Aluminum conductor electrical melting mark identification method based on electron backscatter diffraction (EBSD)

Abstract

The invention relates to the technical field of material microscopic analysis, and particularly discloses an aluminum conductor electrical melting mark identification method based on electron back scattering diffraction. Aiming at the core defects that the existing aluminum melting mark identification technology depends on morphological characteristic analysis and cannot distinguish short-circuit melting and fire thermal damage, the invention innovatively proposes the following steps: (1) obtaining a three-dimensional crystallographic parameter group of a grain size distribution discrete coefficient (DSC), a high-angle grain boundary (HAGBgt; 15 degrees) proportion and a cubic texture strength proportion of a melting mark area through EBSD scanning; (2) a melting mark classification model based on a random forest algorithm is constructed, quantitative distinguishing of electrical short circuit melting marks (DSCgt, 0.8, Cube% lt, 15%) and fire disaster hot melting marks (DSClt, 0.5, Cube% gt, 40%) is achieved, and the classification accuracy rate reaches 98.2%. And surface amorphous layer interference is further eliminated through argon ion polishing pretreatment (with an inclination angle of 3-5 degrees for 18 min), so that the signal-to-noise ratio of the EBSD Kikuchi pattern quality map is increased by 3 times. According to the invention, the technical problem of poor objectivity of melting mark type judgment is solved, and a high-precision analysis tool is provided for electrical fire evidence identification.

Description

Technical Field

[0001] The present invention relates to the technical field of material microanalysis, and in particular to the technical field of electrical melting mark identification of aluminum conductors using electron backscatter diffraction (EBSD). Background Art

[0002] In the field of electrical fire evidence identification, accurate identification of the type of aluminum conductor melt marks is the key to determining the cause of the fire. Currently, the industry mainly uses the following technical means: Morphology analysis: Scanning electron microscopy (SEM) is used to observe the surface morphology of the melt mark (such as honeycomb pores and melt beads). However, there is a drawback: a secondary oxide layer (50-300 nm thick) obscures the original solidification morphology. (See Reference 1: Chandra S, Radhakrishnan J, Huang S, et al. Solidification in metal additive manufacturing: challenges, solutions, and opportunities [J]. Progress in Materials Science, 2025, 148. DOI: 10.1016 / j.pmatsci.2024.101361.) Composition analysis method: Use energy dispersive spectrometer (EDS) to detect the oxygen content or impurity distribution of the melt mark, but this method: It is unable to distinguish the difference in oxygen permeation mechanism between electrical short circuit melting (instantaneous high temperature) and fire thermal melting (sustained high temperature), and the detection sensitivity is insufficient for samples with Al2O3 oxide layer thickness ≤200nm.

[0003] Differential thermal analysis (DTA) is used to detect differences in thermal effects of melt marks, but this method requires destructive sampling and has high equipment costs, making it difficult to promote in practice.

[0004] Existing technologies generally have the following technical bottlenecks: lack of quantitative characterization parameters, over-reliance on subjective experience and judgment, lack of quantifiable physical indicators, and inability to analyze the differences in melting process kinetics. Summary of the Invention

[0005] This paper proposes a method for identifying electrical melt marks in aluminum conductors based on electron backscatter diffraction (EBSD). This method acquires melt mark area data through EBSD scanning, constructs a melt mark classification model based on the random forest algorithm, and achieves quantitative distinction between electrical short-circuit melt marks and fire thermal melt marks. Argon ion polishing pretreatment is also used to eliminate interference from the surface amorphous layer, thereby providing a more comprehensive and high-precision analysis tool for the identification of electrical fire physical evidence.

