Stainless steel pipe fitting material internal defect detection system based on ultrasonic detection technology

Abstract

The invention relates to the technical field of steel pipe defect detection, in particular to a stainless steel pipe material internal defect detection system based on an ultrasonic detection technology, which comprises an initial detection module, a detection module, a detection module and a detection module, the initial defect possibility is obtained; the depth analysis module is used for performing depth analysis on the target echo signal segment to respectively obtain a first real defect coefficient and a second real defect coefficient; fusing the first real defect coefficient and the second real defect coefficient to obtain a comprehensive real defect coefficient; the defect detection module is used for acquiring the defect degree of the position, corresponding to the target echo signal segment, on the pipe fitting by utilizing the initial defect possibility of the target echo signal segment and the comprehensive real defect coefficient; according to the defect degree, whether defect early warning is carried out on the position, corresponding to the target echo signal segment, of the pipe fitting or not is judged. According to the invention, the defects of the pipe fitting can be effectively detected.

Description

Internal Defect Detection System for Stainless Steel Pipe Fittings Based on Ultrasonic Testing Technology Technical Field

[0001] This invention relates to the field of steel pipe defect detection technology, specifically to a stainless steel pipe material internal defect detection system based on ultrasonic testing technology. Background Technology

[0002] With the continuous advancement of industrial technology, stainless steel pipe fittings are increasingly widely used as a key basic material in liquid and gas transportation systems in high-end equipment manufacturing fields such as energy, chemical industry, aerospace, and nuclear power. However, during long-term operation, stainless steel pipe fittings are susceptible to multiple factors such as environment, stress, and corrosion, leading to defects such as cracks, porosity, inclusions, and decarburization layers within the material. These defects not only reduce the service life and performance of the pipe fittings but may also pose a significant threat to the safety of the overall system. Therefore, the detection of internal defects in stainless steel pipe fittings is crucial. Ultrasonic testing technology, as a non-destructive testing method, has been widely used in the quality inspection of industrial products such as metal materials and pipelines due to its advantages of high precision, deep penetration, and high efficiency. This technology effectively identifies internal defects in metal materials such as stainless steel pipe fittings by emitting high-frequency sound waves and receiving echo signals, based on the propagation characteristics of sound waves in different materials. This provides solid technical support for building an internal defect detection system for stainless steel pipe fittings based on ultrasonic testing technology.

[0003] When using ultrasonic technology to detect internal defects in stainless steel pipes, traditional methods typically rely solely on the overall amplitude or simple time-domain characteristics of the echo signal to preliminarily determine the presence of defects. For example, a fixed threshold is set, and when the echo amplitude exceeds this threshold, it is considered a defect. However, pseudo-defects such as structural noise (e.g., "grass waves" caused by grain boundary scattering), structural echoes (e.g., regular reflections from threads or weld reinforcement), and coupling noise (e.g., wide-amplitude echoes caused by poor probe coupling) can also produce similar abnormal echoes, making it difficult for traditional methods to distinguish their sources, thus leading to false alarms. Summary of the Invention

[0004] To address the aforementioned technical problems, the present invention aims to provide a stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology. The specific technical solution adopted is as follows: One embodiment of the present invention provides a stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology. This system includes: an initial detection module, used to detect the pipe fitting using an ultrasonic probe, acquire different echo signal segments, and designate any echo signal segment as a target echo signal segment; and to obtain the initial defect probability of the target echo signal segment based on the number of amplitudes exceeding a defect threshold in the target echo signal segment; and a depth analysis module, used to analyze the rise time of the incident pulse and return signal of the target echo signal segment, the main amplitude of the target echo signal segment, and the depth of the target echo signal segment. The first true defect coefficient is obtained by taking the energy values ​​of the two parts to the left and right of the peak point and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point; the second true defect coefficient is obtained by taking the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the spectral centroid of the target echo signal segment; the first true defect coefficient and the second true defect coefficient are fused to obtain the comprehensive true defect coefficient; the defect detection module is used to obtain the defect degree of the position on the pipe corresponding to the target echo signal segment by using the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient; and to determine whether to issue a defect warning for the position on the pipe corresponding to the target echo signal segment based on the defect degree.

