A method, system and device for identifying corrosion fatigue damage of aluminum alloy

Abstract

The present invention discloses a method, system and device for identifying corrosion fatigue damage of aluminum alloys. The method comprises the following steps: obtaining AE signal parameters in the process of aluminum alloy corrosion fatigue crack propagation, and preprocessing the AE signal parameters; clustering analysis of the preprocessed AE signal parameters using the ACO‑Kmeans++ clustering algorithm to determine the cluster center; testing the cluster center by clustering result evaluation index to determine the optimal number of clusters; determining the important parameter characteristic center value of the cluster center according to the clustering result, and matching the AE signal with the aluminum alloy corrosion fatigue damage pattern in combination with the physical fracture characteristics. The present invention greatly improves the matching accuracy and realizes the effective identification of aviation aluminum alloy corrosion fatigue damage.

Description

Technical Field

[0001] The present invention relates to the technical field of material analysis, and in particular to a method, system and device for identifying corrosion fatigue damage of aluminum alloy. Background Art

[0002] Acoustic emission (AE) is a common physical phenomenon, which refers to the phenomenon that the material releases energy suddenly and quickly due to deformation or damage to generate transient elastic waves. As a non-destructive testing technology that can be monitored in real time, it can directly reflect the damage defects and crack propagation inside the material. From the principle of AE detection, only when the internal damage of the inspected part occurs, the material structure will change and generate AE signals. This feature reflects the dynamic detection nature of AE technology and the high correlation between damage and AE signals.

[0003] With the rapid development of my country's aerospace industry, the research on the monitoring and detection of aircraft components has become more and more in-depth, and aluminum alloys have important applications in aerospace and other fields due to their corrosion resistance, light weight and high strength. After in-depth research, researchers found that aviation aluminum alloys will produce significant AE signals when damaged. Using advanced pattern recognition methods such as multiple clustering, the characteristic parameters of the AE signal are accurately matched with different damage types, thereby achieving accurate identification of the damage type.

[0004] There are many clustering methods at present. K-means clustering has long been the mainstream clustering algorithm because of its simple and practical algorithm. Therefore, K-means clustering is also mainly used for clustering AE signal parameters. However, K-means clustering is sensitive to the initial cluster center. Different selection methods will produce different results. It is easy to fall into the local optimal solution and is sensitive to outliers. Summary of the invention

[0005] In order to solve the above-mentioned problems, the present application provides a method, system, and device for identifying corrosion fatigue damage of aluminum alloys, which greatly improve the matching accuracy and realize the effective identification of corrosion fatigue damage of aviation aluminum alloys.

[0006] The present invention provides a method for identifying corrosion fatigue damage of aluminum alloy, and the method comprises the following steps:

[0007] Acquire AE signal parameters during the aluminum alloy corrosion fatigue crack growth process, and preprocess the AE signal parameters;

[0008] ACO-Kmeans++ clustering algorithm is used to perform cluster analysis on the preprocessed AE signal parameters and determine the cluster center;

[0009] Testing the cluster centers by clustering result evaluation index to determine the optimal number of clusters;

[0010] The characteristic center value of the important parameter of the cluster center is determined according to the clustering results, and the AE signal is matched with the aluminum alloy corrosion fatigue damage pattern in combination with the physical fracture characteristics.

[0011] In some embodiments, cluster analysis is performed on the preprocessed AE signal parameters using the ACO-Kmeans++ clustering algorithm to determine the cluster center, including:

[0012] Initializing clustering parameters, wherein the clustering parameters include at least one of the following: the number of clusters, the number of ant colonies, the number of iterations, the pheromone concentration, and the pheromone volatilization rate;

[0013] Use K-Means++ method to initialize cluster centers;

[0014] Perform ant colony iteration;

[0015] Update the cluster center according to the allocation results of ants in the ant colony iteration;

[0016] Determine whether the new cluster center converges to the old cluster center. If not, perform ant colony iteration;

[0017] Use new cluster centers for clustering.

