A non-destructive flaw detection method and system for metal castings

Abstract

The present invention relates to the field of metal ultrasonic detection, and specifically to a nondestructive flaw detection method and system for metal castings, including: comparing the change difference of ultrasonic signals within and outside the ultrasonic flaw detection time interval at the ultrasonic detection moment to obtain ultrasonic signal sensitivity; based on the ultrasonic signal sensitivity, analyzing the directional distribution of positions between different metal surface detection areas to obtain reference metal defect detection areas; comparing the distribution stability change difference of ultrasonic signals between different reference metal defect detection areas within the ultrasonic flaw detection time interval, and combining the distance distribution between the metal surface detection area and different reference metal defect detection areas to obtain defect signal correlation; according to the defect signal correlation and ultrasonic signal sensitivity, adaptively adjusting the noise threshold, and performing nondestructive flaw detection. The present invention reduces the loss of real damage information of metal castings and improves the accuracy of metal casting flaw detection results.

Description

Technical Field

[0001] The invention relates to the field of metal ultrasonic testing, and in particular to a non-destructive flaw detection method and system for metal castings. Background Art

[0002] Among the existing metal casting inspection technologies, non-destructive testing technology is an important quality control method. This technology can detect defects inside metal castings, such as pores, inclusions, cracks, etc., without destroying the metal castings, thereby ensuring the safety of metal castings. Under normal circumstances, in order to ensure the inspection efficiency of metal castings, certain areas on the surface of metal castings are sampled and screened for ultrasonic flaw detection. When using an ultrasonic flaw detector to inspect the metal surface area, a certain amount of random noise will be generated due to the electromagnetic interference of the electronic components in the instrument. These random noises will interfere with the accurate expression of the ultrasonic signal data on the ultrasonic flaw detector, thereby reducing the accuracy of the metal casting inspection results.

[0003] Existing data denoising algorithms usually preset the denoising threshold of the denoising algorithm based on experience, which may cause the ultrasonic flaw detector to over-denoise the ultrasonic signal data representing the damage of metal castings when detecting some metal surface areas, lose some metal casting damage information, and reduce the accuracy of metal casting flaw detection results. Summary of the invention

[0004] The present invention provides a method and system for nondestructive flaw detection of metal castings to solve the existing problem that the existing data denoising algorithm usually presets the denoising threshold of the denoising algorithm based on experience, and cannot be adjusted according to the actual situation of the surface area of ​​the metal casting, resulting in the loss of effective information of some metal castings.

[0005] A non-destructive flaw detection method and system for metal castings of the present invention adopts the following technical solutions:

[0006] The present invention provides a nondestructive flaw detection method for metal castings, which comprises the following steps:

[0007] Acquire ultrasonic signal data of each metal surface detection area at different ultrasonic detection times;

[0008] The other ultrasonic detection moments on both sides of the ultrasonic detection moment are divided to obtain the ultrasonic flaw detection time interval of the ultrasonic detection moment; on the same metal surface detection area, the difference in ultrasonic signal changes between the ultrasonic detection moment and other ultrasonic detection moments inside and outside the ultrasonic flaw detection time interval is compared to obtain the ultrasonic signal sensitivity of the metal surface detection area;

[0009] Based on the comprehensive analysis of the stable change of ultrasonic signal data within the ultrasonic flaw detection time interval at the ultrasonic detection moment and the sensitivity of the ultrasonic signal, according to the directional distribution of the positions between different metal surface detection areas, several reference metal defect detection areas at each ultrasonic detection moment are screened out from all metal surface detection areas;

[0010] The difference in the distribution stability of ultrasonic signals within the ultrasonic flaw detection time interval between different reference metal defect detection areas is compared, and the defect signal correlation at each ultrasonic detection moment is obtained by combining the distance distribution between the metal surface detection area and different reference metal defect detection areas;

[0011] The defect signal correlation degree is combined with the ultrasonic signal sensitivity to obtain the noise adjustment factor at each ultrasonic detection moment; the noise threshold is adaptively adjusted according to the noise adjustment factor, and non-destructive flaw detection is performed.

[0012] Preferably, the method for acquiring the ultrasonic signal sensitivity is:

[0013] Any ultrasonic detection moment is taken as the target ultrasonic detection moment; in the same metal surface detection area, the difference in ultrasonic signal data between the target ultrasonic detection moment and the overall ultrasonic detection moment is compared to obtain the significance of process traces in the metal surface detection area at different ultrasonic detection moments;

[0014] In the ultrasonic flaw detection time interval of the target ultrasonic detection time, the difference of ultrasonic signal data between the target ultrasonic detection time and other ultrasonic detection times is analyzed to obtain the deviation of ultrasonic flaw detection information of the metal surface detection area at different ultrasonic detection times;

[0015] According to the prominence of process traces and the deviation of ultrasonic information for flaw detection, the ultrasonic signal sensitivity of the metal surface detection area is obtained.

