Method, system and apparatus for toughness testing of steel alloys at low temperatures

Abstract

The application discloses a method, system and device for testing the toughness of steel alloy low-temperature impact, and relates to the related field of toughness testing. The method comprises the following steps: uniformly sampling the steel alloy to be tested according to the production time to obtain a sample set; performing segregation detection by using a scanning electron microscope to generate a segregation result set; performing double distance transformation on the segregation result set and clustering the result set into multiple sample groups; placing the sample groups into a liquid nitrogen cooling device to stand for one hour to obtain a cooled sample group, and using a low-temperature impact testing machine to complete toughness testing to obtain a test result set; and analyzing the result set to obtain a test result. The application solves the technical problems that, in the toughness testing of steel alloy low-temperature impact, the test result deviation is large and the process defects cannot be accurately positioned due to sample segregation and smelting process differences, and achieves the technical effects that the low-temperature toughness of different steel alloys is accurately evaluated through segregation detection and sample analysis, and the stability and reliability of the low-temperature impact toughness testing are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of toughness testing, in particular to a toughness testing method, system and device for low-temperature impact of steel alloy. BACKGROUND

[0002] The impact toughness of steel alloy at low temperature is one of the important indicators to measure its service safety, especially in the fields of low-temperature pressure vessels, marine engineering structures, low-temperature storage and transportation equipment, etc., the low-temperature toughness of the material is particularly strict. The traditional low-temperature impact toughness testing method usually directly tests the low-temperature impact of randomly sampled steel alloy samples, and judges the material performance according to the impact work value. However, in actual production, the steel alloy may produce segregation, composition fluctuation and other phenomena due to differences in smelting process, composition control, cooling method and microstructure uniformity, and these microscopic heterogeneities will significantly affect the low-temperature impact toughness test results, resulting in large fluctuations in test data, which is difficult to accurately reflect the overall material performance. In addition, the existing method lacks systematic and intelligent processing in the aspects of sample selection, microstructure detection and data analysis, cannot effectively cluster analyze the material differences of different batches and different production times, and lacks correlation evaluation of segregation and toughness results, so that the test results have deficiencies in consistency, repeatability and reliability. Especially at very low temperature (such as-70℃), the impact work value is more easily disturbed by sample individual differences, causing quality judgment risk and affecting the safe use of materials in harsh environments. SUMMARY

[0003] The present application provides a toughness testing method, system and device for low-temperature impact of steel alloy, which solves the technical problems of large test result deviation and inability to accurately locate process defects caused by sample segregation and smelting process differences in the low-temperature impact toughness testing of steel alloy, and achieves the technical effect of accurately evaluating the low-temperature toughness of different steel alloys through segregation detection and aggregated sample analysis, and improving the stability and reliability of low-temperature impact toughness testing.

[0004] The application provides a toughness test method for steel alloy low-temperature impact, comprising: obtaining a steel alloy set to be sampled, uniformly sampling the steel alloy according to a production time stamp of the steel alloy, and obtaining a steel alloy sample set; performing segregation detection on the steel alloy sample set respectively by using a scanning electron microscope, and determining a segregation detection result set; performing two-item distance transformation on the segregation detection result set, and performing same-type aggregation to obtain a plurality of aggregated steel alloy sample sets; placing the plurality of aggregated steel alloy sample sets in a liquid nitrogen cooling device respectively, and standing for 1 hour to obtain a plurality of aggregated cooled steel alloy sample sets, and performing toughness test on the plurality of aggregated cooled steel alloy sample sets by using a low-temperature impact testing machine to obtain a plurality of aggregated toughness test result sets; performing same-type longitudinal analysis and different-type transverse analysis on the plurality of aggregated toughness test result sets to obtain a target toughness test result of the steel alloy set to be sampled.

[0005] The method comprises: configuring an acceleration voltage, a working distance and a scanning mode of the scanning electron microscope; placing the steel alloy sample set in a scanning cavity of the scanning electron microscope for electron beam scanning to obtain a steel alloy sample surface image set; identifying the composition of the steel alloy sample set by using energy dispersive X-ray spectroscopy to obtain a steel alloy sample element concentration set; and performing one-to-one mapping on the steel alloy sample surface image set and the steel alloy sample element concentration set to determine the segregation detection result set.

[0006] The method comprises: extracting a first segregation detection result from the segregation detection result set, wherein the first segregation detection result is any one of the segregation detection results in the segregation detection result set; performing two-item distance transformation on the segregation detection result set based on the first segregation detection result from two dimensions of segregation distribution and element concentration to obtain a first transformed segregation detection result set; identifying the first transformed segregation detection result set according to a preset aggregation scale based on the first segregation detection result as an aggregation reference, and adding a first aggregated steel alloy sample set obtained by the identification to the plurality of aggregated steel alloy sample sets.

[0007] The method comprises: using a segregation area identifier to perform binary conversion on the steel alloy sample surface image set in the segregation detection result set to obtain a steel alloy sample segregation area binary image set; from the two dimensions of segregation distribution and element concentration, taking the first steel alloy sample segregation area binary image corresponding to the first segregation detection result and the first steel alloy sample element concentration as a reference, performing similarity identification on the steel alloy sample segregation area binary image set and the steel alloy sample element concentration set in the segregation detection result set to obtain a segregation detection result similarity set; and performing two-item distance transformation on the segregation detection result set based on the segregation detection result similarity set to obtain the first transformed segregation detection result set.

[0008] The method comprises: using a cosine similarity calculation formula to calculate the similarity of the steel alloy sample segregation area binary image set and the steel alloy sample element concentration set in the segregation detection result set with the first steel alloy sample segregation area binary image and the first steel alloy sample element concentration respectively to obtain an image similarity set and an element concentration similarity set; and performing mapping and weighting on the image similarity set and the element concentration similarity set to obtain the segregation detection result similarity set.