[0018] In some implementations, initializing cluster centers using the K-Means++ method includes:

[0019] Randomly select the current cluster center, calculate the shortest distance between the sample and the current cluster center, and probabilistically determine the next cluster center based on the shortest distance;

[0020] Repeat the above steps until all cluster centers are determined.

[0021] In some embodiments, testing the cluster center by a clustering result evaluation index comprises:

[0022] Normalizing the clustering result evaluation indicators, including the silhouette coefficient and the CH index, performing weighted summation according to a preset ratio to determine the comprehensive score of the cluster center;

[0023] The optimal number of clusters is determined based on the comprehensive scores of cluster centers.

[0024] In some implementations, performing ant colony iteration includes:

[0025] In each iteration, the next data point is selected based on the pheromone concentration and the roulette algorithm, and the ants are assigned to the nearest cluster center.

[0026] Update the pheromone matrix to strengthen the paths taken by the ants.

[0027] In some implementations, updating the cluster center according to the allocation results of ants in the ant colony iteration includes:

[0028] Calculate the sum of pheromone concentrations on the paths of all data points in each cluster, count the number of data points contained in each cluster, and update the cluster center based on the number of data points.

[0029] In some embodiments, preprocessing the AE signal parameters includes:

[0030] Selecting AE signal parameters, wherein the AE signal parameters include amplitude, energy, and average frequency;

[0031] AE signal parameters were normalized by z-score.

[0032] In some implementations, determining whether the new cluster center converges to the old cluster center includes:

[0033] The distance between the new cluster center and the old cluster center is calculated. If the distance between the new cluster center and the old cluster center is greater than the convergence threshold, the new cluster center converges to the old cluster center.

[0034] The present invention also provides an aluminum alloy corrosion fatigue damage identification system, comprising:

[0035] An acquisition module is used to acquire AE signal parameters during the expansion of aluminum alloy corrosion fatigue cracks and to preprocess the AE signal parameters;

[0036] The cluster analysis module uses the ACO-Kmeans++ clustering algorithm to perform cluster analysis on the preprocessed AE signal parameters and determine the cluster center;

[0037] A clustering validity test module tests the cluster centers through clustering result evaluation indicators to determine the optimal number of clusters;

[0038] The damage pattern recognition module determines the characteristic center value of the important parameters of the cluster center according to the clustering results, and matches the aluminum alloy corrosion fatigue damage pattern of the AE signal in combination with the physical fracture characteristics.

[0039] The present invention also provides an identification device for aluminum alloy corrosion fatigue damage, comprising:

[0040] Memory;

[0041] processor;

[0042] And a computer program stored in a memory and executable on a processor, wherein when the processor executes the computer program, the method for identifying corrosion fatigue damage of an aluminum alloy as described above is implemented.

[0043] The present application provides a method, system, and device for identifying corrosion fatigue damage of aluminum alloys. The K-means++ algorithm is used to replace the traditional K-means clustering method, which effectively solves the problem of sensitivity to the initial clustering center. At the same time, the present invention also effectively combines the ACO and K-Means++ algorithms to form the ACO-Kmeans++ algorithm, further improving the ability to obtain the global optimal solution and significantly improving the computational efficiency and performance. This method makes the matching of AE signals and damage types more accurate, thereby greatly improving the matching accuracy and realizing the effective identification of corrosion fatigue damage of aviation aluminum alloys. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of the process provided by the present invention;

[0045] Figure 2 Schematic diagram of the AE monitoring system provided by the present invention

[0046] Figure 3 This is a schematic diagram of the process of step S3 provided by the present invention;

[0047] Figure 4 Schematic diagram of the AE parameter ACO-Kmeans++ clustering results provided by the present invention

[0048] Figure 5 This is the change curve of the number of cluster centers of the AE signal Silh index and CH index provided by the present invention. DETAILED DESCRIPTION

[0049] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0050] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0051] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in this article, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this article, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination". Furthermore, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising" and "including" indicate the presence of the described features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the presence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or", "and / or", "including at least one of the following" used in this application, etc. can be interpreted as inclusive, or mean any one or any combination. An exception to this definition will occur only when a combination of elements, functions, steps or operations are inherently mutually exclusive in some manner.

