Automatic flaw detection method, device and system for long-shaft forged rod

Abstract

The invention relates to the technical field of forging rod flaw detection, in particular to an automatic flaw detection method, device and system for a long-axis forging rod, and the method comprises the following steps: collecting ultrasonic data of each detection point of the long-axis forging rod, and calculating the sound path distance, the wave crest and the wave trough of the detection point; wave crests with the same depth in the ultrasonic data are screened, and a to-be-judged detection point is determined; the same crack possibility of the adjacent to-be-judged detection points is determined, the types of crack defects are determined according to the same crack possibility, and the types of the crack defects comprise transverse crack defects and longitudinal crack defects; and according to the same crack possibility of the adjacent to-be-judged detection points corresponding to the transverse crack defects and the distance between the to-be-judged detection points, screening the to-be-judged transverse crack detection points with the same defects, and determining the flaw detection result of the long-axis forging rod according to the positions and the sonic path distances of the to-be-judged detection points corresponding to the types of the crack defects. The flaw detection precision can be improved on the basis of ensuring the flaw detection efficiency and the cost.

Description

Technical Field

[0001] The present application relates to the technical field of forging rod flaw detection, and in particular to a method, device and system for automatic flaw detection of long-axis forging rods. Background Art

[0002] Long-shaft forgings are shaft-like forgings in mechanical systems, with axial dimensions significantly larger than their radial and transverse dimensions. Automated flaw detection of long-shaft forgings can be performed using a variety of methods, including ultrasonic testing, eddy current testing, magnetic flux leakage testing, and infrared testing. Ultrasonic testing is typically used to reduce human error and improve detection accuracy and efficiency.

[0003] Ultrasonic flaw detection of long-axis forging rods requires testing core parts such as nuclear power rotor shafts at a preset step length. When setting testing points on the long-axis forging rods with a preset step length, if the location where the long-axis forging rod is damaged is far away from the adjacent testing points, the flaw detection results of the long-axis forging rods are easily affected, resulting in inaccurate flaw detection or missed detection. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method, device and system for automatic flaw detection of long-axis forged bars. The technical solutions adopted are as follows: In a first aspect, an embodiment of the present application provides a method for automatic flaw detection of a long-axis forged bar, the method comprising the following steps: Collecting ultrasonic data at each detection point of the long-axis forged rod, calculating the acoustic path of each detection point of the long-axis forged rod, and the peak and trough of the ultrasonic data at each detection point, wherein the ultrasonic data corresponds to the propagation time of an ultrasonic wave and the intensity of an ultrasonic signal; Based on the differences between adjacent peaks in the ultrasonic data of the same detection point, peaks at the same depth in the ultrasonic data of the same detection point are screened out; based on the positional relationship between each peak and the adjacent trough in the ultrasonic data of the same detection point and the intensity value of the corresponding ultrasonic signal, whether the detection point is a detection point to be determined is determined; based on the differences between all peaks at the same depth in the ultrasonic data of the adjacent detection points to be determined and the crack probability of all peaks in the ultrasonic data, the probability of the adjacent detection points to be determined being the same crack is determined; based on the probability of the adjacent detection points to be determined being the same crack on the same cross section of the long-axis forging bar, the type of crack defect is determined, and the types of crack defects include transverse crack defects and longitudinal crack defects; According to the possibility of the same cracks in adjacent detection points to be determined corresponding to the transverse crack defect and the distance between the detection points to be determined, the transverse crack detection points to be determined with the same defect are screened, and secondary detection points are set between the transverse crack detection points to be determined with the same defect. The transverse crack detection points to be determined with the same defect in the secondary detection points are screened, and the long-axis forging bar flaw detection results are determined according to the position and sound range of the detection points to be determined corresponding to the type of crack defect.

[0005] Furthermore, the method for screening the peaks at the same depth in the ultrasonic data of the same detection point is: Any peak of the ultrasonic data of the same detection point is recorded as the target peak, the peak immediately preceding the target peak is recorded as the adjacent peak, and the time interval between the propagation time of the ultrasonic waves corresponding to the target peak and the adjacent peak is recorded as the adjacent interval of the target peak; the average of the adjacent intervals of all peaks of the ultrasonic data of the same detection point is recorded as the adjacent average interval of the detection point; the difference between the adjacent interval of the target peak and the adjacent average interval of the detection points corresponding to the target peak is recorded as the first difference of the target peak, and the hyperbolic tangent value of the ratio of the first difference of the target peak to the adjacent interval is recorded as the same defect characteristic value of the target peak; The same defect feature values ​​of all peaks in the ultrasonic data of the same detection point are arranged in sequence to obtain a defect feature sequence. The same defect feature sequence is divided into defect feature subsequences from the position where the defect feature value is less than 0. The peaks corresponding to the same defect feature value contained in the same defect feature subsequence are recorded as peaks of the same depth.