[0016] Preferably, the method for obtaining the process trace prominence is:

[0017] Obtain the mean value of ultrasonic signal data at all ultrasonic detection moments;

[0018] By comparing the difference between the ultrasonic signal data at the target ultrasonic detection time and the mean value of the ultrasonic signal data, the significance of the process traces in the metal surface detection area at the target ultrasonic detection time is obtained.

[0019] Preferably, the method for obtaining the deviation of the ultrasonic flaw detection information is:

[0020] In the ultrasonic flaw detection time interval of the target ultrasonic flaw detection time, other ultrasonic flaw detection times except the target ultrasonic flaw detection time are regarded as the ultrasonic flaw detection time;

[0021] Obtaining the local flaw detection signal change value at the target ultrasonic detection time and each flaw detection ultrasonic detection time;

[0022] The deviation of ultrasonic flaw detection information of each metal surface detection area at different ultrasonic detection times is obtained by integrating the change value of the local flaw detection signal.

[0023] Preferably, the method for obtaining the reference metal defect detection area is:

[0024] For any ultrasonic detection moment, according to the stable change of ultrasonic signal data within the ultrasonic flaw detection time interval and the ultrasonic signal sensitivity at the ultrasonic detection moment, the defect reference detection area at the ultrasonic detection moment is obtained;

[0025] Any metal surface detection area except the defect reference detection area is used as the reference surface area, and a number of reference metal defect detection areas are screened out at each ultrasonic detection moment within the direction range from the defect reference detection area to the reference surface area.

[0026] Preferably, the method for acquiring the defect reference detection area is:

[0027] For any ultrasonic detection moment, within the ultrasonic flaw detection time interval of the ultrasonic detection moment, the standard deviation of the ultrasonic signal of each metal surface detection area is obtained; based on the ultrasonic signal standard deviation and the ultrasonic signal sensitivity, the metal damage reference degree of each metal surface detection area is obtained;

[0028] According to the metal damage reference degree, the defect reference detection area at each ultrasonic detection moment is screened out from all metal surface detection areas.

[0029] Preferably, the method for obtaining the defect signal correlation is:

[0030] For any ultrasonic detection moment, at the ultrasonic detection moment, the change difference of the ultrasonic signal distribution stability in the ultrasonic flaw detection time interval between different reference metal defect detection areas is compared to obtain the defect signal coherence at the ultrasonic detection moment;

[0031] Obtaining detection area distances between different reference metal defect detection areas;

[0032] According to the coherence of the defect signal and the distance of the detection area, the correlation of the defect signal at the time of ultrasonic detection is obtained.

[0033] Preferably, the method for acquiring the coherence of the defect signal is:

[0034] Obtaining the ultrasonic signal stability of each reference metal defect detection area within the ultrasonic flaw detection time interval;

[0035] According to the difference in the stability of ultrasonic signals in different reference metal defect detection areas during the ultrasonic flaw detection time intervals, the coherence of the defect signal at the ultrasonic detection moment is obtained.

[0036] Preferably, the noise adjustment factor is obtained by:

[0037] For any ultrasonic detection moment, the noise adjustment factor of each metal surface detection area at the ultrasonic detection moment is obtained according to the product of the ultrasonic signal sensitivity of each metal surface detection area and the defect signal correlation at the ultrasonic detection moment.

[0038] The present invention also proposes a non-destructive flaw detection system for metal castings, comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the steps of the above-mentioned non-destructive flaw detection method for metal castings.

[0039] The beneficial effects of the technical solution of the present invention are: dividing the ultrasonic detection time into ultrasonic flaw detection time intervals; comparing the difference in the ultrasonic detection time and the ultrasonic signal inside and outside the ultrasonic flaw detection time interval, and obtaining the ultrasonic signal sensitivity of the metal surface detection area; wherein the ultrasonic signal sensitivity better reflects the complex spatial composition of the material inside the casting in the metal surface detection area;

[0040] According to the directional distribution of positions between different metal surface detection areas, several reference metal defect detection areas are obtained; the distribution stability change differences of ultrasonic signals in the ultrasonic detection time interval between different reference metal defect detection areas are compared, and the defect signal correlation at each ultrasonic detection moment is obtained in combination with the distance distribution between the metal surface detection area and different reference metal defect detection areas; wherein the defect signal correlation better reflects the similarity between the internal material structure of the reference metal surface detection area and the structural characteristics in the surrounding space at the ultrasonic detection moment; then, according to the defect signal correlation and the ultrasonic signal sensitivity, the noise threshold is adaptively adjusted, and non-destructive flaw detection is performed; the present invention adaptively adjusts the denoising threshold by analyzing the changing characteristics of internal defects in the surface area of ​​the metal casting and the influence on the internal material space, thereby reducing the loss of real damage information of the metal casting and improving the accuracy of the flaw detection results of the metal casting. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1A flowchart of the steps of a nondestructive flaw detection method for metal castings of the present invention;

