Ultrasonic co-frequency sensitivity difference method for quantitative determination of microdefects

Abstract

The invention discloses an ultrasonic co-frequency sensitivity difference method for quantitatively determining microdefects, which comprises the following steps: S1, preparing a polished sample, and putting the polished sample into an ultrasonic scanning device; s2, an ultrasonic sensor group with the same parameter is adopted to be matched with an ultrasonic scanning device to obtain an imaging result image group; s3, the ultrasonic sensor group scans a position point with a signal exceeding a threshold value in the imaging result image group under the third-order sensitivity; s4, extracting the ultrasonic signals of the position points with all the signals exceeding the threshold value, and intercepting the time domain signal length of the same monitoring depth; s5, grouping the position points of all the signal over-threshold values according to whether the position points of all the signal over-threshold values coincide or not under the third-order sensitivity; s6, comparing the overthresholds at the same detection point under different conditions one by one, and outputting a detection quantitative result according to quantitative analysis; and S7, obtaining the number, the size and the distribution condition of the microdefects of each size level. According to the method, different cleanliness conditions of the steel can be quickly analyzed in batches.

Description

Technical Field

[0001] This invention relates to steel testing technology, and more specifically, to an ultrasonic same-frequency sensitivity difference method for quantitatively determining microscopic defects. Background Technology

[0002] With the development of the iron and steel metallurgy industry, the requirements for the cleanliness of molten steel are becoming increasingly stringent, and the control of microscopic defects in molten steel is becoming increasingly important. The qualified inspection of microscopic defects in steel products is a crucial step in ensuring steel quality. Because microscopic defects are extremely small, especially larger microscopic defect particles, they are randomly distributed within the steel and have a significant impact on the performance of the steel and the safety and quality of the product.

[0003] Currently, techniques for detecting microscopic defects in steel include optical quantitative metallographic analysis, large-sample electrolysis, and fully automated scanning electron microscopy (SEM) analysis of non-metallic microscopic defects in steel. In recent years, although technical means for detecting and analyzing microscopic defects have emerged, they all suffer from low efficiency.

[0004] Optical quantitative metallographic analysis is a traditional method widely used to analyze microscopic defects. It offers precise analysis, and advanced quantitative metallographic devices combine metallographic microscopes with image analyzers, providing advantages such as speed, convenience, and accuracy. However, the equipment is relatively expensive, the results are significantly affected by human factors, and sample preparation is time-consuming. Results also vary between operators, reducing the comparability between batches of samples. Furthermore, the testing time is long, placing a heavy workload on the operator. Therefore, the cost of quantitative metallographic analysis of internal microscopic defects in materials is very high. It is estimated that testing a 1kg sample will take six months.

[0005] Zeiss's newly launched fully automated scanning electron microscope (SEM) system for analyzing microscopic defects in non-metallic steel represents a novel approach developed in recent years. It is a comprehensive analytical system comprised of a scanning electron microscope, a high-speed energy dispersive spectroscopy (EDS) detector, and fully automated microscopic defect analysis software. This system utilizes a large, multi-interface sample chamber and a complete electron optics system, enabling observation of the surface morphology of various metallic materials. It automatically performs comprehensive analysis of microscopic defects, accurately detecting defects down to sub-micrometer levels (below 1 μm) and identifying the phases of complex microscopic defects. Compared to quantitative microsonic ultrasound methods, which only detect sample surfaces, are limited to small samples, are time-consuming, and are expensive, this system offers significant advantages.

[0006] The electrolytic analysis method for large-scale micro-defects in steel, introduced from Germany, involves electrolyzing a small sample in a sulfate solution to obtain micro-defects and anode slime. Carbides are dissolved using acid, and the micro-defect components, such as silica, calcium oxide, aluminum oxide, and manganese dioxide, are then analyzed chemically. This method is primarily used to analyze large micro-defects larger than 50 μm in steel, enabling the grading of micro-defect sizes (approximately 50 μm, 50–100 μm, >100 μm) and the analysis of their composition. The analytical process includes electrolysis, washing, reduction, and separation. The separated micro-defects undergo particle size classification, morphological photography, and electron probe microanalysis for quantitative compositional analysis. Another method uses physical methods to separate micro-defects. Washing removes carbides, while preserving micro-defects and iron compounds. Reduction and magnetic separation are then used to separate the micro-defects. Its disadvantages include a long electrolysis time (days), typically only allowing analysis of about 2 kg of matrix material, lack of representativeness, environmental friendliness, and high cost. To capture more large micro-defects, it is necessary to increase the sample size, with samples weighing 3-5 kg ​​and an electrolysis time of 7-15 days. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the purpose of this invention is to provide a quantitative method for measuring the ultrasonic sensitivity difference of micro-defects. This method enables rapid and accurate quantitative detection of micro-defects within steel, obtaining information on the specific size, quantity, and distribution location of micro-defects at various size levels within the steel. This assists steel smelting, analysis, and research workers in analyzing different cleanliness levels of steel more quickly and in batches.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] A method for quantitatively determining the ultrasonic sensitivity difference of microscopic defects at the same frequency includes the following steps:

[0010] S1, Prepare the polished sample and place it in an ultrasonic scanning device;

[0011] S2, using an ultrasonic sensor group with the same parameters in conjunction with the ultrasonic scanning device to obtain an imaging result group;

[0012] S3, Under third-order sensitivity, the ultrasonic sensor group scans the locations where the signal exceeds the threshold in the imaging result group;

[0013] S4, extract the ultrasonic signals of all locations where the signals exceed the threshold, and extract the time-domain signal length at the same monitoring depth;

[0014] S5, group all signal points exceeding the threshold according to whether the points where all signals exceed the threshold coincide under the third-order sensitivity.

