Method for detecting corrosion resistance of copper-clad aluminum alloy wire

Abstract

The invention is applicable to the technical field of alloy wire corrosion resistance detection, and particularly relates to a copper-clad aluminum alloy wire corrosion resistance detection method which comprises the following steps: acquiring a microscopic image of a to-be-detected copper-clad aluminum alloy wire; according to the microscopic image, identifying and calculating corrosion parameters of the to-be-detected copper-clad aluminum alloy wire; on the basis of the corrosion parameters, physical parameters of the to-be-tested copper-clad aluminum alloy wire are calculated through finite element simulation; based on the corrosion parameters and the physical parameters, the macroscopic corrosion rate of the to-be-tested copper-clad aluminum alloy wire is calculated; the actual corrosion rate is obtained, the service life of the to-be-detected copper-clad aluminum alloy wire is calculated according to the macroscopic corrosion rate and the actual corrosion rate, and the corrosion resistance detection result of the to-be-detected copper-clad aluminum alloy wire is obtained. According to the method, the corrosion resistance of the copper-clad aluminum alloy wire can be efficiently and accurately detected and analyzed, the problems that a traditional method is long in period, insufficient in precision and the like are solved, and the accuracy of corrosion resistance evaluation is improved.

Description

Technical Field

[0001] This application belongs to the technical field of detecting the corrosion resistance of alloy wires, and particularly relates to a method for detecting the corrosion resistance of copper-clad aluminum alloy wires. Background Art

[0002] Copper-clad aluminum alloy wire (CCA wire for short) is a kind of wire composed of copper and aluminum alloy materials. The characteristic of this wire is that the outer layer is covered with copper, while the inner layer uses aluminum alloy. Due to the good electrical conductivity of copper and the low density and good strength of aluminum alloy, copper-clad aluminum alloy wire provides a solution that can both improve electrical conductivity and reduce weight in some applications.

[0003] In the prior art, the traditional salt spray test is a common method for detecting corrosion resistance, but it has problems such as a long test cycle, the need for a large number of samples, and high costs. In order to shorten the cycle of the salt spray test, accelerated tests are carried out to obtain corrosion data in a shorter time, but there are deviations from the corrosion dynamic data under actual use conditions. The corrosion data in the salt spray test is used to evaluate the corrosion resistance of copper-clad aluminum alloy wire, and predicting the corrosion resistance of copper-clad aluminum alloy wire often relies on simple empirical formulas or calculations based on incomplete test data, resulting in insufficient accuracy of corrosion resistance prediction.

[0004] In summary, when detecting the corrosion resistance of copper-clad aluminum alloy wire, there are problems that the corrosion resistance prediction is inaccurate due to a long test cycle and deviation from the actual environment. Summary of the Invention

[0005] The embodiments of this application provide a method for detecting the corrosion resistance of copper-clad aluminum alloy wire, which can solve the problems in the related art that when detecting the corrosion resistance of copper-clad aluminum alloy wire, the corrosion resistance prediction is inaccurate due to a long test cycle and deviation from the actual environment.

[0006] In the first aspect, the embodiments of this application provide a method for detecting the corrosion resistance of copper-clad aluminum alloy wire, including:

[0007] Obtain the microscopic image of the copper-clad aluminum alloy wire to be tested;

[0008] According to the microscopic image, identify and calculate the corrosion parameters of the copper-clad aluminum alloy wire to be tested; wherein, the corrosion parameters include corrosion type, geometric parameters corresponding to the corrosion type, and distribution parameters;

[0009] Based on the corrosion parameters, calculate the physical parameters of the copper-clad aluminum alloy wire to be tested through finite element simulation; wherein, the physical parameters include stress concentration coefficient and corrosion current density distribution;

[0010] Calculate the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be measured based on the corrosion parameters and the physical parameters;

[0011] Obtain the actual corrosion rate, and calculate the service life of the copper-clad aluminum alloy wire to be measured based on the macroscopic corrosion rate and the actual corrosion rate, so as to obtain the detection result of the corrosion resistance of the copper-clad aluminum alloy wire to be measured; wherein, the actual corrosion rate is obtained through a short-term salt spray test on the copper-clad aluminum alloy wire to be measured, and the detection result of the corrosion resistance includes the service life and high-risk areas of the copper-clad aluminum alloy wire to be measured.

[0012] In the embodiments of the present application, the above technical solutions have at least the following technical effects:

[0013] The method for detecting the corrosion resistance of copper-clad aluminum alloy wire provided by the present application first obtains the microscopic image of the copper-clad aluminum alloy wire to be measured, and then based on the microscopic image, identifies and calculates the corrosion parameters (corrosion type, geometric parameters and distribution parameters corresponding to the corrosion type) of the copper-clad aluminum alloy wire to be measured. Then, based on the corrosion parameters, the physical parameters (stress concentration coefficient and corrosion current density distribution) of the copper-clad aluminum alloy wire to be measured are calculated through finite element simulation. Based on the corrosion parameters and physical parameters, the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be measured is calculated. Finally, the actual corrosion rate (obtained through a short-term salt spray test on the copper-clad aluminum alloy wire to be measured) is obtained, and based on the macroscopic corrosion rate and the actual corrosion rate, the service life of the copper-clad aluminum alloy wire to be measured is calculated to obtain the detection result of the corrosion resistance of the copper-clad aluminum alloy wire to be measured (the service life and high-risk areas of the copper-clad aluminum alloy wire to be measured). This method can more accurately reflect the corrosion behavior of the material in the actual environment by calculating the physical parameters of the copper-clad aluminum alloy wire through finite element simulation, and can quantify the influence of the corrosion process on the mechanical properties of the material, thereby helping to evaluate the corrosion resistance of the material. This method can complete the corrosion resistance evaluation in a short time through image analysis and computational simulation, avoiding the long experimental process. At the same time, since the demand for a large number of samples is reduced, the cost of the test is also significantly reduced. By comprehensively considering the macroscopic corrosion rate and the actual corrosion rate, this method can calculate the accurate service life of the copper-clad aluminum alloy wire and can identify the high-risk areas. This method based on the combination of actual tests and computational models helps to provide more scientific and predictable corrosion resistance evaluation results. Compared with the limitations of traditional test methods, this method not only has strong repeatability through calculation and simulation, but also can be flexibly adjusted according to different conditions to adapt to the detection of different types of copper-clad aluminum alloy wire materials, enhancing the universality of the technology.

[0014] In a second aspect, the embodiments of the present application provide a device for detecting the corrosion resistance of copper-clad aluminum alloy wire, including:

[0015] An acquisition unit for acquiring a microscopic image of the copper-clad aluminum alloy wire to be measured;

[0016] A corrosion parameter calculation unit for identifying and calculating the corrosion parameters of the copper-clad aluminum alloy wire to be measured according to the microscopic image; wherein, the corrosion parameters include corrosion types, geometric parameters corresponding to the corrosion types, and distribution parameters;

[0017] A physical parameter calculation unit for calculating the physical parameters of the copper-clad aluminum alloy wire to be measured through finite element simulation based on the corrosion parameters; wherein, the physical parameters include stress concentration coefficient and corrosion current density distribution;

[0018] A macroscopic corrosion rate prediction unit for calculating the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be measured based on the corrosion parameters and the physical parameters;

[0019] A detection result obtaining unit for obtaining the actual corrosion rate, calculating the lifespan of the copper-clad aluminum alloy wire to be measured according to the macroscopic corrosion rate and the actual corrosion rate, and obtaining the corrosion resistance performance detection result of the copper-clad aluminum alloy wire to be measured; wherein, the actual corrosion rate is obtained by performing a short-term salt spray test on the copper-clad aluminum alloy wire to be measured, and the corrosion resistance performance detection result includes the lifespan and high-risk areas of the copper-clad aluminum alloy wire to be measured.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in any one of the embodiments in the first aspect is implemented.

[0021] It can be understood that the beneficial effects of the above second aspect to the third aspect can refer to the relevant descriptions in the first aspect above, and will not be elaborated here. Description of the Drawings

[0022] 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 use in the embodiments or the description of the prior art. Obviously, the drawings in the following description 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 creative efforts.

[0023] Figure 1 It is a schematic flowchart of a method for detecting the corrosion resistance performance of a copper-clad aluminum alloy wire provided by an embodiment of the present application;

[0024] Figure 2 It is a schematic overall implementation flowchart for predicting the lifespan in the method for detecting the corrosion resistance performance of a copper-clad aluminum alloy wire provided by an embodiment of the present application;

[0025] Figure 3 It is a schematic flowchart of the implementation process for selecting a life calculation method in the corrosion resistance performance detection method of the copper-clad aluminum alloy wire provided by the embodiments of the present application;

[0026] Figure 4 It is a schematic structural diagram of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0027] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures, technologies, etc. are proposed to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0028] It should be understood that when used in the specification of the present application and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0029] It should also be understood that the term "and / or" used in the specification of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0030] As used in the specification of the present application and the appended claims, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" according to the context.

[0031] In addition, in the description of the specification of the present application and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0032] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that specific features, structures, or characteristics described in connection with that embodiment are included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but rather mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.

[0033] In the related art, the time of salt spray testing is usually long, usually ranging from several days to several weeks, or even longer. This is because the salt spray corrosion process is a relatively slow process, where the material surface gradually corrodes until the corrosion is significant enough. Therefore, the test time period directly affects the accuracy and integrity of the test results. To obtain reliable corrosion results, a long experimental period and a large number of samples are often required for repeated experiments, which not only increases the workload of the laboratory and the complexity of sample preparation, but also results in high costs that can quickly accumulate, imposing an additional economic burden on enterprises.

