A Method and System for Intelligent Analysis of Connector Data

Abstract

The present invention relates to the technical field of data processing, and particularly relates to a method and system for intelligent analysis of connector data. The method includes the steps of: processing the parameter items of each data point in the connector data set through a mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, and recording the data points within the same clustering cluster as similar data points; obtaining the neighborhood of the data point with the data point as the center; obtaining the central directivity of the data point through the numerical difference between each parameter item of the data point before and after each iteration and the corresponding parameter items of the clustering center; calculating the abnormality degree of the data point; and obtaining the analysis result of the connector data through the comparison result between the abnormality degree of the data point and a preset threshold, effectively improving the accuracy of the obtained analysis result of the connector.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a method and system for intelligent analysis of connector data. Background Art

[0002] A connector is an electronic component for current or signal transmission and exchange between electronic system devices. As a node, it realizes connection through the plugging and unplugging of a plug and a socket. For example, in a plug-and-play connector, a USB (Universal Serial Bus) connector, as a standard interface for connecting between a computer and peripheral devices, can not only realize data transmission but also provide power support for the connected devices.

[0003] In the prior art, there is a lot of research and application content on USB connectors. For example, the patent application document with the publication number CN115934604A discloses a USB circuit, a switching method and device for a USB controller. The USB circuit in this application can accurately and quickly determine the problematic device through the mutual operation and influence of the enumeration situation of the USB device and the power detection circuit, saving the time for determining the problematic device and effectively improving the reliability of the USB interface.

[0004] The above prior art can determine the problematic device by setting a USB circuit. However, if there are problems with the quality of the connector itself, it will also affect the quality judgment result of the related connected devices, reducing the accuracy of the quality judgment. Moreover, in daily use, the quality of the connector also determines the service life of the related devices.

[0005] Based on this, how to accurately obtain the quality detection result of the connector is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0006] To solve the technical problem of how to accurately obtain the quality detection result of the connector, the present invention provides a method and system for intelligent analysis of connector data.

[0007] In the first aspect, the present invention provides a method for intelligent analysis of connector data, adopting the following technical scheme:

[0008] Process the parameter items of each data point in the connector dataset through the mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, and record the data points within the same clustering cluster as the same type of data points; obtain the neighborhood of the data point with the data point as the center; obtain the central direction degree of the data point through the numerical difference between each parameter item of the data point before and after each iteration and the corresponding parameter items of the clustering center. ; is the degree of abnormality of the i-th data point. is the Euclidean distance between the i-th data point and the corresponding cluster center, is the average Euclidean distance between all data points in the cluster where the i-th data point is located and the cluster center, and are the central direction degrees of the i-th data point and the -th data point of the same type within the neighborhood of the i-th data point respectively, is the number of data points of the same type within the neighborhood of the i-th data point, is the Euclidean distance between the i-th data point and the -th data point of the same type within its neighborhood, is the exponential function with base e; the analysis result of the connector data is obtained through the comparison result between the abnormality degree of the data point and the preset threshold.

[0009] When analyzing the data of the connector, the present invention clusters each data point in the connector dataset through the mean shift clustering algorithm, and can accurately identify the abnormal data points deviating from the cluster, so as to accurately obtain the connectors with quality abnormalities. In this process, the present invention considers that although some data points can be divided into the cluster, they do not always tend to the cluster center during the clustering process and may be abnormal data points; based on this, the present invention accurately obtains the abnormality degree of each data point by obtaining the tendency degree of each data point to the cluster center during the iteration process and the difference between the Euclidean distance between the initial position of the data point and the cluster center, so as to accurately identify the connector data with quality abnormalities in the cluster, reduce the influence of such abnormal data points on the detection result, and effectively improve the accuracy of the obtained analysis result of the connector.

[0010] According to an intelligent analysis method for connector data provided by the present invention, before the parameter items of each data point in the connector dataset are processed by the mean shift clustering algorithm, it further includes: forming a data point with the values of the parameter items corresponding to each connector, and obtaining the connector dataset after preprocessing.

[0011] The present invention considers that there may be data missing etc. in the originally collected connector data, so the overall quality of the data is improved through data preprocessing.

[0012] According to an intelligent analysis method for connector data provided by the present invention, obtaining the cluster center of the cluster where each data point is located by processing the parameter items of each data point in the connector dataset through the mean shift clustering algorithm includes: calculating the weighted average within the preset bandwidth of the data point through the Gaussian kernel function, and obtaining the mean shift vector of the data point according to the difference between the weighted average and the data point; moving the data point along the corresponding mean shift vector direction; repeating the position movement step until the preset iteration termination condition is reached, and obtaining the cluster center corresponding to each data point.