[0015] S6, compare the threshold exceedances under different conditions at the same detection point one by one, and output the quantitative detection results based on quantitative analysis;

[0016] S7. Based on the quantitative detection results, and combined with the quantitative evaluation model for micro-defects, the quantity, size, and distribution of micro-defects at each size level are obtained.

[0017] Preferably, in step S1, the thickness of the polished sample is 10-15 mm, and the surface roughness is below 5 μm.

[0018] Preferably, in step S2, three ultrasonic channels with the same frequency are used, which are the ultrasonic sensor groups with low, medium and high third-order gains, respectively.

[0019] Preferably, the frequency of the ultrasonic sensor group is 25MHz to 50MHz.

[0020] Preferably, step S3 specifically includes:

[0021] In step S2, the locations where the signals exceed the threshold are determined from the imaging result groups measured by the ultrasonic sensor groups with low, medium, and high gains, respectively. Then, the locations are compared and it is determined whether each location overlaps in the imaging result group.

[0022] Preferably, in step S5, the location points where all signals exceed the threshold are grouped according to the following rules:

[0023] 1) For a certain same scanning position point, the high-order gain channel detection result shows obvious reflected signal echo, but below the set threshold, while the low and medium-order gain channel detection results show no obvious echo, then the corresponding micro defect is of size level V or above.

[0024] 2) If the detection result of the high-order gain channel exceeds the threshold for a certain same scanning position point, while the detection results of the low- and mid-order gain channels do not exceed the threshold, then the corresponding size level is a micro defect of level IV or above.

[0025] 3) If the detection results of the high-order and mid-order gain channels both exceed the threshold for a certain same scanning position, while the detection results of the low-order gain channel do not exceed the threshold, then the corresponding size level is a micro defect of level III or above.

[0026] 4) In the mid-order gain channel scan map, there are adjacent points that exceed the threshold, while at this point, the detection result of the low-order gain channel does not exceed the threshold, corresponding to micro-defects of size level II or above.

[0027] 5) If there is a point in the low-order gain channel scan that exceeds the threshold, and the detection results of both the high-order and mid-order gain channels exceed the threshold at this point, then the corresponding size level is a micro defect of grade I or above.

[0028] Preferably, in step S6, the different cases of the same detection point are specifically as follows:

[0029] Only high-sensitivity threshold exceeding;

[0030] The low, medium, and high-order sensitivity results all exceeded the threshold;

[0031] When middle and high-order threshold points are connected, does the low-order sensitivity exceed the threshold?

[0032] This invention provides a quantitative ultrasonic sensitivity difference method for measuring microscopic defects. Employing focused ultrasonic sensitivity difference, it quantitatively detects microscopic defects within steel, enabling rapid and hierarchical quantitative detection of all microscopic defects in steel samples. The method visually characterizes the quantity, density, morphology, and location of microscopic defects at various points on the tested steel plate using microscopic defect distribution images with different sensitivities for each gain channel. This method is highly beneficial for steel smelting, analysis, and research workers in rapidly and extensively analyzing steel at different cleanliness levels, as well as in material analysis during failure incidents. It also provides a highly sensitive and quantitative analytical technique for determining the quality of high-quality steel. Attached Figure Description

[0033] Figure 1 This is a schematic flowchart of the ultrasonic same-frequency sensitivity difference method of the present invention;

[0034] Figure 2 This is a schematic diagram illustrating the distribution of micro-defects of the smallest size level (Level V) and above in a steel plate sample quantitatively detected using the ultrasonic same-frequency sensitivity difference method of the present invention.

[0035] Figure 3 This is a schematic diagram illustrating the distribution of micro-defects of smaller size (level IV) and above in a steel plate sample quantitatively detected using the ultrasonic same-frequency sensitivity difference method of the present invention.

[0036] Figure 4 This is a schematic diagram illustrating the distribution of micro-defects of smaller size (level III) and above in a steel plate sample quantitatively detected using the ultrasonic same-frequency sensitivity difference method of the present invention.

[0037] Figure 5 This is a schematic diagram illustrating the distribution of micro-defects of smaller size (Level II) and above in a steel plate sample quantitatively detected using the ultrasonic same-frequency sensitivity difference method of the present invention.

[0038] Figure 6 This is a schematic diagram illustrating the distribution of micro-defects of smaller size (Level I) and above in a steel plate sample in the ultrasonic same-frequency sensitivity difference method of the present invention.