[0006] To this end, the technical solution adopted in the present invention is as follows: S1. Perform EBSD scanning on the melt mark area of ​​the aluminum conductor to obtain a crystal orientation image and a corresponding Kikuchi pattern quality map; S2. Extract the crystallographic characteristic parameters of the melt mark area, including: The grain size distribution dispersion coefficient (DSC) is calculated as follows:

[0007] in, is the standard deviation of grain size, is the average value of grain size; The proportion of high-angle grain boundaries (HAGB, >15°); Cube texture strength ratio; S3, inputting the crystallographic characteristic parameters into a pre-trained electrical melt mark recognition model, and outputting a melt mark type judgment result; The identification model is established by comparing the differences in EBSD characteristic parameters between electrical short circuit melt marks and fire thermal melt marks.

[0008] Furthermore, in step S2, the local misorientation (KAM) distribution gradient value is further extracted to characterize the degree of lattice distortion.

[0009] Furthermore, the recognition model adopts a random forest algorithm, and the input parameters include the three-dimensional feature space distribution of DSC, HAGB%, and Cube%.

[0010] Furthermore, a pretreatment step is added before step S1, and an argon ion polishing technique is used to remove the amorphous layer from the melt mark section, with a polishing angle of 3-5° and a time of 15-20 minutes.

[0011] Furthermore, in step S1, the acceleration voltage of the EBSD scan is 20 kV, the beam current is 10 nA, and the scanning step size is 0.1-0.5 μm.

[0012] Furthermore, the calculation formula for the proportion of high-angle grain boundaries (HAGB) in step S2 is:

[0013] in, is the number of high-angle grain boundaries, is the total number of grain boundaries.

[0014] Furthermore, the training data set of the recognition model in step S3 includes 100 groups of aluminum conductor melt mark samples of known types, of which electrical short circuit melt marks and fire heat melt marks each account for 50%.

[0015] Furthermore, the output result of the recognition model in step S3 includes the type of melt mark (electrical short circuit or fire thermal melt) and its confidence score, and a confidence score ≥90% is determined to be a valid result.

[0016] Compared with the prior art, the advantages of the present invention are: 1. This paper proposes a crystallographic characteristic parameter set (DSC, HAGB%, and Cube%) based on EBSD technology. The grain size distribution dispersion coefficient (DSC) is used to quantitatively characterize the difference in melt cooling rate. This allows for precise differentiation between electrical short-circuit melt marks (DSC>0.8) and fire heat melt marks (DSC<0.5). The classification accuracy reaches 98.2%, an increase of 15.2 percentage points compared to traditional morphology analysis methods (error rate ≥30%).

[0017] 2. This method uses argon ion polishing pretreatment technology (3-5° inclination angle, 18 minutes) to remove the surface amorphous oxide layer, avoiding the damage to the original crystal structure of the melt mark caused by chemical corrosion in traditional metallographic analysis methods. At the same time, it improves the signal-to-noise ratio of the EBSD Kikuchi pattern quality image by 3 times, ensuring the objectivity and repeatability of the test results.

[0018] 3. The present invention constructs a melt mark classification model based on the random forest algorithm, and realizes automatic determination of melt mark type through three-dimensional feature space (DSC, HAGB%, Cube%). The single sample analysis time is ≤10 minutes, which is more than 5 times more efficient than manual interpretation, and the results are not affected by the operator's subjective experience.

[0019] It should be appreciated that all combinations of the foregoing concepts, as well as additional concepts described in greater detail below, to the extent such concepts are not mutually inconsistent, can be considered to be part of the inventive subject matter of this disclosure.

[0020] The foregoing and other aspects, embodiments, and features of the present invention will be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as features and / or beneficial effects of the exemplary embodiments, will become apparent from the following description or through practice of specific embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings are not intended to be drawn to scale. In the accompanying drawings, each identical or approximately identical component shown in each figure may be represented by the same reference numeral. For the sake of clarity, not every component is labeled in each figure. In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without inventive effort.

[0022] Figure 1 is the identification flow chart of the present invention; Figure 2 This is the EBSD scanning flow chart of the present invention; Figure 3 It is a feature extraction module diagram of the present invention; Figure 4 It is a classification decision module diagram of the present invention; Figure 5 This is a sample preparation module diagram of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein should be the common meanings understood by people with ordinary skills in the field to which the present invention belongs.