[0006] Furthermore, the method for determining the detection point to be determined is: Select an adjacent point at half the distance between the target peak and the adjacent trough, take the target peak as the vertex, and record the angle formed by the rays starting from the vertex and passing through the two adjacent points as the peak angle of the target peak; The mean of the intensities of the ultrasonic signals corresponding to the adjacent troughs of the target peak is recorded as the mean of the adjacent troughs of the target peak; the absolute value of the difference in the intensities of the ultrasonic signals corresponding to the adjacent troughs of the target peak is recorded as the adjacent trough difference of the target peak; the cumulative sum of the products of the mean of the adjacent troughs of all peaks at the same depth as the target peak and the adjacent trough differences is recorded as the first cumulative sum of the target peak; the normalized value of the product of the peak angle of the target peak and the first cumulative sum is recorded as the crack possibility of the target peak; When there is a peak in the wave peak of the ultrasonic data of the detection point whose crack possibility is greater than or equal to a preset first crack evaluation threshold, the detection point is recorded as a detection point to be determined.

[0007] Furthermore, the process of determining the possibility of the same crack of the adjacent detection points to be determined is as follows: Select peaks at the same depth in the ultrasonic data of adjacent detection points to be determined. For each detection point to be determined, arrange the intensities of the ultrasonic signals of the peaks selected at the detection point to be determined in sequence to obtain a first vector of the detection point to be determined; record the dot product of the first vectors of the adjacent detection points to be determined as the first dot product of the adjacent detection points to be determined at the depth corresponding to the first vector; and record the normalized value of the cumulative sum of the first dot products of all adjacent detection points to be determined at the same depth as the defect waveform similarity of the adjacent detection points to be determined; The average of the crack probabilities of all peaks of the ultrasonic data of the detection point to be determined is recorded as the average crack probability of the detection point to be determined, and the absolute value of the difference between the average crack probabilities of adjacent detection points to be determined is recorded as the first absolute value of the adjacent detection points to be determined; the normalized value of the ratio of the defect waveform similarity of adjacent detection points to be determined to the first absolute value is recorded as the same crack probability of the adjacent detection points to be determined.

[0008] Furthermore, the method of determining the type of crack defect based on the possibility of the same crack at adjacent detection points to be determined on the same cross section of the long-axis forged bar includes the following specific processes: Adjacent detection points to be determined on the same cross section of the long-axis forged bar, whose probability of the same crack is greater than or equal to a preset second crack evaluation threshold, are all recorded as detection points to be determined for the same defect; When adjacent detection points to be determined on the same cross section of the long-axis forged bar are detection points to be determined for the same defect, the crack defects corresponding to the adjacent detection points to be determined are recorded as transverse crack defects; When adjacent detection points to be determined on the same cross section of the long-axis forged bar do not have detection points to be determined with the same defect, the crack defects corresponding to the adjacent detection points to be determined are recorded as longitudinal crack defects.

[0009] Furthermore, the method of screening the transverse crack detection points to be determined with the same defect according to the likelihood of the same crack of adjacent detection points to be determined corresponding to the transverse crack defect and the distance between the detection points to be determined includes the following specific process: The ratio of the likelihood of the same crack of adjacent detection points to be determined corresponding to the transverse crack defect to the Euclidean distance between the adjacent detection points to be determined is recorded as the first ratio of the adjacent detection points to be determined corresponding to the transverse crack defect, and the hyperbolic tangent of the first ratio of the adjacent detection points to be determined corresponding to the transverse crack defect is recorded as the defect elongation of the adjacent detection points to be determined corresponding to the transverse crack defect; Adjacent detection points to be determined corresponding to transverse crack defects with a defect elongation greater than or equal to a preset third crack evaluation threshold are all recorded as transverse crack detection points to be determined of the same defect.

[0010] Furthermore, the method for setting the secondary detection point is: The midpoint of the line segment of the two identical transverse cracks to be determined detection points is used as the secondary detection point.

[0011] Furthermore, the method of determining the flaw detection result of the long-axis forged bar according to the position and sound path of the detection point to be determined corresponding to the type of the crack defect includes the following specific steps: When a transverse crack defect is identified, the triangulation method is used to determine the defect position of the transverse crack defect in the long axis forging bar based on the sound path and position of all the transverse crack detection points corresponding to the same defect; When a longitudinal crack defect is identified, the triangulation method is used to determine the defect position of the longitudinal crack defect in the long axis forging bar based on the sound path and position of all the detection points to be determined corresponding to the longitudinal crack defect; When the detection point to be determined is not identified, the flaw detection result of the long-axis forging bar is that there is no defect.

[0012] In a second aspect, an embodiment of the present application provides an automatic flaw detection device for a long-axis forged bar, the automatic flaw detection device comprising: a flaw detection data acquisition module, a defect category identification module, and a flaw detection result determination module.

