the rth data point of the projection curve of probe No. 2 on the A side of the weld corresponding to the i-th weld width interval.

[0066] The same calculation method as the correlation evaluation values ​​of probe No. 1 and probe No. 2 on the weld A side corresponding to the i-th weld width interval is used to obtain the correlation evaluation values ​​of any two probes on either side of the weld among all the probes corresponding to the i-th weld width interval.

[0067] It should be noted that the greater the difference in the cumulative sum of the fitting polynomial coefficients of the echo signals of the two different probes on one side of the weld corresponding to each weld width interval, the smaller the corresponding correlation evaluation value of the two probes on one side of the weld; the greater the goodness of fit of the echo signals of the two probes on one side of the weld, the greater the difference between the coefficients of the polynomials after fitting and the credibility of the impact on the relevant evaluation, and the greater the relevant evaluation value; the greater the difference in the slopes of the two probes at the same position in each projection curve on one side of the weld, the greater the difference in the trend of the corresponding projection curves of the two probes, and thus the smaller the relevant evaluation value.

[0068] It should be further explained that the slope is used instead of the signal amplitude in the formula of the relevant evaluation value because the detection positions of different probes may be slightly different, which will lead to differences in signal amplitude, thus affecting the accuracy of the results. In the case of actual defects, although the signal amplitudes of different probes may be different, the waveforms they produce are similar. Therefore, using the slope for comparison can more effectively reflect the actual situation inside the weld and reduce the error caused by amplitude differences.

[0069] The inverse proportional mapping value of each probe corresponding to each weld width interval and the relevant evaluation values ​​of the remaining probes on one side of the weld is obtained, and combined with the transverse pseudo-defect evaluation corresponding to each weld width interval, the comprehensive pseudo-defect evaluation of each probe corresponding to each weld width interval on one side of the weld is obtained.

[0070] Specifically, the comprehensive pseudo defect evaluation of probe No. 1 on the weld A side corresponding to the i-th weld width interval is recorded as , its formula form is: ; In the formula, represents the Sigmoid function, represents the transverse pseudo-defect evaluation corresponding to the i-th weld width interval, It represents the correlation evaluation value of probe No. 1 and probe No. 2 on the weld A side corresponding to the i-th weld width interval, Represents the correlation evaluation values ​​of probe No. 1 and probe No. 3 on the weld A side corresponding to the i-th weld width interval.

[0071] The same calculation method as the relevant evaluation value of probe No. 1 on the weld A side corresponding to the i-th weld width interval is used to obtain the comprehensive pseudo-defect evaluation of any probe on either side of the weld corresponding to the i-th weld width interval.

[0072] It should be noted that It represents the longitudinal correlation evaluation between probe No. 1 and other probes on the weld A side corresponding to the i-th weld width interval, The value reflects the possibility of a pseudo defect wave appearing in the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval. The larger the longitudinal correlation evaluation, the smaller the possibility of a pseudo defect wave being generated longitudinally, that is, the smaller the comprehensive pseudo defect evaluation.

[0073] S5: Based on the comprehensive pseudo-defect evaluation, determine whether there is a pseudo-defect wave in the projection curve of each probe; reconstruct the projection curves of all probes using a signal decomposition algorithm to obtain an optimized echo curve, and combine it with a neural network to obtain the inspection result after the weld heat treatment.

[0074] When the comprehensive pseudo-defect evaluation is greater than or equal to the preset pseudo-defect threshold, it is determined that the corresponding echo signal is affected by the pseudo-defect wave, and the projection curve corresponding to the same weld width interval without the pseudo-defect wave is used for replacement. The first priority is to replace it with the projection curve of the same probe on the other side of the weld, the second priority is to replace it with the projection curve of the probe with the smallest angle difference with the currently used probe on the same side of the weld, and the third priority is to replace it with the projection curve of the probe with the smallest angle difference with the currently used probe on the other side of the weld. For example, if the projection curve of probe No. 1 on the weld A side corresponding to the i-th weld width interval is affected by the pseudo-defect wave, the projection curve of probe No. 1 on the weld B side corresponding to the i-th weld width interval is replaced by the projection curve of the pseudo-defect wave; if the projection curve of probe No. 1 on both sides of the weld corresponding to the i-th weld width interval is affected by the pseudo-defect wave, the projection curve of probe No. 2 on the weld A side corresponding to the i-th weld width interval is replaced by the projection curve of the pseudo-defect wave.

[0075] The processed echo signals of all probes on both sides of the weld are decomposed using the Empirical Mode Decomposition (EMD) algorithm to obtain several IMF components and a residual signal. For the IMF components decomposed by the EMD algorithm, the two IMF components with the highest frequency and the lowest frequency are removed, and the other IMF components are added to the residual signal for reconstruction, and the reconstructed signal is recorded as the optimized echo signal. Among them, the EMD algorithm is an existing well-known technology, and this application will not elaborate on it.