[0013] The present invention obtains the clustering results of each data point in the connector dataset through the mean shift clustering algorithm, so as to accurately identify the abnormal data points deviating from the clustering clusters.

[0014] According to an intelligent analysis method for connector data provided by the present invention, the parameters of each data point in the connector dataset are processed through the mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, and further includes: removing the data points without corresponding clustering clusters as abnormal data points.

[0015] According to an intelligent analysis method for connector data provided by the present invention, obtaining the neighborhood of the data point with the data point as the center includes: presetting the neighborhood radius of the data point; obtaining other data points within the neighborhood radius of the data point as the neighborhood of the data point.

[0016] According to an intelligent analysis method for connector data provided by the present invention, the central directivity of the data point satisfies the relational expression:

[0017] ;

[0018] is the central directivity of the i-th data point, is the number of iterations of the mean shift clustering algorithm, is the number of parameter items in the data point, is the weight of the j-th parameter item, 、 are respectively the values of the j-th parameter item before and after the -th iteration of the i-th data point, is the value of the j-th parameter item of the clustering center corresponding to the i-th data point.

[0019] The present invention provides an accurate calculation method for the central directivity of data points. By analyzing the deviation degree between the data points before and after iteration and the parameter items of the clustering center, the displacement difference between the data points before and after iteration and the clustering center is obtained, so as to accurately obtain the tendency degree of each data point to the clustering center.

[0020] According to an intelligent analysis method for connector data provided by the present invention, the method for obtaining the weight of parameter items includes: obtaining the weight of each parameter item through the analytic hierarchy process.

[0021] According to an intelligent analysis method for connector data provided by the present invention, obtaining the analysis result of the connector data through the comparison result between the abnormal degree of the data point and the preset threshold includes: if the abnormal degree of the data point is greater than the preset threshold, the analysis result of the connector corresponding to the data point is abnormal; otherwise, the analysis result of the connector corresponding to the data point is normal.

[0022] A method for intelligent analysis of connector data provided by the present invention, after obtaining the analysis result of the connector data, further includes: in response to the analysis result of the connector corresponding to the data point being abnormal, marking the abnormal connector and the value of its corresponding parameter item, and then storing them in the abnormal database.

[0023] Considering the importance of connector data analysis in daily life and production, the present invention separately creates a database for abnormal data to facilitate subsequent data analysis.

[0024] In a second aspect, the present invention provides an intelligent analysis system for connector data, adopting the following technical solution:

[0025] An intelligent analysis system for connector data includes: a processor and a memory, and the memory stores computer program instructions, which implement the above-mentioned method for intelligent analysis of connector data when executed by the processor.

[0026] By adopting the above technical solution, the above-mentioned method for intelligent analysis of connector data is generated into a computer program and stored in the memory to be loaded and executed by the processor, so as to manufacture a terminal device according to the memory and the processor, which is convenient to use.

[0027] The present invention has the following technical effects:

[0028] Based on the above technical solution, when analyzing the data of the connector, the present invention clusters each data point in the connector data set through the mean shift clustering algorithm, and can accurately identify the abnormal data points deviating from the clustering cluster, so as to accurately obtain the connectors with quality abnormalities. In this process, the present invention considers that although some data points can be divided into the clustering cluster, they do not tend to the clustering center during the clustering process; based on this, the present invention accurately obtains the abnormality degree of each data point by obtaining the tendency degree of each data point to the clustering center during the iteration process and the difference in the Euclidean distance between the initial position of the data point and the clustering center, so as to accurately identify the connector data with quality abnormalities in the clustering cluster, reduce the influence of such abnormal data points on the detection result, and effectively improve the accuracy of the obtained analysis result of the connector. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become easily understood. In the drawings, several embodiments of the present invention are shown in an exemplary rather than restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0030] Figure 1Schematic flowchart of a method for intelligent analysis of connector data provided by an embodiment of the present invention. Detailed implementation manners

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0032] It should be understood that when terms such as "first" and "second" are used in the claims, the description, and the drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "comprising" and "including" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0033] A connector is an electronic component for current or signal transmission and exchange between electronic system devices. As a node, it realizes connection through the plugging and unplugging of a plug and a socket. For example, in a plug-and-play connector, a USB connector, as a standard interface for connecting a computer and peripheral devices, can not only transmit data but also provide power support for the connected devices.