[0034] To shorten the cycle of salt spray testing, some laboratories conduct accelerated tests by increasing environmental parameters such as salt spray concentration and temperature. Although such accelerated tests can obtain corrosion data in a shorter time, their results do not always accurately reflect the corrosion behavior of materials in the actual use environment. The corrosion rate in accelerated tests is usually high, and there are differences in environmental conditions such as humidity, temperature, salt concentration, and air flow in actual use, so there are deviations between the test results and the corrosion dynamic data under real use conditions. Such deviations may lead to an overly optimistic evaluation of the corrosion resistance of materials or misjudgments of the corrosion resistance performance of materials in actual use, limiting the practical significance and application value of the accelerated test results.

[0035] The corrosion data in salt spray testing is used to evaluate the corrosion resistance of copper-clad aluminum alloy wires, and the prediction of the corrosion resistance of copper-clad aluminum alloy wires often relies on simple empirical formulas or calculations based on incomplete test data, resulting in insufficient accuracy of the corrosion resistance prediction. In actual use, materials may be exposed to various corrosion environments, such as temperature and humidity changes, the presence of pollutants, mechanical stress, etc., and salt spray testing cannot simulate all possible environmental factors. In addition, due to the large differences between the accelerated test conditions and the actual environment, the prediction results often have great uncertainties, which pose potential risks to the application safety and reliability of materials.

[0036] To solve the above problems, an embodiment of the present application provides a method for detecting the corrosion resistance of copper-clad aluminum alloy wires. In this method, first, a microscopic image of the copper-clad aluminum alloy wire to be tested is obtained. Then, based on the microscopic image, the corrosion parameters (corrosion type, geometric parameters and distribution parameters corresponding to the corrosion type) of the copper-clad aluminum alloy wire to be tested are identified and calculated. Next, based on the corrosion parameters, the physical parameters (stress concentration factor and corrosion current density distribution) of the copper-clad aluminum alloy wire to be tested are calculated through finite element simulation. Based on the corrosion parameters and physical parameters, the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be tested is calculated. Finally, the actual corrosion rate (obtained by conducting a short-term salt spray test on the copper-clad aluminum alloy wire to be tested) is obtained, and based on the macroscopic corrosion rate and the actual corrosion rate, the service life of the copper-clad aluminum alloy wire to be tested is calculated, and the detection result of the corrosion resistance of the copper-clad aluminum alloy wire to be tested (the service life and high-risk areas of the copper-clad aluminum alloy wire to be tested) is obtained. By calculating the physical parameters of the copper-clad aluminum alloy wire through finite element simulation, this method can more accurately reflect the corrosion behavior of the material in the actual environment, can quantify the influence of the corrosion process on the mechanical properties of the material, thereby helping to evaluate the corrosion resistance of the material. By means of image analysis and computational simulation, this method can complete the corrosion resistance evaluation in a relatively short time, avoiding the long experimental process. At the same time, since the demand for a large number of samples is reduced, the cost of the test is also significantly reduced. By integrating the macroscopic corrosion rate and the actual corrosion rate, this method can calculate the accurate service life of the copper-clad aluminum alloy wire and can identify high-risk areas. This way of combining actual tests and computational models helps to provide a more scientific and predictable corrosion resistance evaluation result. Compared with the limitations of traditional test methods, through calculation and simulation, this method not only has strong repeatability but also can be flexibly adjusted according to different conditions, adapting to the detection of different types of copper-clad aluminum alloy wire materials, enhancing the universality of the technology.

[0037] The method for detecting the corrosion resistance of copper-clad aluminum alloy wires provided by the embodiment of the present application can be applied to electronic devices. At this time, the electronic device is the execution subject of the method for detecting the corrosion resistance of copper-clad aluminum alloy wires provided by the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of the electronic device.

[0038] For example, the electronic device may be a tablet computer, a wearable device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a desktop computer, a computer, a laptop computer, a handheld computing device, a customer premise equipment (CPE), and / or other devices for communicating on a wireless system, as well as a next-generation communication system, for example, a mobile terminal in a 5G network or a mobile terminal in a future evolved Public Land Mobile Network (PLMN).

[0039] To better understand the method for detecting the corrosion resistance of copper-clad aluminum alloy wire provided by the embodiments of the present application, the following provides an exemplary introduction to the specific implementation process of the method for detecting the corrosion resistance of copper-clad aluminum alloy wire provided by the embodiments of the present application.

[0040] Figure 1 FIG. shows a schematic flowchart of the method for detecting the corrosion resistance of copper-clad aluminum alloy wire provided by the embodiments of the present application. The method for detecting the corrosion resistance of copper-clad aluminum alloy wire includes:

[0041] S100, obtaining a microscopic image of the copper-clad aluminum alloy wire to be tested.

[0042] It can be understood that the microscopic image is obtained through microscopic techniques (such as scanning electron microscopy (SEM), optical microscopy, etc.), and is an image that can display the structural characteristics of the surface or cross-section of a material or object at a microscopic scale (from nanometers to micrometers). The microscopic image can help researchers observe in detail the microscopic structure, defects, corrosion types, grain distribution, cracks, pores, surface morphology and other details of the material, and thus provide a basis for the performance analysis and optimization of the material.

[0043] Scanning electron microscopy (SEM) is a technique that uses an electron beam to scan the surface of a sample and collects the reflected or secondary electron signals of the sample to form a high-resolution image. The SEM image can provide very clear surface morphology, details and elemental composition distribution, with a resolution reaching the nanometer level. The SEM image is suitable for observing the surface, corrosion area, cracks, particles and other characteristics of metal materials.

[0044] An optical microscope uses visible light to magnify and observe a sample. Its resolution is generally low and it is suitable for observing thick structures or large defects. It can display the surface structure and cross-section of the sample. However, its resolution is generally in the micron range and it is suitable for observing structures at a larger scale.

[0045] Exemplarily, the surface of the copper-clad aluminum alloy wire to be measured can be treated, such as cleaning, fixing, etc., to ensure that there are no contaminants and oxide layers on the surface. The treated copper-clad aluminum alloy wire to be measured is placed in a scanning electron microscope (SEM), and the voltage of the electron beam, the scanning mode, etc. are adjusted. Through the SEM display screen, an enlarged microscopic image is obtained.

[0046] Pre-treatments such as cleaning and cutting can be performed on the copper-clad aluminum alloy wire to be measured to ensure that the surface of the copper-clad aluminum alloy wire to be measured under the microscope is smooth and free of impurities. The copper-clad aluminum alloy wire to be measured is placed under an optical microscope, and the magnification, focal length, etc. are adjusted to observe the surface or cross-section of the copper-clad aluminum alloy wire to be measured. A digital camera or the image capture system of the optical microscope itself can be used to obtain microscopic images.

[0047] Microscopic images of the copper-clad aluminum alloy wire to be measured can be obtained through a scanning electron microscope (SEM) or an optical microscope, providing support for subsequent in-depth corrosion analysis and performance evaluation.

[0048] S200, according to the microscopic image, identify and calculate the corrosion parameters of the copper-clad aluminum alloy wire to be measured. Among them, the corrosion parameters include the corrosion type, the geometric parameters and distribution parameters corresponding to the corrosion type.

[0049] It can be understood that the corrosion type refers to the different forms of corrosion shown on the material surface. The corrosion types can include: uniform corrosion, where the corrosion is evenly distributed on the material surface, manifested as the gradual thinning or loss of luster of the metal surface; pitting corrosion, where small holes or pits appear in local areas, occurring in places with severe local corrosion; crevice corrosion, occurring at the joints, gaps or interfaces of the material, manifested as the corrosion spreading along the crevice, causing local thickness changes; stress corrosion cracking, where under the action of external stress, corrosion and cracks occur in coupling, manifested as the crack propagation of the material; intergranular corrosion, where the corrosion occurs at the grain boundaries, affecting the overall structural integrity of the material.

[0050] Exemplarily, techniques such as edge detection algorithms (such as Canny edge detection), texture analysis (such as gray-level co-occurrence matrix), etc. can be used to extract the morphological features of the corrosion area from the image. Through machine learning models (such as support vector machine (SVM), convolutional neural network (CNN), etc.), the morphological features of the extracted corrosion area are classified to identify the corrosion type in the image.

[0051] It can be understood that geometric parameters describe the geometric characteristics of the corrosion morphology, including the size, shape, depth, etc. of the corrosion area. Geometric parameters can include the corrosion area used to describe the area of the material surface lost due to corrosion; the corrosion depth, which is the depth of corrosion penetration into the metal substrate. Especially in pitting corrosion and crevice corrosion, the depth is an important geometric parameter; the pore diameter or crack width. For pitting corrosion or stress corrosion cracking, the corrosion pore diameter or crack width is a key geometric parameter; the aspect ratio of the corrosion morphology is used to describe the expansion direction of the corrosion morphology and can be quantified by the aspect ratio.

[0052] Exemplarily, image segmentation techniques (such as thresholding methods, clustering algorithms (such as K-means clustering), etc.) can be used to separate the corrosion area from the background, and the contour of the corrosion area can be extracted through edge detection or region growing algorithms. Morphological operations (such as erosion, dilation, opening operation, and closing operation, etc.) can be used to process the corrosion area, and geometric features such as the area, perimeter, aspect ratio, curvature, etc. of the corrosion area are calculated. These geometric parameters help to describe the shape of the corrosion area.

[0053] It can be understood that distribution parameters are used to describe the distribution characteristics of corrosion on the material surface, including corrosion density, the distribution pattern of corrosion (uniform, non-uniform, concentrated, etc.), and the expansion direction of corrosion. Corrosion density represents the number of corrosion points per unit area and can be determined by calculating the number of points or the area distribution in the corrosion area of the image. The expansion direction of corrosion (such as along grain boundaries, longitudinal or transverse expansion, etc.) can be analyzed as a distribution parameter; the distribution pattern of corrosion shows whether the corrosion is concentrated in certain areas or is more evenly distributed.