[0076] The CNN convolutional neural network model is pre-trained using the existing crack defect wave image, the obtained optimized echo curve is input into the pre-trained CNN model, and the neural network is used to determine whether the waveform in the optimized echo curve meets the crack defect waveform characteristics, and finally the inspection result after the weld heat treatment is output. Among them, the CNN convolutional neural network is an existing well-known technology.

[0077] The embodiment of the present application provides a method for inspecting welds after heat treatment in a large evaporator heating chamber, the method comprising: first, using a probe with a preset angle to obtain echo signals on a corresponding detection path on both sides of a weld in the evaporator heating chamber, which helps to obtain echo signals from multiple directions, so that the quality of the weld can be more comprehensively evaluated and missed detection caused by angle problems can be reduced; the echo signals of each probe on one side of the weld are rearranged to obtain reverse echo signals of each probe, which helps to correspond one-to-one to the echo signals of each probe on both sides of the weld based on the position of the same weld width; the weld width is evenly divided, and the echo signals are curve-fitted according to the projection of the detection path corresponding to each probe on the weld width to obtain the projection curve of each probe on both sides of the weld corresponding to each weld width interval, which helps to unify the different numbers of echo signals generated on the inspection paths corresponding to probes at different angles, and then analyze the shape characteristics of the echo signals corresponding to each section of the weld width; the echo signals of each probe and the data points in the reverse echo signals are aligned according to the projection curve, which helps to subsequently align the echo signals of different probes and the reverse echo signals. The echo signals obtained by the probes at the same angle are compared and analyzed to obtain more accurate results; the distribution characteristics and numerical differences between the echo signals of all the aligned probes corresponding to each weld width interval and the aligned reverse echo signals are integrated to obtain the transverse pseudo-defect evaluation of each weld width interval, and its beneficial effect is to compare the projection curves on different sides of the same angle, and to preliminarily judge whether the pseudo-defect wave exists by transverse comparison; according to the fitting results of the projection curves of the echo signals of each probe corresponding to each weld width interval and the remaining probes on one side of the weld and the change characteristics of the slope of each data point, combined with the transverse pseudo-defect evaluation, the comprehensive pseudo-defect evaluation of the echo signals of each probe corresponding to each weld width interval on one side of the weld is obtained, and its beneficial effect is to further combine and compare the longitudinal comparison of the projection curves on the same side of different angles, so that the subsequent judgment of the pseudo-defect wave is more accurate, and the projection curve without pseudo-defect waves is reconstructed to obtain the optimized echo curve, which is helpful to improve the echo signal quality and the accuracy of the detection results after weld heat treatment in the heating chamber of a large evaporator.

[0078] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two continuous boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two continuous operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0079] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for inspecting welds after heat treatment in a large evaporator heating chamber, characterized in that: The method comprises the following steps: Use multiple probes with different preset angles to obtain echo signals on the corresponding detection path on both sides of the weld in the evaporator heating chamber; The echo signals of each probe on one side of the weld are rearranged to obtain the reverse echo signals of each probe; the weld width is evenly divided, and according to the projection of the detection path corresponding to each probe on the weld width, the echo signals and reverse echo signals of each probe corresponding to each weld width interval are curve fitted to obtain the projection curves of each probe on both sides of the weld corresponding to each weld width interval; the data points in the echo signals and reverse echo signals of each probe are aligned according to the projection curves; The evaluation of transverse pseudo-defects in each weld width interval is obtained by combining the distribution characteristics and numerical differences between the echo signals of all probes after alignment and the reverse echo signals after alignment corresponding to each weld width interval; According to the fitting results of the projection curves of the echo signals of each probe corresponding to each weld width interval and the remaining probes on one side of the weld and the variation characteristics of the slope of each data point, combined with the transverse pseudo-defect evaluation, a comprehensive pseudo-defect evaluation of the echo signals of each probe corresponding to each weld width interval on one side of the weld is obtained; Based on the comprehensive pseudo-defect evaluation, it is judged whether there is a pseudo-defect wave in the projection curve of each probe; the projection curves of all probes without pseudo-defect waves are reconstructed using a signal decomposition algorithm to obtain an optimized echo curve, and combined with a neural network, the inspection result after weld heat treatment is obtained; The step of acquiring the reverse echo signal of each probe includes: after sorting the echo signals of each probe on one side of the weld to obtain an echo signal sequence, sorting the elements in the echo signal sequence in reverse order according to time sequence to obtain the reverse echo signal of each probe.

2. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 1, characterized in that: The abscissa of the projection curve is the projection length of the ultrasonic detection path of each probe on the width of the weld, and the ordinate is the amplitude of the echo signal.

3. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 1, characterized in that: The data points in the echo signals and the reverse echo signals of each probe are aligned according to the projection curve, specifically: Based on the number of echo signals of the probe corresponding to the longest angle of the detection path, the echo signals obtained by the remaining probes and the reverse echo signals are supplemented in combination with the projection curve.

4. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 1, characterized in that: The steps of obtaining the transverse pseudo defect evaluation of each weld width interval are: Based on the echo signals of each probe after alignment corresponding to each weld width interval and the difference between adjacent maximum value points in the aligned reverse echo signals, the distribution difference of each weld width interval in each probe is obtained; The difference in the echo signal amplitudes of all data points in the projection curves of each probe on both sides of the weld corresponding to each weld width interval is merged, and combined with the distribution difference, a difference evaluation of the echo signals of each probe on both sides of the weld corresponding to each weld width interval is obtained; The difference evaluation of the echo signals of all probes on both sides of the weld corresponding to each weld width interval is comprehensively evaluated to obtain the transverse pseudo-defect evaluation corresponding to each weld width interval.

5. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 4, characterized in that: The specific process of obtaining the distribution difference of each weld width interval in each probe is as follows: The average level of the difference between all adjacent maximum points in the echo signals of the probes corresponding to each weld width interval after alignment is recorded as the first difference; the average level of the difference between all adjacent maximum points in the reverse echo signals of the probes corresponding to each weld width interval after alignment is recorded as the second difference; based on the change between the first difference and the second difference of each weld width interval in each probe, the distribution difference of each weld width interval in each probe is obtained.

6. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 1, characterized in that: The comprehensive pseudo defect evaluation of the echo signals of each probe on one side of the weld corresponding to each weld width interval is specifically as follows: According to the fitting result of the projection curve of the echo signals of any two probes on one side of the weld corresponding to each weld width interval, combined with the change characteristics of the slope of each data point in the projection curve corresponding to the echo signals of the any two probes, the relevant evaluation value of the echo signals of the any two probes on one side of the weld is obtained; The inverse proportional mapping value of each probe corresponding to each weld width interval and the relevant evaluation values ​​of the remaining probes on one side of the weld is obtained, and combined with the transverse pseudo-defect evaluation corresponding to each weld width interval, the comprehensive pseudo-defect evaluation of each probe corresponding to each weld width interval on one side of the weld is obtained.

7. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 6, characterized in that: The specific process of obtaining the correlation evaluation value of the echo signals of any two probes on one side of the weld is as follows: The probes at least include: probe No. 1 and probe No. 2; The relevant evaluation values ​​of probe No. 1 and probe No. 2 on the weld A side corresponding to the i-th weld width interval are recorded as , and its calculation formula is: ; In the formula, represents an exponential function with a natural constant as base, It represents the cumulative sum of all coefficients of the fitting polynomial of the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the cumulative sum of all coefficients of the fitting polynomial of the echo signal of probe No. 2 on the weld A side corresponding to the i-th weld width interval, It represents the goodness of fit of the echo signal of probe No. 1 on the weld A side corresponding to the i-th weld width interval, represents the goodness of fit of the echo signal of probe No. 2 on the weld A side corresponding to the i-th weld width interval, represents the number of data points in the echo signal corresponding to the i-th weld width interval, It represents the tangent slope of the rth data point of the projection curve of probe No. 1 on the weld A side corresponding to the i-th weld width interval, It represents the tangent slope of the rth data point of the projection curve of probe No. 2 on the weld A side corresponding to the i-th weld width interval; The same calculation method as the correlation evaluation values ​​of probe No. 1 and probe No. 2 on the weld A side corresponding to the i-th weld width interval is used to obtain the correlation evaluation values ​​of any two probes on either side of the weld among all the probes corresponding to the i-th weld width interval.

8. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 1, characterized in that: The specific process of judging whether there is a pseudo defect wave in the projection curve of each probe is as follows: When the comprehensive pseudo defect evaluation is greater than or equal to the preset pseudo defect threshold, it is determined that the corresponding projection curve has a pseudo defect wave and the projection curve is processed; otherwise, it is determined that the corresponding projection curve does not have a pseudo defect wave.

9. The method for inspecting welds after heat treatment for a large evaporator heating chamber according to claim 8, characterized in that: The processing of the projection curve is specifically as follows: The projection curve with pseudo defect waves is replaced with the projection curve without pseudo defect waves corresponding to the same weld width interval. The first priority is to replace it with the projection curve of the same probe on the other side of the weld. The second priority is to replace it with the projection curve of the probe with the smallest angle difference from the currently used probe on the same side of the weld. The third priority is to replace it with the projection curve of the probe with the smallest angle difference from the currently used probe on the other side of the weld.

Images