[0034] If there are problems with the quality of the connector itself, it will affect the quality judgment results and daily use of related connected devices. However, there are a large number of connector types and a large number of related parameter items for evaluating the quality of connectors. If abnormal data in each parameter item is directly identified through threshold judgment, normal connectors may be identified as abnormal, thereby reducing the accuracy of abnormal identification.

[0035] The mean shift clustering algorithm is a density-based non-parametric clustering algorithm. This algorithm iteratively updates the positions of data points by obtaining the fastest direction in which the density of data points increases, causing the data points to move towards regions with higher density until they converge to the local density maximum. In this way, the algorithm can automatically discover the cluster structure of the data without the need to pre-set the number of clusters, thereby accurately identifying abnormal data points that do not belong to any cluster class.

[0036] Based on this, an embodiment of the present invention discloses a method for intelligent analysis of connector data. This method processes connector data through the mean shift clustering algorithm, so that connectors with poor quality can be accurately obtained from a large number of connector data.

[0037] Specifically, refer to Figure 1 as shown.Figure 1 The flowchart shows a method for intelligent analysis of connector data provided by an embodiment of the present invention. The method specifically includes the following steps.

[0038] S1: Obtain the parameter items corresponding to each data point in the connector dataset.

[0039] Among them, the parameter items corresponding to each data point can be the physical dimensions of the connector, temperature resistance data, humidity resistance data, shock resistance, and service life, etc., which can be specifically set according to actual needs.

[0040] It should be noted that the physical dimensions of the connector determine the volume and shape of the connector, which in turn affect its insertion and extraction force and transmission performance; the operating temperature range of the connector refers to the temperature range in which it can operate normally. Good temperature resistance can ensure that the connector operates normally in various temperature environments; the intrusion of moisture has a great impact on the insulation performance of the connector, easily corrodes metal parts, and increases the risk of leakage and short circuit. The better the humidity resistance of the connector, the higher the corresponding quality; shock resistance is the key for the connector to maintain a stable connection in complex application environments. Good shock resistance can increase the service life of the connector and reduce the maintenance and replacement frequency; the service life of the connector is the primary indicator to measure its performance reliability. The longer the service life, the better the quality of the connector.

[0041] Based on this, the embodiment of the present invention uses the mean shift clustering algorithm to obtain the clustering result of the parameter items of each data point to evaluate the quality of the connector, so as to accurately identify the abnormal data points that do not belong to any clustering cluster, and the connectors with quality abnormalities corresponding to the abnormal data points.

[0042] Exemplarily, in the embodiment of the present invention, before processing the parameter items of each data point in the connector dataset by the mean shift clustering algorithm, it further includes: forming a data point with the values of the parameter items corresponding to each connector, and obtaining the connector dataset after preprocessing.

[0043] Among them, the preprocessing method can be missing data interpolation, data format conversion, etc., which can be specifically set according to actual needs. The embodiment of the present invention does not limit this too much here.

[0044] It can be understood that a data point is the parameter items corresponding to a connector, and the connector dataset is a set of numerous connectors and their corresponding parameter items.

[0045] After obtaining the parameter items corresponding to each data point based on the above steps, for the convenience of data processing, the parameter items can also be normalized, which can be specifically set according to actual needs.

[0046] S2: Process the parameter items of each data point in the connector dataset through the mean shift clustering algorithm to obtain the clustering center of the cluster where each data point is located, and record the data points within the same cluster as data points of the same type; obtain the neighborhood of the data point with the data point as the center.

[0047] Exemplarily, in the embodiment of the present invention, processing the parameter items of each data point in the connector dataset through the mean shift clustering algorithm to obtain the clustering center of the cluster where each data point is located includes: calculating the weighted average within the preset bandwidth of the data point through the Gaussian kernel function, and obtaining the mean shift vector of the data point according to the difference between the weighted average and the data point; moving the data point along the corresponding mean shift vector direction; repeating the position movement step until the preset iteration termination condition is reached to obtain the clustering center corresponding to each data point.

[0048] Among them, the preset bandwidth can be specifically set according to actual needs.

[0049] Exemplarily, the preset iteration termination condition can be that the change amount of the clustering center position is less than the preset threshold and / or the preset number of iterations is reached.

[0050] Among them, the preset threshold and the preset number of iterations can be specifically set according to actual needs. The specific steps of obtaining the clustering center of the cluster where each data point is located through the mean shift clustering algorithm can be implemented according to the prior art, and the embodiments of the present invention will not elaborate herein.