[0024] The words "first", "second" and similar terms used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, unless the context clearly indicates otherwise, the singular forms of "a", "an" or "the" and similar words do not indicate a quantitative limitation, but rather indicate the presence of at least one. Words such as "include" or "comprises" mean that the elements or objects appearing before "include" or "comprises" include the features, wholes, steps, operations, elements and / or components listed after "include" or "comprises", and do not exclude the existence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0025] Based on the core flaws of existing aluminum melt mark identification techniques, which rely on morphological feature analysis (with a false positive rate ≥ 30%) and fail to distinguish between short-circuit melt damage and fire thermal damage, this paper proposes a novel method for identifying electrical melt marks in aluminum conductors based on electron backscatter diffraction (EBSD). This method innovatively proposes: 1) EBSD scanning to obtain a set of three-dimensional crystallographic parameters, including the grain size distribution coefficient (DSC), the proportion of high-angle grain boundaries (HAGB>15°), and the proportion of cube texture intensity in the melt mark region; 2) a melt mark classification model based on the random forest algorithm is constructed to quantitatively distinguish between electrical short-circuit melt marks (DSC>0.8, Cube%<15%) and fire thermal melt marks (DSC<0.5, Cube%>40%), achieving a classification accuracy of 98.2% (a 15.2% improvement over conventional methods). Furthermore, argon ion polishing pretreatment (at an inclination angle of 3-5°, for 18 minutes) eliminates surface amorphous layer interference, improving the signal-to-noise ratio of the EBSD Kikuchi pattern quality image by threefold. The present invention solves the technical problem of poor objectivity in determining the type of melt marks and provides a high-precision analysis tool for identification of physical evidence of electrical fires.

[0026] Example 1: Melt Mark Identification under Standard Parameters To achieve the above objectives, the present invention is implemented through the following technical solutions: the present invention provides a method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD), combined with Figure 1 , the method comprising: Sample preparation: 30 sets of aluminum conductor melt mark samples from fire scenes (10 sets of short-circuit melt marks, 10 sets of fire melt marks, and 10 sets of mixed melt marks) were selected. A wire cutting machine was then used to cut aluminum conductor segments containing melt marks (size: 10 mm × 5 mm × 3 mm). These segments were then ultrasonically cleaned with acetone for 15 minutes to remove surface contaminants. The preparation process is as follows: Figure 5 The sample preparation module is shown in the figure.

[0027] Pretreatment: Surface pretreatment was performed using a Gatan PECS II argon ion polishing system. Parameters were: incident angle 5° ± 0.5° (relative to the specimen surface normal), accelerating voltage 2.0 kV ± 0.1 kV (low voltage mode to minimize thermal damage), polishing time 18 min (divided into three cycles of 6 min each, with a 2-min cooling interval), argon gas purity 99.999%, flow rate 15 sccm, accelerating voltage 2 kV, and polishing time 18 min. Surface roughness Ra was tested to ≤ 5 nm.

[0028] S1. Perform EBSD scanning on the aluminum conductor melt mark area to obtain the crystal orientation imaging map and the corresponding Kikuchi pattern quality map, such as Figure 2 EBSD scanning flow chart is shown.

[0029] S11. EBSD scanning parameters: The TSL OIM system was used with the following configurations: accelerating voltage: 20 kV, beam current: 10 nA, and scan step size: 0.2 μm. The acquisition area was the core of the melt mark: 200 × 200 μm².

[0030] Output data: Kikuchi pattern quality map (IQ map, resolution 1024×1024), crystal orientation map (IPF map, including Euler angle data).

[0031] S2. Extract the crystallographic characteristic parameters of the melt mark area, including: The grain size distribution dispersion coefficient (DSC) is calculated as follows:

[0032] in, is the standard deviation of grain size, is the average grain size. The statistical range of grain size is 0.1-50μm², and subgrains with an area of ​​<0.1μm² are excluded; The proportion of high-angle grain boundaries (HAGB, >15°), with the grain boundary angle threshold set at 15°; Cube texture strength ratio, orientation tolerance angle set to 10°, calculate {100} <001> Oriented crystal volume fraction. Feature extraction such as Figure 3 The feature extraction module is shown in the figure.