[0013] The flaw detection data acquisition module is used to collect ultrasonic data of each detection point of the long-axis forging bar, calculate the sound path of each detection point of the long-axis forging bar, and the peak and trough of the ultrasonic data of each detection point; A defect category identification module is used to screen out peaks of the same depth in the ultrasonic data of the same detection point based on the differences between adjacent peaks in the ultrasonic data of the same detection point, and determine whether the detection point is a detection point to be determined based on the positional relationship between each peak and the adjacent trough in the ultrasonic data of the same detection point and the intensity value of the corresponding ultrasonic signal; determine the probability of the adjacent detection points to be determined having the same crack based on the differences between all peaks of the same depth in the ultrasonic data of the adjacent detection points to be determined and the crack probability of all peaks in the ultrasonic data; and determine the type of crack defect based on the probability of the adjacent detection points to be determined having the same crack on the same cross section of the long-axis forging bar. The types of crack defects include transverse crack defects and longitudinal crack defects. The flaw detection result determination module is used to screen the transverse crack detection points to be determined with the same defect based on the possibility of the same cracks in adjacent detection points to be determined corresponding to the transverse crack defects and the distance between the detection points to be determined, set secondary detection points between the transverse crack detection points to be determined with the same defect, screen the transverse crack detection points to be determined with the same defect among the secondary detection points, and determine the flaw detection results of the long-axis forging bar based on the position and sound range of the detection points to be determined corresponding to the type of crack defects.

[0014] In a third aspect, an embodiment of the present application also provides an automatic flaw detection system for long-axis forging bars, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of any one of the above methods when executing the computer program.

[0015] As can be seen from the above embodiments, the method, device, and system for automatic flaw detection of long-axis forged bars provided in the embodiments of the present application have at least the following beneficial effects: The present application determines whether there is a defect within the detection range of the detection point based on the positional relationship between adjacent wave crests and wave troughs and the corresponding intensity values ​​of the ultrasonic signal, and records the detection point with a defect within the detection range as the detection point to be determined; the depth of the internal defect of the long-axis forging rod is different, and the peaks of the same depth are determined based on the difference between adjacent wave crests in the ultrasonic data of the same detection point, and the wave crests corresponding to the same depth in the ultrasonic data are analyzed to evaluate the possibility that the adjacent detection points to be determined correspond to the same defect, obtain the possibility of the same crack of the adjacent detection points to be determined, and determine whether the crack defect is a transverse crack defect or a longitudinal crack defect based on the possibility of the same crack of the adjacent detection points to be determined on the same cross section of the long-axis forging rod; when a transverse crack defect is detected, the transverse crack defect will cross multiple cross sections of multiple long-axis forging rods. In order to further determine the precise defect location, further evaluate the possibility that adjacent detection points to be determined correspond to the same transverse crack defect, record the detection points to be determined corresponding to the same transverse crack defect as the transverse crack detection points to be determined with the same defect, and set secondary detection points to evaluate the possibility that the secondary detection points and adjacent detection points to be determined correspond to the same transverse crack defect under a smaller measurement range, and record the secondary detection points corresponding to the same transverse crack defect as the transverse crack detection points to be determined with the same defect; finally, determine the long-axis forging bar flaw detection results according to the position and sound range of the detection points to be determined corresponding to the type of crack defect, achieve accurate long-axis forging bar flaw detection, avoid cost waste caused by over-fine detection and missed detection caused by insufficient detection coverage, and improve detection accuracy while ensuring detection efficiency and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1 A flowchart of a method for automatically detecting flaws in a long-axis forged bar provided in one embodiment of the present application; Figure 2A schematic structural diagram of an automatic flaw detection device for a long-axis forged bar provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] To further illustrate the technical means and effects employed by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effects of a method, device, and system for automatic flaw detection of long-axis forged bars proposed in this application. All technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which this application belongs.

[0019] The following describes in detail the specific solutions of the method, device and system for automatic flaw detection of long-axis forged bars provided by the present application with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flowchart of a method for automatic flaw detection of a long-axis forged bar provided by an embodiment of the present application, the method comprising the following steps: Step S001: collecting ultrasonic data of each detection point of the long-axis forged rod, calculating the sound path of each detection point of the long-axis forged rod, and calculating the peak and trough of the ultrasonic data of each detection point.

[0021] Measure the diameter of the long axis forging bar.

[0022] In this embodiment, an ultrasonic flaw detector is selected as the ultrasonic testing equipment, and a long-axis forged bar is horizontally mounted on a testing table to perform flaw detection on the long-axis forged bar. A position is randomly selected on the edge of the cross section of the long-axis forged bar as a testing point, and two testing points are set at equal distances from the randomly selected position. Therefore, there are a total of three testing points on the edge of the cross section of the long-axis forged bar, and the distances between adjacent testing points are equal. Three testing points are also set at the corresponding positions on the edge of the cross section on the other side of the long-axis forged bar. The testing points corresponding to the edges of the cross sections on both sides of the long-axis forged bar are respectively determined on three line segments on the side of the long-axis forged bar. On each of the determined line segments, a testing point is set starting from one end of the long-axis forged bar and with a preset step size.