[0051] Exemplarily, in the embodiment of the present invention, obtaining the neighborhood of the data point with the data point as the center includes: presetting the neighborhood radius of the data point; obtaining other data points within the neighborhood radius of the data point as the neighborhood of the data point.

[0052] Among them, the size of the data point neighborhood radius can be set to half of the average Euclidean distance between all data points and the clustering center in the cluster where the data point is located; the size of the neighborhood radius can be specifically set according to actual needs, and the embodiments of the present invention will not impose too many restrictions herein.

[0053] It can be understood that obtaining other data points within the neighborhood radius of the data point as the neighborhood of the data point above means the data points whose Euclidean distance from the data point is less than or equal to the neighborhood radius, and the obtained neighborhood range of the data point includes at least the data point itself.

[0054] Based on the above steps, using the mean shift clustering algorithm to process the parameter items of each data point in the connector dataset, multiple clusters and data points that do not belong to any cluster can be obtained. The data points that do not belong to any cluster are connectors with abnormal quality.

[0055] By way of example, in an embodiment of the present invention, the parameter items of each data point in the connector data set are processed by a mean shift clustering algorithm to obtain the cluster center of the cluster where each data point is located, and also includes: eliminating data points without corresponding clusters as abnormal data points.

[0056] In this way, the embodiment of the present invention processes each data point in the connector data set through the mean shift clustering algorithm, and can accurately filter out abnormal data points that do not belong to any clustering cluster. The identification result of the connector corresponding to the abnormal data point is quality abnormality.

[0057] It should be noted that the abnormal data points identified by the mean shift clustering algorithm are data points that are obviously discrete and free from the clusters. Normal data points will gradually tend to their cluster centers during the iteration process. However, although some data points can be classified into clusters, they do not always tend to their cluster centers during the process of iteratively obtaining cluster centers, but go back and forth during the iteration process. Although such data points are classified into the corresponding clusters, they may also be abnormal data.

[0058] Based on this, the embodiment of the present invention further analyzes the degree of tendency between each data point and the cluster center during the process of iteratively obtaining its cluster center, thereby accurately identifying abnormal data points in the cluster cluster, that is, continuing to perform the following steps.

[0059] S3: The central directivity of the data point is obtained by the numerical difference between each parameter item of the data point and each parameter item of the corresponding cluster center before and after each iteration.

[0060] It should be noted that by analyzing the displacement of the data point with respect to the cluster center during the iterative clustering process, the degree to which the data point points to the cluster center can be obtained. However, the importance of the parameter items corresponding to the data points for the connector quality assessment is different, and the degree of influence on the calculation of the abnormality degree during the iteration process is also different.

[0061] Based on this, the embodiment of the present invention obtains the weight of each parameter item when calculating the degree of abnormality, and weights the degree of deviation between each parameter item of the data point and the cluster center parameter item based on the weight, so as to accurately obtain the center directivity of each data point. The higher the center directivity, the higher the possibility that the data point is a normal data point.

[0062] For example, in an embodiment of the present invention, a method for obtaining weights of parameter items includes: obtaining the weights of each parameter item by using a hierarchical analysis method.

[0063] Specifically, when obtaining the weights of each parameter item through the analytic hierarchy process, the connector quality assessment can be used as the target layer, each parameter item as the criterion layer, and the specific values of each parameter item as the parameter layer to construct a hierarchical structure model; the 1-9 scale method is used to represent the relative importance between each parameter item, each pair of parameter items is compared, and corresponding scores are set according to their influence on the connector quality to construct a judgment matrix; each column of the judgment matrix is normalized and summed by row, and the sum of each row is divided by the sum of the sums of all rows to finally obtain the weights of each parameter item.

[0064] It can be understood that the specific steps of obtaining the weights of each parameter item through the analytic hierarchy process can be implemented by the prior art, and the embodiments of the present invention will not be elaborated herein.

[0065] Exemplarily, in the embodiments of the present invention, to determine the central directionality of a data point, reference can be specifically made to the following relational expression:

[0066] ;

[0067] is the central directionality of the i-th data point, is the number of iterations of the mean shift clustering algorithm, is the number of parameter items in the data point, is the weight of the j-th parameter item, is the value of the j-th parameter item of the i-th data point before the -th iteration, is the value of the j-th parameter item of the i-th data point after the -th iteration, is the value of the j-th parameter item of the clustering center corresponding to the i-th data point.