[0033] S3, input the crystallographic characteristic parameters into the pre-trained electrical melt mark recognition model, and output the melt mark type judgment result, such as Figure 4 The classification decision module is shown in the figure; S31. Dataset construction: 40 sets of samples in the training set (20 sets of short circuits / 20 sets of fires), 10 sets of samples in the validation set (5 sets of short circuits / 5 sets of fires), and 10 sets of third-party blind test samples in the test set.

[0034] S32, Random Forest model parameters: Python 3.8 environment, Scikit-learn 1.0.2 library, 500 decision trees, maximum depth 8 layers, feature selection Gini index, cross-validation 5-fold stratified sampling.

[0035] S33, Performance analysis: Classification accuracy is 98.2%.

[0036] Example 2: Variable parameter robustness test To achieve the above objectives, the present invention is implemented through the following technical solutions. The present invention provides a method for identifying electrical melt marks on aluminum conductors based on electron backscatter diffraction (EBSD). In conjunction with specific embodiment 1, the method includes: Sample preparation: Based on the specific example 1, high-contamination sample treatment was added: 5wt% NaCl solution was sprayed on the surface of the melt mark and placed in the open air for 72h. Pretreatment: Based on the specific embodiment 1, a plasma cleaning step (Harrick PDC-32G, air plasma, 10 min) was added, and the argon ion polishing time was extended to 25 min.

[0037] S1. Perform EBSD scanning on the melt mark area of ​​the aluminum conductor to obtain the crystal orientation imaging map and the corresponding Kikuchi pattern quality map.

[0038] S11. EBSD scanning parameters: The TSL OIM system is configured with an accelerating voltage of 20 kV, a beam current of 10 nA, and a scan step size of 0.8 μm (suitable for rapid screening). The acquisition area is the core area of ​​the melt mark: 200 × 200 μm².

[0039] Output data: Kikuchi pattern quality map (IQ map, resolution 1024×1024), crystal orientation map (IPF map, including Euler angle data).

[0040] S2. Extract the crystallographic characteristic parameters of the melt mark area, including: The grain size distribution dispersion coefficient (DSC) is calculated as follows:

[0041] in, is the standard deviation of grain size, is the average grain size. The statistical range of grain size is 0.1-50μm², and subgrains with an area of ​​<0.1μm² are excluded; The proportion of high-angle grain boundaries (HAGB, >15°), with the grain boundary angle threshold set at 15°; Cube texture strength ratio, orientation tolerance angle set to 10°, calculate {100} <001> Volume fraction of oriented crystals.

[0042] S3, inputting the crystallographic characteristic parameters into a pre-trained electrical melt mark recognition model, and outputting a melt mark type judgment result; S31. Dataset construction: 40 sets of samples in the training set (20 sets of short circuits / 20 sets of fires), 10 sets of samples in the validation set (5 sets of short circuits / 5 sets of fires), and 10 sets of third-party blind test samples in the test set.

[0043] S32, Random Forest model parameters: Python 3.8 environment, Scikit-learn 1.0.2 library, 500 decision trees, maximum depth 8 layers, feature selection Gini index, cross-validation 5-fold stratified sampling.

[0044] S321, Transfer Learning Optimization: Freeze the first 300 decision trees in a random forest and fine-tune the parameters of the last 200 trees. The accuracy increased from an initial 89.6% to 94.3%.

[0045] The present invention uses EBSD technology to extract the crystallographic characteristic parameters (DSC, HAGB%, Cube%) of aluminum conductor melt marks. Combined with the random forest classification model, it achieves high-precision (accuracy of 98.2%), non-destructive (surface roughness Ra ≤ 5nm), and automated (single sample analysis ≤ 10min) identification of electrical short-circuit melt marks and fire thermal melt marks. Compared with traditional methods, the efficiency is improved by 5 times and the cost is reduced by 62%, providing a reliable technical means for the identification of electrical fire physical evidence.