[0052] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0053] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0054] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] like Figure 1 As shown, the present application provides a method for identifying corrosion fatigue damage of aluminum alloys, which effectively solves the problem of sensitivity to the initial clustering center, makes the matching of AE signals and damage types more accurate, and greatly improves the matching accuracy.

[0056] Specifically, the method comprises the following steps:

[0057] Step S1, obtaining AE signal parameters during the aluminum alloy corrosion fatigue crack growth process, and preprocessing the AE signal parameters.

[0058] Aluminum alloy corrosion fatigue crack growth AE acquisition system was built, and compact tensile specimens were used to conduct Figure 2 The aluminum alloy corrosion fatigue crack growth experiment shown is used to obtain the AE signal parameter data during the aluminum alloy corrosion fatigue crack growth process.

[0059] Preprocessing the AE signal parameters includes:

[0060] Selecting AE signal parameters, wherein the AE signal parameters include amplitude, energy, and average frequency;

[0061] The AE signal parameters were normalized by z-score to eliminate the influence of different AE signal parameters.

[0062] Step S2: Use the ACO-Kmeans++ clustering algorithm to perform cluster analysis on the preprocessed AE signal parameters to determine the cluster center.

[0063] ACO stands for Ant Colony Optimization. Its basic principle is to simulate the behavior of ants in the process of searching for food, by releasing and sensing pheromones (a volatile substance) to choose paths. Ants leave pheromones on the road they walk on, and other ants choose paths by sensing the concentration of pheromones. The higher the pheromone concentration, the greater the probability of the path being selected. In this way, the ant colony can find the shortest path from the nest to the food source. The ACO algorithm has a strong global search capability and can search in the entire solution space, making it more likely to find the global optimal solution. It can also perform distributed computing to find the optimal clustering center and has better robustness.

[0064] Compared with the random selection of initial centers of K-means clustering, K-Means++ can select initial centers more evenly, reducing the risk of the algorithm converging to a local optimal solution and accelerating the convergence speed of the algorithm. Therefore, the combination of ACO and K-Means++ in this application can significantly optimize the selection of initial cluster centers, while improving the global optimization capability of the algorithm. By introducing the search mechanism of ACO, the combined algorithm can effectively solve the problem that the K-Means algorithm is sensitive to the initial cluster center, and reduce the risk of falling into a local optimal solution, thereby showing higher stability and accuracy in cluster analysis. The ACO-Kmeans++ algorithm is formed by combining the ACO and K-Means++ algorithms, which overcomes the shortcomings of the typical K-means algorithm clustering process that is prone to local optimal solutions and is sensitive to the initial cluster centers. An AE signal pattern recognition method for corrosion fatigue crack extension of aluminum alloys based on the ACO-Kmeans++ algorithm is constructed, which realizes the effective identification of corrosion fatigue damage of aviation aluminum alloys.

[0065] Specifically, Figure 3 As shown, step S2 includes:

[0066] Step S21, initializing clustering parameters, wherein the clustering parameters include at least one of the following: the number of clusters, the number of ant colonies, the number of iterations, the pheromone concentration, and the pheromone volatilization rate.

[0067] Determine the number of clusters K that need to be clustered. The range of K is 2 to the set value. The set value is the type of damage mode that exists in the corrosion fatigue crack propagation of aluminum alloy. The number of ants and the number of iterations depend on the scale and complexity of the problem. For small-scale simple problems, the number of ants is generally set to 10-50. The data collected in the experiment of this application did not reach a large scale, so the number of ants was set to 30 and the number of iterations was first set to 100. The initial pheromone concentration is usually set to a small positive value. This experiment is set to 0.1. The pheromone volatilization rate is usually in the range of [0,1]. A smaller volatilization rate will cause the pheromone to dissipate more slowly, which helps to maintain global exploration. The value of this experiment is 0.3.