[0005] Preferably, obtaining the initial defect probability of the target echo signal segment based on the number of amplitudes exceeding the defect threshold in the target echo signal segment includes: subtracting each amplitude exceeding the defect threshold in the target echo signal segment from the defect threshold and then summing the results to obtain the degree of exceeding the limit; multiplying the normalized value of the degree of exceeding the limit with the normalized value of the number of amplitudes exceeding the defect threshold in the target echo signal segment to obtain the initial defect probability of the target echo signal segment.

[0006] Preferably, the first true defect coefficient is obtained based on the rise time of the incident pulse and the return signal of the target echo signal segment, the energy values ​​of the left and right parts of the main peak point of the target echo signal segment, and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point. This includes: dividing the target echo signal segment into left and right parts with the main peak point of the target echo signal segment as the center; obtaining the waveform symmetry by summing the absolute value of the difference in energy values ​​between the left and right parts with the hyperparameter and taking the reciprocal; obtaining the rise time proportionality coefficient by subtracting the ratio of the rise time of the incident pulse to the rise time of the return signal of the target echo signal segment from a first preset value and taking the absolute value; and setting the rise time... The rise time consistency is obtained by adding the proportional coefficient and the hyperparameter and taking their reciprocal. The reciprocal of the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point in the target echo signal segment is obtained and recorded as the peak characteristic coefficient. The first characteristic coefficient is obtained by multiplying the normalized value of waveform symmetry, the normalized value of rise time consistency and the normalized value of peak characteristic coefficient. The peak influence coefficient is obtained. If the number of peak points in the target echo signal segment is greater than 1, the peak influence coefficient is recorded as 0. When the number of peak points is equal to 1, the peak influence coefficient is 1. The first true defect coefficient is obtained by adding the normalized value of the first characteristic coefficient and the normalized value of the peak influence coefficient and then normalizing them.

[0007] Preferably, the second true defect coefficient is obtained based on the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the spectral centroid of the target echo signal segment. This includes: adding the absolute value of the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse to the hyperparameter and taking its reciprocal to obtain the bandwidth characteristic coefficient; obtaining the reciprocal of the entropy value of the power spectrum of the target echo signal segment to obtain the entropy characteristic coefficient; using the hyperparameter to add the absolute value of the difference between the theoretical spectral centroid and the spectral centroid of the target echo signal segment and taking its reciprocal to obtain the spectral centroid characteristic coefficient; and multiplying and normalizing the normalized values ​​of the bandwidth characteristic coefficient, the entropy characteristic coefficient, and the spectral centroid characteristic coefficient to obtain the second true defect coefficient of the target echo signal segment.

[0008] Preferably, the first true defect coefficient and the second true defect coefficient are fused to obtain a comprehensive true defect coefficient, which includes: calculating the average of the first true defect coefficient and the second true defect coefficient of the target echo signal segment to obtain the comprehensive true defect coefficient of the target echo signal segment.

[0009] Preferably, obtaining the defect level at the location on the pipe corresponding to the target echo signal segment using the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient includes: multiplying the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient to obtain the defect level at the location on the pipe corresponding to the target echo signal segment.

[0010] Preferably, determining whether to issue a defect warning for the location on the pipe corresponding to the target echo signal segment based on the degree of defect includes: if the degree of defect on the location on the pipe corresponding to the target echo signal segment is greater than or equal to a judgment threshold, then issue a defect warning for the location on the pipe corresponding to the target echo signal segment.

[0011] The embodiments of the present invention have at least the following beneficial effects: First, this application uses an ultrasonic probe to detect pipe fittings and obtain different echo signal segments. Any echo signal segment is designated as the target echo signal segment. Then, based on the number of amplitudes exceeding the defect threshold in the target echo signal segment, the initial defect probability of the target echo signal segment is obtained. Further, a first true defect coefficient is obtained based on the rise time of the incident pulse and the return signal of the target echo signal segment, the energy values ​​of the left and right parts of the main peak point of the target echo signal segment, and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point. Finally, based on the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the target echo signal segment... The second true defect coefficient is obtained by analyzing the spectral centroid. The first and second true defect coefficients are then fused to obtain a comprehensive true defect coefficient. Here, the comprehensive true defect coefficient is obtained by combining the time-domain waveform dimension, frequency-domain dimension, and statistical stability dimension of the target echo signal segment. Finally, the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient are used to obtain the defect degree at the location on the pipe corresponding to the target echo signal segment, thereby completing the detection of internal defects in stainless steel pipes. This improves the traditional echo signal anomaly detection method, avoiding the problem of inaccurate detection of internal defects in pipes caused by relying solely on the overall amplitude or simple time-domain characteristics of the echo signal to preliminarily determine the existence of defects, thus improving the accuracy of internal defect detection in pipes. Attached Figure Description