[0054] Exemplarily, the distribution density of the corrosion area on the surface of the copper-clad aluminum alloy wire to be measured can be calculated through statistical analysis. For example, calculate the number of corrosion spots per unit area or the spatial distribution of the corrosion area. The expansion morphology of the corrosion area (such as the spread of pitting corrosion, the growth of cracks, etc.) can be measured through fractal analysis or texture analysis. The aggregation of the corrosion area (such as whether pitting corrosion is concentrated in certain specific areas) can be calculated to evaluate the distribution characteristics of corrosion.

[0055] Based on microscopic image analysis, combined with different corrosion parameters (type, geometric parameters, distribution parameters), a more comprehensive description of corrosion characteristics can be obtained. These parameters will help to understand the corrosion behavior of the copper-clad aluminum alloy wire and evaluate the corrosion resistance of the copper-clad aluminum alloy wire.

[0056] In a possible implementation, please refer to Figure 2 , S200, according to the microscopic image, identify and calculate the corrosion parameters of the copper-clad aluminum alloy wire to be measured, including:

[0057] S210. Based on the corrosion recognition model, extract the corrosion features to be measured of the microscopic image, and identify the corrosion type of the microscopic image according to the corrosion features to be measured. Among them, the corrosion recognition model is constructed based on the transfer learning method.

[0058] It can be understood that the corrosion features to be measured of the microscopic image are extracted by the underlying layer of the corrosion recognition model. The underlying layer of the corrosion recognition model can be composed of a deep convolutional neural network (CNN) or other modern feature extraction architectures (such as Vision Transformer). The task of the underlying network is to extract general features from the input microscopic image. The general features are not specific to a certain corrosion type, but capture the underlying patterns of the image, including basic information such as edges, textures, and shapes.

[0059] Exemplarily, if CNN is used, multiple convolutional layers can perform filtering operations on the microscopic image, and extract features from simple to complex layer by layer. For example, the early convolutional layers extract low-level features (such as edges, corners, etc.), while the deep convolutional layers will learn more complex patterns (such as the shape and size of the corrosion area, etc.).

[0060] If Vision Transformer is used, the image can be cut into small pieces, and the relationships between different regions can be captured through the self-attention mechanism, and then global features can be extracted.

[0061] After being processed by the underlying network, the output is a high-dimensional feature map, which contains important corrosion information in the microscopic image, such as the texture, edges, and shapes of the corrosion area. After extracting the preliminary features (high-dimensional feature map), the deep network of the corrosion recognition model (such as Residual Network ResNet or other structures) can further process the preliminary features to capture complex corrosion patterns from a higher level. The deep network can help the model learn the subtle differences between different corrosion types.

[0062] It can be understood that the corrosion type of the microscopic image is identified by the classification head in the top layer of the corrosion recognition model. The classification head in the top layer can be a fully connected layer or a multi-layer perceptron (MLP), which is used to map the features learned by the deep network to the categories of different corrosion types.

[0063] Exemplarily, the classification head can use the Softmax function to probabilistically process the extracted features, and output the category probabilities of each corrosion type. The Softmax function converts the output of each category into a probability value, and the sum of the probability values of each category is 1. Suppose there are K corrosion types, the classification head can output a vector of length K: P = [P1, P2,..., P K , where P irepresents the probability that the image belongs to the corrosion type i, and P1 + P2 +... + P K = 1. For example, there are three corrosion types (pitting corrosion, uniform corrosion, crack corrosion), and the probability vector output by the classification head is P = [0.7, 0.2, 0.1]. Then the image is most likely to belong to pitting corrosion, with a 70% probability.

[0064] The underlying network can effectively extract physically meaningful features from microscopic images, capture the patterns of different types of corrosion. The classification head outputs the probability of each corrosion type through the Softmax activation function, enabling the model to identify the most likely corrosion type based on the corrosion features in the image. The corrosion recognition model can efficiently identify and classify different types of corrosion, providing reliable data support for the evaluation of the corrosion resistance of materials.

[0065] S220, according to the corrosion characteristics to be measured and the corrosion type, mark the position of the corrosion type in the microscopic image and calculate the geometric parameters of the corrosion type.

[0066] Exemplarily, preliminary positioning can be performed according to the rough position of the corrosion area provided by the classification head of the corrosion recognition model (such as a rectangular box or a bounding box). For example, the category output by the classification head is pitting corrosion, and the area position where this category is located is provided, such as a bounding box or a rough area range. The regional coordinates of the preliminary positioning can be used for subsequent refinement processing in Mask R-CNN to obtain a more accurate segmentation of the corrosion area.

[0067] It can be understood that the geometric parameters of the corrosion type can be calculated by the detection head in the top layer of the corrosion recognition model (Mask R-CNN and geometric parameter calculation).

[0068] Mask R-CNN is a region proposal-based object detection and segmentation network that can perform pixel-level segmentation on each object in the image. The main task in this step is to generate an accurate segmentation mask for each marked corrosion area.

[0069] Exemplarily, for each corrosion area (such as a rectangular box) identified by the classification head, Mask R-CNN can use RoI Align (Region of Interest Align) to map these corrosion areas to a specific layer of the network and perform feature alignment. The goal of RoI Align is to obtain a high-resolution feature representation of the region by precisely extracting features and spatially aligning the input region, ensuring that the details of the corrosion area are not lost and allowing for more accurate segmentation in subsequent steps.

[0070] After RoI Align processing, Mask R-CNN can generate a binary segmentation mask for each corrosion area, which represents the specific shape of the corrosion area. The mask for each corrosion area is a binary image of the same size as the original image (microscopic image), where the part corresponding to the corrosion area is 1 (or True), and the other parts are 0 (or False). The mask can precisely define the boundary of the corrosion area. For example, for pitting corrosion, the generated mask will only contain the specific location and shape of the pitting corrosion; for uniform corrosion, the mask may cover a larger area and reflect the uniformity of the corrosion.

[0071] Based on the segmentation mask of each corrosion area, geometric parameters of the corresponding corrosion area can be calculated, including area, boundary, shape, centroid, and perimeter. The area is the number of pixels in the corrosion area, representing the size of the corrosion area, which can be obtained by calculating the number of pixels with a value of 1 in the mask; the boundary is the contour or boundary of the corrosion area, which can be calculated by extracting the edge pixels of the mask; the shape can include aspect ratio, roundness, and ellipticity, which can be obtained by calculating the major axis direction and the lengths of the major and minor axes of the corrosion area; the centroid is the geometric center or centroid of the corrosion area, which can be obtained by calculating the weighted average position of all pixels in the mask; the perimeter is the edge length of the corrosion area, which can be obtained by calculating the number of pixels on the mask edge.

[0072] After Mask R-CNN and geometric parameter calculation, the geometric features of the corrosion area are finally output, including parameters such as area, shape, centroid, boundary, and perimeter. These geometric parameters can be further used to evaluate the severity, uniformity, and morphology of the corrosion.

[0073] By using Mask R-CNN to perform pixel-level segmentation on the corrosion area, a more accurate position of the corrosion area can be obtained, avoiding the rough positioning in traditional methods. The calculation of geometric parameters can quantify the degree and morphology of the corrosion, providing more accurate data support for subsequent corrosion assessment and material life prediction.

[0074] S230, according to the geometric parameters of the microscopic image and the corrosion type, calculate the continuous features related to the corrosion type distribution to obtain the distribution parameters of the corrosion type.

[0075] It can be understood that the distribution parameters of the corrosion type are calculated through the regression head in the top layer of the corrosion recognition model. The task of the regression head is to calculate the continuous features related to the corrosion type distribution, such as aggregation degree, uniformity, and expansibility, based on the characteristics of the corrosion area. These continuous features describe the spatial distribution and morphology of the corrosion area.

[0076] The regression head can map the features in the image to continuous numerical values related to the distribution of the corrosion areas by learning the spatial patterns of the corrosion areas. The function of the regression head is different from that of the classification head because the classification head outputs discrete class labels, while the regression head outputs continuous numerical features.

[0077] The aggregation degree refers to the degree of concentration of the corrosion areas, indicating whether the corrosion areas are more dispersed or aggregated. A higher aggregation degree means that multiple corrosion points are closely distributed, while a lower aggregation degree indicates that the corrosion areas are dispersed.

[0078] The uniformity refers to whether the distribution of the corrosion areas in the image is uniform. A corrosion area with high uniformity indicates that the corrosion is more evenly distributed, while low uniformity indicates that the corrosion areas exhibit concentrated hot spot characteristics.

[0079] The expansibility is used to describe the degree of expansion of the corrosion areas in the image, indicating the growth trend and potential expansion of the corrosion areas. High expansibility of corrosion means that the corrosion area is large and may affect more areas, while low expansibility indicates that the corrosion area is small and highly localized.

[0080] Exemplarily, the centroid of each corrosion area can be calculated based on mask segmentation, the distance between the centroids of different corrosion areas can be calculated, and the aggregation degree can be calculated based on the distance between the centroids of different corrosion areas. The calculation formula is where D1 represents the aggregation degree, n represents the total number of corrosion areas, and d ij represents the distance between the centroid of corrosion area i and the centroid of corrosion area j.

[0081] The microscopic image can be divided into small blocks (such as grids), and according to the area of the corrosion areas, the area ratio of the corrosion areas in each small block can be calculated, and then the uniformity can be calculated through the standard deviation. The calculation formula is where D2 represents the uniformity, A represents the area of the corrosion areas in each small block, μ(A) represents the mean value of the corrosion area, and σ(A) represents the standard deviation of the corrosion area.

[0082] The total area of all corrosion areas can be calculated, and the expansibility can be represented by the ratio of the corrosion area to the entire microscopic image. The calculation formula is where D3 represents the expansibility, A1 represents the total area of all corrosion areas, and A 总 represents the total area of the microscopic image.

[0083] The regression head can return continuous features (aggregation degree, uniformity, expansibility) related to the corrosion distribution. These continuous features can be used for further analysis of the corrosion distribution pattern and its impact on the material, providing basic data support for subsequent corrosion analysis, life prediction, etc.