[0068] The pheromone matrix is ​​used to record the pheromone concentration between data points. The formula for initializing the pheromone matrix is ​​as follows:

[0069] τ(x i ,x j )=τ0

[0070] In the formula, τ(x i ,x j ) represents the sample x i To sample x j The pheromone concentration of τ0 is a constant value.

[0071] Step S22, using the K-Means++ method to initialize the cluster center;

[0072] First, a cluster center is randomly selected, the shortest distance between the sample and the current cluster center is calculated, and the next cluster center is probabilistically determined based on the shortest distance.

[0073] The shortest distance between the sample and the current cluster center is calculated by Euclidean distance:

[0074]

[0075] In the formula, c k is the current cluster center data point, x i For sample.

[0076] Secondly, calculate the probability P(x) of each sample being selected as the next cluster center:

[0077]

[0078] X is the number of samples contained in the cluster under the current K value. Finally, the roulette wheel method is used to select the next cluster center. In the roulette wheel method, the probability is accumulated and summed to form a probability sequence, and a random number r between 0 and 1 is generated. According to the random number r, the corresponding sample is selected in the cumulative probability sequence as the next cluster center. Repeat the above steps until all cluster centers are determined.

[0079] Step S23, performing ant colony iteration; comprising:

[0080] Step S231: In each iteration, the next data point is selected according to the pheromone concentration and the roulette algorithm, and the ants are assigned to the nearest cluster center.

[0081] For each iteration, each ant selects the next data point based on the pheromone concentration and the roulette algorithm and assigns it to the cluster center closest to it. The probability of selecting the next data point is as follows:

[0082]

[0083] In the formula, τ ij is the residual pheromone concentration on the path from data point i to j, η ij is the heuristic information, α is the relative importance of the residual information, and β is the relative importance of the expected value. The values ​​of α and β can be adjusted according to experience or problem characteristics. ij is the probability of an ant going from data point i to data point j. X is the data point that the current ant has not reached. Based on the calculated probability distribution, the ant selects the next data point according to the roulette method.

[0084] Step S232: Update the pheromone matrix to strengthen the path that the ants pass through.

[0085] Update the pheromone matrix, specifically:

[0086] τ(x,c)=(1-ρ)×τ(x,c)+Δτ(x,c)

[0087] Where τ(x,c) represents the pheromone value from data point x to cluster center c, ρ is the pheromone volatility, and △τ(x,c) is the pheromone value updated in the current iteration.

[0088] Use (1-ρ)×τ(x,c) to weaken the previous path pheromone, and add it to the pheromone value of △τ(x,c) in the current iteration to get the pheromone concentration of the new path. Use this pheromone concentration to update the pheromone matrix.

[0089] Each ant completes a complete data point assignment process.

[0090] Step S24, updating the cluster center according to the allocation result of the ants in the ant colony iteration; judging whether the new cluster center converges to the old cluster center, if not, performing ant colony iteration.

[0091] Calculate the sum of pheromone concentrations on the paths of all data points in each cluster, count the number of data points contained in each cluster, and update the cluster center based on the number of data points.

[0092] For all data points in each cluster, calculate the sum of the pheromone concentrations on their paths and count the number of data points n contained in each cluster. For each cluster, update its cluster center c to the average value of the sum of the pheromone concentrations on the paths in the cluster. The calculation formula is as follows:

[0093]

[0094] That is, c k is the new cluster center.

[0095] Determine whether the new cluster center converges to the old cluster center. Specifically, calculate the distance D between the new cluster center and the old cluster center. (i,h) , if the distance between the new cluster center and the old cluster center is greater than the convergence threshold, the new cluster center converges to the old cluster center. Generally, the convergence threshold is set to a very small positive number d. If the convergence threshold d <D (i,j) , it converges, otherwise, the ant colony iteration continues.

[0096] Step S25: Use the new cluster center to perform clustering. Figure 4 As shown, the new cluster centers are substituted into the K-means clustering for clustering.

[0097] Step S3: testing the cluster centers by clustering result evaluation indicators to determine the optimal number of clusters.

[0098] Step S31, normalizing clustering result evaluation indicators, including silhouette coefficient (Silh index) and CH index (CH index), performing weighted summation according to a preset ratio to determine a comprehensive score of the cluster center.