[0023] Among them, one tenth of the length of the long axis forging rod is recorded as the first parameter, and the minimum value between the first parameter and 10 cm is recorded as the second parameter. In this embodiment, the value of the preset step length is the second parameter.

[0024] Ultrasonic data is collected at each inspection point of the long-axis forging rod to obtain the peak, trough and sound path of the ultrasonic data.

[0025] Among them, it can be understood that the ultrasonic data is expressed as a data curve, the horizontal axis of the data curve is the propagation time of the ultrasonic wave, and the vertical axis is the intensity of the ultrasonic signal; according to the ultrasonic data and the diameter of the long-axis forging rod, determining the peaks, troughs and sound path of the ultrasonic data are all well-known technologies and will not be repeated here.

[0026] At this point, the diameter of the long-axis forging rod, the sound path of each detection point of the long-axis forging rod, and the peaks and troughs of the ultrasonic data of each detection point of the long-axis forging rod are obtained.

[0027] Step S002: based on the difference between adjacent peaks in the ultrasonic data of the same detection point, filter out the peaks of the same depth in the ultrasonic data of the same detection point, and judge whether the detection point is a detection point to be determined based on the positional relationship between each peak and the adjacent trough in the ultrasonic data of the same detection point and the intensity value of the corresponding ultrasonic signal; based on the difference between all peaks of the same depth in the ultrasonic data of the adjacent detection points to be determined and the crack probability of all peaks in the ultrasonic data, determine the same crack probability of the adjacent detection points to be determined; based on the same crack probability of the adjacent detection points to be determined on the same cross section of the long-axis forging bar, determine the type of crack defect, which includes two types of crack defects: transverse crack defects and longitudinal crack defects.

[0028] Small and scattered defects inside long-axis forging bars have little impact on the quality of long-axis forging bars, while long and dense cracks, pores and other defects inside long-axis forging bars have a greater impact on the quality of long-axis forging bars. There may be multiple defects at different depths inside the long-axis forging bars. Therefore, it is necessary to divide the depths of suspected defects.

[0029] Any peak of the ultrasonic data of the same detection point is recorded as the target peak, the previous adjacent peak of the target peak is recorded as the adjacent peak, and the time interval between the propagation time of the ultrasonic waves corresponding to the target peak and the adjacent peak is recorded as the adjacent interval of the target peak; the average of the adjacent intervals of all peaks of the ultrasonic data of the same detection point is recorded as the adjacent average interval of the detection point; the difference between the adjacent interval of the target peak and the adjacent average interval of the detection point corresponding to the target peak is recorded as the first difference of the target peak, and the hyperbolic tangent value of the ratio of the first difference of the target peak to the adjacent interval is recorded as the same defect characteristic value of the target peak.

[0030] It should be noted that the first peak of the ultrasonic data of the same detection point does not have the previous adjacent peak. Therefore, when calculating the adjacent average interval of the detection points, the adjacent interval of the first peak of the ultrasonic data of the same detection point is calculated.

[0031] The same defect characteristic value of any peak in the ultrasonic data of the same detection point can be obtained in the same way.

[0032] The same defect feature values ​​of all peaks in the ultrasonic data of the same detection point are arranged in sequence to obtain a defect feature sequence. The same defect feature sequence is divided into defect feature subsequences from the position where the defect feature value is less than 0. The peaks corresponding to the same defect feature value contained in the same defect feature subsequence are recorded as peaks of the same depth.

[0033] The crack possibility of the target wave peak is determined based on the positional relationship between the target wave peak and the wave trough adjacent to the target wave peak and the corresponding ultrasonic signal intensity value.

[0034] Select an adjacent position point at half the distance between the target peak and the adjacent trough, take the target peak as the vertex, and record the angle formed by the rays starting from the vertex and passing through the two adjacent position points as the peak angle of the target peak.

[0035] Among them, the adjacent troughs of the target peak are the adjacent troughs before and after the target peak, that is, there are two adjacent troughs of the target peak, and there are also two adjacent position points selected from the target peak; when there are less than two adjacent position points selected for a peak, the peak is not analyzed.

[0036] The surface of the crack inside the long-axis forged rod is rough and irregular in shape, which will cause the reflection and scattering characteristics of different crack positions to be different from those at the position where there is no crack inside the long-axis forged rod. Specifically, the ripple echo at the crack position usually presents a multi-peak structure, with multiple small peaks next to the main peak, showing a peak branching phenomenon. The fluctuation falls from the peak to the trough. Due to the chaotic reflection, the waveform echo has multiple peaks and continues to propagate longitudinally before finally falling back. Therefore, the height and number of adjacent troughs can be analyzed to evaluate the possibility of cracks inside the long-axis forged rod.