[0009] The method comprises: calculating the ratio of any one of the segregation detection result similarities in the segregation detection result similarity set to the sum of the similarities in the segregation detection result similarity set, and taking the difference between the ratio and 1 as a two-item distance transformation coefficient to obtain a two-item distance transformation coefficient set; and adjusting the segregation detection result set in the direction away from the first segregation detection result based on the two-item distance transformation coefficient set to obtain the first transformed segregation detection result set.

[0010] The method comprises: performing intra-set homogenous longitudinal analysis on the multiple aggregated toughness test result sets to obtain multiple homogenous aggregated toughness test results; randomly extracting one aggregated toughness test result from each of the multiple aggregated toughness test result sets without replacement, performing mean shift screening on the multiple heterogeneous aggregated toughness test results obtained through multiple extractions to obtain multiple heterogeneous aggregated toughness test results; and performing weighted calculation on the multiple homogenous aggregated toughness test results and the multiple heterogeneous aggregated toughness test results to obtain the target toughness test result.

[0011] The method comprises: performing mean value calculation on the multiple aggregated toughness test result sets respectively to determine multiple aggregated toughness test mean values; taking the multiple aggregated toughness test mean values as initial mean shift centers, and iteratively performing mean shift screening on the multiple aggregated toughness test result sets according to a preset mean shift bandwidth until a preset iteration number is met to obtain the multiple homogenous aggregated toughness test results.

[0012] The application also provides a toughness test system for low-temperature impact of steel alloy, comprising: a uniform sampling module: obtaining a set of steel alloy to be sampled, uniformly sampling the set of steel alloy according to a production timestamp of the steel alloy, and obtaining a set of steel alloy samples; a segregation detection module: detecting the set of steel alloy samples respectively by using a scanning electron microscope, and determining a set of segregation detection results; a two-item distance transformation module: performing two-item distance transformation on the set of segregation detection results, and performing same-type aggregation to obtain a plurality of sets of aggregated steel alloy samples; a toughness test module: placing the plurality of sets of aggregated steel alloy samples respectively in a liquid nitrogen cooling device, standing for 1 hour, obtaining a plurality of sets of aggregated cooled steel alloy samples, and performing toughness test on the plurality of sets of aggregated cooled steel alloy samples by using a low-temperature impact testing machine, and obtaining a plurality of sets of aggregated toughness test results; and a test result analysis module: performing same-type longitudinal analysis and different-type transverse analysis on the plurality of sets of aggregated toughness test results, and obtaining a target toughness test result of the set of steel alloy to be sampled.

[0013] The application also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the toughness test method for low-temperature impact of steel alloy.

[0014] The toughness test method, system and device for low-temperature impact of steel alloy provided by the application first uniformly sample a set of samples from a set of steel alloy to be sampled according to a production timestamp. Then, the samples are detected by using a scanning electron microscope to obtain segregation results. Subsequently, a plurality of sets of aggregated steel alloy samples are formed by two-item distance transformation and same-type aggregation. Then, the samples are cooled in liquid nitrogen for 1 hour, and toughness test is performed by using a low-temperature impact testing machine to obtain a plurality of test results. Finally, the test results are analyzed longitudinally and transversely to obtain a target toughness test result of the steel alloy to be sampled, thereby improving the stability and reliability of low-temperature impact toughness test. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings of the embodiments of the application will be briefly introduced as follows. In the present application, a flowchart is used to illustrate the operations performed by the system according to the embodiments of the application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously according to needs. Meanwhile, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0016] Figure 1A flowchart of a toughness test method for low-temperature impact of steel alloy provided by an embodiment of the present application.

[0017] Figure 2 A structural diagram of a toughness test system for low-temperature impact of steel alloy provided by an embodiment of the present application.

[0018] Figure 3 A structural diagram of an electronic device provided by an embodiment of the present application.

[0019] Legend: uniform sampling module 11, segregation detection module 12, two-item distance transformation module 13, toughness test module 14, test result analysis module 15, processor 21, memory 22, input device 23, output device 24. DETAILED DESCRIPTION

[0020] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clear, the embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0021] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will further describe the present application in combination with the drawings, and the described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without making creative labor are within the scope of protection of the present application.

[0022] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subset of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to only distinguishes similar objects, and does not represent a specific order of the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art of the technology to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiments of the present application provide a toughness test method for low-temperature impact of steel alloy, as shown in Figure 1 The method comprises:

[0024] Obtain a steel alloy set to be sampled, and uniformly sample the steel alloy set according to a production timestamp of the steel alloy to obtain a steel alloy sample set.

[0025] Specifically, in order to ensure the representativeness and comprehensiveness of the steel alloy sample, the steel alloy set to be sampled is uniformly sampled according to the production timestamp, that is, a certain number of samples are selected according to the interval of the production timestamp, so as to ensure that the selected samples can cover various production batches and are representative. The sampling interval is determined according to the total size of the sample, the test requirement and the selected time period. Each sampled sample is numbered and labeled, and is stored to form a steel alloy sample set, which provides a reliable sample basis for subsequent segregation detection and low-temperature toughness test, and further improves the accuracy and reliability of the test result.

[0026] The steel alloy sample set is subjected to segregation detection by using a scanning electron microscope to determine a segregation detection result set.

[0027] Specifically, after obtaining the steel alloy sample set, each sample in the steel alloy sample set is scanned by using a pre-configured scanning electron microscope, and the microstructure image of the alloy surface is recorded. Meanwhile, the energy dispersive X-ray spectroscopy (EDS) function of the scanning electron microscope is used to quantitatively analyze the composition of each sample, and the element composition and concentration of each region are identified. By mapping the image and element concentration information obtained by the scanning electron microscope, a segregation detection result set is constructed, which records the segregation region, the variation range of the element concentration and the distribution characteristics of each sample. The segregation detection result set will serve as the basic data for subsequent aggregation analysis, low-temperature toughness test and other analysis, and ensures the accuracy of the low-temperature impact toughness test result.