[0012] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 is a system block diagram of a stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology provided in an embodiment of the present invention. Detailed Implementation

[0014] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0016] The following description, in conjunction with the accompanying drawings, details a specific scheme for an internal defect detection system for stainless steel pipe fittings based on ultrasonic testing technology provided by the present invention.

[0017] Example: The main application scenario of this invention is to detect defects inside pipe fittings using ultrasonic waves, and then analyze the returned signals to obtain the defects inside the pipe fittings more accurately.

[0018] Please refer to Figure 1, which shows a system block diagram of an internal defect detection system for stainless steel pipe fittings based on ultrasonic testing technology provided by an embodiment of the present invention. The system includes the following modules: an initial detection module, used to detect the pipe fittings using an ultrasonic probe, obtain different echo signal segments, and record any echo signal segment as a target echo signal segment; and obtain the initial defect probability of the target echo signal segment based on the number of amplitudes exceeding the defect threshold in the target echo signal segment.

[0019] In the detection of internal defects in stainless steel pipe fittings based on ultrasonic testing technology, echo signal acquisition is a crucial step. First, a suitable high-frequency ultrasonic probe is selected, with parameters such as frequency and crystal size chosen according to the pipe fitting specifications and testing requirements to ensure effective ultrasonic wave transmission into the pipe. Next, the probe is well coupled to the pipe surface, typically using a coupling agent to eliminate air gaps and ensure smooth ultrasonic wave transmission. Then, an excitation pulse is applied to the probe using a pulse generator, causing it to generate ultrasonic waves that penetrate the pipe. The ultrasonic waves propagate inside the pipe, generating reflected echoes when they encounter defects or the bottom surface. These echo signals are then captured in real time using a high-precision signal receiving device, such as an oscilloscope or a dedicated data acquisition card. During acquisition, parameters such as sampling frequency (typically in the 0.5-15MHz range in ultrasonic testing) and sampling duration (which depends on the propagation time of the ultrasonic wave within the pipe and the potential range of the defect to be detected) must be set appropriately to ensure complete recording of the echo characteristics.

[0020] When performing defect detection on stainless steel pipes, this application takes a straight pipe as an example. The ultrasonic probe is placed vertically on the outer surface of the straight pipe, so that the ultrasonic wave is incident in a direction perpendicular to the axis of the pipe, and the first echo signal segment is obtained. After that, the ultrasonic probe is moved to obtain the second, third, and nth echo signal segments in sequence. Each echo signal segment is analyzed in sequence. For the convenience of subsequent analysis, any echo signal segment is recorded as the target echo signal segment, and subsequent analysis is performed based on a target echo signal segment.

[0021] Furthermore, the amplitude corresponding to each data point on the target echo signal segment is obtained, which is the signal strength; then, based on the reference values ​​or judgment criteria for defect echo amplitude given by international standards (such as ASME, ISO) or domestic standards (such as GB / T), the upper limit of defect echo amplitude corresponding to the current working pressure and material of the stainless steel pipe fitting is obtained, which is the defect threshold.

[0022] Then, the initial defect probability of the target echo signal segment is obtained based on the number of amplitudes exceeding the defect threshold in the target echo signal segment. Specifically, the degree of exceeding the threshold is obtained by subtracting each amplitude exceeding the defect threshold in the target echo signal segment from the defect threshold and then summing the results; the normalized value of the degree of exceeding the threshold is multiplied by the normalized value of the number of amplitudes exceeding the defect threshold in the target echo signal segment to obtain the initial defect probability of the target echo signal segment.