[0039] Figure 7This is a schematic diagram of the signal intensity characteristics of the smallest size level (level V) micro-defects in a steel plate sample quantitatively detected in an embodiment of the ultrasonic same frequency sensitivity difference method of the present invention. (a) shows a strong reflection signal in the high-order gain channel, (b) shows a weak reflection signal in the mid-order gain channel, and (c) shows no obvious reflection signal in the low-order gain channel.

[0040] Figure 8 This is a schematic diagram of the signal intensity characteristics of the quantitative detection of small-size (Level IV) micro-defects in a steel plate sample in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows a strong reflection signal in the high-order gain channel, (b) shows a relatively strong reflection signal in the mid-order gain channel, and (c) shows no obvious reflection signal in the low-order gain channel.

[0041] Figure 9 This is a schematic diagram of the signal intensity characteristics of the quantitative detection of intermediate-sized (Level III) micro-defects in a steel plate sample in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows a strong reflection signal in the high-order gain channel, (b) shows a strong reflection signal in the mid-order gain channel, and (c) shows no obvious reflection signal in the low-order gain channel.

[0042] Figure 10 This is a schematic diagram of the signal intensity characteristics of the quantitative detection of large-size (Level II) micro-defects in a steel plate sample in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows a strong reflection signal in the high-order gain channel, (b) shows a strong reflection signal in the mid-order gain channel, and (c) shows a weak reflection signal in the low-order gain channel.

[0043] Figure 11 This is a schematic diagram of the signal intensity characteristics of the quantitative detection of large-size (Level I) micro-defects in a steel plate sample in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows a strong reflection signal in the high-order gain channel, (b) shows a strong reflection signal in the mid-order gain channel, and (c) shows a reflection signal exceeding the threshold in the low-order gain channel.

[0044] Figure 12 This is a schematic diagram of the anatomical morphology of the smallest size level (level V) micro-defects in the steel plate sample quantitatively measured in the embodiment of the ultrasonic same frequency sensitivity difference method of the present invention. (a) is a micro-defect within 50 μm (level V), and (b) is a micro-defect within 50 μm (level V).

[0045] Figure 13 This is a schematic diagram of the anatomical morphology of small-size (Level IV) micro-defects in a steel plate sample quantitatively measured in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows micro-defects within 100 μm (Level IV), and (b) shows micro-defects within 100 μm (Level IV).

[0046] Figure 14This is a schematic diagram of the anatomical morphology of medium-sized (Level III) micro-defects in a steel plate sample quantitatively measured in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows micro-defects within 200 μm (Level III), and (b) shows micro-defects within 200 μm (Level III).

[0047] Figure 15 This is a schematic diagram of the anatomical morphology of micro-defects of larger size (Level II) and large size (Level I) in a steel plate sample quantitatively measured in an embodiment of the ultrasonic same-frequency sensitivity difference method of the present invention. (a) shows micro-defects within 500 μm (Level II), and (b) shows micro-defects above 500 μm (Level I). Detailed Implementation

[0048] To better understand the above-mentioned technical solutions of the present invention, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0049] This invention employs a three-channel mechanical scanning detection method to detect microscopic defects within a certain depth range below the surface of a steel plate. Specifically, (1) all three channels use ultrasonic sensors with identical ultrasonic pulse acoustic field characteristics at a specific frequency within the high-frequency range (25MHz to 50MHz), but the gains of the three detection channels are different, divided into low, medium, and high orders; (2) the physical positions of the grids used for mechanical scanning detection by the three channels are completely overlapping. The scanning tasks can be performed in the same batch or sequentially, or in batches. However, the measured scanning data results must be saved for subsequent comparison to obtain quantitative evaluation results.

[0050] Following the above detection method, ultrasonic reflection signals from microscopic defects within steel plates with different detection sensitivities can be obtained under identical detection conditions, such as high-frequency ultrasonic waves. During detection, it is crucial to ensure that the scanning step size and starting position of the mechanical scanner are precisely identical (scanning step: 0.5mm; the method for obtaining three sets of signal data with different detection sensitivities through the detection system can be achieved by repeating the detection three times at each of the three sensitivities. Since the mechanical repeatability of the detection system is very high (≤±0.01mm), the starting position can be guaranteed to be precisely identical. Alternatively, it can be achieved by simultaneously operating three identical sensors at the three sensitivities, with their measurement point positions also precisely overlapping). The purpose of this is to allow for comparison of ultrasonic signals with different sensitivity levels measured at the same location using low, medium, and high-order gains. By comparing and analyzing the amplitude of the reflection signal from micro-defects at the same location, and combining this with the actual size of the corresponding micro-defects obtained from actual dissection verification, the combined characteristics of the third-order gain conditions for each sensitivity level can be obtained for quantitative determination of micro-defects. Therefore, in the subsequent detection process, based on the signal amplitude characteristics of the detection result at a certain detection point under the third-order sensitivity condition, the measured super-threshold micro-defect reflection signal can be quickly quantitatively graded and processed to obtain the size, quantity, and distribution location of micro-defects at different size levels.