[0084] In a possible implementation, the corrosion resistance detection method for copper-clad aluminum alloy wire further includes:

[0085] S201, obtaining a public corrosion dataset and sample data of the copper-clad aluminum alloy wire. Among them, the sample data includes 50 to 100 SEM or optical microscopic images of the surface / cross-section of the copper-clad aluminum alloy wire, and the corrosion type, geometric parameters, and distribution parameters of the corrosion type are manually labeled.

[0086] Exemplarily, the database of the NIST website can be accessed or contacted with the NIST Corrosion Research Center to obtain relevant datasets. NIST has released multiple public datasets related to material corrosion, covering the corrosion behaviors of different materials; relevant datasets shared by authors can be obtained by searching relevant papers through Google Scholar or ResearchGate; the public corrosion dataset can be obtained from a public corrosion dataset repository. For example, there are some public datasets related to corrosion (image datasets for corrosion detection or corrosion impact analysis) on the Kaggle platform.

[0087] Corrosion experiments (such as salt spray tests, immersion corrosion, etc.) can be carried out using copper-clad aluminum alloy wires (samples of different specifications), and SEM or optical microscope images of the sample surface and cross-section are taken. Approximately 50 to 100 images in different corrosion states are taken for manual annotation and classification. Each image is manually annotated to indicate the corrosion type (such as pitting corrosion, uniform corrosion, crack corrosion, etc.), geometric parameters of the corrosion morphology (such as corrosion depth, width, shape, etc.), and distribution parameters (such as the aggregation degree and uniformity of corrosion). Image annotation tools such as Labelbox, VGG Image Annotator, or LabelImg can be used for image annotation. The images and annotation results are stored in a standard format (such as JSON, XML, CSV) for subsequent analysis, model training, or research.

[0088] S202, based on the public corrosion dataset, pre-training the underlying layer of the corrosion recognition model so that the underlying layer of the corrosion recognition model learns general corrosion features. Among them, the underlying layer of the corrosion recognition model is a convolutional neural network.

[0089] It can be understood that ResNet or Vision Transformer can be used for the underlying layer of the corrosion recognition model. ResNet is suitable for most computer vision tasks, such as image classification and object detection, has a good residual connection structure, and can capture complex image features; Vision Transformer (ViT) is suitable for tasks with global context information, processes images through the self-attention mechanism, and is suitable for modeling large-scale image data.

[0090] Exemplarily, before inputting into the model, the image can be normalized, such as scaled to a fixed size (e.g., 224x224 pixels), and normalized by mean and standard deviation (e.g., the mean and standard deviation of ImageNet), to ensure the stability during model training. The corrosion images may vary under different environments and conditions, so data augmentation techniques (such as rotation, flipping, scaling, shearing, etc.) can be used to enable the model to better generalize to different corrosion images.

[0091] If ResNet is used, a pre-trained model on ImageNet can be used to initialize the weights of ResNet. By pre-training with the ImageNet dataset, ResNet can learn general visual features (such as edges, textures, shapes, etc.) in the convolutional layer, and then fine-tune on the public corrosion dataset to enable ResNet to focus on learning general corrosion features.

[0092] ViT can also adopt a similar approach, pre-train on a large-scale image dataset (such as ImageNet). Through the self-attention mechanism during training, ViT can learn the global relationships of corrosion images, capture the long-range dependencies in corrosion images, and further extract higher-level features (general corrosion features) of corrosion images.

[0093] S203, freeze the bottom layer of the corrosion recognition model, and based on the sample data, fine-tune the top layer of the corrosion recognition model so that the top layer of the corrosion recognition model can learn the specific corrosion features of copper-clad aluminum alloy wires, obtaining the corrosion recognition model. Among them, the top layer of the corrosion recognition model includes a classification head, a detection head, and a regression head.

[0094] Exemplarily, in order to enable the corrosion recognition model to extract the corrosion type and the corresponding geometric parameters and distribution parameters according to the input microscopic image, a top layer structure can be constructed. The task of the classification head in the top layer is to identify the corrosion type of copper-clad aluminum alloy wires, which consists of a fully connected layer and outputs the probability of each corrosion type through the Softmax activation function; the task of the detection head in the top layer is to detect the location of the corrosion area and calculate the geometric parameters corresponding to the corrosion type, and the detection head is composed of Mask R-CNN; the task of the regression head in the top layer is to calculate the distribution parameters of the corrosion type, which consists of a fully connected layer and outputs one or more continuous values to describe the spatial distribution characteristics of the corrosion.

[0095] The pre-trained bottom layer has learned to extract general corrosion features, so the feature extraction module (convolutional layer) of the bottom layer can be frozen, thus retaining the general corrosion feature extraction ability of the bottom layer, and only fine-tuning the top layer of the corrosion recognition model, that is, the classification head, the detection head, and the regression head, so as to make the top layer adapt to the specific corrosion features of copper-clad aluminum alloy wires.

[0096] The classification head learns the specific corrosion types of copper-clad aluminum alloy wires through sample data. The cross-entropy loss function can be used for training, and the classification performance is optimized by minimizing the cross-entropy loss.

[0097] The annotated geometric parameters of the corrosion area can be used to train Mask R-CNN to learn the geometric parameters of corrosion. The bounding box regression and mask segmentation tasks of Mask R-CNN can be optimized through objective functions (such as loss functions).

[0098] The weights of the regression head can be optimized through the distribution parameters of the sample data. The mean squared error (MSE) loss function can be used to minimize the gap between the distribution parameters predicted by the regression head and the true values.

[0099] A separate validation set can be used to evaluate the fine-tuned top layer and verify its performance in corrosion type classification, geometric parameter detection, and distribution parameter regression. The accuracy of corrosion type classification can be evaluated through classification accuracy and F1 score; the detection performance of geometric parameters can be calculated by evaluating the overlap between the masks generated by Mask R-CNN and the true labels through IoU; the accuracy of the corrosion distribution parameters predicted by the regression head can be evaluated through the mean squared error.

[0100] By utilizing pre-trained models and fine-tuning, only a small number of annotated images (50 - 100) of copper-clad aluminum alloy wires are needed to train a powerful corrosion recognition model. This model can not only identify corrosion types but also locate defect positions, extract geometric parameters, and distribution parameters. Through transfer learning, the existing large-scale corrosion datasets can be fully utilized to achieve efficient and accurate corrosion recognition under the specific corrosion patterns of copper-clad aluminum alloy wires, solving the problem that the lack of sufficient corrosion data for copper-clad aluminum alloy wires makes it difficult to train AI models.

[0101] S300, based on the corrosion parameters, calculates the physical parameters of the copper-clad aluminum alloy wire to be measured through finite element simulation. Among them, the physical parameters include the stress concentration coefficient and the corrosion current density distribution.

[0102] It can be understood that finite element simulation is a numerical method widely used to solve complex engineering problems, including the mechanical response, electrical characteristics, and thermal problems of materials. By decomposing materials or structures into small, computable units, the finite element method can accurately simulate the behavior of the entire material or structure. In corrosion analysis, finite element simulation can be used to evaluate the impact of corrosion on material properties and calculate parameters such as stress, deformation, and corrosion current density.

[0103] The stress concentration coefficient refers to the amplification coefficient of the local area stress when there are geometric defects (such as holes, cracks, corrosion pits, etc.) in the material. The corrosion current density refers to the amount of current flowing due to the corrosion reaction per unit area.

[0104] Exemplarily, according to the dimensions of the copper-clad aluminum alloy wire to be measured (including diameter, length, etc.), a three-dimensional geometric model of the copper-clad aluminum alloy wire to be measured can be created in finite element software (such as ANSYS, ABAQUS, etc.), and corresponding corrosion defects can be created in the geometric model according to the corrosion parameters. The geometric model is divided into a finite number of small elements, and appropriate mesh generation techniques (such as tetrahedral meshes, hexahedral meshes, etc.) are used. The accuracy of mesh generation can affect the accuracy of the simulation. The mesh in the corrosion area can be made more detailed to obtain accurate results. The material properties of the geometric model can be set, including the mechanical properties of aluminum and copper (such as elastic modulus, Poisson's ratio, yield strength, etc.) and electrical characteristics (such as electrical conductivity).

[0105] In order to simulate the working state of the copper-clad aluminum alloy wire to be measured under actual working conditions, appropriate boundary conditions and external loads can be applied. The loads can be tensile, compressive or bending stresses, and the boundary conditions can reflect the actual support and fixation conditions. Run the finite element analysis to calculate the stress distribution in different regions. By analyzing the stress distribution, identify the stress concentration regions (corrosion regions), and calculate the corresponding stress concentration coefficients.

[0106] In order to simulate the distribution of corrosion current density, the boundary conditions of the electrochemical reaction can be set. The boundary conditions can include the initial distribution of current density, the distribution of corrosion potential, the distribution of corrosion current, etc. Solve the distribution of the electric field through finite element analysis, and calculate the corrosion current density (corrosion current density distribution) in different regions according to the electric field distribution and electrochemical models (such as Tafel equation or Faraday's law).

[0107] Through finite element simulation combined with corrosion parameters, the stress concentration coefficient and the distribution of corrosion current density of the copper-clad aluminum alloy wire to be measured can be accurately calculated, so as to better evaluate the corrosion resistance of the copper-clad aluminum alloy wire to be measured in the actual use environment.

[0108] In one possible implementation, please refer to Figure 2 , S300, calculate the physical parameters of the copper-clad aluminum alloy wire to be measured through finite element simulation based on corrosion parameters, including:

[0109] S310, construct a finite element model of the copper-clad aluminum alloy wire to be measured according to the corrosion parameters.

[0110] Exemplarily, three-dimensional modeling software such as COMSOL Multiphysics, ABAQUS, or ANSYS can be used to create a three-dimensional geometric model (finite element model) of the copper-clad aluminum alloy wire to be measured according to the outer diameter, inner diameter, length, and the thicknesses of the copper layer and aluminum layer. Different material properties can be assigned to the copper layer and aluminum layer of the finite element model to ensure that the physical properties (such as elastic modulus, Poisson's ratio, density, conductivity, etc.) of the copper layer and aluminum layer match those of the actual copper-clad aluminum alloy wire to be measured.