[0028] In one possible implementation, the steel alloy sample set is subjected to segregation detection by using a scanning electron microscope to determine a segregation detection result set, and the method comprises the following steps:

[0029] The acceleration voltage, working distance and scanning mode of the scanning electron microscope are configured; the steel alloy sample set is placed in the scanning cavity of the scanning electron microscope for electron beam scanning to obtain a steel alloy sample surface image set; the composition of the steel alloy sample set is identified by using energy dispersive X-ray spectroscopy to obtain a steel alloy sample element concentration set; and the steel alloy sample surface image set and the steel alloy sample element concentration set are one-to-one mapped to determine the segregation detection result set.

[0030] Specifically, before starting the scanning, first, according to the type of steel alloy sample, alloy composition and surface characteristics of the sample, the working parameters of the scanning electron microscope are configured, including the accelerating voltage, working distance and scanning mode of the scanning electron microscope, wherein the accelerating voltage is usually between 15 kV and 30 kV to ensure that high-resolution surface images and good element analysis results are obtained, higher accelerating voltage is suitable for thicker samples, and lower accelerating voltage is suitable for thinner samples or occasions requiring higher image clarity; the working distance is usually between 10 and 20 mm to ensure that the electron beam can be accurately focused and effectively scan the sample surface, a shorter working distance can obtain higher resolution, but will limit the maximum observation area of the sample; the scanning mode is selected according to the characteristics of the steel alloy sample, including point scanning, line scanning and area scanning, point scanning can obtain higher resolution and is suitable for detail analysis, line scanning is suitable for analyzing the change of element distribution, and area scanning can quickly obtain information of a larger area. After the scanning electron microscope is configured, the steel alloy sample is placed in the scanning cavity of the scanning electron microscope. During the scanning process, the electron beam scans the sample surface to generate secondary electron signals, and the scanning electron microscope converts these signals into images to obtain detailed images of the microstructure of the steel alloy sample surface, forming a set of steel alloy sample surface images. Each surface image in this set of steel alloy sample surface images can clearly show the microstructure, cracks, phase interface and segregation area of the alloy. The scanning electron microscope is usually equipped with an energy dispersive X-ray spectrum detector for analyzing the element composition of the steel alloy sample. During the electron beam scanning process, when the electron beam interacts with the sample, it will excite the elements in the sample to emit X-rays. By capturing these X-rays with the energy dispersive X-ray spectrum detector, the characteristic X-ray spectra of different elements can be obtained, which show the intensity and distribution of each element in the sample. By identifying the characteristic peak energy values of the characteristic X-ray spectra of each element, the type of each element can be determined, such as iron, carbon, manganese, silicon, chromium, etc. By calculating the spectral peak intensity of the characteristic X-ray spectra of each element, the concentration of each element can be determined, thereby forming a set of steel alloy sample element concentration. Finally, the set of steel alloy sample surface images and the set of steel alloy sample element concentrations are mapped one-to-one, i.e., the steel alloy sample surface images are used as a reference to label each sample area, indicating the microstructure of each area, and then the steel alloy sample element concentration is mapped onto the corresponding surface image to show the element distribution in that area, further revealing the segregation area in the alloy. After the mapping is completed, a set of segregation detection results can be obtained, which includes the type of each area, the concentration change and distribution of elements, providing an important data basis for subsequent aggregation analysis and low-temperature toughness testing.

[0031] The segregation detection result set is subjected to two-item distance transformation, and the same type of aggregation is performed to obtain a plurality of aggregated steel alloy sample sets.

[0032] Specifically, after obtaining the segregation detection result set, first, the segregation detection result set is subjected to two-item distance transformation from the two dimensions of segregation distribution and element concentration to determine a first transformed segregation detection result set. Then, taking the currently analyzed segregation detection result as an aggregation reference, the first transformed segregation detection result set is subjected to aggregation identification according to a preset aggregation scale, and regions with a similarity to the reference exceeding a preset threshold are included in the same aggregation set. Finally, the same operation is performed on other segregation detection results in the segregation detection result set, and all obtained aggregation sets are integrated to obtain a plurality of aggregated steel alloy sample sets. These aggregated steel alloy sample sets provide more uniform and representative sample data for subsequent low-temperature impact toughness tests, ensuring the accuracy and reliability of the test results.

[0033] In a possible implementation, the segregation detection result set is subjected to two-item distance transformation, and the same type of aggregation is performed to obtain a plurality of aggregated steel alloy sample sets, including:

[0034] A first segregation detection result is extracted from the segregation detection result set, wherein the first segregation detection result is any one of the segregation detection results in the segregation detection result set; the segregation detection result set is subjected to two-item distance transformation from the two dimensions of segregation distribution and element concentration, taking the first segregation detection result as a reference, to obtain a first transformed segregation detection result set; the first transformed segregation detection result set is subjected to identification according to a preset aggregation scale, taking the first segregation detection result as an aggregation reference, and a first aggregated steel alloy sample set obtained by identification is added to the plurality of aggregated steel alloy sample sets.