[0051] Combination Figure 1 As shown, the present invention provides a method for quantitatively measuring the ultrasonic sensitivity difference of micro-defects at the same frequency, comprising the following steps:

[0052] S1. Prepare polished samples according to standards, place them in an ultrasonic scanning device, and adjust all testing conditions to maintain consistency.

[0053] Step S1 involves processing the extracted steel raw material for analysis. On one hand, the raw material is processed to a thickness suitable for detection (10-15 mm). On the other hand, the surface of a batch of samples is subjected to the same polishing process. The surface polishing process can be carried out on a specially configured sample grinding machine. The surface roughness is controlled at the micron level after processing. In the experiment of this patent, a grinding machine is used to grind both the upper and lower surfaces of the sample flat, and the roughness reaches below 5 μm.

[0054] The purpose of the processing is to ensure that the surface roughness of the sample does not have different effects on the test results or lead to spurious results. This also ensures the consistency of the test conditions, so as not to affect the comparative effect of performing the same test task using low, medium, and high gain, thus impacting quantitative analysis.

[0055] Then, the technical parameters of the ultrasound scanning device are adjusted to remain consistent, specifically as follows:

[0056] Detection method: C-image scanning covering the sample; (repeatability ≤ ±0.01 mm);

[0057] Scan step distance: 1mm

[0058] Scanning speed: 100mm / s

[0059] Focusing sensor frequency: 25MHz, focusing length: 80mm

[0060] Operating temperature: 15℃~35℃

[0061] The system has a 30MHz receiving bandwidth.

[0062] Output impedance: 50 ohms

[0063] Input impedance: 100 ohms

[0064] Gain ranging from -20 to +80 dB, adjustable by ±0.1 dB

[0065] High-pass filter: 22.5MHz

[0066] Adjust the low-pass filter to 30MHz.

[0067] Pulse voltage: 350 volts (<5 nsec rise time), ±10 volt adjustable

[0068] Pulse width: 25MHz

[0069] Emission energy: 470uJ

[0070] Ultrasonic receiver input impedance: 100 ohms

[0071] Maximum input voltage range: + / - 1 volt

[0072] Quantization resolution: 12 bits

[0073] Signal acquisition rate: 100M / s

[0074] Internal trigger mode.

[0075] S2 uses three ultrasonic sensor groups with the same parameters (25MHz~50MHz) in conjunction with an ultrasonic scanning device to obtain an imaging result set;

[0076] Step S2 involves using three ultrasonic channels with the same frequency (25MHz to 50MHz) and sensor groups with low, medium, and high third-order gains, respectively, in conjunction with mechanical scanning to obtain imaging result sets. For example, in a subsequent embodiment, the present invention uses a combination of three ultrasonic sensors with a frequency of 25MHz and gains of 49dB (lower sensitivity), 61dB (medium sensitivity), and 79dB (higher sensitivity), respectively. After scanning the processed sample, three microscopic defect distribution maps with different sensitivity levels are obtained. In the scanning detection, the step size and starting position of the mechanical scanning are ensured to be precisely the same.

[0077] The purpose of this is to avoid the quantitative results being interfered with or affected by other factors in the ultrasonic testing system. By ensuring that the scanning step length and starting position are accurate, it can be guaranteed that the measured signals of the three channels will not be affected by the differences in the detected target due to positional deviations, which would produce inconsistent results.

[0078] S3, the locations where the signal exceeds the threshold in the ultrasound sensor group scanning imaging results group under third-order sensitivity. The threshold used in each image is obtained based on anatomical verification.

[0079] Step S3 involves identifying the signal exceeding the threshold points in the scan images obtained from the low, medium, and high gain measurements, and then comparing and determining whether each point overlaps in the image. The effect of this is that the low-sensitivity scan image can detect a very small number of micro-defects, which is determined by its low sensitivity.

[0080] When ultrasound propagates through steel, the propagation speed of ultrasound remains constant for a given material. If a specific high-frequency ultrasound band is used for detection, its wavelength will also be the corresponding specific wavelength. For a specific micro-defect in the steel, in the second step, it has been ensured that the reflection surface of the ultrasound waves to the three channels of the same frequency is identical, so the reflected energy received by the sensors is also the same. Given a constant propagation speed and frequency of ultrasound in steel, and a consistent wavelength, the detectable micro-defect targets become smaller as the gain increases. Therefore, when using low, medium, and high-order gains, the size of the detectable micro-defect targets differs. At the same threshold height in the time domain, the low-order gain channel detects relatively larger micro-defects, while the high-order gain channel detects even smaller micro-defects at the defect location point in the scan image, which are undetectable by the low and medium-order gain channels. Thus, the high-order gain channel detects the largest number of micro-defects with the smallest size, which is determined by its correspondingly higher sensitivity.