[0043] Figure 2 is a schematic diagram of an ultrasonic flaw detector of the present invention;

[0044] Figure 3 It is a schematic diagram of the flaw detection process of the metal flaw detector of the present invention;

[0045] Figure 4 A schematic diagram of internal defects in a metal surface detection area according to the present invention;

[0046] Figure 5 A schematic diagram showing the comparison between the conventional process traces of the present invention and the real defects of metal castings;

[0047] Figure 6 It is a schematic diagram of cracks around internal defects of metal castings of the present invention. DETAILED DESCRIPTION

[0048] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a non-destructive flaw detection method and system for metal castings proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0050] The specific scheme of the non-destructive flaw detection method and system for metal castings provided by the present invention is described in detail below with reference to the accompanying drawings.

[0051] See also Figure 1 , which shows a flowchart of a method for nondestructive flaw detection of metal castings provided by an embodiment of the present invention, the method comprising the following steps:

[0052] Step S001: Acquire ultrasonic signal data of each metal surface detection area at different ultrasonic detection times.

[0053] It should be noted that the existing data denoising algorithms usually preset the denoising threshold of the denoising algorithm based on experience, which may cause the ultrasonic flaw detector to over-denoise the ultrasonic signal data representing the damage of metal castings when detecting some metal surface areas, lose some metal casting damage information, and reduce the accuracy of metal casting flaw detection results. Figure 2 , which shows a schematic diagram of an ultrasonic flaw detector. Figure 3 , which shows a schematic diagram of the flaw detection process of a metal flaw detector, wherein a gray rectangle represents an ultrasonic flaw detector, a dotted arrow represents an ultrasonic wave emitted by the ultrasonic flaw detector, and the rectangular edge pointed by the arrow on the surface of a metal casting represents the surface of the metal casting.

[0054] In a specific implementation of the embodiment of the present invention, the method for obtaining ultrasonic signal data is as follows: a region number is preset , the surface of the metal casting is evenly divided into metal surface detection area; take any metal surface detection area as an example, use the ultrasonic flaw detector to detect the metal surface in the detection area The ultrasonic flaw detector performs flaw detection at a uniform speed, and records the ultrasonic signal data displayed on the ultrasonic flaw detector once every 1 second as an ultrasonic detection moment, until the ultrasonic flaw detector detects the entire metal surface detection area, and obtains the ultrasonic signal data of the metal surface detection area at all ultrasonic detection moments; obtains the ultrasonic signal data of each metal surface detection area at all ultrasonic detection moments.

[0055] It should be noted that before the ultrasonic flaw detector inspects the next new metal surface detection area, the present embodiment will re-record the order of ultrasonic detection moments so that the number of ultrasonic detection moments in each metal surface detection area is consistent. When the present embodiment subsequently analyzes any ultrasonic detection moment, there is an ultrasonic detection moment corresponding to each metal surface detection area.

[0056] It is particularly noted that in this embodiment This example is described as an example, and this embodiment is not specifically limited. , the moving speed of the ultrasonic flaw detector and the time interval for recording data can be determined according to the specific implementation situation.

[0057] So far, the ultrasonic signal data of each metal surface detection area at different ultrasonic detection times are obtained through the above method.

[0058] Step S002: Divide the other ultrasonic detection moments on both sides of the ultrasonic detection moment to obtain the ultrasonic flaw detection time interval of the ultrasonic detection moment; in the same metal surface detection area, compare the difference in change of the ultrasonic signal between the ultrasonic detection moment and other ultrasonic detection moments inside and outside the ultrasonic flaw detection time interval to obtain the ultrasonic signal sensitivity of the metal surface detection area.

[0059] It should be noted that when using an ultrasonic flaw detector to detect metal castings, the ultrasonic flaw detector will move to different metal surface locations as the ultrasonic detection time progresses; under normal circumstances, if there are defects inside the metal casting, then this part of the defect sometimes covers the metal surface locations at multiple adjacent ultrasonic detection times; please refer to Figure 4 , which shows a schematic diagram of internal defects in the metal surface detection area, where t1, t2, and t3 respectively represent the surface positions of metal castings at the t1th ultrasonic detection moment, the t2th ultrasonic detection moment, and the t3th ultrasonic detection moment when the ultrasonic flaw detector detects the same metal surface detection area, and the rectangular side pointed by the arrow on the metal casting surface represents the metal casting surface. Therefore, the other ultrasonic detection moments on both sides of the ultrasonic detection moment can be divided to obtain the ultrasonic flaw detection time interval of the ultrasonic detection moment, which is conducive to the subsequent distinction of metal surface areas with defects.