[0111] According to the corrosion parameters, local corrosion can be created in the three-dimensional geometric model (finite element model), and the corrosion can be simulated by subtracting or changing the shape of the corrosion area. For example, pitting corrosion can be simulated by creating small holes or depressions on the surface of the copper layer in the finite element model; uniform corrosion can be simulated by uniformly reducing the thicknesses of the copper layer and aluminum layer on the surface of the finite element model.

[0112] Material removal (or corrosion) techniques can be used to simulate the generation of corrosion defects. By adjusting the material properties of certain elements (such as material removal, thickness change, etc.) during the mesh generation stage, corrosion defects can be created.

[0113] Through this step, a finite element model of the copper-clad aluminum alloy wire to be measured can be created, and corresponding corrosion can be created in the model according to the corrosion parameters, which can help better simulate the actual corrosion process, and then predict and optimize the life, corrosion rate, etc. of the copper-clad aluminum alloy wire to be measured.

[0114] S320, according to the load conditions and the finite element model, simulate the effects of the load and corrosion on the copper-clad aluminum alloy wire to be measured, and calculate the stress concentration factor of the copper-clad aluminum alloy wire to be measured. Among them, the load conditions are the actual working condition loads applied to the finite element model.

[0115] It can be understood that the load conditions can include tensile load, bending load, compressive load (if the alloy wire is in a compressed state, the corresponding compressive load can be applied), electric field, and thermal effects (if the copper-clad aluminum alloy wire to be measured bears current, the thermal effects generated due to the current passing through the copper-clad aluminum alloy wire can be simulated. Due to different resistances, copper and aluminum will have different thermal distributions under the electric field, which may affect the stress distribution of the material; the temperature distribution and stress effects caused by the current can be simulated by combining the relationship between current and resistance and Joule heating).

[0116] Exemplarily, the above load conditions can be applied to a three-dimensional geometric model (finite element model), and a finite element analysis software (such as ABAQUS, ANSYS, COMSOL) can be used to perform a static analysis on the three-dimensional geometric model (finite element model) to obtain a stress distribution diagram of the three-dimensional geometric model (finite element model) under the action of the load. By analyzing the stress distribution diagram, the regions with higher stress concentration can be identified, and the stress concentration coefficient can be calculated through the stress gradient, that is where SCF represents the stress concentration coefficient, and σ max represents the maximum stress value in the corroded area, and σ0 represents the reference stress in the area without corrosion.

[0117] The stress concentration coefficient can help to understand how corrosion exacerbates the stress state of the copper-clad aluminum alloy wire to be tested, and further affects its corrosion resistance, which can provide a reference for design optimization and improvement of corrosion resistance.

[0118] S330. According to the boundary conditions and the finite element model, simulate the diffusion path of the corrosive medium and calculate the corrosion current density distribution of the copper-clad aluminum alloy wire to be tested. Among them, the boundary conditions include the interface between the copper layer and the aluminum core and the concentration of the corrosive medium.

[0119] It can be understood that the interface between the copper layer and the aluminum core is a key part of the electrochemical reaction. The electrochemical reaction at the interface between the copper layer and the aluminum core may lead to different corrosion rates, and appropriate electrochemical boundary conditions can be defined at this interface. In the finite element simulation, the concentration distribution of the corrosive medium can be set. For example, the concentration of the corrosive medium is the highest in the surface area of the three-dimensional geometric model (finite element model), and then gradually decreases with the increase of depth.

[0120] Exemplarily, on the surface of the copper-clad aluminum alloy wire to be tested, the corrosive medium will diffuse to the interface between the copper layer and the aluminum core and react with the metal. According to the concentration change of the corrosive medium in different regions, an initial distribution of the concentration of the corrosive medium can be established in the finite element simulation. The concentration of the corrosive medium is higher on the surface and gradually decreases with the increase of depth. The diffusion path of the corrosive medium can be solved through the diffusion equation (Fick's Law), and the concentration distribution of the corrosive medium on the metal surface and inside can be calculated. The electrochemical equation (Tafel equation or Butler-Volmer equation) can be applied to the spatial distribution of the concentration of the corrosive medium, and the corrosion current density of each region can be calculated by solving the electrochemical equation, that is, the corrosion current density distribution.

[0121] After finite element simulation, the corrosion current density distribution map on the surface or inside of the entire copper-clad aluminum alloy wire to be measured can be obtained. This map shows the current density intensity in different regions and helps identify regions with a higher corrosion rate. This step helps provide data support for the subsequent evaluation of the corrosion resistance of the copper-clad aluminum alloy wire to be measured and provides important reference data for the design and use of the copper-clad aluminum alloy wire.

[0122] S400, calculate the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be measured based on corrosion parameters and physical parameters.

[0123] Exemplarily, according to the current density of each corrosion region in the corrosion current density distribution, the local corrosion rate of each corrosion region can be calculated using Faraday's law. The calculation formula for the local corrosion rate is where, v 局部 represents the local corrosion rate of a certain corrosion region (unit: m / s or mm / year), M represents the molar mass of the metal (copper or aluminum), i represents the current density of this corrosion region, n represents the number of electrons transferred in the corrosion reaction (such as for aluminum: n = 3, for copper: n = 2), F represents the Faraday constant (96485 C / mol), and ρ represents the metal density.

[0124] Since stress concentration will accelerate local corrosion (such as stress corrosion cracking, deepening of local corrosion), therefore, according to the stress concentration coefficient of each corrosion region, the local corrosion rate of the corresponding corrosion region can be corrected. The correction formula is v 修正 = v 局部 ×(1 + α×(SCF - 1)), where, v 修正 represents the corrected local corrosion rate of a certain corrosion region, α represents the stress influence coefficient (which can be determined by empirical data or experiments, such as in the range of 0.1 - 0.5), SCF represents the stress concentration coefficient of this corrosion region. It can be seen from the correction formula that the higher the local stress, the more obvious the corrosion acceleration and the increase in the corrosion rate.

[0125] According to the geometric parameters of the corrosion type and the corrected local corrosion rate, the corrosion rates of all corrosion regions can be weighted averaged according to the corresponding area to obtain a macroscopic corrosion rate representing the whole, that is where, v 初 represents the macroscopic corrosion rate, A1 represents the corrosion area corresponding to a certain corrosion region, and A represents the surface area of the copper-clad aluminum alloy wire to be measured.

[0126] If the volume loss caused by corrosion (such as the depth and volume of pits) can be calculated based on the geometric parameters of the corrosion type, the macroscopic corrosion rate can also be calculated using the change in corrosion volume, that is where, V 总Let \(V\) represent the total corrosion volume of all corrosion areas, and \(t\) represent the corrosion time (which can be the short-term salt spray test or the simulation time).

[0127] Since different corrosion types have different characteristics on the corrosion rate, in the calculation of the macroscopic corrosion rate, a type factor \(k\) can also be introduced for further correction, that is, \(v\) 宏 \(=\) \(v\) 初 \(\times k\), where \(v\) 宏 represents the corrected macroscopic corrosion rate. \(k\) can be set according to historical experimental data. For example, in pitting corrosion leading to local accelerated corrosion, \(k>1\); for uniform corrosion, approximately \(k = 1\); for crevice corrosion, there may be local maximum corrosion, and \(k\) can be determined according to the shape of the crevice.

[0128] This step can accurately quantify the overall corrosion rate of the copper-clad aluminum alloy wire to be tested in the corrosion environment, reflect the comprehensive effects of corrosion defects, stress concentration, and electrochemical reactions on the macroscopic corrosion behavior, and thus provide reliable data support for life prediction and high-risk area identification.

[0129] In one possible implementation, please refer to Figure 2 , S400, calculate the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be tested based on the corrosion parameters and physical parameters, including:

[0130] S410, extract the first eigenvector corresponding to the corrosion parameters and the second eigenvector corresponding to the physical parameters.

[0131] Exemplarily, each corrosion type can be converted into a binary vector through one-hot encoding. For example, if there are three corrosion types (pitting corrosion, uniform corrosion, and crack corrosion), then their encodings are pitting corrosion [1, 0, 0], uniform corrosion [0, 1, 0], and crack corrosion [0, 0, 1]. The geometric parameters are continuous numerical values and can be directly used as elements of the eigenvector without any additional processing. The distribution parameters are discrete or continuous numerical values and can be directly used as eigenvectors. For example, the uniformity is a discrete value and can be represented by calibration (e.g., 0 represents non-uniform, 1 represents uniform); the aggregation is a continuous value and can be directly used as a feature. The vectors corresponding to the corrosion type, the vectors corresponding to the geometric parameters, and the vectors corresponding to the distribution parameters can be concatenated to obtain the first eigenvector.

[0132] The stress concentration factor and the corrosion current density distribution are both numerical values. All the corrosion current densities in the stress concentration factor and the corrosion current density distribution can be directly concatenated to obtain the second eigenvector.

[0133] Through this step, complex corrosion parameters and physical parameters can be converted into numerical features available for machine learning and provide input data for further tasks such as corrosion rate prediction.

[0134] S420. Calculate the average corrosion current density based on the corrosion current density distribution, and calculate the ratio between the maximum stress and the average corrosion current density to obtain the coupling factor. Among them, the maximum stress is obtained through finite element simulation.

[0135] Exemplarily, the average corrosion current density can be obtained by integrating the corrosion current density distribution, that is Among them, represents the average corrosion current density, A f represents the total area of the corrosion area, j i represents the corrosion current density of the corrosion area i.

[0136] The maximum stress value can be extracted from the stress distribution obtained from the finite element simulation, and the ratio between the maximum stress and the average corrosion current density is calculated to obtain the coupling factor, that is Among them, α represents the coupling factor, σ max represents the maximum stress.