[0037] The mean of the intensities of the ultrasonic signals corresponding to the adjacent troughs of the target peak is recorded as the mean of the adjacent troughs of the target peak; the absolute value of the difference in the intensities of the ultrasonic signals corresponding to the adjacent troughs of the target peak is recorded as the adjacent trough difference of the target peak; the cumulative sum of the products of the mean of the adjacent troughs of all peaks at the same depth as the target peak and the adjacent trough differences is recorded as the first cumulative sum of the target peak; the normalized value of the product of the peak angle of the target peak and the first cumulative sum is recorded as the crack possibility of the target peak.

[0038] It should be noted that this embodiment uses the Z-Score standard normalization method to calculate the normalized value. In actual application, the implementer can use other existing methods such as maximum and minimum value normalization method, sigmoid function, etc. to calculate the normalized value, which is not limited here.

[0039] The crack probability of any peak in the ultrasonic data of the same detection point can be obtained using the same method.

[0040] When there is a peak in the ultrasonic data of the detection point with a crack possibility greater than or equal to the first crack evaluation threshold, it is determined that there is a defect within the detection range of the detection point, and the detection point is recorded as a detection point to be determined.

[0041] The first crack evaluation threshold is a preset parameter value. In this embodiment, the first crack evaluation threshold is set to 0.6.

[0042] In order to avoid the accidental influence of the detection results of a single detection point on the flaw detection results and to improve the accuracy of the flaw detection of the long-axis forged rod, this embodiment sets multiple detection points with close spatial positions on the long-axis forged rod, and the detection areas of different detection points overlap. Therefore, the same defect in the long-axis forged rod is often detected at the same time at different detection points. In order to improve the accuracy of the flaw detection of the long-axis forged rod, the adjacent detection points to be determined on the same cross section of the long-axis forged rod are further analyzed.

[0043] The detection points to be determined on the same cross-sectional edge of the long-axis forged rod are recorded as adjacent detection points to be determined. The detection points corresponding to the cross-sectional edges on both sides of the long-axis forged rod are adjacent to the detection points to be determined on each of the three line segments respectively determined on the side of the long-axis forged rod. The adjacent detection points to be determined on each line segment are also recorded as adjacent detection points to be determined.

[0044] Select the peaks at the same depth in the ultrasonic data of adjacent detection points to be determined. For each detection point to be determined, arrange the intensities of the ultrasonic signals of the peaks selected by the detection point to be determined in sequence to obtain the first vector of the detection point to be determined; record the dot product of the first vectors of the adjacent detection points to be determined as the first dot product of the adjacent detection points to be determined at the depth corresponding to the first vector, and record the normalized value of the cumulative sum of the first dot products of the adjacent detection points to be determined at the same depth as the defect waveform similarity of the adjacent detection points to be determined.

[0045] It should be noted that when calculating the dot product of two first vectors, if the lengths of the two first vectors are unequal, align the first values ​​of the vectors and intercept the first vector with the longest length to make it equal to the length of the other first vector; when there is only one first dot product at the same depth, the depth is not calculated when calculating the defect waveform similarity of the detection point to be determined.

[0046] The average of the crack probabilities of all peaks of the ultrasonic data of the detection point to be determined is recorded as the average crack probability of the detection point to be determined, and the absolute value of the difference between the average crack probabilities of adjacent detection points to be determined is recorded as the first absolute value of the adjacent detection points to be determined; the normalized value of the ratio of the defect waveform similarity of adjacent detection points to be determined to the first absolute value is recorded as the same crack probability of the adjacent detection points to be determined.

[0047] Adjacent detection points to be determined on the same cross-section of a long-axis forged bar, where the probability of the same crack is greater than or equal to the second crack evaluation threshold, are all recorded as detection points to be determined for the same defect. When adjacent detection points to be determined on the same cross-section of a long-axis forged bar are detection points to be determined for the same defect, the crack defects corresponding to the adjacent detection points to be determined are recorded as transverse crack defects; when adjacent detection points to be determined on the same cross-section of a long-axis forged bar do not have detection points to be determined for the same defect, the crack defects corresponding to the adjacent detection points to be determined are recorded as longitudinal crack defects.

[0048] The second crack evaluation threshold is a preset parameter value. In this embodiment, the value of the second crack evaluation threshold is 0.8.

[0049] At this point, the type of crack defect is determined.

[0050] Step S003: Based on the possibility of the same cracks in adjacent detection points to be determined corresponding to the transverse crack defect and the distance between the detection points to be determined, the transverse crack detection points to be determined with the same defect are screened, secondary detection points are set between the transverse crack detection points to be determined with the same defect, the transverse crack detection points to be determined with the same defect among the secondary detection points are screened, and the long-axis forging bar flaw detection results are determined based on the position and sound range of the detection points to be determined corresponding to the type of crack defect.