[0099] Silh is an indicator used to evaluate clustering quality, which takes into account both the compactness of clustering (similarity of samples within a cluster) and the separation (distance between samples in different clusters). The calculation formula of the silhouette coefficient is as follows:

[0100]

[0101] In the formula, a(i) represents the similarity between the sample and other samples of the same class, and b(i) represents the dissimilarity between the sample and samples of other classes. The value range of the silhouette coefficient s(i) is between [-1,1]. Generally, the closer the silhouette coefficient is to 1, the better the clustering effect is; the closer it is to -1, the worse the clustering effect is.

[0102] The CH index is an indicator used to evaluate clustering quality, which is based on the compactness within the cluster and the separation between clusters. The larger the CH index, the smaller the difference within the cluster group, the larger the difference between the groups, and the better the clustering effect. The formula is as follows:

[0103]

[0104] Where Bk is the inter-cluster scatter matrix, which represents the differences between different clusters; Wk is the intra-cluster scatter matrix, which represents the differences between samples in the same cluster; Tr is the trace operation of the matrix, that is, the sum of the elements on the diagonal of the matrix; N is the total number of samples; and k is the number of clusters.

[0105] The clustering result evaluation index (CH index, Silh index) is used to judge the clustering results of different cluster center numbers K (2≤K≤set value), normalize the silhouette coefficient and CH index to the same [0-1] interval range, and perform weighted summation according to a certain ratio to obtain a comprehensive score. The calculation formula for the comprehensive score is as follows:

[0106]

[0107] In the formula, ω1+ω2=1 is satisfied, and the two are adjusted according to experience. In the formula, max(Silh) and min(Silh) represent the maximum and minimum values ​​of the silhouette coefficient, respectively, and max(CH) and min(CH) represent the maximum and minimum values ​​of the CH index, respectively.

[0108] Step S32: Determine the optimal number of clusters based on the comprehensive score of the cluster center. The cluster with the highest score is determined as the optimal number of clusters K. The curve of the cluster evaluation index changing with K is as follows: Figure 5 shown.

[0109] Step S4: determine the important parameter characteristic center value of the cluster center according to the clustering result, and match the aluminum alloy corrosion fatigue damage pattern to the AE signal in combination with the physical fracture characteristics.

[0110] The output parameters in the clustering results include the central values ​​of important signal parameters. The data characteristics of each type of AE signal represented by the central value of the signal parameter are analyzed. Combined with previous research results, the damage mode that the data characteristics of each type of AE signal may represent is inferred.

[0111] Furthermore, the fracture characteristics of the samples were observed using a scanning electron microscope to obtain the actual damage pattern, which was then compared with the characteristics obtained by cluster analysis. According to the corresponding relationship, different damage patterns were matched with the characteristics of various AE signals. The characteristic center value of the important parameter of the cluster center was obtained through the clustering results, and the corrosion fatigue damage pattern of aluminum alloys was matched for various AE signals in combination with the physical fracture characteristics.

[0112] Based on the same inventive concept, the present application also provides an aluminum alloy corrosion fatigue damage identification system, comprising:

[0113] An acquisition module is used to acquire AE signal parameters during the expansion of aluminum alloy corrosion fatigue cracks and to preprocess the AE signal parameters;

[0114] The cluster analysis module uses the ACO-Kmeans++ clustering algorithm to perform cluster analysis on the preprocessed AE signal parameters and determine the cluster center;

[0115] A clustering validity test module tests the cluster centers through clustering result evaluation indicators to determine the optimal number of clusters;

[0116] The damage pattern recognition module determines the characteristic center value of the important parameters of the cluster center according to the clustering results, and matches the aluminum alloy corrosion fatigue damage pattern of the AE signal in combination with the physical fracture characteristics.