[0081] Through extensive comparative analysis of detection signals and microscopic defects, this patent has determined the following quantitative technical parameters: Under the aforementioned detection technology conditions, the corresponding thresholds for comparative quantitative analysis at three different gain levels (high (79dB), medium (61dB), and low (49dB) are 90% of 0.125V, 70% of 0.25V, and 50%, respectively.

[0082] S4, extract the ultrasonic signals of all locations where the signal exceeds the threshold, and extract the time domain signal length at the same monitoring depth;

[0083] Step S4 involves extracting the ultrasonic signals from all mechanical location points exceeding the threshold, and extracting signal segments at the same monitoring depth. The purpose of this is to compare whether two or more connected (i.e., adjacent detection mechanical location points) signals exceeding the threshold are at the same depth within the sample when large micro-defects are encountered. If the depths of two connected signals exceeding the threshold are different, it means that the signals at these two locations are likely generated by reflections from two different, relatively large micro-defects. If the depths of two connected signals exceeding the threshold are the same, it means that the signals at these two locations are generated by reflections from the same large micro-defect.

[0084] S5. Based on whether the locations where all signals exceed the threshold coincide under the third-order sensitivity, group all the locations where signals exceed the threshold.

[0085] S6, compare one by one the following situations: only the high sensitivity exceeds the threshold at the same detection point; low, medium and high order sensitivity results all exceed the threshold; when the medium and high order thresholds are connected; whether the low order sensitivity exceeds the threshold; and output the detection quantitative result based on quantitative analysis.

[0086] Step S6 uses low, medium, and high gains to measure scan images at different sensitivity levels. It compares whether the threshold points at the same location in each image overlap. All threshold points are grouped according to the following rules: For example, in this embodiment of the invention, the three ultrasonic channels have the same frequency (25MHz) and are configured with low (49dB), medium (61dB), and high (79dB) gains to form sensor groups, which are then used in conjunction with mechanical scanning for imaging detection.

[0087] 1) For a certain same scanning position point, the high-order gain channel detection result shows obvious reflected signal echo, but below the set threshold, while the low and medium-order gain channel detection results show no obvious echo, then the corresponding micro defect is of size level V or above.

[0088] 2) If the detection result of the high-order gain channel exceeds the threshold for a certain same scanning position point, while the detection results of the low- and mid-order gain channels do not exceed the threshold, then the corresponding size level is a micro defect of level IV or above.

[0089] 3) If the detection results of the high-order and mid-order gain channels both exceed the threshold for a certain same scanning position, while the detection results of the low-order gain channel do not exceed the threshold, then the corresponding size level is a micro defect of level III or above.

[0090] 4) In the mid-order gain channel scan map, there are adjacent points that exceed the threshold, while at this point, the detection result of the low-order gain channel does not exceed the threshold, corresponding to micro-defects of size level II or above.

[0091] 5) If there is a point in the low-order gain channel scan that exceeds the threshold, and the detection results of both the high-order and mid-order gain channels exceed the threshold at this point, then the corresponding size level is a micro defect of grade I or above.

[0092] S7, based on the quantitative detection results, and combined with the quantitative evaluation model of micro-defects, the quantity, size, and distribution of micro-defects at each size level are obtained, such as... Figures 2 to 6 As shown.

[0093] The levels I to V mentioned above refer to inclusion size levels, with units of μm.

[0094] Step S1 of the ultrasonic same-frequency sensitivity difference method of the present invention involves surface processing of the extracted steel raw material to be analyzed. This includes processing the raw material to a thickness suitable for detection, and performing a uniform surface polishing process on the surface of, for example, all samples in a batch. This surface polishing process can be performed on a specially configured sample grinding machine, and the surface roughness is controlled at the micrometer level after processing. The purpose of this processing is to ensure that the surface roughness of the sample does not have different effects on the detection results or lead to artifacts in the detection results. This also ensures the consistency of the detection conditions, so as not to affect the comparative effect of performing the same detection task using low, medium, and high gain, thus impacting quantitative analysis.

[0095] Step S2 of the ultrasonic same-frequency sensitivity difference method of the present invention involves using three ultrasonic channels with the same frequency (25MHz~50MHz) and sensor groups with low, medium, and high third-order gains, respectively, in conjunction with mechanical scanning to obtain imaging result maps. For example, in a subsequent embodiment of the present invention, a combination of three-channel ultrasonic sensors with a frequency of 25MHz and gains of 49dB (lower sensitivity), 61dB (medium sensitivity), and 79dB (higher sensitivity) are used. After scanning the processed sample, three micro-defect distribution maps with different sensitivity levels are obtained. In the scanning detection, the step size and starting position of the mechanical scanning are ensured to be precisely the same. From the relationship between the size and volume of micro-defects, larger micro-defects are fewer in number and require correspondingly larger volumes to be detected.