[0035] Specifically, one segregation detection result is randomly selected from the segregation detection result set as a first segregation detection result, and this first segregation detection result is taken as a starting point of transformation and aggregation. Subsequently, according to the segregation distribution characteristics and element concentration information of the first segregation detection result, the other segregation detection results are compared with the selected first segregation detection result, the similarity in the dimensions of segregation distribution and element concentration is calculated, and then the segregation detection results are subjected to two-item distance transformation according to the calculated similarity, so as to adjust the spatial position and element concentration, ensure that the segregation detection results similar to the first segregation detection result are relatively closer, and the segregation detection results with large differences are relatively farther away, so as to increase the distinction degree, thereby obtaining a first transformed segregation detection result set. Then, a preset aggregation scale is used to aggregate the first transformed segregation detection result set of the same kind, that is, the two-item distance coefficient of all segregation regions in the first transformed segregation detection result set and the reference region is compared with the preset aggregation scale, if the coefficient is less than the preset aggregation scale, the sample is classified into the same kind, aggregated, and the first aggregated steel alloy sample set obtained after aggregation is added to the multiple aggregated steel alloy sample sets. These aggregated steel alloy sample sets represent the aggregation results of different segregation characteristic regions, each aggregated sample set has similar composition distribution and segregation characteristics, and can provide more accurate test samples for subsequent low-temperature impact toughness test, so as to ensure that more consistent and more representative test results are obtained in the low-temperature impact test.

[0036] In a possible implementation, from the two dimensions of segregation distribution and element concentration, the two-item distance transformation is performed on the segregation detection result set based on the first segregation detection result, to obtain a first transformed segregation detection result set, including:

[0037] The segregation region identifier is used to perform binary conversion on the steel alloy sample surface image set in the segregation detection result set, to obtain a steel alloy sample segregation region binary image set; from the two dimensions of segregation distribution and element concentration, the similarity of the steel alloy sample segregation region binary image set and the steel alloy sample element concentration set in the segregation detection result set is identified based on the first steel alloy sample segregation region binary image corresponding to the first segregation detection result and the first steel alloy sample element concentration as a reference, to obtain a segregation detection result similarity set; and the two-item distance transformation is performed on the segregation detection result set based on the segregation detection result similarity set, to obtain the first transformed segregation detection result set.

[0038] Specifically, in order to accurately identify and aggregate the segregation area in the steel alloy sample, first, the segregation area identifier is used to perform binary conversion of the segregation area for the surface image of each steel alloy sample in the segregation detection result set, that is, the segregation separation threshold value in the segregation area identifier is compared with the element concentration of each area, if the element concentration of an area in the image exceeds the segregation separation threshold value, the area is identified as a segregation area, at this time, the pixel value of the area is set to 1, and the pixel value of other areas (areas with low or uniform element concentration) meeting the segregation separation threshold value requirement is set to 0. After comparing all areas of all images, a set of binary images of the segregation area of the steel alloy sample can be obtained, which is convenient for subsequent analysis and processing. Subsequently, starting from the two dimensions of segregation distribution and element concentration, the first steel alloy sample segregation area binary image corresponding to the first segregation detection result and the element concentration of the first steel alloy sample are used as the reference to identify the similarity of all samples in the segregation detection result set. In this process, the similarity in the segregation distribution and the similarity in the element concentration are obtained by the cosine similarity calculation formula, and the two similarities are weighted to obtain the segregation detection result similarity of each sample to the reference sample, and form a segregation detection result similarity set to reflect the similarity of all samples to the reference sample. Then, according to the segregation detection result similarity set, the two-item distance transformation coefficient between each sample and the reference sample is calculated, and then the calculated two-item distance transformation coefficient is adjusted in the direction away from the first segregation detection result, so that the samples with higher similarity are closer in space, and the samples with lower similarity are farther away from the reference sample in space, thereby obtaining a first transformed segregation detection result set, which provides a more accurate sample set for subsequent aggregation analysis.

[0039] In a possible implementation, the method comprises:

[0040] The cosine similarity calculation formula is used to calculate the similarity of the set of binary images of the segregation area of the steel alloy sample and the set of element concentrations of the steel alloy sample in the segregation detection result set and the binary image of the segregation area of the first steel alloy sample and the element concentration of the first steel alloy sample, to obtain a set of image similarity and a set of element concentration similarity; and the set of image similarity and the set of element concentration similarity are mapped and weighted to obtain the set of segregation detection result similarity.

[0041] Specifically, in the similarity recognition, first, each steel alloy sample segregation area binary image in the segregation detection result set is converted into a binary image vector with a length of n, where n is the total number of pixels of the steel alloy sample segregation area binary image, and each pixel value (0 or 1) in the image corresponds to an element in the vector, and then the first steel alloy sample segregation area binary image is converted into a first binary image vector. Subsequently, the element concentration of each steel alloy sample in the segregation detection result set is converted into a concentration vector, and each value in the concentration vector represents the concentration of an element, and then the element concentration of the first steel alloy sample is converted into a first concentration vector. Then, the similarity of the first binary image vector and each binary image vector is calculated using the cosine similarity calculation formula, and the similarity of the first concentration vector and each concentration vector is calculated, forming an image similarity set and an element concentration similarity set. Then, for the same segregation detection result, the image similarity and the element concentration similarity of the segregation detection result are mapped to a preset weight, and the image similarity and the element concentration are weighted and summed based on the mapped weight, to obtain the segregation detection result similarity of each segregation detection result. By integrating these segregation detection result similarities, a final segregation detection result similarity set is constructed, which provides strong support for subsequent aggregation analysis and low-temperature impact toughness testing.

[0042] In a possible implementation, the segregation detection result set is subjected to a two-item distance transformation based on the segregation detection result similarity set, to obtain the first transformed segregation detection result set, including:

[0043] The ratio of any one segregation detection result similarity in the segregation detection result similarity set to the sum of the similarities in the segregation detection result similarity set is calculated, and the difference between the ratio and 1 is taken as a two-item distance transformation coefficient to obtain a two-item distance transformation coefficient set; and the segregation detection result set is adjusted in a direction away from the first segregation detection result based on the two-item distance transformation coefficient set, to obtain the first transformed segregation detection result set.