[0023] Where Ai represents the initial defect probability of the i-th target echo signal segment, Si represents the number of amplitudes exceeding the defect threshold in the i-th target echo signal segment, Hij represents the j-th amplitude exceeding the defect threshold in the i-th target echo signal segment, and H represents the defect threshold. It is the sum of the differences between the amplitudes of each target echo signal segment that exceed the defect threshold and the defect threshold, that is, the degree to which the target echo signal segment exceeds the defect threshold. The higher the degree of exceedance, the greater the possibility of defect.

[0024] This allows us to determine the initial defect probability of the target echo signal segment.

[0025] The deep analysis module is used to obtain a first true defect coefficient based on the rise time of the incident pulse and the returned signal of the target echo signal segment, the energy values ​​of the two parts to the left and right of the main peak point of the target echo signal segment, and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point; to obtain a second true defect coefficient based on the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the spectral centroid of the target echo signal segment; and to fuse the first true defect coefficient and the second true defect coefficient to obtain a comprehensive true defect coefficient.

[0026] After obtaining the probability of defects in the target echo signal segment from the above, since the amplitude anomaly may not only be a real defect, but also a data anomaly caused by noise due to normal structure or improper operation, it is necessary to further combine the characteristics of local data to rule out the possibility that it is noise data and obtain the possibility that it is a real defect.

[0027] Echoes from genuine defects (such as cracks) typically maintain the shape of the incident pulse, while those from spurious defects (such as structural noise or coupling noise) cause waveform distortion. First, the incident ultrasonic pulse is usually symmetrical, and genuine defect reflections maintain this symmetry. However, structural echoes from complex geometries (such as threads) may cause asymmetrical mode transitions, and structural noise is inherently asymmetrical. Second, genuine defect echoes have steep leading edges and rise times consistent with the incident pulse, while poor coupling leads to beam diffusion, resulting in a slower rise edge. The rise edge of structural noise also slows down due to superposition. Furthermore, genuine defect echoes typically have a single main peak due to reflection from a single interface, while structural noise may have multiple smaller peaks at the main peak position (scattering interference), resulting in multiple peaks in the structural echo due to multiple reflecting surfaces. Finally, genuine defect echoes generally do not introduce additional overshoot, while coupled vibrations or circuit interference can cause significant overshoot or oscillations.

[0028] Therefore, the first true defect coefficient is obtained based on the rise time of the incident pulse and the return signal of the target echo signal segment, the energy values ​​of the two parts to the left and right of the main peak point of the target echo signal segment, and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point.

[0029] Specifically, the target echo signal segment is divided into left and right parts centered on the main peak point. The absolute value of the energy difference between the left and right parts is obtained, and the sum of the hyperparameters is calculated and their reciprocal is taken to obtain the waveform symmetry. The rise time ratio coefficient is obtained by subtracting the ratio of the rise time of the incident pulse to the rise time of the returned signal in the target echo signal segment from the first preset value and taking the absolute value. The rise time ratio coefficient is added to the hyperparameters and its reciprocal is taken to obtain the rise time consistency. The reciprocal of the difference between the peak value of the main peak point in the target echo signal segment and the nearest valley value to the right of the main peak point is obtained and recorded as the peak characteristic coefficient. The normalized value of the waveform symmetry, the normalized value of the rise time consistency, and the normalized value of the peak characteristic coefficient are multiplied to obtain the first characteristic coefficient. The peak influence coefficient is obtained. If the number of peak points in the target echo signal segment is greater than 1, the peak influence coefficient is recorded as 0. When the number of peak points is equal to 1, the peak influence coefficient is 1. The normalized value of the first characteristic coefficient and the normalized value of the peak influence coefficient are added together and then normalized to obtain the first true defect coefficient.