[0068] In the above formula, represents the difference between the j-th parameter item of the i-th data point before the -th iteration and the j-th parameter item of its clustering center, represents the difference between the j-th parameter item of the i-th data point after the -th iteration and the j-th parameter item of its clustering center. The larger the ratio of the two, the more the j-th parameter item of the i-th data point after the -th iteration tends to the j-th parameter item of the clustering center corresponding to the i-th data point, the greater the displacement before and after iteration, and the higher the possibility that the i-th data point tends to the clustering center after the -th iteration, and the higher the corresponding central directionality.

[0069] The weights of the parameter items are used to weight the tendency degrees of each parameter item and the clustering center when determining the central directionality of the data point, so as to accurately obtain the degree to which the current data point tends to the clustering center.

[0070] After obtaining the central orientation degrees of each data point based on the above steps, continue to execute the following steps.

[0071] S4: Correct the central orientation degree of the data point through the Euclidean distance difference between the initial position of the data point and the corresponding cluster center, and obtain the abnormality degree of the data point.

[0072] It should be noted that through the above steps, by analyzing the displacement difference between the parameter items of the data point before and after iteration and the parameter items of the cluster center, the central orientation degree of the data point can be obtained. The data points near the cluster center have a relatively high probability of being normal data points. However, the displacement difference between such data points before and after iteration and the cluster center will be relatively small, resulting in a lower central orientation degree of the obtained data points.

[0073] Based on this, the embodiments of the present invention obtain the Euclidean distance between the initial position of each data point and the cluster center, and correct the central orientation degree of the data point through the Euclidean distance, so as to accurately obtain the true abnormality degree of each data point in the cluster.

[0074] Exemplarily, in the embodiments of the present invention, to determine the abnormality degree of a data point, specifically, the following relational expression can be referred to:

[0075] ;

[0076] is the abnormality degree of the i-th data point, is the Euclidean distance between the i-th data point and the corresponding cluster center, is the average value of the Euclidean distances between all data points in the cluster where the i-th data point is located and the cluster center, is the central orientation degree of the i-th data point, is the central orientation degree of the -th similar data point in the neighborhood of the i-th data point, is the number of similar data points in the neighborhood of the i-th data point, is the Euclidean distance between the i-th data point and the -th similar data point in its neighborhood, is the exponential function with e as the base.

[0077] In the above formula, represents the ratio of the Euclidean distance between the i-th data point and the corresponding cluster center to the average value of the Euclidean distances between all data points in the cluster where the i-th data point is located and the cluster center. The larger this value is, it indicates that compared with the data points in the cluster where the i-th data point is located, the Euclidean distance between the i-th data point and the corresponding cluster center is farther, that is, the probability that the i-th data point is an abnormal data point is higher.

[0078] It represents the ratio of the centrality of the j-th data point of the same type within the neighborhood of the i-th data point to the centrality of the i-th data point. The larger this value is, it indicates that the centrality of the i-th data point is smaller relative to the data points of the same type within the neighborhood of the i-th data point, and the higher the probability that the i-th data point is an abnormal data point, and the greater the corresponding degree of abnormality. It represents the weighted average of the ratio of the centrality between the i-th data point and each data point of the same type within the neighborhood through the Euclidean distance between data points. The larger this value is, it indicates that the difference between the i-th data point and each data point of the same type within the neighborhood is greater, and the greater the probability that this data point is an abnormal data point, and the greater the corresponding degree of abnormality.

[0079] After obtaining the degree of abnormality of each data point in the clustering cluster based on the above steps, the abnormal data in the clustering cluster can be accurately identified based on the degree of abnormality of the data points.

[0080] S5: Obtain the analysis result of the connector data through the comparison result between the degree of abnormality of the data point and the preset threshold.

[0081] Among them, the preset threshold can be set to 0.9; the preset threshold can be specifically set according to actual needs, and the embodiments of the present invention do not limit this too much here.

[0082] Exemplarily, in the embodiments of the present invention, the analysis result of the connector data is obtained through the comparison result between the degree of abnormality of the data point and the preset threshold, including: if the degree of abnormality of the data point is greater than the preset threshold, the analysis result of the connector corresponding to this data point is abnormal; otherwise, the analysis result of the connector corresponding to this data point is normal.

[0083] It can be understood that the degree of abnormality of the data point being greater than the preset threshold indicates that the quality of the connector corresponding to this data point is poor. In order to facilitate the analysis of the parameter items of the connector with poor quality, the abnormal connector with poor quality and the corresponding data can be associated and stored for subsequent data analysis.