[0044] Specifically, after obtaining the segregation detection result similarity set, the ratio between each segregation detection result similarity and the sum of all similarities in the segregation detection result similarity set is calculated, and 1 is subtracted from the calculated ratio to obtain the double distance transformation coefficient of each segregation detection result, forming a double distance transformation coefficient set, which is used to adjust the position of the corresponding segregation detection result. Then, the segregation detection result set is adjusted in the direction away from the first segregation detection result, and the product of the double distance transformation coefficient and the reference offset is used as the adjustment amplitude. In this process, for the double distance transformation coefficient exceeding the preset distance, the adjustment amplitude is directly added to the segregation detection result, and conversely, the segregation detection result is directly subtracted by the adjustment amplitude, so that the spatial relationship between the samples is more clear, and the samples with high similarity remain closer in space. After the above adjustment, each sample in the segregation detection result set is adjusted in space according to the double distance transformation coefficient, thereby forming a first transformed segregation detection result set. The samples in the first transformed segregation detection result set are rearranged in space, the distance relationship between the samples is more reasonable, and the spatial distribution of the samples is consistent with their segregation characteristics and similarity, providing a more reasonable data structure for subsequent aggregation analysis and testing.

[0045] The plurality of aggregated steel alloy sample sets are placed in a liquid nitrogen cooling device for 1 hour to obtain a plurality of aggregated cooled steel alloy sample sets, and a toughness test is performed on the plurality of aggregated cooled steel alloy sample sets using a low-temperature impact testing machine to obtain a plurality of aggregated toughness test result sets.

[0046] Specifically, after determining the multiple aggregated steel alloy sample sets, the aggregated steel alloy samples corresponding to the multiple aggregated steel alloy sample sets are placed in a liquid nitrogen cooling device in turn. The temperature of the liquid nitrogen is about -196°C, which can rapidly cool the steel alloy samples to an extremely low temperature. During the cooling process, the liquid nitrogen rapidly lowers the temperature of the steel alloy samples by contacting the surface of the steel alloy samples. To ensure the stability and uniformity of the cooling process, each sample needs to be kept stationary in the liquid nitrogen cooling device for at least 1 hour to ensure that the sample reaches a low temperature state and avoids any temperature gradient. The multiple steel alloy samples processed by the liquid nitrogen cooling device constitute multiple aggregated cooled steel alloy sample sets. These cooled samples can simulate the toughness performance of steel alloys at low temperatures in actual use environments and provide representative experimental samples for subsequent testing. Subsequently, the cooled steel alloy samples corresponding to the multiple aggregated cooled steel alloy sample sets are taken out of the liquid nitrogen cooling device and immediately subjected to impact toughness testing using a low-temperature impact testing machine. The low-temperature impact testing machine can perform standardized impact testing on the samples under low-temperature conditions. During the testing process, the deformation amount and impact absorbed energy of the samples under impact load are accurately recorded, forming multiple aggregated toughness test result sets. Each toughness test result in the multiple aggregated toughness test result sets reflects the toughness performance of different aggregated cooled steel alloy samples under low-temperature conditions, including impact work value, fracture mode, and energy absorption performance, which can provide necessary data support for subsequent material performance optimization, quality control, and safety evaluation in low-temperature environments, and help further analyze the applicability of steel alloys at extremely low temperatures.

[0047] Performing same-class longitudinal analysis and different-class transverse analysis on the multiple aggregated toughness test result sets to obtain the target toughness test result of the steel alloy set to be sampled.

[0048] Specifically, after obtaining the multiple aggregated toughness test result sets, first, the toughness test results of each aggregated steel alloy sample are subjected to same-class longitudinal analysis, i.e., the test results in each aggregated steel alloy sample set are compared to identify the performance trend within the same class or the same aggregation range, obtaining multiple same-class aggregated toughness test results. Subsequently, the toughness test results of the multiple aggregated toughness test result sets are subjected to different-class transverse analysis, i.e., the test results from different aggregated sample sets are compared to identify the performance differences between different steel alloy types or different production batches, obtaining multiple different-class aggregated toughness test results. Then, the multiple same-class aggregated toughness test results and the multiple different-class aggregated toughness test results are subjected to weighted calculation to obtain the final target toughness test result of the steel alloy set to be sampled. This target toughness test result not only reflects the impact resistance performance of steel alloys under low-temperature conditions, but also provides a scientific basis for subsequent production process optimization and quality control.

[0049] In a possible implementation, the same-class longitudinal analysis and the different-class transverse analysis are performed on the plurality of aggregated toughness test result sets to obtain the target toughness test result of the steel alloy to be sampled, including:

[0050] The same-class longitudinal analysis is performed on the plurality of aggregated toughness test result sets to obtain a plurality of same-class aggregated toughness test results; one aggregated toughness test result is randomly extracted from each of the plurality of aggregated toughness test result sets without replacement, the plurality of different-class aggregated toughness test result sets extracted by multiple times are subjected to mean shift screening to obtain a plurality of different-class aggregated toughness test results; and the plurality of same-class aggregated toughness test results and the plurality of different-class aggregated toughness test results are subjected to weighted calculation to obtain the target toughness test result.