[0030] The specific calculation model for the first true defect coefficient is as follows: Where Bi represents the first true defect coefficient of the i-th target echo signal segment; norm represents the normalization operation; To determine waveform symmetry, the target data segment is divided into two parts, left and right, centered on the main peak (the largest peak point). Eli is the energy value of the left half, and Eri is the energy value of the right half. The smaller the difference in energy between the two parts, the higher the waveform symmetry of the target echo signal segment, and the more likely it is to be a real defect. This is the rise time scaling factor. Rise time consistency refers to the consistency between the rise time of the incident pulse and the rise time of the returned signal in the target echo signal segment (the rise time is the time required from 10% to 90% of the main peak value). Tri is the rise time of the incident pulse, and Tfi is the rise time of the returned signal. The higher the consistency, the closer this ratio is to 1. That is, the larger this value is, the more the target data segment matches the characteristics of high rise time consistency of real defect data. Fzi-Gzi represents the peak characteristic coefficient, indicating the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point, i.e., the degree of echo overshoot. The smaller this value, the less significant the echo overshoot, and the more consistent it is with the actual defect echo, which generally does not introduce additional overshoot characteristics. Fzi is the peak value of the main peak point, and Gzi is the nearest valley value to the right of the main peak point. Fi is the peak influence coefficient of the i-th target echo signal segment. When the number of peak points in the i-th target echo signal segment is greater than 1, the peak influence coefficient is 0, meaning that the data segment has many peaks, which does not conform to the actual defect characteristics. When the number of peak points is equal to 1, the peak influence coefficient is 1, meaning that the data segment is a single peak, which conforms to the actual defect characteristics. c is a hyperparameter, with the numerator set to 0, so it is taken as 0.01; the first preset value is 1. Thus, the first actual defect coefficient of each target echo signal segment can be obtained.

[0031] Meanwhile, the spectral characteristics of real defect echoes are stable and have a definite transformation relationship with the incident pulse spectrum (mainly affected by material attenuation), while the spectrum of pseudo-defects exhibits randomness, unpredictability, or abnormal changes. First, real defect reflections do not significantly change the pulse bandwidth, while strong scattering (structural noise) causes dispersion, leading to a significant increase in bandwidth, and poor coupling also changes the bandwidth. At the same time, the spectrum of real defect echoes is concentrated (low entropy) because the main reflection frequency components are concentrated, while the spectrum of structural and coupling noise is dispersed (high entropy). Finally, real defect echoes come from a fixed depth, and their spectral centroid shift is predictable, while structural noise (grain scattering) causes irregular fluctuations in the spectral centroid.

[0032] Therefore, the second true defect coefficient is obtained based on the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the spectral centroid of the target echo signal segment.

[0033] Specifically, the bandwidth characteristic coefficient is obtained by adding the absolute value of the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse to the hyperparameter and taking its reciprocal; the entropy characteristic coefficient is obtained by taking the reciprocal of the entropy value of the power spectrum of the target echo signal segment; the spectral centroid characteristic coefficient is obtained by adding the absolute value of the difference between the theoretical spectral centroid and the spectral centroid of the target echo signal segment to the hyperparameter and taking its reciprocal; the second true defect coefficient of the target echo signal segment is obtained by multiplying the normalized values ​​of the bandwidth characteristic coefficient, the entropy characteristic coefficient, and the spectral centroid characteristic coefficient and normalizing the result.

[0034] The specific calculation model for the second true defect coefficient is as follows: Where Ci represents the second true defect coefficient of the i-th target echo signal segment; norm represents the normalization operation; The bandwidth characteristic coefficient indicates that the smaller the difference between the bandwidth of the target echo signal segment (the frequency width where the power spectrum drops to half of the peak value) and the bandwidth of the incident pulse, the more consistent it is with the fact that the reflection of a real defect will not significantly change the bandwidth characteristics of the pulse. Bwi is the bandwidth of the i-th target echo signal segment, and Bw0i is the bandwidth of the incident pulse corresponding to the i-th target echo signal segment. is the entropy characteristic coefficient. The smaller the entropy value, the more it conforms to the concentrated characteristics of the real defect echo spectrum. Sei is the bandwidth of the power spectrum of the i-th target echo signal segment. Here, fc0i represents the spectral centroid characteristic coefficient, indicating the difference between the theoretical spectral centroid (the center frequency of the energy distribution of the signal spectrum, directly calculated using existing techniques) and the spectral centroid of the i-th target echo signal segment. The smaller the difference, the more it conforms to the predictable characteristics of the true defect spectral centroid. fc0i represents the theoretical spectral centroid, and fci represents the spectral centroid of the i-th target echo signal segment. The theoretical spectral centroid can be obtained based on the attenuation coefficient. fi represents the spectral centroid of the incident pulse in the i-th target echo signal segment, a represents the material attenuation coefficient, and d is the propagation depth, which is obtained based on the spectral centroid of the incident pulse and the current stainless steel pipe material attenuation coefficient (the coefficient of energy attenuation with distance when ultrasound propagates in a material, which is known through material properties or obtained through experimental calibration) and depth (obtained based on echo time).