[0060] It should be further explained that, under normal circumstances, the defective part inside the metal casting will destroy the uniform distribution of the material inside the normal metal casting, so that the defective part itself has a certain irregular surface area; when the metal flaw detector passes through the corresponding metal casting surface above the defective part, the ultrasonic information received by the metal flaw detector will fluctuate greatly, so that the corresponding ultrasonic signal data detected will change greatly. Therefore, the difference in the change of the ultrasonic signal between the ultrasonic detection moment and other ultrasonic detection moments inside and outside the ultrasonic flaw detection time interval can be compared to obtain the ultrasonic signal sensitivity of the metal surface detection area; among them, if the ultrasonic signal sensitivity is greater, it means that the material space structure inside the casting in the metal surface detection area is more complex, reflecting that the metal surface detection area is more likely to have internal damage.

[0061] The process of obtaining the ultrasonic flaw detection time interval is:

[0062] Preset an ultrasonic detection time quantity , taking any ultrasonic detection time as an example, The ultrasonic detection time and the time after the ultrasonic detection time The time period formed by the ultrasonic detection moments is taken as the ultrasonic flaw detection time interval of the ultrasonic detection moment. The ultrasonic flaw detection time interval of each ultrasonic detection moment is obtained. This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0063] It is particularly noted that if the actual number of ultrasonic detection moments before and after the ultrasonic detection moment does not meet the preset , then the ultrasonic flaw detection time interval of the ultrasonic detection moment is obtained based on the actual number of ultrasonic detection moments before and after the ultrasonic detection moment.

[0064] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the ultrasonic signal sensitivity is: taking any ultrasonic detection moment as the target ultrasonic detection moment; in the same metal surface detection area, comparing the difference in ultrasonic signal data between the target ultrasonic detection moment and the overall ultrasonic detection moment, and obtaining the process trace significance of the metal surface detection area at different ultrasonic detection moments; within the ultrasonic flaw detection time interval of the target ultrasonic detection moment, analyzing the difference in ultrasonic signal data between the target ultrasonic detection moment and other ultrasonic detection moments, and obtaining the flaw detection ultrasonic information deviation of the metal surface detection area at different ultrasonic detection moments; obtaining the ultrasonic signal sensitivity of the metal surface detection area according to the process trace significance and the flaw detection ultrasonic information deviation. The specific process is as follows:

[0065] 1. Obtain the prominence of process traces.

[0066] It should be noted that in the same metal surface detection area, in addition to the internal defects of metal castings, there may also be conventional process traces produced by the metal during the casting process. For example, the fusible crystals inside the metal castings may not be completely melted, leaving a certain amount of crystalline particles. This part of the conventional process traces does not affect the normal use of metal castings, so it is usually classified as an allowable defect range. This part of the conventional process traces will also cause the corresponding ultrasonic signal data to fluctuate to a certain extent, and because the crystalline particles themselves are small, the corresponding number of ultrasonic detection moments is small. Please refer to Figure 5 , which shows a schematic diagram of the comparison between conventional process traces and real defects of metal castings, where the rectangle pointed by the arrow on the surface of the metal casting represents the surface of the metal casting. Therefore, in the same metal surface detection area, the difference in ultrasonic signal data between the target ultrasonic detection moment and the overall ultrasonic detection moment can be compared to obtain the significance of process traces in the metal surface detection area at different ultrasonic detection moments. The greater the significance of process traces, the more obvious the traces of conventional processes in the metal surface detection area, reflecting that the metal surface detection area is more likely to have internal damage that is misidentified.

[0067] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the process trace significance is: obtaining the mean value of the ultrasonic signal data at all ultrasonic detection moments; comparing the difference between the ultrasonic signal data at the target ultrasonic detection moment and the mean value of the ultrasonic signal data, and obtaining the process trace significance of the metal surface detection area at the target ultrasonic detection moment. The specific process is:

[0068] Take any metal surface detection area as an example. In the metal surface detection area, the mean of the ultrasonic signal data at all ultrasonic detection moments is taken as the mean of the ultrasonic signal data. The absolute value of the difference between the ultrasonic signal data at the target ultrasonic detection moment and the mean of the ultrasonic signal data is taken as the process trace significance of the metal surface detection area at the target ultrasonic detection moment. The process trace significance of each metal surface detection area at different ultrasonic detection moments is obtained.