[0051] When a transverse crack defect is detected, the transverse crack defect will cross multiple cross sections of multiple long axis forging bars. In order to further determine the precise defect location, it is necessary to set a second detection point between the detection points to be determined.

[0052] The defect extension of the adjacent to-be-determined detection points corresponding to the transverse crack defect is determined according to the same crack possibility of the adjacent to-be-determined detection points corresponding to the transverse crack defect and the distance between the to-be-determined detection points.

[0053] The ratio of the probability of the same crack of adjacent detection points to be determined corresponding to the transverse crack defect to the Euclidean distance between the adjacent detection points to be determined is recorded as the first ratio of the adjacent detection points to be determined corresponding to the transverse crack defect, and the hyperbolic tangent value of the first ratio of the adjacent detection points to be determined corresponding to the transverse crack defect is recorded as the defect elongation of the adjacent detection points to be determined corresponding to the transverse crack defect.

[0054] Adjacent detection points to be determined corresponding to transverse crack defects with a defect elongation greater than or equal to the third crack evaluation threshold are all recorded as transverse crack detection points to be determined of the same defect.

[0055] The third crack evaluation threshold is a preset parameter value. In this embodiment, the value of the third crack evaluation threshold is 0.7.

[0056] A secondary detection point is set between every two transverse crack detection points of the same defect. The midpoint of the line segment defining the two transverse crack detection points of the same defect is used as the secondary detection point. The defect elongation between each secondary detection point and its corresponding transverse crack detection point of the same defect is calculated, and the transverse crack detection points of the same defect are marked based on the defect elongation. Specifically, secondary detection points with a defect elongation greater than or equal to the third crack evaluation threshold and their corresponding transverse crack detection points of the same defect are both marked as transverse crack detection points of the same defect.

[0057] When a transverse crack defect is detected, the triangulation positioning method is used to determine the defect position of the transverse crack defect in the long-axis forged rod based on the sound range and position of all the transverse crack detection points corresponding to the same defect. The defect position of the transverse crack defect in the long-axis forged rod is the transverse crack defect detection result of the long-axis forged rod; when a longitudinal crack defect is detected, the triangulation positioning method is used to determine the defect position of the longitudinal crack defect in the long-axis forged rod based on the sound range and position of all the detection points corresponding to the longitudinal crack defect. The defect position of the longitudinal crack defect in the long-axis forged rod is the longitudinal crack defect detection result of the long-axis forged rod; when the detection point to be determined is not identified, the detection result of the long-axis forged rod is that there is no defect.

[0058] At this point, automatic flaw detection of long-axis forging bars is realized.

[0059] See also Figure 2 , Figure 2 This is a schematic diagram of the structure of an automatic flaw detection device for a long-axis forged bar provided by an embodiment of the present application. The various units included in the device in this embodiment are used to perform the various steps in the embodiment corresponding to the automatic flaw detection method for a long-axis forged bar. Figure 2 The automatic flaw detection device includes: a flaw detection data acquisition module, a defect category identification module, and a flaw detection result determination module.

[0060] The flaw detection data acquisition module is used to collect ultrasonic data of each detection point of the long-axis forging bar, calculate the sound path of each detection point of the long-axis forging bar, and the peak and trough of the ultrasonic data of each detection point; A defect category identification module is used to screen out peaks of the same depth in the ultrasonic data of the same detection point based on the differences between adjacent peaks in the ultrasonic data of the same detection point, and determine whether the detection point is a detection point to be determined based on the positional relationship between each peak and the adjacent trough in the ultrasonic data of the same detection point and the intensity value of the corresponding ultrasonic signal; determine the probability of the adjacent detection points to be determined having the same crack based on the differences between all peaks of the same depth in the ultrasonic data of the adjacent detection points to be determined and the crack probability of all peaks in the ultrasonic data; and determine the type of crack defect based on the probability of the adjacent detection points to be determined having the same crack on the same cross section of the long-axis forging bar. The types of crack defects include transverse crack defects and longitudinal crack defects. The flaw detection result determination module is used to screen the transverse crack detection points to be determined with the same defect based on the possibility of the same cracks in adjacent detection points to be determined corresponding to the transverse crack defects and the distance between the detection points to be determined, set secondary detection points between the transverse crack detection points to be determined with the same defect, screen the transverse crack detection points to be determined with the same defect among the secondary detection points, and determine the flaw detection results of the long-axis forging bar based on the position and sound range of the detection points to be determined corresponding to the type of crack defects.

[0061] Based on the same inventive concept as the above-mentioned method, an embodiment of the present application also provides an automatic flaw detection system for long-axis forged bars, comprising a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned methods for automatic flaw detection of long-axis forged bars are implemented.