[0096] Step S3 of the ultrasonic same-frequency sensitivity difference method of the present invention involves determining the signal exceeding the threshold position points in the scan images measured by the low, medium, and high-order gains, and then comparing and determining whether each point overlaps in the image. The effect of this is that the low-sensitivity scan image can detect very few micro-defects, which is determined by its low sensitivity. When ultrasound propagates in steel, the propagation speed of ultrasound in a specific material is constant. If a certain high-frequency ultrasonic wave is used for detection, its wavelength is also the corresponding specific wavelength. For a specific micro-defect in the steel, its reflecting surface to the three same-frequency ultrasonic channels is the same, so the reflected energy received by the sensor is also the same. With a constant propagation speed and frequency of ultrasound in steel, the wavelength is also the same. Therefore, as the gain increases, the smaller the micro-defect target that can be detected becomes. Therefore, when using low, medium, and high gain, the size of the micro-defects that can be detected varies. In the time domain signal with the same threshold height, the micro-defects detected by the low-order gain channel are relatively larger, while the scan image obtained by using the high-order gain channel can detect even smaller micro-defects that cannot be detected by the low and medium-order gain channels. Therefore, the number of micro-defects is the largest, and the size can also be the smallest. This is determined by the relatively high sensitivity of the high-order gain.

[0097] Finally, the ultrasonic sensitivity difference method of this patent extracts the ultrasonic signals from the mechanical location points of all over-threshold signals, and extracts signal segments at the same monitoring depth. The purpose of this is to compare whether two or more connected over-threshold micro-defect signals are at the same depth within the sample when large micro-defects are encountered. If the depth positions of two connected over-threshold micro-defect signals are different, it means that the signals at these two locations are generated by reflections from two different, relatively large micro-defects; if the depth positions of two connected over-threshold micro-defect signals are the same, it means that the over-threshold signals at these two locations are generated by reflections from the same large micro-defect.

[0098] Based on the above steps, scan images with different sensitivity levels are obtained using low, medium, and high gains. The overlap of threshold points at the same location in each image is compared, and all threshold points are grouped according to the following rules: For example, in this embodiment of the invention, the three ultrasonic channels have the same frequency (25MHz), and sensor groups are formed with low (49dB), medium (61dB), and high (79dB) gains, respectively, to perform imaging detection in conjunction with mechanical scanning.

[0099] (1) For a certain same scanning position point, the high-order gain channel detection results show obvious reflected signal echo, but below the set threshold, the low and medium-order gain channel detection results show no obvious echo.

[0100] (2) For a certain same scanning position point, the detection result of the high-order gain channel exceeds the threshold, while the detection results of the low- and medium-order gain channels do not exceed the threshold.

[0101] (3) For a certain same scanning position point, the detection results of the high-order and mid-order gain channels both exceed the threshold, while the detection results of the low-order gain channel do not exceed the threshold.

[0102] (4) In the high- and mid-order gain channel scanning maps, there are adjacent points that exceed the threshold, while at these points, the detection result of the low-order gain channel does not exceed the threshold.

[0103] (5) In the low-order gain channel scan map, there is a point where the threshold is exceeded, and at this point, the detection results of the high, medium and low-order gain channels all exceed the threshold.

[0104] Based on the above classification, and after comparing the detection results and performing dissection and calibration on the corresponding locations, the micro-defect size level under each type of condition can be determined. In subsequent detections, the comprehensive characteristics of the gain channel signals of each order using the above-mentioned same-frequency sensitivity difference method can be used to compare the exceedance of threshold values ​​for each gain channel at the same detection point, as well as whether the exceedance points are connected, thus forming quantitative analysis criteria.

[0105] Through extensive comparative analysis of detection signals and microscopic defects, this invention has determined the following quantitative technical parameters:

[0106] Under the aforementioned detection technology conditions,

[0107] For detection at high (79dB) gain, the corresponding threshold for comparative quantitative analysis was set to 90% of 0.125V.

[0108] For detection at medium (61dB) gain, the corresponding threshold for comparative quantitative analysis is set to 70% of 0.25V;

[0109] For detection at low (49dB) gain, the corresponding threshold for comparative quantitative analysis was set to 50% of 0.25V.

[0110] In summary, this invention enables rapid and quantitative detection of all microscopic defects within steel samples at different levels. It visually characterizes the quantity, density, morphology, and location of microscopic defects at various points on the tested steel plate using microscopic defect distribution images with different sensitivities across various gain channels. This method facilitates rapid and large-scale analysis of steel samples at different cleanliness levels, as well as material analysis in failure incidents, for workers in steel smelting, analysis, and research. It also provides a highly sensitive and quantitatively effective detection technique for determining the quality of high-quality steel.

[0111] Example

[0112] In this embodiment, an ultrasonic sensitivity difference method for quantitatively measuring micro-defects was used to quantitatively detect and analyze a steel plate that had been processed according to testing requirements.

[0113] In quantitative detection and analysis, the ultrasonic mechanical scanning equipment used includes a mechanical scanning system, a signal transmission / reception and processing section, a signal digitization system, an ultrasonic probe, and detection operation procedures. The system employs built-in low-pass and high-pass filters, with an amplitude quantization accuracy of 12 bits; the maximum data acquisition speed is 200MHz. A short focal length focusing sound field is used, employing three sensor groups with the same frequency of 25MHz. Detection is performed at three different gains: low (49dB), medium (61dB), and high (79dB), with corresponding threshold settings of 90%, 70%, and 50%. The imaging detection scan step size is 1mm × 1mm; the pulse voltage is 350V; the pulse width is 25MHz; and the emission energy is 470uJ.