[0069] It should be noted that if the significance of the process traces is greater, it means that the traces of conventional processes in the metal surface detection area are more obvious, which reflects that the metal surface detection area is more likely to have internal damage that is misidentified.

[0070] 2. Obtain the deviation of ultrasonic flaw detection information.

[0071] It should be noted that if there is a defective area inside a metal casting, there may be a certain amount of cracks around this defective area. The fluctuation of these cracks in the corresponding ultrasonic signal data is small. If random noise is distributed on this part of the ultrasonic signal data, the defect information represented by these cracks will be lost in the denoising process; and because these cracks are distributed around the defective area, they are usually also included in the metal surface corresponding to the ultrasonic flaw detection time interval. Please refer to Figure 6 , which shows a schematic diagram of cracks around internal defects of metal castings. Therefore, within the ultrasonic flaw detection time interval of the target ultrasonic detection time, the difference in ultrasonic signal data between the target ultrasonic detection time and other ultrasonic detection times can be analyzed to obtain the flaw detection ultrasonic information deviation of the metal surface detection area at different ultrasonic detection times. The larger the flaw detection ultrasonic information deviation, the more likely it is that there are subtle internal cracks in the metal surface detection area.

[0072] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the deviation of the ultrasonic information of the flaw detection is: within the ultrasonic flaw detection time interval of the target ultrasonic detection time, other ultrasonic detection times except the target ultrasonic detection time are used as the ultrasonic flaw detection time; the local flaw detection signal change value between the target ultrasonic detection time and each flaw detection ultrasonic detection time is obtained; the local flaw detection signal change value is integrated to obtain the deviation of the ultrasonic information of the flaw detection of each metal surface detection area at different ultrasonic detection times. The specific process is:

[0073] In the metal surface detection area, the absolute value of the difference between the ultrasonic signal data at the target ultrasonic detection time and each flaw detection ultrasonic detection time is used as the local flaw detection signal change value at the target ultrasonic detection time and each flaw detection ultrasonic detection time; the average value of the local flaw detection signal change value at the target ultrasonic detection time and all flaw detection ultrasonic detection times is used as the flaw detection ultrasonic information deviation degree of the metal surface detection area at the target ultrasonic detection time. The flaw detection ultrasonic information deviation degree of each metal surface detection area at different ultrasonic detection times is obtained.

[0074] It should be noted that the greater the deviation of the ultrasonic flaw detection information, the more likely it is that there are tiny internal cracks in the detection area of ​​the metal surface.

[0075] 3. Obtain the sensitivity of ultrasonic signals.

[0076] As an example, the sensitivity of ultrasound signal can be calculated by the following formula:

[0077]

[0078] In the formula, Indicates Ultrasonic signal sensitivity of each metal surface detection area; Indicated in The number of all ultrasonic detection moments in a metal surface detection area; Indicates The metal surface detection area is in The prominence of process traces at each ultrasonic detection moment; Indicates The metal surface detection area is in The deviation of ultrasonic detection information at each ultrasonic detection moment.

[0079] It should be noted that the greater the sensitivity of the ultrasonic signal, the more complex the material space structure inside the casting in the metal surface detection area, which means that the metal surface detection area is more likely to have internal damage.

[0080] So far, the ultrasonic signal sensitivity of each metal surface detection area is obtained through the above method.

[0081] Step S003: Based on a comprehensive analysis of the stable change of ultrasonic signal data within the ultrasonic flaw detection time interval at the ultrasonic detection moment and the sensitivity of the ultrasonic signal, according to the directional distribution of positions between different metal surface detection areas, several reference metal defect detection areas at each ultrasonic detection moment are screened out from all metal surface detection areas.

[0082] It should be noted that the defects inside the metal casting usually extend in a certain direction, making the ultrasonic signal data on the corresponding area on the surface of the metal casting more prone to large fluctuations, and the data is more unstable. Therefore, based on the comprehensive analysis of the stable changes in the ultrasonic signal data within the ultrasonic flaw detection time interval at the ultrasonic detection moment and the sensitivity of the ultrasonic signal, according to the directional distribution of the positions between different metal surface detection areas, several reference metal defect detection areas at each ultrasonic detection moment are selected from all metal surface detection areas.

[0083] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the reference metal defect detection area is as follows: for any ultrasonic detection moment, according to the stable change of ultrasonic signal data in the ultrasonic flaw detection time interval at the ultrasonic detection moment and the ultrasonic signal sensitivity, the defect reference detection area at the ultrasonic detection moment is obtained; any metal surface detection area other than the defect reference detection area is used as the reference surface area, and within the direction range from the defect reference detection area to the reference surface area, several reference metal defect detection areas at each ultrasonic detection moment are screened out. The specific process is as follows:

[0084] 1. Obtain the defect baseline detection area.