[0035] The attenuation coefficient is mainly obtained through experimental calibration. The main steps are: 1. Measure the bottom echo amplitude of samples with different thicknesses (or the bottom echo of the same sample at different distances).

[0036] 2. According to the ultrasonic attenuation formula: A=A0*exp(-2αd), where A: echo amplitude, representing the signal amplitude returned by the ultrasonic wave after propagation distance d; A0: initial amplitude, representing the initial signal amplitude of the ultrasonic wave at the time of transmission; α: attenuation coefficient, representing the rate of energy attenuation when the ultrasonic wave propagates in the material; d: propagation distance, representing the one-way distance of the ultrasonic wave from the transmitting probe to the bottom surface (or defect) of the material.

[0037] 3. Fit the results of multiple measurements to obtain the attenuation coefficient α. The main steps for determining the depth are: 1. Determine the arrival time t of the defect echo (the time from the start of the emission pulse to the peak value of the defect echo).

[0038] 2. The speed of sound v is known (the speed at which ultrasonic waves propagate in a material; for stainless steel, it is usually around 5900 m / s, but the exact value needs to be determined based on the material).

[0039] 3. Depth d = v * t / 2 (because the sound wave travels back and forth).

[0040] This allows us to obtain the second true defect coefficient for each target echo signal segment.

[0041] After obtaining the first and second true defect coefficients of the target echo signal segment respectively, the combined true defect coefficient can be obtained by combining the two. Specifically, the combined true defect coefficient of the target echo signal segment is obtained by averaging the first and second true defect coefficients.

[0042] The defect detection module is used to obtain the defect level of the location on the pipe corresponding to the target echo signal segment by using the initial defect probability of the target echo signal segment and the comprehensive real defect coefficient; and to determine whether to issue a defect warning for the location on the pipe corresponding to the target echo signal segment based on the defect level.

[0043] The initial defect probability and comprehensive true defect coefficient of the target echo signal segment were obtained above. By combining the two, the defect degree of the position on the pipe corresponding to the target echo signal segment can be obtained. Specifically, the defect degree of the position on the pipe corresponding to the target echo signal segment is obtained by multiplying the initial defect probability and comprehensive true defect coefficient of the target echo signal segment.

[0044] After obtaining the defect level of the corresponding position on the stainless steel pipe fitting from the target echo signal segment of the target location, it is possible to determine whether there is a real defect inside the stainless steel pipe fitting based on the corresponding defect level. That is, when the ultrasonic probe moves to a certain position on the stainless steel pipe fitting, and the defect level of the corresponding position on the pipe fitting is greater than or equal to the judgment threshold (the reference value of the judgment threshold in this application is 0.5, which is set by relevant personnel according to the working requirements and materials of different stainless steel pipe fittings), a defect warning is issued for the position on the pipe fitting corresponding to the target echo signal segment, indicating that there is a high probability of a defect at the position on the pipe fitting corresponding to the target echo signal segment, and the current position is marked. After the entire pipe fitting inspection is completed, the marked position can remind relevant personnel to repair or further process the defect area inside the stainless steel pipe fitting.

[0045] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0046] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. The above descriptions are merely preferred embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A system for detecting internal defects in stainless steel pipe fittings based on ultrasonic testing technology, characterized in that, The system includes: an initial detection module for detecting pipe fittings using an ultrasonic probe, acquiring different echo signal segments, and designating any one echo signal segment as a target echo signal segment; obtaining the initial defect probability of the target echo signal segment based on the number of amplitudes exceeding a defect threshold in the target echo signal segment; a depth analysis module for obtaining a first true defect coefficient based on the rise time of the incident pulse and return signal of the target echo signal segment, the energy values ​​of the left and right parts of the main peak point of the target echo signal segment, and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point; obtaining a second true defect coefficient based on the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the spectral centroid of the target echo signal segment; fusing the first and second true defect coefficients to obtain a comprehensive true defect coefficient; and a defect detection module for obtaining the defect severity at the location on the pipe fitting corresponding to the target echo signal segment using the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient; and determining whether to issue a defect warning for the location on the pipe fitting corresponding to the target echo signal segment based on the defect severity.