[0137] S430. Based on the first eigenvector, the second eigenvector, and the coupling factor, use the correlation model to calculate the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be measured. Among them, the correlation model is a machine learning model.

[0138] Exemplarily, the sample data set (each sample data includes the corresponding first eigenvector, second eigenvector, and coupling factor, and each sample data is labeled with the corresponding corrosion rate) can be divided into a training set and a test set. The training set is used to train the model, and the test set is used to verify the prediction ability of the model.

[0139] A suitable machine learning model can be selected, such as random forest, gradient boosting tree, neural network, etc., and the selected machine learning model is trained using the training set. During the training process, the model can learn the relationship between each sample data and its corrosion rate. For a linear model (such as linear regression), the model can estimate the weight of each feature; for a non-linear model (such as decision tree, random forest, SVR), the model can learn complex mapping relationships based on the data. The training set can be further split into multiple subsets for training and verification to help reduce the risk of model overfitting and evaluate the generalization ability of the model.

[0140] After training the model, the performance of the model can be evaluated using the test set. The evaluation metrics can include mean squared error (MSE), root mean squared error (RMSE), R 2 score. By adjusting the hyperparameters of the model (such as the depth of the decision tree, the number of trees in the random forest, the kernel function of SVR, etc.), the performance of the model can be further optimized. If the training set is small, regularization techniques (such as L1 / L2 regularization) can be considered to prevent overfitting.

[0141] When the associated model training is completed and the evaluation performance is good, the model can be used to predict new sample data. The first feature vector, the second feature vector, and the coupling factor of the copper-clad aluminum alloy wire to be measured can be input, and the associated model can output the predicted macroscopic corrosion rate.

[0142] The associated model realizes the accurate prediction from microscopic corrosion to macroscopic corrosion rate by integrating multi-source data through feature engineering and capturing non-linear relationships through machine learning. Its core value lies in breaking the boundary between data and physics in traditional methods, providing an efficient and high-precision intelligent solution for the corrosion resistance performance detection of copper-clad aluminum alloy wires.

[0143] S500, obtain the actual corrosion rate. According to the macroscopic corrosion rate and the actual corrosion rate, calculate the service life of the copper-clad aluminum alloy wire to be measured, and obtain the corrosion resistance performance detection result of the copper-clad aluminum alloy wire to be measured. Among them, the actual corrosion rate is obtained by conducting a short-term salt spray test on the copper-clad aluminum alloy wire to be measured, and the corrosion resistance performance detection result includes the service life and high-risk areas of the copper-clad aluminum alloy wire to be measured.

[0144] It can be understood that in the short-term salt spray test, the main purpose is to evaluate the corrosion behavior of the copper-clad aluminum alloy wire to be measured in a salt spray environment. By shortening the test time, preliminary data on the corrosion rate and corrosion path can be quickly obtained.

[0145] Exemplarily, a 5% NaCl solution can be used to simulate the corrosive environment in the atmosphere; the test temperature can be set at 35°C to simulate warm and humid climatic conditions; the traditional salt spray test takes 500 hours, while this test is designed for 24 hours, aiming to accelerate the corrosion reaction by shortening the time and facilitating the observation of early corrosion behavior.

[0146] Remove the surface impurities of the copper-clad aluminum alloy wire to be measured to ensure that the results of the corrosion test are not affected by other factors. A standard salt spray test chamber can be used, set to an environmental temperature of 35°C, and the 5% NaCl solution is sprayed onto the surface of the copper-clad aluminum alloy wire to be measured. Before the test starts, accurately measure the initial mass of the copper-clad aluminum alloy wire to be measured. After the test ends (24 hours), take out the copper-clad aluminum alloy wire to be measured from the salt spray test chamber, clean and dry it to remove the surface salt and moisture, and then measure the final mass of the copper-clad aluminum alloy wire to be measured. The actual corrosion rate is calculated by the weight loss method, and the formula is where v 实际 represents the actual corrosion rate, Δm represents the mass loss of the copper-clad aluminum alloy wire to be measured, A represents the surface area of the copper-clad aluminum alloy wire to be measured, and t represents the short-term salt spray test time.

[0147] It can be understood that since the actual corrosion rate (short-term) and the macroscopic corrosion rate (predicted) respectively reflect different corrosion characteristics, directly using one of them may not be sufficient to accurately describe the long-term corrosion changes. Therefore, the actual corrosion rate and the macroscopic corrosion rate can be combined to accurately calculate the lifespan.

[0148] Exemplarily, the actual corrosion rate gradually approaches the long-term corrosion rate over time, while the macroscopic corrosion rate may be relatively stable in the long term. Therefore, an exponential decay model can be used to describe this transition process, i.e., v 长 (t) = v 实际 ·e -λt + v 宏 ·(1 - e -λt ), where v 长 (t) is the long-term corrosion rate after time t, and λ is the decay coefficient of the corrosion rate transition (reflecting the speed of the transition from short-term to long-term).

[0149] In some cases, the long-term corrosion rate can also be estimated by weighted average. In the initial stage, the actual corrosion rate contributes more to the total corrosion rate, while the contribution of the macroscopic corrosion rate to the total corrosion rate gradually increases over time, and the weight coefficient can be dynamically adjusted according to time. That is, v 长 (t) = ω(t)·v 实际 + (1 - ω(t))·v 宏 , where ω(t) represents the weight coefficient.

[0150] After obtaining the long-term corrosion rate through the above method, the lifespan of the copper-clad aluminum alloy wire to be measured can be calculated according to the allowable corrosion depth. The calculation formula is where L represents the lifespan of the copper-clad aluminum alloy wire to be measured, and Th represents the allowable corrosion depth.

[0151] The actual corrosion rate provides direct support for experimental data, while the macroscopic corrosion rate takes into account the overall corrosion behavior of the material. By combining the actual corrosion rate and the macroscopic corrosion rate and adopting a suitable transition model (exponential decay model or weighted average), the long-term corrosion rate of the copper-clad aluminum alloy wire to be measured can be calculated more accurately, and then its service life can be predicted.

[0152] A risk matrix can be used to evaluate the combination of the corrosion rate and the stress concentration coefficient in each corrosion area. The risk matrix converts the combination of the corrosion rate and the stress concentration coefficient into a risk level. Grades can be defined for the corrosion rate and the stress concentration coefficient respectively, and different combinations of the corrosion rate and the stress concentration coefficient are mapped to a risk level. The grades of the corrosion rate and the stress concentration coefficient are combined, and a risk level is specified for each combination to obtain the risk matrix.

[0153] An example of the risk matrix is shown in the following table:

[0154]

[0155] According to the corrosion rate and stress concentration coefficient of each corrosion area, the corresponding risk level can be found from the risk matrix, so as to obtain the high-risk area of the copper-clad aluminum alloy wire to be tested.

[0156] Combining the lifespan of the copper-clad aluminum alloy wire to be tested with the high-risk area, the detection result of the corrosion resistance of the copper-clad aluminum alloy wire to be tested is obtained.

[0157] By combining the macroscopic corrosion rate and the actual corrosion rate, the accuracy and reliability of the results are improved. This step can achieve a comprehensive evaluation of the corrosion resistance of the copper-clad aluminum alloy wire, providing a scientific and accurate basis for the design, use and maintenance of the copper-clad aluminum alloy wire.

[0158] In a possible implementation, please refer to Figure 3 , in step S500, according to the macroscopic corrosion rate and the actual corrosion rate, calculate the lifespan of the copper-clad aluminum alloy wire to be tested, including:

[0159] S510, extract the peak value of the corrosion current density from the corrosion current density distribution, and calculate the ratio between the peak value of the corrosion current density and the average corrosion current density to obtain the proportionality factor.

[0160] Exemplarily, the maximum value (peak value) of the corrosion current density can be extracted from the corrosion current density distribution, and the ratio between the peak value of the corrosion current density and the average corrosion current density is calculated. This ratio (proportionality factor) represents the degree of concentration of the corrosion current density. The larger the ratio, the greater the gap between the current density in the local corrosion area and the overall average value, that is where β represents the proportionality factor, and j max represents the peak value of the corrosion current density.

[0161] S520, compare the proportionality factor with the proportionality threshold. If the proportionality factor is greater than the proportionality threshold, calculate the lifespan of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the proportionality factor.

[0162] It can be understood that the proportionality threshold is a preset standard for judging the severity of the corrosion area. The proportionality threshold can be determined through experimental data or experience, such as the proportionality threshold is 1.5.

[0163] Exemplarily, if the scale factor is greater than the scale threshold, it indicates that there is a relatively concentrated corrosion phenomenon in the corrosion area, which may have a greater impact on the life of the material. In this case, the damage rate can be used to calculate the life of the copper-clad aluminum alloy wire to be measured. The damage rate is the product of the macroscopic corrosion rate and the scale factor. By integrating the damage rate, the total damage can be calculated. When the total damage accumulates to the damage critical value, the copper-clad aluminum alloy wire to be measured fails, and the time when the total damage reaches the damage critical value is the life of the copper-clad aluminum alloy wire to be measured. The damage critical value represents the maximum damage value that the material can withstand before irreversible damage, failure, or destruction, and can be obtained through experimental methods such as fatigue tests and corrosion tests, or set according to experience. The calculation formula for the total damage is where D(t) represents the total damage.

[0164] Optionally, in step S520, calculating the life of the copper-clad aluminum alloy wire to be measured according to the macroscopic corrosion rate and the scale factor includes:

[0165] S521, calculating the local corrosion rate according to the macroscopic corrosion rate and the scale factor.

[0166] Exemplarily, the scale factor being greater than the scale threshold can indicate that the local corrosion is relatively concentrated and the corrosion process is accelerated, which may lead to faster failure of the material. The local corrosion rate can be used to calculate the life. Calculating the product between the macroscopic corrosion rate and the scale factor is the local corrosion rate.