[0085] It should be noted that in the process of extending the defects inside the metal casting, they usually extend in a fluctuating manner from a certain area within a certain direction range, and the damage of the defects to the material space inside the metal casting will gradually decrease; therefore, the defect reference detection area at each ultrasonic detection moment can be obtained based on the stable change of the ultrasonic signal data within the ultrasonic flaw detection time interval at the ultrasonic detection moment and the ultrasonic signal sensitivity; in order to characterize the starting metal surface area where the defect begins to extend.

[0086] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the defect reference detection area is: for any ultrasonic detection moment, within the ultrasonic flaw detection time interval of the ultrasonic detection moment, obtain the ultrasonic signal standard deviation of each metal surface detection area; obtain the metal damage reference degree of each metal surface detection area according to the ultrasonic signal standard deviation and the ultrasonic signal sensitivity; and screen out the defect reference detection area at each ultrasonic detection moment from all metal surface detection areas according to the metal damage reference degree. The specific process is:

[0087] Taking any metal surface detection area as an example, the standard deviation of all ultrasonic signal data in the metal surface detection area during the ultrasonic flaw detection time interval at the ultrasonic detection moment is used as the standard deviation of the ultrasonic signal in the metal surface detection area at the ultrasonic detection moment; the product of the ultrasonic signal standard deviation of the metal surface detection area and the ultrasonic signal sensitivity is used as the metal damage reference degree in the metal surface detection area at the ultrasonic detection moment.

[0088] Furthermore, the metal damage reference degree in all metal surface detection areas at the ultrasonic detection moment is obtained; the metal surface detection area with the largest metal damage reference degree is used as the defect reference detection area at the ultrasonic detection moment.

[0089] 2. Obtain a reference metal defect detection area.

[0090] Taking any reference surface area as an example, the point on the defect reference detection area at the time of ultrasonic detection is taken as the defect starting point; the point on the reference surface area at the time of ultrasonic detection is taken as the defect ending point; the line segment between the defect starting point and the defect ending point is taken as the defect direction reference line.

[0091] Furthermore, a preset degree , taking the defect starting point as the starting point, rotate the defect direction reference line clockwise The ray after is taken as the clockwise boundary; the defect direction reference line is rotated counterclockwise The ray after the ultrasonic detection is taken as the counterclockwise boundary; the angle area formed by the clockwise boundary and the counterclockwise boundary is taken as the defect reference extension area; each metal surface detection area passed by the defect reference extension area is taken as the reference metal defect detection area at the ultrasonic detection moment. A number of reference metal defect detection areas are obtained at each ultrasonic detection moment. This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0092] So far, several reference metal defect detection areas at each ultrasonic detection moment are obtained through the above method.

[0093] Step S004: Compare the distribution stability changes of ultrasonic signals within the ultrasonic flaw detection time interval between different reference metal defect detection areas, and combine the distance distribution between the metal surface detection area and different reference metal defect detection areas to obtain the defect signal correlation at each ultrasonic detection moment.

[0094] It should be noted that for any defect damage inside the metal casting, this defect damage destroys the structural distribution of the material space inside the metal casting, changes the stress condition of the internal space of the metal casting where the defect damage is located, and makes the internal material space around the defect damage more prone to defect damage. Therefore, the difference in the stable change of the distribution of ultrasonic signals in the ultrasonic flaw detection time interval between different reference metal defect detection areas can be compared, and the defect signal correlation at each ultrasonic detection moment can be obtained by combining the distance distribution between the metal surface detection area and different reference metal defect detection areas. The greater the defect signal correlation, the more similar the internal material structure at the surface position in the reference metal surface detection area at the corresponding ultrasonic detection moment is to the structural characteristics in the surrounding space, reflecting that the internal defect is more likely to exist at the surface position in the reference metal surface detection area at the corresponding ultrasonic detection moment.

[0095] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the defect signal correlation is: for any ultrasonic detection moment, at the ultrasonic detection moment, compare the change difference of the ultrasonic signal distribution stability in the ultrasonic flaw detection time interval between different reference metal defect detection areas to obtain the defect signal coherence at the ultrasonic detection moment; obtain the detection area distance between different reference metal defect detection areas; according to the defect signal coherence and the detection area distance, obtain the defect signal correlation at the ultrasonic detection moment. The specific process is as follows:

[0096] 1. Obtain the coherence of defect signals.

[0097] It should be noted that the defects inside metal castings are usually a closed coherent area, so the difference in the change of the stability of the ultrasonic signal distribution in the ultrasonic flaw detection time interval between different reference metal defect detection areas can be compared in the metal surface detection area to obtain the coherence of the defect signal at the ultrasonic detection time. The greater the coherence of the defect signal, the more coherent the defect area at the corresponding ultrasonic detection time.