[0117] This application provides a method for identifying corrosion fatigue damage of aluminum alloys. The K-means++ algorithm is used to replace the traditional K-means clustering method, which effectively solves the problem of sensitivity to the initial cluster center. At the same time, the present invention also effectively combines the ACO and K-Means++ algorithms to form the ACO-Kmeans++ algorithm, further improving the ability to obtain the global optimal solution and significantly improving the computational efficiency and performance. This method makes the matching of AE signals and damage types more accurate, thereby greatly improving the matching accuracy and realizing the effective identification of corrosion fatigue damage of aviation aluminum alloys.

[0118] Based on the same inventive concept, the present application also provides a device, comprising: a memory storing executable program code;

[0119] a processor coupled to the memory;

[0120] A transceiver used to communicate with other devices or communication networks and to receive or send network messages;

[0121] A bus used to connect memory, processors, and transceivers for internal communication.

[0122] The transceiver receives messages transmitted over the network and passes them to the processor through the bus. The processor calls the executable program code stored in the memory through the bus for processing, and passes the processing results to the transceiver through the bus for transmission, thereby realizing a method for identifying corrosion fatigue damage of aluminum alloy provided in an embodiment of the present application.

[0123] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the described method for identifying corrosion fatigue damage of an aluminum alloy.

[0124] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the described method for identifying corrosion fatigue damage of an aluminum alloy.

[0125] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0126] In the present application, the same or similar terminology concepts, technical solutions and / or application scenario descriptions are generally described in detail only the first time they appear. When they appear again later, they are generally not repeated for the sake of brevity. When understanding the technical solutions and other contents of the present application, for the same or similar terminology concepts, technical solutions and / or application scenario descriptions that are not described in detail later, reference can be made to the previous related detailed descriptions.

[0127] In the present application, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0128] The various technical features of the technical solution of the present application can be arbitrarily combined. In order to make the description concise, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present application.

[0129] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as above, and includes a number of instructions for a terminal device (which can be an electrical device or a network device, etc.) to execute the method of each embodiment of the present application.

[0130] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for identifying corrosion fatigue damage of aluminum alloy, characterized in that: The method comprises the following steps: Acquire AE signal parameters during the aluminum alloy corrosion fatigue crack growth process, and preprocess the AE signal parameters; ACO-Kmeans++ clustering algorithm is used to perform cluster analysis on the preprocessed AE signal parameters and determine the cluster center; Testing the cluster centers by clustering result evaluation index to determine the optimal number of clusters; According to the clustering results, the characteristic center value of the important parameter of the cluster center is determined, and the AE signal is matched with the aluminum alloy corrosion fatigue damage pattern in combination with the physical fracture characteristics. Among them, the ACO-Kmeans++ clustering algorithm is used to perform cluster analysis on the preprocessed AE signal parameters to determine the cluster center, including: Initialize clustering parameters, which include at least one of the following: number of clusters, number of ant colonies, number of iterations, pheromone concentration, and pheromone volatilization rate; the range of the number of clusters is 2 to a set value, and the set value is the type of damage mode existing in the corrosion fatigue crack propagation of aluminum alloy; the pheromone volatilization rate is in the interval [0, 1]; the pheromone matrix is ​​used to record the pheromone concentration between data points; the formula for initializing the pheromone matrix is ​​as follows: In the formula, Indicates that from the sample To the sample pheromone concentration; is a constant value; Initialize the cluster center using the K-Means++ method, including: randomly selecting a cluster center, calculating the shortest distance between the sample and the current cluster center, and probabilistically determining the next cluster center based on the shortest distance; calculating the probability of each sample being selected as the next cluster center; and selecting the next cluster center using the roulette wheel method until all cluster centers are determined; Perform ant colony iteration, including: in each iteration, select the next data point according to the pheromone concentration and roulette algorithm, and assign the ants to the nearest cluster center; update the pheromone matrix and strengthen the path passed by the ants; The cluster center is updated according to the distribution results of the ants in the ant colony iteration; whether the new cluster center converges to the old cluster center is determined, and if not, the ant colony iteration is performed, including: calculating the sum of the pheromone concentrations on the paths of all data points in each cluster, counting the number of data points contained in each cluster, updating the cluster center according to the number of data points, and for each cluster, updating its cluster center to the average value of the sum of the pheromone concentrations on the paths in the cluster; Use new cluster centers for clustering; The step of testing the cluster centers by using a clustering result evaluation index to determine the optimal number of clusters includes: The clustering result evaluation indicators including the silhouette coefficient and the CH index are normalized, weighted summed according to a preset ratio, and the comprehensive score of the cluster center is determined. The optimal number of clusters is determined according to the comprehensive score of the cluster center.