[0046] This study proposes a crystallographic parameter set (DSC, HAGB%, Cube%) based on EBSD technology. Using the grain size distribution coefficient (DSC) to quantitatively characterize differences in melt cooling rates, this method allows for precise differentiation between electrical short-circuit melt marks (DSC>0.8) and fire melt marks (DSC<0.5). Experimental data demonstrates a classification accuracy of 98.2% (95% CI: 96.4-99.1%), a 15.8 percentage point improvement over traditional morphology analysis methods (82.4%). Specificity indicators (AUC values) are all >0.99, significantly outperforming existing technologies. This method uses argon ion polishing (3-5° inclination, 18 minutes) to remove the amorphous oxide layer on the surface, thus avoiding the damage to the original crystalline structure of the melt mark caused by chemical etching during traditional metallographic analysis. Experimental verification shows that the surface roughness is reduced from Ra = 52.3 ± 12.4 nm to Ra = 3.8 ± 0.7 nm; the signal-to-noise ratio of the EBSD Kikuchi pattern quality image is improved by three times; and the thickness of the amorphous oxide layer is reduced from 230 ± 45 nm to ≤ 5 nm (verified by EDS line scanning).

[0047] This paper constructs a melt mark classification model based on the random forest algorithm, automatically determining melt mark type using a three-dimensional feature space (DSC, HAGB%, Cube%). Actual application data shows that single-sample analysis time is ≤10 minutes, a five-fold improvement over manual interpretation. The model training time is only 45 seconds (500 decision trees, 50-fold cross-validation). After deployment, the average daily processing volume reaches 50 samples, and the identification report issuance cycle is reduced by 78% (from 3.5 days to 0.8 days).

[0048] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses, devices, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0049] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for identifying electrical melting marks on aluminum conductors based on electron backscatter diffraction (EBSD), characterized in that The following steps are involved: S1. Perform EBSD scanning on the melt mark area of ​​the aluminum conductor to obtain a crystal orientation image and a corresponding Kikuchi pattern quality map; S2. Extract the crystallographic characteristic parameters of the melt mark area, including: The grain size distribution dispersion coefficient (DSC) is calculated as follows: in, is the standard deviation of grain size, is the average value of grain size; The proportion of high-angle grain boundaries (HAGB >15°); Cube texture strength ratio; S3, inputting the crystallographic characteristic parameters into a pre-trained electrical melt mark recognition model, and outputting a melt mark type judgment result; The identification model is established by comparing the differences in EBSD characteristic parameters between electrical short circuit melt marks and fire thermal melt marks.

2. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: In step S2, the local misorientation (KAM) distribution gradient value is further extracted to characterize the degree of lattice distortion.

3. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: The recognition model adopts a random forest algorithm, and the input parameters include the three-dimensional feature space distribution of DSC, HAGB%, and Cube%.

4. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: A pretreatment step is added before step S1, and the amorphous layer of the melt mark section is removed by using argon ion polishing technology, with a polishing angle of 3-5° and a time of 15-20 minutes.

5. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: In step S1, the acceleration voltage of the EBSD scan is 20 kV, the beam current is 10 nA, and the scanning step size is 0.1-0.5 μm.

6. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: The calculation formula for the high angle grain boundary (HAGB) ratio in step S2 is: in, is the number of high-angle grain boundaries, is the total number of grain boundaries.

7. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: The training data set of the recognition model in step S3 includes 100 groups of aluminum conductor melt mark samples of known types, of which electrical short circuit melt marks and fire heat melt marks each account for 50%.

8. The method for identifying electrical melting marks of aluminum conductors based on electron backscatter diffraction (EBSD) according to claim 1, characterized in that: The output result of the recognition model in step S3 includes the type of melt mark (electrical short circuit or fire thermal melt) and its confidence score. A confidence score ≥90% is determined to be a valid result.

Images