2. The stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology according to claim 1, characterized in that, The method of obtaining the initial defect probability of the target echo signal segment based on the number of amplitudes exceeding the defect threshold in the target echo signal segment includes: subtracting each amplitude exceeding the defect threshold in the target echo signal segment from the defect threshold and then summing the results to obtain the degree of exceeding the limit; and multiplying the normalized value of the degree of exceeding the limit by the normalized value of the number of amplitudes exceeding the defect threshold in the target echo signal segment to obtain the initial defect probability of the target echo signal segment.

3. The stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology according to claim 1, characterized in that, The method of obtaining the first true defect coefficient based on the rise time of the incident pulse and the return signal of the target echo signal segment, the energy values ​​of the left and right parts of the main peak point of the target echo signal segment, and the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point includes: dividing the target echo signal segment into left and right parts with the main peak point as the center; obtaining the absolute value of the difference in energy values ​​between the left and right parts and the sum of the hyperparameters, and taking the reciprocal to obtain the waveform symmetry; subtracting the ratio of the rise time of the incident pulse to the rise time of the return signal of the target echo signal segment from a first preset value and taking the absolute value to obtain the rise time ratio coefficient; and then applying the rise time ratio coefficient to the ratio of the incident pulse to the return signal of the target echo signal segment. Example: Add the coefficient to the hyperparameter and take the reciprocal to obtain the rise time consistency; obtain the reciprocal of the difference between the peak value of the main peak point and the nearest valley value to the right of the main peak point in the target echo signal segment, and record it as the peak characteristic coefficient; multiply the normalized value of waveform symmetry, the normalized value of rise time consistency and the normalized value of peak characteristic coefficient to obtain the first characteristic coefficient; obtain the peak influence coefficient. If the number of peak points in the target echo signal segment is greater than 1, the peak influence coefficient is recorded as 0. When the number of peak points is equal to 1, the peak influence coefficient is 1; add the normalized value of the first characteristic coefficient and the normalized value of the peak influence coefficient and then normalize to obtain the first true defect coefficient.

4. The stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology according to claim 1, characterized in that, The method for obtaining the second true defect coefficient based on the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse, the entropy value of the power spectrum of the target echo signal segment, the theoretical spectral centroid, and the spectral centroid of the target echo signal segment includes: adding the absolute value of the difference between the bandwidth of the target echo signal segment and the bandwidth of the incident pulse to the hyperparameter and taking its reciprocal to obtain the bandwidth characteristic coefficient; obtaining the reciprocal of the entropy value of the power spectrum of the target echo signal segment to obtain the entropy characteristic coefficient; using the hyperparameter to add the absolute value of the difference between the theoretical spectral centroid and the spectral centroid of the target echo signal segment and taking its reciprocal to obtain the spectral centroid characteristic coefficient; and multiplying and normalizing the normalized values ​​of the bandwidth characteristic coefficient, the entropy characteristic coefficient, and the spectral centroid characteristic coefficient to obtain the second true defect coefficient of the target echo signal segment.

5. The stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology according to claim 1, characterized in that, The step of fusing the first true defect coefficient and the second true defect coefficient to obtain the comprehensive true defect coefficient includes: calculating the average of the first true defect coefficient and the second true defect coefficient of the target echo signal segment to obtain the comprehensive true defect coefficient of the target echo signal segment.

6. The stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology according to claim 1, characterized in that, The method of obtaining the defect level of the pipe fitting corresponding to the target echo signal segment by using the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient includes: multiplying the initial defect probability of the target echo signal segment and the comprehensive true defect coefficient to obtain the defect level of the pipe fitting corresponding to the target echo signal segment.

7. The stainless steel pipe fitting internal defect detection system based on ultrasonic testing technology according to claim 1, characterized in that, The step of determining whether to issue a defect warning for the location on the pipe corresponding to the target echo signal segment based on the degree of defect includes: if the degree of defect on the location on the pipe corresponding to the target echo signal segment is greater than or equal to the judgment threshold, then issue a defect warning for the location on the pipe corresponding to the target echo signal segment.

Images