2. The method for identifying corrosion fatigue damage of aluminum alloy according to claim 1, characterized in that: Preprocessing the AE signal parameters includes: Selecting AE signal parameters, wherein the AE signal parameters include amplitude, energy, and average frequency; AE signal parameters were normalized by z-score.

3. The method for identifying corrosion fatigue damage of aluminum alloy according to claim 1, characterized in that: Determine whether the new cluster center converges to the old cluster center, including: The distance between the new cluster center and the old cluster center is calculated. If the distance between the new cluster center and the old cluster center is greater than the convergence threshold, the new cluster center converges to the old cluster center.

4. A system for identifying corrosion fatigue damage of aluminum alloy, characterized in that: include An acquisition module is used to acquire AE signal parameters during the expansion of aluminum alloy corrosion fatigue cracks and to preprocess the AE signal parameters; The cluster analysis module uses the ACO-Kmeans++ clustering algorithm to perform cluster analysis on the preprocessed AE signal parameters and determine the cluster center; A clustering validity test module tests the cluster centers through clustering result evaluation indicators to determine the optimal number of clusters; The damage pattern recognition module determines the important parameter characteristic center value of the cluster center based on the clustering results, and matches the aluminum alloy corrosion fatigue damage pattern of the AE signal in combination with the physical fracture characteristics; Among them, the cluster analysis module uses the ACO-Kmeans++ clustering algorithm to perform cluster analysis on the preprocessed AE signal parameters and determine the cluster center, including: Initialize clustering parameters, which include at least one of the following: number of clusters, number of ant colonies, number of iterations, pheromone concentration, and pheromone volatilization rate; the range of the number of clusters is 2 to a set value, and the set value is the type of damage mode existing in the corrosion fatigue crack propagation of aluminum alloy; the pheromone volatilization rate is in the interval [0, 1]; the pheromone matrix is ​​used to record the pheromone concentration between data points; the formula for initializing the pheromone matrix is ​​as follows: In the formula, Indicates that from the sample To the sample pheromone concentration; is a constant value; Initialize the cluster center using the K-Means++ method, including: randomly selecting a cluster center, calculating the shortest distance between the sample and the current cluster center, and probabilistically determining the next cluster center based on the shortest distance; calculating the probability of each sample being selected as the next cluster center; and selecting the next cluster center using the roulette wheel method until all cluster centers are determined; Perform ant colony iteration, including: in each iteration, select the next data point according to the pheromone concentration and roulette algorithm, and assign the ants to the nearest cluster center; update the pheromone matrix and strengthen the path passed by the ants; The cluster center is updated according to the distribution results of the ants in the ant colony iteration; it is determined whether the new cluster center converges to the old cluster center. If not, the ant colony iteration is performed, including: calculating the sum of the pheromone concentrations on the paths of all data points in each cluster, counting the number of data points contained in each cluster, updating the cluster center according to the number of data points, and for each cluster, updating its cluster center to the average value of the sum of the pheromone concentrations on the paths in the cluster. Use new cluster centers for clustering; The step of testing the cluster centers by using a clustering result evaluation index to determine the optimal number of clusters includes: The clustering result evaluation indicators including the silhouette coefficient and the CH index are normalized, weighted summed according to a preset ratio, and the comprehensive score of the cluster center is determined. The optimal number of clusters is determined according to the comprehensive score of the cluster center.

5. A device for identifying corrosion fatigue damage of aluminum alloy, characterized in that: include: Memory; processor; And a computer program stored in a memory and executable on a processor, wherein when the processor executes the computer program, the method for identifying corrosion fatigue damage of an aluminum alloy as described in any one of claims 1 to 3 is implemented.

Images