[0114] The specific steps of the ultrasonic same-frequency sensitivity difference method in this embodiment are as described above. Figure 1 As shown, the results obtained through experiments Figures 2 to 6 Distribution diagram of micro-defects at various size levels.

[0115] The specific technical parameters of the ultrasound scanning device are as follows:

[0116] Detection method: C-image scanning covering the sample; (repeatability ≤ ±0.01 mm);

[0117] Scan step distance: 1mm

[0118] Scanning speed: 100mm / s

[0119] Focusing sensor frequency: 25MHz, focusing length: 80mm

[0120] Operating temperature: 15℃~35℃

[0121] The system has a 30MHz receiving bandwidth.

[0122] Output impedance: 50 ohms

[0123] Input impedance: 100 ohms

[0124] Gain ranging from -20 to +80 dB, adjustable by ±0.1 dB

[0125] High-pass filter: 22.5MHz

[0126] Adjust the low-pass filter to 30MHz.

[0127] Pulse voltage: 350 volts (<5 nsec rise time), ±10 volt adjustable

[0128] Pulse width: 25MHz

[0129] Emission energy: 470uJ

[0130] Ultrasonic receiver input impedance: 100 ohms

[0131] Maximum input voltage range: + / - 1 volt

[0132] Quantization resolution: 12 bits

[0133] Signal acquisition rate: 100M / s

[0134] Internal trigger mode.

[0135] Figure 2 This is a distribution map of micro-defects at the minimum size level and above, which is also the distribution map of all micro-defects that can be detected by the high (79dB) order gain channel. Further analysis determined that this sensitivity level corresponds to micro-defect sizes within 50μm (temporarily designated as level V in this embodiment).

[0136] Figure 3 This is a distribution map of micro-defects at the smaller size level and above. This level also represents the distribution map of all micro-defects exceeding the threshold that can be detected by the high (79dB) order gain channel, and the distribution map of all micro-defects below the threshold that can be detected by the mid (61dB) order gain channel. Further analysis determined that this sensitivity level corresponds to micro-defect sizes in the range of 50μm to 100μm (temporarily designated as level IV in this embodiment). In this embodiment, the number of micro-defects at levels V and IV is relatively large.

[0137] Figure 4This is a distribution map of micro-defects at the medium to large size level. This level also represents the distribution map of all micro-defects exceeding the threshold that can be detected by both high- and mid-order gain channels, and the distribution map of all micro-defects below the threshold that can be detected by the low (49dB) order gain channel. Further analysis determined that this sensitivity level corresponds to micro-defect sizes in the range of 100μm to 200μm (temporarily designated as Level III in this embodiment). In this embodiment, there are 8 such defects.

[0138] Figure 5 This is a distribution map of micro-defects at the larger size level and above. This level also includes the distribution map of all connected over-threshold micro-defects that can be detected by both mid- and high-order gain channels, and the distribution map of all micro-defects below the threshold that can be detected by the low-order gain channel. Further analysis determined that this sensitivity level corresponds to micro-defect sizes in the range of 200 μm to 500 μm (temporarily designated as Level II in this embodiment). In this embodiment, there are two such defects.

[0139] Figure 6 This is a distribution map of micro-defects at the large-size level and above. This level also includes the distribution map of all connected super-threshold micro-defects that can be detected by both mid- and high-order gain channels, and the distribution map of all micro-defects above the threshold that can be detected by low-order gain channels. Further analysis confirmed that the micro-defect sizes obtained at this sensitivity level are all micro-defects in the range of 500 μm and above (temporarily designated as Level I in this embodiment). None are present in this embodiment.

[0140] exist Figures 2 to 6 There are four detection points labeled A, B, C, and D, which are all locations where the sensitivity of a certain gain channel exceeds the threshold. Figures 7 to 10 The paper presents the signals obtained from reflections through microscopic defects at corresponding locations and with three low, medium, and high-order gain channels. A method for grouping signals is described based on whether the threshold points at the same location in the scan diagrams of each gain channel overlap.

[0141] Figure 7 The high-order gain channel detection results show obvious reflected signal echoes below the threshold, while the mid- and low-order gain channel detection results show no obvious echoes.

[0142] Figure 8 The detection results of the high-order gain channel exceed the threshold, while the detection results of the medium and low-order gain channels do not exceed the threshold. Figure 2 and Figure 3 Points A and B are marked in the middle.

[0143] Figure 9 The detection results for both high- and mid-order gain channels exceed the threshold, while the detection results for the low-order gain channel do not exceed the threshold. Figures 2 to 6 The location is marked as C.

[0144] Figure 10 In the high- and mid-order gain channel scan maps, there are adjacent points exceeding the threshold, and at these points, the low-order gain channel results exceed the threshold.