[0062] It should be understood that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0063] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0064] The above content is only an implementation method of the present application and is not intended to limit the scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the scope of protection of the present application.

Claims

1. A method for automatic flaw detection of long-axis forging bars, characterized in that: The method comprises the following steps: Collecting ultrasonic data at each detection point of the long-axis forged rod, calculating the acoustic path of each detection point of the long-axis forged rod, and the peak and trough of the ultrasonic data at each detection point, wherein the ultrasonic data corresponds to the propagation time of an ultrasonic wave and the intensity of an ultrasonic signal; Based on the differences between adjacent peaks in the ultrasonic data of the same detection point, peaks at the same depth in the ultrasonic data of the same detection point are screened out; based on the positional relationship between each peak and the adjacent trough in the ultrasonic data of the same detection point and the intensity value of the corresponding ultrasonic signal, whether the detection point is a detection point to be determined is determined; based on the differences between all peaks at the same depth in the ultrasonic data of the adjacent detection points to be determined and the crack probability of all peaks in the ultrasonic data, the probability of the adjacent detection points to be determined being the same crack is determined; based on the probability of the adjacent detection points to be determined being the same crack on the same cross section of the long-axis forging bar, the type of crack defect is determined, and the types of crack defects include transverse crack defects and longitudinal crack defects; According to the possibility of the same cracks in adjacent detection points to be determined corresponding to the transverse crack defect and the distance between the detection points to be determined, the transverse crack detection points to be determined with the same defect are screened, and secondary detection points are set between the transverse crack detection points to be determined with the same defect. The transverse crack detection points to be determined with the same defect in the secondary detection points are screened, and the long-axis forging bar flaw detection results are determined according to the position and sound range of the detection points to be determined corresponding to the type of crack defect.

2. The automatic flaw detection method for a long-axis forged bar according to claim 1, characterized in that: The method for screening the peaks at the same depth in the ultrasonic data of the same detection point is: Any peak of the ultrasonic data of the same detection point is recorded as the target peak, the peak immediately preceding the target peak is recorded as the adjacent peak, and the time interval between the propagation time of the ultrasonic waves corresponding to the target peak and the adjacent peak is recorded as the adjacent interval of the target peak; the average of the adjacent intervals of all peaks of the ultrasonic data of the same detection point is recorded as the adjacent average interval of the detection point; The difference between the adjacent intervals of the target peak and the adjacent average intervals of the detection points corresponding to the target peak is recorded as the first difference of the target peak, and the hyperbolic tangent value of the ratio of the first difference of the target peak to the adjacent interval is recorded as the same defect characteristic value of the target peak; The same defect feature values ​​of all peaks in the ultrasonic data of the same detection point are arranged in sequence to obtain a defect feature sequence. The same defect feature sequence is divided into defect feature subsequences from the position where the defect feature value is less than 0. The peaks corresponding to the same defect feature value contained in the same defect feature subsequence are recorded as peaks of the same depth.

3. The automatic flaw detection method for a long-axis forged bar according to claim 1, characterized in that: The method for determining the detection point to be determined is: Select an adjacent point at half the distance between the target peak and the adjacent trough, take the target peak as the vertex, and record the angle formed by the rays starting from the vertex and passing through the two adjacent points as the peak angle of the target peak; The average value of the ultrasonic signal intensity corresponding to the adjacent troughs of the target peak is recorded as the average value of the adjacent troughs of the target peak; The absolute value of the difference in the intensity of the ultrasonic signal corresponding to the adjacent troughs of the target peak is recorded as the adjacent trough difference of the target peak; The cumulative sum of the product of the mean value of the adjacent troughs and the difference of the adjacent troughs of all the peaks at the same depth as the target peak is recorded as the first cumulative sum of the target peak, and the normalized value of the product of the peak angle of the target peak and the first cumulative sum is recorded as the crack possibility of the target peak; When there is a peak in the wave peak of the ultrasonic data of the detection point whose crack possibility is greater than or equal to a preset first crack evaluation threshold, the detection point is recorded as a detection point to be determined.

4. The method for automatic flaw detection of a long-axis forged bar according to claim 1, wherein: The process of determining the possibility of the same crack of the adjacent detection points to be determined is as follows: Selecting peaks at the same depth in the ultrasonic data of adjacent detection points to be determined, and for each detection point to be determined, sequentially arranging the intensities of the ultrasonic signals of the peaks selected for the detection point to be determined, to obtain a first vector of the detection point to be determined; The dot product of the first vectors of the adjacent detection points to be determined is recorded as the first dot product of the adjacent detection points to be determined at the depth corresponding to the first vector, and the normalized value of the cumulative sum of the first dot products of the adjacent detection points to be determined at all the same depths is recorded as the defect waveform similarity of the adjacent detection points to be determined; The average of the crack probabilities of all peaks of the ultrasonic data of the detection point to be determined is recorded as the average crack probability of the detection point to be determined, and the absolute value of the difference between the average crack probabilities of adjacent detection points to be determined is recorded as the first absolute value of the adjacent detection points to be determined; the normalized value of the ratio of the defect waveform similarity of adjacent detection points to be determined to the first absolute value is recorded as the same crack probability of the adjacent detection points to be determined.