[0167] S522, calculating the life of the copper-clad aluminum alloy wire to be measured according to the local corrosion rate and the allowable corrosion depth. Among them, the allowable corrosion depth is determined by the material standard or the application scenario.

[0168] It can be understood that the allowable corrosion depth refers to the maximum corrosion depth allowed before the material completely fails. The allowable corrosion depth is determined by the initial thickness, design requirements, and application environment of the copper-clad aluminum alloy wire to be measured.

[0169] Exemplarily, calculating the ratio between the allowable corrosion depth and the local corrosion rate is the life of the copper-clad aluminum alloy wire to be measured.

[0170] This step accurately calculates the life of the copper-clad aluminum alloy wire to be measured by considering the intensifying effect of local corrosion, and can better reflect the true performance of the copper-clad aluminum alloy wire to be measured in the application.

[0171] S530, if the scale factor is less than or equal to the scale threshold, calculating the life of the copper-clad aluminum alloy wire to be measured according to the macroscopic corrosion rate and the actual corrosion rate.

[0172] Exemplarily, when the scale factor is less than or equal to the scale threshold, it can be shown that the corrosion is relatively uniform and the damage to the material will not increase rapidly. Therefore, the actual corrosion rate can be directly used to calculate the life. That is, the life of the copper-clad aluminum alloy wire to be measured is the ratio of the difference between the initial thickness and the minimum remaining thickness before failure to the actual corrosion rate.

[0173] Optionally, in step S530, to calculate the life of the copper-clad aluminum alloy wire to be measured according to the macroscopic corrosion rate and the actual corrosion rate, it includes:

[0174] S531, calculate the long-term corrosion rate according to the actual corrosion rate.

[0175] Exemplarily, multiple groups of short-term corrosion and long-term corrosion experimental data can be collected, and the corrosion rates (short-term corrosion rate and long-term corrosion rate) at different times can be recorded. Environmental data related to corrosion, such as temperature, humidity, corrosion medium concentration, etc., which may affect the long-term corrosion rate, can be collected.

[0176] Based on the trend of the experimental data, the performance of different non-linear models in fitting the experimental data can be compared, and the most suitable model (such as the exponential decay model, Weibull distribution model, etc.) can be selected. Methods such as non-linear regression or least squares method are used to fit the long-term corrosion rate model, and the parameters of the model are calculated from the experimental data to ensure the minimum error during the fitting process and that the model can reasonably reflect the true situation of the corrosion rate changing with time.

[0177] After obtaining the parameters by fitting, test data or other experimental data are used to verify the accuracy of the model. If the prediction of the model is consistent with the experimental data, then this model can be used to estimate the long-term corrosion rate.

[0178] Among them, the long-term corrosion rate may gradually decrease over time. After a stable oxide layer is formed on the material surface, the corrosion rate shows a decaying trend. In this case, the corrosion rate may follow the exponential decay model. The exponential decay model is v 长 (t) = v 短 ·e -kt , where v 长 represents the long-term corrosion rate, v 短 represents the short-term corrosion rate, k is the decay constant that controls the speed of the corrosion rate decay, and t is the time.

[0179] The Weibull distribution model is a statistical model for life analysis, which can be used to describe the change of the corrosion rate with time, and can adapt to various different types of corrosion rate changes, especially in the case where the material properties change greatly with time. The Weibull distribution model is v 长 = v 短 ·(1 + βt)α , α and β are parameters fitted from experimental data and are used to control the variation of the corrosion rate with time.

[0180] After the fitting is completed, the long-term corrosion rate model can be used to calculate the long-term corrosion rate based on the actual corrosion rate.

[0181] S532, perform a weighted average of the long-term corrosion rate and the macroscopic corrosion rate to obtain the comprehensive corrosion rate.

[0182] Exemplarily, a weighted average can be performed on the long-term corrosion rate and the macroscopic corrosion rate to calculate the comprehensive corrosion rate, that is, v 综合 = ω·v 宏 +(1 - ω)·v 长 , where v 综合 represents the comprehensive corrosion rate, ω represents the weight of the macroscopic corrosion rate, and 1 - ω represents the weight of the long-term corrosion rate.

[0183] S533, calculate the life of the copper-clad aluminum alloy wire to be tested according to the comprehensive corrosion rate and the allowable corrosion depth.

[0184] Exemplarily, calculate the ratio between the allowable corrosion depth and the comprehensive corrosion rate, which is the life of the copper-clad aluminum alloy wire to be tested.

[0185] By using the actual corrosion data for long-term prediction, the limitations of relying only on short-term experimental data are avoided, the reliability and practicality of the prediction are improved, and at the same time, the influence of the long-term corrosion rate and the macroscopic corrosion rate is balanced, enabling a more comprehensive calculation of the life of the copper-clad aluminum alloy wire to be tested in the actual environment.

[0186] In a possible implementation manner, please refer to Figure 3 , the method for detecting the corrosion resistance of copper-clad aluminum alloy wire further includes:

[0187] S10, calculate the absolute difference between the macroscopic corrosion rate and the actual corrosion rate, and calculate the ratio between the absolute difference and the actual corrosion rate to obtain the rate deviation; S20, compare the rate deviation with the deviation threshold. If the rate deviation is greater than the deviation threshold, trigger the calibration correlation model and update the parameters of the correlation model according to the actual corrosion rate; S30, if the rate deviation is less than or equal to the deviation threshold, calculate the life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the actual corrosion rate.

[0188] Exemplarily, the absolute difference between the macroscopic corrosion rate and the actual corrosion rate can be calculated, and the ratio between the absolute difference and the actual corrosion rate can be determined as the rate deviation, that is, , where η represents the rate deviation.

[0189] The deviation threshold is a preset standard value used to determine whether the rate deviation is within an acceptable range. The deviation threshold can be set according to experimental data, material properties, and application scenarios. For example, the deviation threshold can be set to 5% or 10%.

[0190] If the rate deviation is greater than the deviation threshold, it indicates that the associated model prediction is not accurate enough, and the associated model can be calibrated. The actual corrosion rate can be used to refit the associated model, adjust the model parameters, so that the model output is closer to the actual corrosion rate. The model can be updated through methods such as regression analysis or the least squares method to ensure that the new model can better match the actual corrosion rate. The macroscopic corrosion rate is recalculated through the updated associated model to make it more consistent with the data of the actual corrosion rate.

[0191] When the rate deviation is less than or equal to the deviation threshold, the calculation method of the life of the copper-clad aluminum alloy wire to be measured can be determined according to the scale factor (i.e., step S510, step S520, step S530).

[0192] Through these steps above, it is beneficial to use appropriate corrosion rates for life prediction in different situations and improve the accuracy of prediction by calibrating the model.

[0193] In a possible implementation, in step S500, the corrosion resistance detection results of the copper-clad aluminum alloy wire to be measured are obtained, including:

[0194] S501, comparing each stress concentration coefficient of the copper-clad aluminum alloy wire to be measured with the stress threshold, and determining the area where the stress concentration coefficient is greater than the stress threshold as the mechanical weak area; S502, comparing each corrosion current density in the corrosion current density distribution with twice the average corrosion current density, and determining the area where the corrosion current density is greater than twice the average corrosion current density as the electrochemical weak area; S503, calculating the intersection area of the mechanical weak area and the electrochemical weak area to obtain the high-risk area; S504, integrating the high-risk area and the life of the copper-clad aluminum alloy wire to be measured to obtain the corrosion resistance detection results of the copper-clad aluminum alloy wire to be measured.

[0195] Exemplarily, the stress threshold is a preset critical value representing the maximum stress concentration coefficient that the material can withstand. The stress concentration coefficient of each corrosion area can be compared with the stress threshold. If the stress concentration coefficient of a certain corrosion area is greater than the stress threshold, then this corrosion area is the mechanical weak area.

[0196] The corrosion current density of each corrosion area can be compared with twice the average corrosion current density. If the corrosion current density of a certain corrosion area is greater than twice the average corrosion current density, then this area is the electrochemical weak area.

[0197] Determine the intersection of the mechanically weak area and the electrochemically weak area as the high-risk area.

[0198] The lifespan of the copper-clad aluminum alloy wire to be measured can be integrated with the high-risk area to obtain the corrosion resistance performance detection result of the copper-clad aluminum alloy wire to be measured. The corrosion resistance performance detection result can include the position and size of the high-risk area and the lifespan calculation result of the entire copper-clad aluminum alloy wire to be measured.

[0199] The above steps comprehensively consider mechanical and electrochemical factors, can more accurately evaluate the overall performance of the copper-clad aluminum alloy wire to be measured in a corrosive environment, and highlight the high-risk area, which is helpful for engineering design and material optimization. It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not imply the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0200] Corresponding to the copper-clad aluminum alloy wire corrosion resistance performance detection method described in the above embodiments, the embodiments of the present application also provide a copper-clad aluminum alloy wire corrosion resistance performance detection device, and each unit of the device can implement each step of the copper-clad aluminum alloy wire corrosion resistance performance detection method.

[0201] The device includes:

[0202] An acquisition unit for acquiring a microscopic image of the copper-clad aluminum alloy wire to be measured.

[0203] A corrosion parameter calculation unit for identifying and calculating the corrosion parameters of the copper-clad aluminum alloy wire to be measured according to the microscopic image. Among them, the corrosion parameters include the corrosion type, the geometric parameters and distribution parameters corresponding to the corrosion type.

[0204] A physical parameter calculation unit for calculating the physical parameters of the copper-clad aluminum alloy wire to be measured through finite element simulation based on the corrosion parameters. Among them, the physical parameters include the stress concentration coefficient and the corrosion current density distribution.

[0205] A macroscopic corrosion rate prediction unit for calculating the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be measured based on the corrosion parameters and physical parameters.