[0051] Specifically, after obtaining the plurality of aggregated toughness test result sets, in order to accurately evaluate the target toughness test result of the steel alloy to be sampled, first, the same-class longitudinal analysis is performed on the toughness test result sets of the plurality of aggregated steel alloy samples, in this process, the mean value of the test results in each aggregated steel alloy sample set is calculated to obtain the toughness performance of the sample under low-temperature impact, and then the mean shift algorithm is used for iteration to obtain a plurality of same-class aggregated toughness test results, wherein the mean shift algorithm is a density-based non-parametric estimation method and can be used to identify the clustering pattern in the data set. Subsequently, the different-class transverse analysis is performed on the toughness test result sets of the plurality of aggregated steel alloy samples, in this process, one test result is randomly extracted from each of the plurality of aggregated toughness test result sets without replacement, and through repeated extraction, a plurality of different-class aggregated toughness test result sets can be obtained. Then, the mean shift algorithm is used for iterative analysis on the extracted different-class test results to obtain a plurality of different-class aggregated toughness test results. After the same-class longitudinal analysis and the different-class transverse analysis are completed, the same-class aggregated toughness test results and the different-class aggregated toughness test results are subjected to weight distribution according to a preset weight, generally, the same-class aggregated toughness test results have a higher weight in the class group to which they belong, and the weight of the different-class aggregated toughness test results is adjusted according to their contribution to the overall result. Finally, the plurality of same-class aggregated toughness test results and the plurality of different-class aggregated toughness test results are subjected to weighted fusion according to the distributed weights to obtain the target toughness test result, which can accurately reflect the comprehensive performance of the steel alloy to be sampled under low-temperature impact, and provide data support for subsequent quality control, production process optimization and performance verification.

[0052] In a possible implementation, the same-class longitudinal analysis is performed on the plurality of aggregated toughness test result sets to obtain a plurality of same-class aggregated toughness test results, including:

[0053] The mean value calculation is performed on each of the plurality of sets of aggregate toughness test results to determine a plurality of aggregate toughness test result means; the plurality of aggregate toughness test result means are taken as initial mean shift centers, and the plurality of sets of aggregate toughness test results are iterated according to a preset mean shift bandwidth until a preset iteration number is met, to obtain the plurality of homogeneous aggregate toughness test results.

[0054] Specifically, in the homogeneous longitudinal analysis of the plurality of sets of aggregate toughness test results, first, the mean value calculation is performed on each set of aggregate toughness test results to understand the average toughness of each aggregate steel alloy sample under low-temperature impact test, thereby obtaining a plurality of aggregate toughness test result means. Subsequently, the plurality of aggregate toughness test result means obtained are taken as initial mean shift centers for subsequent iterative adjustment, and a mean shift bandwidth set in advance according to actual business requirements is obtained again. The preset mean shift bandwidth is used to control the range of each iteration, and determines the search distance of each center point in each iteration. The greater the preset mean shift bandwidth, the greater the search range allowed in each iteration; the smaller the preset mean shift bandwidth, the narrower the search range. Then, the Euclidean distance is used to calculate the distance between each aggregate toughness test result and the current center point through the current mean shift center point, and all test results within the search bandwidth are searched according to the preset mean shift bandwidth, and the average of these results is calculated as a new mean shift center. This process is iterated until the aggregate results no longer change significantly (the difference with the result of the previous iteration is less than a preset threshold) or the preset iteration number is reached. After the iterative processing of the mean shift algorithm, the plurality of aggregate toughness test results finally obtained are the plurality of homogeneous aggregate toughness test results, which represent the toughness performance of the homogeneous steel alloy samples under low-temperature impact, and can help to more accurately evaluate and compare the performance of different aggregate steel alloy samples.

[0055] In the foregoing, with reference to Figure 1 The toughness test method for low-temperature impact of steel alloy according to the embodiments of the present application is described in detail. Next, with reference to Figure 2 The toughness test system for low-temperature impact of steel alloy according to the embodiments of the present application is described.

[0056] The toughness test system for low-temperature impact of steel alloy according to the embodiment of the present application is used to solve the technical problem that the test result deviation is large and the process defects cannot be accurately positioned due to sample segregation and smelting process difference in the low-temperature impact toughness test of steel alloy, so as to achieve the technical effect of accurately evaluating the low-temperature toughness of different steel alloys through segregation detection and aggregated sample analysis, and improving the stability and reliability of the low-temperature impact toughness test. The toughness test system for low-temperature impact of steel alloy comprises: a uniform sampling module 11, a segregation detection module 12, a two-item distance transformation module 13, a toughness test module 14, and a test result analysis module 15.

[0057] The uniform sampling module 11: acquires a set of steel alloy to be inspected, uniformly samples the set of steel alloy according to a production timestamp of the steel alloy, and obtains a set of steel alloy samples; the segregation detection module 12: respectively performs segregation detection on the set of steel alloy samples by using a scanning electron microscope, and determines a set of segregation detection results; the two-item distance transformation module 13: performs two-item distance transformation on the set of segregation detection results, and obtains a plurality of sets of aggregated steel alloy samples through homogenous aggregation; the toughness test module 14: respectively places the plurality of sets of aggregated steel alloy samples in a liquid nitrogen cooling device, and stands for 1 hour to obtain a plurality of sets of aggregated cooled steel alloy samples, and performs toughness test on the plurality of sets of aggregated cooled steel alloy samples by using a low-temperature impact testing machine to obtain a plurality of sets of aggregated toughness test results; and the test result analysis module 15: respectively performs homogenous longitudinal analysis and heterogeneous transverse analysis on the plurality of sets of aggregated toughness test results, and obtains a target toughness test result of the set of steel alloy to be inspected.

[0058] In the following, the specific configuration of the segregation detection module 12 will be described in detail. As described above, the set of steel alloy samples is respectively subjected to segregation detection by using a scanning electron microscope to determine a set of segregation detection results, and the segregation detection module 12 can further comprise: configuring the acceleration voltage, working distance and scanning mode of the scanning electron microscope; respectively placing the set of steel alloy samples in the scanning cavity of the scanning electron microscope for electron beam scanning to obtain a set of steel alloy sample surface images; identifying the composition of the set of steel alloy samples by using energy dispersive X-ray spectroscopy to obtain a set of steel alloy sample element concentrations; and performing one-to-one mapping on the set of steel alloy sample surface images and the set of steel alloy sample element concentrations to determine the set of segregation detection results.