[0098] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the coherence of the defect signal is: obtaining the stability of the ultrasonic signal of each reference metal defect detection area within the ultrasonic flaw detection time interval; according to the difference in the stability of the ultrasonic signal between different reference metal defect detection areas in the ultrasonic flaw detection time interval, obtaining the coherence of the defect signal at the ultrasonic detection moment. The specific process is:

[0099] Taking any reference metal defect detection area as an example, in the reference metal defect detection area, the inverse proportional normalized value of the variance of all ultrasonic signal data within the ultrasonic flaw detection time interval at the ultrasonic detection moment is used as the ultrasonic signal stability in the reference metal defect detection area at the ultrasonic detection moment.

[0100] It is particularly noted that the embodiment adopts Model to present inverse proportional relationship and normalization, As the input of the model, the implementer can choose the inverse proportional function and the normalization function according to the actual situation.

[0101] Furthermore, as an example, the defect signal coherence can be calculated by the following formula:

[0102]

[0103] In the formula, Indicates The continuity of defect signal at each ultrasonic detection moment; Indicates The number of all reference metal defect detection areas at each ultrasonic detection moment; Indicates The ultrasonic detection time is Ultrasonic signal stability in a reference metal defect detection area; Indicates The ultrasonic detection time is Ultrasonic signal stability in a reference metal defect detection area; Indicates the preset hyperparameters. In this example, , used to prevent the denominator from being 0; Indicates taking the absolute value.

[0104] It should be noted that the greater the coherence of the defect signal, the The more continuous the defect area is at each ultrasonic detection moment.

[0105] 2. Obtain the correlation of defect signals.

[0106] At the ultrasonic detection moment, the distance between the centers of any two reference metal defect detection areas is taken as the detection area distance; the mean of all detection area distances is taken as the mean defect distance at the ultrasonic detection moment; the product of the mean defect distance at the ultrasonic detection moment and the coherence of the defect signal is taken as the defect signal correlation at the ultrasonic detection moment.

[0107] It should be noted that if the correlation degree of the defect signal is greater, it means that the internal material structure at the surface position in the reference metal surface detection area at the corresponding ultrasonic detection time is more similar to the structural characteristics in the surrounding space, which reflects that it is more likely that an internal defect exists at the surface position in the reference metal surface detection area at the corresponding ultrasonic detection time.

[0108] So far, the defect signal correlation degree at each ultrasonic detection moment is obtained through the above method.

[0109] Step S005: combining the defect signal correlation with the ultrasonic signal sensitivity to obtain a noise adjustment factor at each ultrasonic detection moment; adaptively adjusting the noise threshold according to the noise adjustment factor, and performing nondestructive flaw detection.

[0110] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the noise adjustment factor is: for any ultrasonic detection moment, the noise adjustment factor of each metal surface detection area at the ultrasonic detection moment is obtained according to the product of the ultrasonic signal sensitivity of each metal surface detection area and the defect signal correlation at the ultrasonic detection moment. The specific process is:

[0111] Taking any metal surface detection area as an example, the normalized value of the product between the ultrasonic signal sensitivity of the metal surface detection area and the defect signal correlation at the ultrasonic detection moment is used as the noise adjustment factor of the metal surface detection area at the ultrasonic detection moment.

[0112] It is particularly noted that in this embodiment A linear normalization function is used to perform normalization processing, wherein the normalization function may be determined according to specific implementation conditions.

[0113] Furthermore, the noise adjustment factor of the metal surface detection area at the time of ultrasonic detection is used as a weight and compared with the preset denoising threshold The multiplied value is used as the adaptive denoising threshold of the metal surface detection area at the ultrasonic detection time; and the ultrasonic signal data of the metal surface detection area at the ultrasonic detection time is denoised by the adaptive denoising threshold to obtain the denoised ultrasonic signal data of the metal surface detection area at the ultrasonic detection time. The denoised ultrasonic signal data of each metal surface detection area at each ultrasonic detection time is obtained.

[0114] It is particularly noted that the process of denoising data according to a denoising threshold is a well-known content of the denoising algorithm. This embodiment is described by taking the wavelet denoising algorithm as an example, wherein the denoising algorithm may be determined according to the specific implementation situation.

[0115] Furthermore, the amplitude threshold of the ultrasonic signal data is obtained from the metal casting damage detection platform , the value is greater than The denoised ultrasonic signal data is used as the real damaged ultrasonic signal data; the position of each real damaged ultrasonic signal data on the metal casting at the corresponding ultrasonic detection moment is marked for damage, thereby completing the damage detection.