[0206] A detection result obtaining unit for obtaining the actual corrosion rate, calculating the lifespan of the copper-clad aluminum alloy wire to be measured according to the macroscopic corrosion rate and the actual corrosion rate, and obtaining the corrosion resistance performance detection result of the copper-clad aluminum alloy wire to be measured. Among them, the actual corrosion rate is obtained by performing a short-term salt spray test on the copper-clad aluminum alloy wire to be measured, and the corrosion resistance performance detection result includes the lifespan of the copper-clad aluminum alloy wire to be measured and the high-risk area.

[0207] It should be noted that for the content such as information interaction and execution process between the above units, since it is based on the same concept as the method embodiment of the present application, for its specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details will not be repeated here.

[0208] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working process of the units and modules in the above device can refer to the corresponding process in the foregoing method embodiment, and details will not be repeated here.

[0209] The embodiment of the present application also provides an electronic device. Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 4 shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 4 only one is shown in the figure), at least one memory 61 ( Figure 4 only one is shown in the figure), and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the electronic device 6 implements the steps in any of the above-mentioned embodiments of the method for detecting the corrosion resistance of copper-clad aluminum alloy wires, or the functions of each unit in the above-mentioned device embodiments.

[0210] Exemplarily, the computer program 62 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing specific functions, and these instruction segments are used to describe the execution process of the computer program 62 in the electronic device 6.

[0211] The electronic device 6 can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The electronic device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Figure 4This is only an example of the electronic device 6 and does not constitute a limitation on the electronic device 6. It may include more or fewer components than those shown in the figure, or combine some components, or different components. For example, it may also include input / output devices, network access devices, buses, etc.

[0212] The processor 60 may be a central processing unit (CPU), and the processor 60 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0213] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as the hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may also be an external storage device of the electronic device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 6. Further, the memory 61 may also include both the internal storage unit and the external storage device of the electronic device 6. The memory 61 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory 61 may also be used to temporarily store data that has been output or will be output.

[0214] The embodiments of the present application also provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0215] The embodiments of the present application provide a computer program product, and when the computer program product runs on an electronic device, the electronic device implements the steps in any of the above method embodiments.

[0216] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of this application, a computer program can be used to instruct relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can at least include: any entity or device capable of carrying the computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electrical carrier signal, a telecommunications signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunications signal.

[0217] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0218] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0219] In the embodiments provided in this application, it should be understood that the disclosed copper-clad aluminum alloy wire corrosion resistance detection device, method, and electronic device can be implemented in other ways. For example, the copper-clad aluminum alloy wire corrosion resistance detection device and electronic device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in an electrical, mechanical, or other form.

[0220] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0221] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A method for detecting the corrosion resistance of copper-clad aluminum alloy wire, characterized in that, Including: Obtain the microscopic image of the copper-clad aluminum alloy wire to be tested; Based on the microscopic image, identify and calculate the corrosion parameters of the copper-clad aluminum alloy wire to be tested; wherein, the corrosion parameters include corrosion type, geometric parameters corresponding to the corrosion type, and distribution parameters; Based on the corrosion parameters, calculate the physical parameters of the copper-clad aluminum alloy wire to be tested through finite element simulation; wherein, the physical parameters include stress concentration coefficient and corrosion current density distribution; Based on the corrosion parameters and the physical parameters, calculate the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be tested; Obtain the actual corrosion rate, and based on the macroscopic corrosion rate and the actual corrosion rate, calculate the service life of the copper-clad aluminum alloy wire to be tested, and obtain the corrosion resistance performance detection result of the copper-clad aluminum alloy wire to be tested; wherein, the actual corrosion rate is obtained through a short-term salt spray test on the copper-clad aluminum alloy wire to be tested, and the corrosion resistance performance detection result includes the service life and high-risk areas of the copper-clad aluminum alloy wire to be tested.

2. The corrosion resistance detection method of the copper-clad aluminum alloy wire according to claim 1, characterized in that, The step of identifying and calculating the corrosion parameters of the copper-clad aluminum alloy wire to be tested based on the microscopic image includes: Based on the corrosion identification model, extract the corrosion features to be tested in the microscopic image, and based on the corrosion features to be tested, identify the corrosion type of the microscopic image; wherein, the corrosion identification model is constructed based on the transfer learning method; Based on the corrosion features to be tested and the corrosion type, mark the position of the corrosion type in the microscopic image, and calculate the geometric parameters of the corrosion type; Based on the microscopic image and the geometric parameters of the corrosion type, calculate the continuous features related to the distribution of the corrosion type, and obtain the distribution parameters of the corrosion type.

3. The method for detecting the corrosion resistance of the copper-clad aluminum alloy wire according to claim 2, wherein, The method further includes: Obtain a public corrosion data set and sample data of copper-clad aluminum alloy wires; wherein, the sample data includes 50-100 SEM or optical microscopic images of the surface / cross-section of copper-clad aluminum alloy wires, and the corrosion type, geometric parameters of the corrosion type, and distribution parameters are manually marked; Based on the public corrosion data set, pre-train the underlying layer of the corrosion identification model so that the underlying layer of the corrosion identification model learns general corrosion features; wherein, the underlying layer of the corrosion identification model is a convolutional neural network; Freeze the underlying layer of the corrosion identification model, and based on the sample data, fine-tune the top layer of the corrosion identification model so that the top layer of the corrosion identification model learns the unique corrosion features of copper-clad aluminum alloy wires, and obtain the corrosion identification model; wherein, the top layer of the corrosion identification model includes a classification head, a detection head, and a regression head.

4. The method for detecting the corrosion resistance of copper-clad aluminum alloy wire according to claim 1, characterized in that, The step of calculating the physical parameters of the copper-clad aluminum alloy wire to be tested through finite element simulation based on the corrosion parameters includes: Based on the corrosion parameters, construct a finite element model of the copper-clad aluminum alloy wire to be tested; Based on the load conditions and the finite element model, simulate the influence of load and corrosion on the copper-clad aluminum alloy wire to be tested, and calculate the stress concentration coefficient of the copper-clad aluminum alloy wire to be tested; wherein, the load conditions are the actual working condition loads applied to the finite element model; According to the boundary conditions and the finite element model, simulate the diffusion path of the corrosive medium and calculate the corrosion current density distribution of the copper-clad aluminum alloy wire to be tested; wherein, the boundary conditions include the interface between the copper layer and the aluminum core and the concentration of the corrosive medium.

5. The method for detecting the corrosion resistance of copper-clad aluminum alloy wire according to claim 1, characterized in that, Calculating the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be tested based on the corrosion parameters and the physical parameters includes: Extracting a first eigenvector corresponding to the corrosion parameters and a second eigenvector corresponding to the physical parameters; According to the corrosion current density distribution, calculate the average corrosion current density, and calculate the ratio between the maximum stress and the average corrosion current density to obtain a coupling factor; wherein, the maximum stress is obtained through the finite element simulation; Based on the first eigenvector, the second eigenvector and the coupling factor, use the correlation model to calculate the macroscopic corrosion rate of the copper-clad aluminum alloy wire to be tested; wherein, the correlation model is a machine learning model.

6. The method for detecting the corrosion resistance of the copper-clad aluminum alloy wire according to claim 5, characterized in that, Calculating the service life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the actual corrosion rate includes: Extract the peak value of the corrosion current density from the corrosion current density distribution, and calculate the ratio between the peak value of the corrosion current density and the average corrosion current density to obtain a proportionality factor; Compare the proportionality factor with a proportionality threshold. If the proportionality factor is greater than the proportionality threshold, calculate the service life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the proportionality factor; If the proportionality factor is less than or equal to the proportionality threshold, calculate the service life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the actual corrosion rate.

7. The method for detecting the corrosion resistance of the copper-clad aluminum alloy wire according to claim 6, wherein Calculating the service life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the proportionality factor includes: Calculate the local corrosion rate according to the macroscopic corrosion rate and the proportionality factor; Calculate the service life of the copper-clad aluminum alloy wire to be tested according to the local corrosion rate and the allowable corrosion depth; wherein, the allowable corrosion depth is determined by the material standard or the application scenario.

8. The method for detecting the corrosion resistance of the copper-clad aluminum alloy wire according to claim 7, characterized in that, Calculating the service life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the actual corrosion rate includes: Calculate the long-term corrosion rate according to the actual corrosion rate; Perform a weighted average on the long-term corrosion rate and the macroscopic corrosion rate to obtain a comprehensive corrosion rate; Calculate the service life of the copper-clad aluminum alloy wire to be tested according to the comprehensive corrosion rate and the allowable corrosion depth.

9. The method for detecting the corrosion resistance of copper-clad aluminum alloy wire according to claim 5, wherein The method further includes: Calculate the absolute difference between the macroscopic corrosion rate and the actual corrosion rate, and calculate the ratio between the absolute difference and the actual corrosion rate to obtain a rate deviation; Compare the rate deviation with a deviation threshold. If the rate deviation is greater than the deviation threshold, trigger calibration of the correlation model and update the parameters of the correlation model according to the actual corrosion rate; If the rate deviation is less than or equal to the deviation threshold, calculate the service life of the copper-clad aluminum alloy wire to be tested according to the macroscopic corrosion rate and the actual corrosion rate.

10. The method for detecting the corrosion resistance of copper-clad aluminum alloy wire according to claim 5, wherein, Obtaining the corrosion resistance performance detection result of the copper-clad aluminum alloy wire to be tested includes: Compare each stress concentration coefficient of the copper-clad aluminum alloy wire to be tested with the stress threshold, and determine the area where the stress concentration coefficient is greater than the stress threshold as the mechanical weak area; Compare each corrosion current density in the corrosion current density distribution with twice the average corrosion current density, and determine the area where the corrosion current density is greater than twice the average corrosion current density as the electrochemical weak area; Calculate the intersection area of the mechanical weak area and the electrochemical weak area to obtain the high-risk area; Integrate the high-risk area and the lifespan of the copper-clad aluminum alloy wire to be tested to obtain the corrosion resistance performance test result of the copper-clad aluminum alloy wire to be tested.

Images