[0059] The specific configuration of the two-item distance transformation module 13 will be described in detail below. As described above, the two-item distance transformation is performed on the segregation detection result set, and the same type aggregation is performed to obtain a plurality of aggregated steel alloy sample sets, and the two-item distance transformation module 13 can further include: extracting a first segregation detection result from the segregation detection result set, wherein the first segregation detection result is any one of the segregation detection result set; performing two-item distance transformation on the segregation detection result set from the two dimensions of segregation distribution and element concentration, taking the first segregation detection result as a reference, to obtain a first transformed segregation detection result set; taking the first segregation detection result as an aggregation reference, identifying the first transformed segregation detection result set according to a preset aggregation scale, and adding a first aggregated steel alloy sample set obtained by identification to the plurality of aggregated steel alloy sample sets.

[0060] Wherein, performing two-item distance transformation on the segregation detection result set from the two dimensions of segregation distribution and element concentration, taking the first segregation detection result as a reference, to obtain a first transformed segregation detection result set, the two-item distance transformation module 13 can further include: performing segregation region binary conversion on the steel alloy sample surface image set in the segregation detection result set by using a segregation region identifier to obtain a steel alloy sample segregation region binary image set; from the two dimensions of segregation distribution and element concentration, according to the first steel alloy sample segregation region binary image corresponding to the first segregation detection result and the first steel alloy sample element concentration as a reference, performing similarity identification on the steel alloy sample segregation region binary image set and the steel alloy sample element concentration set in the segregation detection result set to obtain a segregation detection result similarity set; performing two-item distance transformation on the segregation detection result set based on the segregation detection result similarity set to obtain the first transformed segregation detection result set.

[0061] Wherein, the two-item distance transformation module 13 can further include: calculating the similarity of the steel alloy sample segregation region binary image set and the steel alloy sample element concentration set in the segregation detection result set with the first steel alloy sample segregation region binary image and the first steel alloy sample element concentration respectively by using a cosine similarity calculation formula to obtain an image similarity set and an element concentration similarity set; mapping and weighting the image similarity set and the element concentration similarity set to obtain the segregation detection result similarity set.

[0062] The double-item distance transformation module 13 can further include: calculating the ratio of any one segregation detection result similarity in the segregation detection result similarity set to the sum of the segregation detection result similarities, and taking the difference between the ratio and 1 as a double-item distance transformation coefficient to obtain a double-item distance transformation coefficient set; and adjusting the segregation detection result set in a direction away from the first segregation detection result based on the double-item distance transformation coefficient set to obtain the first transformed segregation detection result set.

[0063] The specific configuration of the test result analysis module 15 will be described in detail below. As described above, the same-class longitudinal analysis and different-class transverse analysis are performed on the plurality of aggregated toughness test result sets to obtain the target toughness test result of the steel alloy to be sampled, and the test result analysis module 15 can further include: performing same-class longitudinal analysis within the set on the plurality of aggregated toughness test result sets to obtain a plurality of same-class aggregated toughness test results; randomly sampling one aggregated toughness test result from each of the plurality of aggregated toughness test result sets without replacement, performing mean shift screening on the plurality of different-class aggregated toughness test result sets sampled multiple times to obtain a plurality of different-class aggregated toughness test results; and performing weighted calculation on the plurality of same-class aggregated toughness test results and the plurality of different-class aggregated toughness test results to obtain the target toughness test result.

[0064] The test result analysis module 15 can further include: performing mean value calculation on the plurality of aggregated toughness test result sets respectively to determine a plurality of aggregated toughness test result means; and taking the plurality of aggregated toughness test result means as initial mean shift centers, and iteratively performing the plurality of aggregated toughness test result sets according to a preset mean shift bandwidth until a preset iteration number is met to obtain the plurality of same-class aggregated toughness test results.

[0065] The toughness test system for low-temperature impact of steel alloy provided in the embodiments of the present application can perform the toughness test method for low-temperature impact of steel alloy provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0066] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, the various units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific name of each functional unit is only for the convenience of mutual differentiation, and does not limit the protection scope of the present application.

[0067] Based on the foregoing embodiments, the embodiments of the present application also provide an electronic device. Figure 3 is a structural schematic diagram of the electronic device provided by the embodiments of the present application, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present application. Figure 3 The electronic device shown is only an example, and should not bring any limitation to the function and use range of the embodiments of the present application, which is in the form of a general computing device, and its components can include but are not limited to a processor 21, a memory 22, an input device 23, and an output device 24. Among them, the processor 21 can be one or more; the memory 22 can include a computer readable medium and at least one program product, the program product has a set of (at least one) program modules, which are configured to perform the functions of the embodiments of the present application.

[0068] The memory 22 shown in the embodiments of the present application can adopt any combination of one or more computer readable media; the computer readable storage medium can be but is not limited to an infrared ray, a semiconductor system, a device or a component, or any combination of the above, for storing software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the toughness test method for steel alloy low temperature impact in the embodiments of the present application, and the processor 21 performs various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 22, that is, realizes the toughness test method for steel alloy low temperature impact.

[0069] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application. In some cases, the actions or steps described in the present application can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

Claims

1. A method for testing the toughness of steel alloys at low temperatures by impact, characterized in that, The method comprises: acquiring a steel alloy set to be inspected, uniformly sampling the steel alloy set according to a production timestamp of the steel alloy, and obtaining a steel alloy sample set; performing segregation detection on the steel alloy sample set by using a scanning electron microscope, and determining a segregation detection result set; performing two-item distance transformation on the segregation detection result set, and performing same-type aggregation to obtain a plurality of aggregated steel alloy sample sets; placing the plurality of aggregated steel alloy sample sets in a liquid nitrogen cooling device respectively, and standing for 1 hour to obtain a plurality of aggregated cooled steel alloy sample sets, and performing toughness testing on the plurality of aggregated cooled steel alloy sample sets by using a low-temperature impact testing machine to obtain a plurality of aggregated toughness test result sets; performing same-type longitudinal analysis and different-type horizontal analysis on the plurality of aggregated toughness test result sets to obtain a target toughness test result of the steel alloy set to be inspected.

2. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 1, characterized in that, The method comprises: configuring an acceleration voltage, a working distance and a scanning mode of the scanning electron microscope; placing the steel alloy sample set in a scanning cavity of the scanning electron microscope for electron beam scanning to obtain a steel alloy sample surface image set; identifying the composition of the steel alloy sample set by using energy dispersive X-ray spectroscopy to obtain a steel alloy sample element concentration set; mapping the steel alloy sample surface image set and the steel alloy sample element concentration set one by one to determine the segregation detection result set.

3. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 2, characterized in that, The method comprises: extracting a first segregation detection result from the segregation detection result set, wherein the first segregation detection result is any one of the segregation detection results in the segregation detection result set; performing two-item distance transformation on the segregation detection result set based on the first segregation detection result from the segregation distribution and element concentration dimensions to obtain a first transformed segregation detection result set; identifying the first transformed segregation detection result set according to a preset aggregation scale based on the first segregation detection result, and adding a first aggregated steel alloy sample set obtained by the identification to the plurality of aggregated steel alloy sample sets.

4. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 3, characterized in that, The method comprises: performing segregation area binary conversion on the steel alloy sample surface image set in the segregation detection result set by using a segregation area identifier to obtain a steel alloy sample segregation area binary image set; performing similarity identification on the steel alloy sample segregation area binary image set and the steel alloy sample element concentration set in the segregation detection result set based on the first steel alloy sample segregation area binary image corresponding to the first segregation detection result and the first steel alloy sample element concentration as the reference from the segregation distribution and element concentration dimensions to obtain a segregation detection result similarity set; and Performing bivariate distance transformation on the segregation detection result set based on the segregation detection result similarity set to obtain the first transformed segregation detection result set.

5. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 4, characterized in that, It comprises: Using the cosine similarity calculation formula, the similarity of the segregation detection result set, the steel alloy sample segregation area binary image set, and the steel alloy sample element concentration set is calculated respectively with the first steel alloy sample segregation area binary image and the first steel alloy sample element concentration, and the image similarity set and the element concentration similarity set are obtained. Map and weight the image similarity set and the element concentration similarity set to obtain the segregation detection result similarity set.

6. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 5, characterized in that, Based on the segregation detection result similarity set, the segregation detection result set is transformed by bivariate distance transformation to obtain the first transformed segregation detection result set, which comprises: Calculate the ratio of any one segregation detection result similarity in the segregation detection result similarity set to the sum of the similarity in the segregation detection result similarity set, and take the difference between the ratio and 1 as the bivariate distance transformation coefficient to obtain the bivariate distance transformation coefficient set. Based on the bivariate distance transformation coefficient set, adjust the segregation detection result set in the direction away from the first segregation detection result to obtain the first transformed segregation detection result set.

7. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 1, characterized in that, Performing same kind longitudinal analysis and different kind transverse analysis on the multiple aggregated toughness test result sets to obtain the target toughness test result of the steel alloy set to be sampled, comprising: Performing same kind longitudinal analysis on the multiple aggregated toughness test result sets to obtain multiple same kind aggregated toughness test results; Randomly extract one aggregated toughness test result from multiple aggregated toughness test result sets without replacement, and perform mean shift filtering on multiple different kind aggregated toughness test result sets to obtain multiple different kind aggregated toughness test results; Weighting the multiple same kind aggregated toughness test results and the multiple different kind aggregated toughness test results to obtain the target toughness test result.

8. The method for testing the toughness of steel alloys at low temperatures by impact as claimed in claim 7, characterized in that, Performing same kind longitudinal analysis on the multiple aggregated toughness test result sets to obtain multiple same kind aggregated toughness test results, comprising: Performing mean value calculation on the multiple aggregated toughness test result sets to determine multiple aggregated toughness test result means; Taking the multiple aggregated toughness test result means as the initial mean shift center, and iterating the multiple aggregated toughness test result sets according to the preset mean shift bandwidth until the preset iteration number is met to obtain the multiple same kind aggregated toughness test results.

9. A system for testing the toughness of steel alloys at low temperatures, characterized in that, The system is used to implement the toughness test method for steel alloy low temperature impact according to any one of claims 1-8, and the system comprises: Uniform sampling module: obtaining a steel alloy set to be sampled, and performing uniform sampling on the steel alloy set according to the production time stamp of the steel alloy to obtain a steel alloy sample set; Segregation detection module: performing segregation detection on the steel alloy sample set by using a scanning electron microscope to determine a segregation detection result set; Bivariate distance transformation module: toughness test; The toughness test module: the plurality of polymerization steel alloy sample sets are respectively placed in a liquid nitrogen cooling device, and are left for 1 hour to obtain a plurality of polymerization cooling steel alloy sample sets, and a low-temperature impact testing machine is used to test the toughness of the plurality of polymerization cooling steel alloy sample sets, and a plurality of polymerization toughness test result sets are obtained; The test result analysis module: the plurality of polymerization toughness test result sets are analyzed longitudinally and transversely, and a target toughness test result of the steel alloy to be sampled is obtained.

10. An electronic device, comprising: The electronic device comprises: A memory for storing executable instructions; A processor for executing the executable instructions stored in the memory, and implementing the toughness test method for low-temperature impact of steel alloy according to any one of claims 1 to 8.

Images