[0145] Figure 11 In the high- and mid-order gain channel scan maps, there are adjacent points exceeding the threshold, and at these points, the low-order gain channel results exceed the threshold.

[0146] Based on the above tests, the results of the dissection and verification of locations where a certain sensitivity level exceeded the threshold were obtained, such as... Figures 11 to 14 As shown, the calibration method for quantitatively determining microscopic defects in steel using the focused ultrasonic wave sensitivity difference method also employs a combination of detection and dissection. Figure 11 Two microscopic defects within 50 μm are given (in this embodiment, they can be tentatively designated as level V). Figure 12 Two microscopic defects in the range of 50μm to 100μm are given (in this embodiment, they can be tentatively designated as level IV). Figure 13 Two microscopic defects in the range of 100μm to 200μm are given (in this embodiment, they can be tentatively designated as level III). Figure 14 (a) is a microscopic defect in the range of 200μm to 500μm (which can be tentatively defined as Class II in this embodiment), i.e. Figures 2 to 6 Microscopic defects at location C. Figure 14 (b) refers to microscopic defects in the range of 500 μm or larger (which can be tentatively defined as Level I in this embodiment). At this detection point, the detection result at a frequency of 15 MHz exceeds the threshold.

[0147] Those skilled in the art should recognize that the above embodiments are merely illustrative of the present invention and are not intended to limit the present invention. Any variations or modifications to the above embodiments that are within the spirit and essence of the present invention will fall within the scope of the claims of the present invention.

Claims

1. A method for quantitatively determining the ultrasonic sensitivity difference of microscopic defects, characterized in that, Includes the following steps: S1, Prepare the polished sample and place it in an ultrasonic scanning device; S2, using an ultrasonic sensor group with the same parameters in conjunction with the ultrasonic scanning device to obtain an imaging result group; S3, Under third-order sensitivity, the ultrasonic sensor group scans the locations where the signal exceeds the threshold in the imaging result group; S4, extract the ultrasonic signals of all locations where the signals exceed the threshold, and extract the time-domain signal length at the same monitoring depth; S5, group all signal points exceeding the threshold according to whether the points where all signals exceed the threshold coincide under the third-order sensitivity. S6, compare the threshold exceedances under different conditions at the same detection point one by one, and output the quantitative detection results based on quantitative analysis; S7. Based on the quantitative detection results, and combined with the quantitative evaluation model for micro-defects, the quantity, size, and distribution of micro-defects at each size level are obtained.

2. The method for quantitatively determining the ultrasonic sensitivity difference of micro-defects according to claim 1, characterized in that: In step S1, the thickness of the polished sample is 10-15 mm, and the surface roughness is below 5 μm.

3. The method for quantitatively determining the ultrasonic sensitivity difference of micro-defects according to claim 1, characterized in that: In step S2, three ultrasonic channels with the same frequency are used, which are the ultrasonic sensor groups with low, medium and high third-order gains, respectively.

4. The method for quantitatively determining the ultrasonic sensitivity difference of micro-defects according to claim 3, characterized in that: The frequency of the ultrasonic sensor group is 25MHz to 50MHz.

5. The method for quantitatively determining the ultrasonic sensitivity difference of micro-defects according to claim 3, characterized in that, Step S3 specifically includes: In step S2, the locations where the signals exceed the threshold are determined from the imaging result groups measured by the ultrasonic sensor groups with low, medium, and high gains, respectively. Then, the locations are compared and it is determined whether each location overlaps in the imaging result group.

6. The method for quantitatively determining the ultrasonic sensitivity difference of micro-defects according to claim 3, characterized in that, In step S5, all the location points where the signal exceeds the threshold are grouped according to the following rules: 1) For a certain same scanning position point, the high-order gain channel detection result shows obvious reflected signal echo, but below the set threshold, while the low and medium-order gain channel detection results show no obvious echo, then the corresponding micro defect is of size level V or above. 2) If the detection result of the high-order gain channel exceeds the threshold for a certain same scanning position point, while the detection results of the low- and mid-order gain channels do not exceed the threshold, then the corresponding size level is a micro defect of level IV or above. 3) If the detection results of the high-order and mid-order gain channels both exceed the threshold for a certain same scanning position, while the detection results of the low-order gain channel do not exceed the threshold, then the corresponding size level is a micro defect of level III or above. 4) In the mid-order gain channel scan map, there are adjacent points that exceed the threshold, while at this point, the detection result of the low-order gain channel does not exceed the threshold, corresponding to micro-defects of size level II or above. 5) If there is a point in the low-order gain channel scan that exceeds the threshold, and the detection results of both the high-order and mid-order gain channels exceed the threshold at this point, then the corresponding size level is a micro defect of grade I or above.

7. The method for quantitatively determining the ultrasonic sensitivity difference of micro-defects according to claim 3, characterized in that, In step S6, the different cases of the same detection point are specifically as follows: Only high-sensitivity threshold exceeding; The low, medium, and high-order sensitivity results all exceeded the threshold; When middle and high-order threshold points are connected, does the low-order sensitivity exceed the threshold?

Images