[0116] Through the above steps, the nondestructive flaw detection method of metal castings is completed.

[0117] Another embodiment of the present invention provides a non-destructive flaw detection system for metal castings, the system comprising a memory and a processor, and when the processor executes the computer program stored in the memory, the above method steps S001 to S005 are performed.

[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A nondestructive flaw detection method for metal castings, characterized in that: The method comprises the following steps: Acquire ultrasonic signal data of each metal surface detection area at different ultrasonic detection times; The other ultrasonic detection moments on both sides of the ultrasonic detection moment are divided to obtain the ultrasonic flaw detection time interval of the ultrasonic detection moment; on the same metal surface detection area, the difference in ultrasonic signal changes between the ultrasonic detection moment and other ultrasonic detection moments inside and outside the ultrasonic flaw detection time interval is compared to obtain the ultrasonic signal sensitivity of the metal surface detection area; Based on the comprehensive analysis of the ultrasonic signal standard deviation and ultrasonic signal sensitivity of the ultrasonic signal data within the ultrasonic flaw detection time interval at the ultrasonic detection moment, according to the directional distribution of the positions between different metal surface detection areas, several reference metal defect detection areas at each ultrasonic detection moment are screened out from all metal surface detection areas; The ultrasonic signal stability of the ultrasonic signal within the ultrasonic flaw detection time interval between different reference metal defect detection areas is compared, and the defect signal correlation at each ultrasonic detection moment is obtained by combining the distance distribution between the metal surface detection area and different reference metal defect detection areas; The defect signal correlation degree is combined with the ultrasonic signal sensitivity to obtain the noise adjustment factor at each ultrasonic detection moment; the noise threshold is adaptively adjusted according to the noise adjustment factor, and non-destructive flaw detection is performed; The method for obtaining the ultrasonic signal sensitivity is: Obtain the mean value of ultrasonic signal data at all ultrasonic detection moments; Compare the difference between the ultrasonic signal data at the target ultrasonic detection time and the mean value of the ultrasonic signal data to obtain the significance of the process traces in the metal surface detection area at the target ultrasonic detection time; In the ultrasonic flaw detection time interval of the target ultrasonic flaw detection time, other ultrasonic flaw detection times except the target ultrasonic flaw detection time are regarded as the ultrasonic flaw detection time; Obtaining the local flaw detection signal change value at the target ultrasonic detection time and each flaw detection ultrasonic detection time; The deviation of ultrasonic flaw detection information of each metal surface detection area at different ultrasonic detection times is obtained by integrating the change value of local flaw detection signal; According to the prominence of process traces and the deviation of ultrasonic information for flaw detection, the ultrasonic signal sensitivity of the metal surface detection area is obtained; The method for obtaining the reference metal defect detection area is: For any ultrasonic detection moment, within the ultrasonic flaw detection time interval of the ultrasonic detection moment, the standard deviation of all ultrasonic signal data in each metal surface detection area is used as the ultrasonic signal standard deviation in each metal surface detection area; according to the ultrasonic signal standard deviation and ultrasonic signal sensitivity, the metal damage reference degree of each metal surface detection area is obtained; According to the metal damage reference degree, the defect reference detection area at each ultrasonic detection moment is screened out from all metal surface detection areas; Any metal surface detection area other than the defect reference detection area is used as a reference surface area, and a number of reference metal defect detection areas are screened out at each ultrasonic detection moment within the direction range from the defect reference detection area to the reference surface area; The method for obtaining the defect signal correlation degree is: The inversely proportional normalized value of the variance of all ultrasonic signal data in each reference metal defect detection area within the ultrasonic flaw detection time interval at the ultrasonic detection moment is used as the ultrasonic signal stability of each reference metal defect detection area within the ultrasonic flaw detection time interval at the ultrasonic detection moment; According to the difference in the stability of ultrasonic signals in different reference metal defect detection areas during ultrasonic flaw detection time intervals, the coherence of defect signals at the ultrasonic detection time is obtained; At the time of ultrasonic detection, the distance between the centers of any two reference metal defect detection areas is taken as the detection area distance; According to the coherence of the defect signal and the distance of the detection area, the correlation of the defect signal at the time of ultrasonic detection is obtained.

2. A non-destructive flaw detection method for metal castings according to claim 1, characterized in that: The method for obtaining the noise adjustment factor is: For any ultrasonic detection moment, the noise adjustment factor of each metal surface detection area at the ultrasonic detection moment is obtained according to the product of the ultrasonic signal sensitivity of each metal surface detection area and the defect signal correlation at the ultrasonic detection moment.

3. A non-destructive flaw detection system for metal castings, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of a method for non-destructive flaw detection of metal castings as described in any one of claims 1 to 2 are implemented.

Images