5. The method for automatic flaw detection of a long-axis forged bar according to claim 1, wherein: The method of determining the type of crack defect based on the possibility of the same crack at adjacent detection points to be determined on the same cross section of the long-axis forged bar includes the following specific steps: Adjacent detection points to be determined on the same cross section of the long-axis forged bar, whose probability of the same crack is greater than or equal to a preset second crack evaluation threshold, are all recorded as detection points to be determined for the same defect; When adjacent detection points to be determined on the same cross section of the long-axis forged bar are detection points to be determined for the same defect, the crack defects corresponding to the adjacent detection points to be determined are recorded as transverse crack defects; When adjacent detection points to be determined on the same cross section of the long-axis forged bar do not have detection points to be determined with the same defect, the crack defects corresponding to the adjacent detection points to be determined are recorded as longitudinal crack defects.

6. The method for automatic flaw detection of a long-axis forged bar according to claim 1, wherein: The method of screening the transverse crack detection points to be determined with the same defect according to the likelihood of the same crack of adjacent detection points to be determined corresponding to the transverse crack defect and the distance between the detection points to be determined includes the following specific process: The ratio of the likelihood of the same crack of adjacent detection points to be determined corresponding to the transverse crack defect to the Euclidean distance between the adjacent detection points to be determined is recorded as the first ratio of the adjacent detection points to be determined corresponding to the transverse crack defect, and the hyperbolic tangent of the first ratio of the adjacent detection points to be determined corresponding to the transverse crack defect is recorded as the defect elongation of the adjacent detection points to be determined corresponding to the transverse crack defect; Adjacent detection points to be determined corresponding to transverse crack defects with a defect elongation greater than or equal to a preset third crack evaluation threshold are all recorded as transverse crack detection points to be determined of the same defect.

7. The method for automatic flaw detection of a long-axis forged bar according to claim 1, wherein: The method for setting the secondary detection point is: The midpoint of the line segment of the two identical transverse cracks to be determined detection points is used as the secondary detection point.

8. The method for automatic flaw detection of a long-axis forged bar according to claim 1, wherein: The method of determining the flaw detection result of the long-axis forged bar according to the position and sound path of the detection point to be determined corresponding to the type of the crack defect includes the following specific steps: When a transverse crack defect is identified, the triangulation method is used to determine the defect position of the transverse crack defect in the long axis forging bar based on the sound path and position of all the transverse crack detection points corresponding to the same defect; When a longitudinal crack defect is identified, the triangulation method is used to determine the defect position of the longitudinal crack defect in the long axis forging bar based on the sound path and position of all the detection points to be determined corresponding to the longitudinal crack defect; When the detection point to be determined is not identified, the flaw detection result of the long-axis forging bar is that there is no defect.

9. An automatic flaw detection device for a long-axis forged bar, implementing the method as claimed in claim 1, characterized in that: The automatic flaw detection device comprises: The flaw detection data acquisition module is used to collect ultrasonic data of each detection point of the long-axis forging bar, calculate the sound path of each detection point of the long-axis forging bar, and the peak and trough of the ultrasonic data of each detection point; A defect category identification module is used to screen out peaks of the same depth in the ultrasonic data of the same detection point based on the differences between adjacent peaks in the ultrasonic data of the same detection point, and determine whether the detection point is a detection point to be determined based on the positional relationship between each peak and the adjacent trough in the ultrasonic data of the same detection point and the intensity value of the corresponding ultrasonic signal; determine the probability of the adjacent detection points to be determined having the same crack based on the differences between all peaks of the same depth in the ultrasonic data of the adjacent detection points to be determined and the crack probability of all peaks in the ultrasonic data; and determine the type of crack defect based on the probability of the adjacent detection points to be determined having the same crack on the same cross section of the long-axis forging bar. The types of crack defects include transverse crack defects and longitudinal crack defects. The flaw detection result determination module is used to screen the transverse crack detection points to be determined with the same defect based on the possibility of the same cracks in adjacent detection points to be determined corresponding to the transverse crack defects and the distance between the detection points to be determined, set secondary detection points between the transverse crack detection points to be determined with the same defect, screen the transverse crack detection points to be determined with the same defect among the secondary detection points, and determine the flaw detection results of the long-axis forging bar based on the position and sound range of the detection points to be determined corresponding to the type of crack defects.

10. An automatic flaw detection system for long-axis forging bars, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images