[0084] Exemplarily, in the embodiments of the present invention, after obtaining the analysis result of the connector data, it further includes: in response to the analysis result of the connector corresponding to the data point being abnormal, after marking the abnormal connector and the value of its corresponding parameter item, store them in the abnormal database.

[0085] Among them, the specific steps of marking and associating and storing the abnormal connector and the value of its corresponding parameter item can be implemented through existing technologies, and the embodiments of the present invention do not elaborate here.

[0086]

[0087] ​It can be seen that in the embodiment of the present invention, when analyzing the data of the connector, the parameter items of each data point in the connector dataset can be processed by the mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, and the data points within the same clustering cluster are recorded as similar data points; a neighborhood of the data point is obtained with the data point as the center; the central directivity of the data point is obtained through the numerical difference between each parameter item of the data point before and after each iteration and the corresponding parameter items of the clustering center.

[0088] ; is the degree of abnormality of the i-th data point, is the Euclidean distance between the i-th data point and the corresponding clustering center, is the average Euclidean distance between all data points in the clustering cluster where the i-th data point is located and the clustering center, and are the central directivities of the i-th data point and the -th similar data point within the neighborhood of the i-th data point respectively, is the number of similar data points within the neighborhood of the i-th data point, is the Euclidean distance between the i-th data point and the -th similar data point within its neighborhood, is the exponential function with e as the base; the analysis result of the connector data is obtained through the comparison result between the degree of abnormality of the data point and the preset threshold.

[0089] In this way, in the embodiment of the present invention, each data point in the connector dataset is clustered by the mean shift clustering algorithm, and the abnormal data points deviating from the clustering cluster can be accurately identified, so as to accurately obtain the connectors with abnormal quality. In this process, the embodiment of the present invention takes into account that although some data points can be divided into the clustering cluster, they do not tend to the clustering center during the clustering process; based on this, the embodiment of the present invention accurately obtains the degree of abnormality of each data point by obtaining the degree of tendency of each data point to the clustering center during the iteration process and the difference in the Euclidean distance between the initial position of the data point and the clustering center, so that the abnormal connector data with quality problems can be accurately identified in the clustering cluster, reducing the influence of such abnormal data points on the detection result, and effectively improving the accuracy of the obtained analysis result of the connector.

[0090] The embodiment of the present invention also discloses an intelligent analysis system for connector data, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, an intelligent analysis method for connector data provided by the present invention is implemented.

[0091] The above system also includes other components well known to those skilled in the art such as a communication bus and a communication interface, and their settings and functions are known in the art, so they will not be described in detail here.

[0092] In the present invention, the foregoing memory may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as, resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0093] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many variations, changes, and alternative ways will occur to those skilled in the art without departing from the spirit and scope of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

[0094] The above are all the preferred embodiments of the present invention, and the protection scope of the present invention is not limited hereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A method for intelligent analysis of connector data, characterized in that Including: Processing the parameter items of each data point in the connector dataset through the mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, and recording the data points within the same clustering cluster as the same type of data points; obtaining the neighborhood of the data point with the data point as the center; the parameter items of each data point are the physical size, temperature resistance data, humidity resistance data, seismic resistance, and service life of the connector. Obtaining the central directivity of the data point through the numerical difference between each parameter item of the data point before and after each iteration and the corresponding parameter items of the clustering center. ； is the anomaly degree of the i-th data point, is the Euclidean distance between the i-th data point and the corresponding cluster center, is the average Euclidean distance between all data points in the cluster where the i-th data point is located and the cluster center, and are the centrality degrees of the i-th data point and the -th data point of the same class within the neighborhood of the i-th data point respectively, is the number of data points of the same class within the neighborhood of the i-th data point, is the Euclidean distance between the i-th data point and the -th data point of the same class within its neighborhood, is the exponential function with base e; Obtaining the analysis result of the connector data through the comparison result between the abnormal degree of the data point and the preset threshold. The central directivity of the data point satisfies the relational expression: ； is the centrality of the $i$-th data point, is the number of iterations of the mean-shift clustering algorithm, is the number of parameter terms in the data point, is the weight of the $j$-th parameter term, 、 are respectively the values of the $j$-th parameter term before and after the -th iteration of the $i$-th data point, is the value of the $j$-th parameter term of the cluster center corresponding to the $i$-th data point; By obtaining the weight of each parameter item when calculating the abnormal degree and weighting the deviation degree between each parameter item of the data point and the parameter items of the clustering center based on the weight, the central directivity of each data point can be accurately obtained. The higher the central directivity, the higher the possibility that the data point is a normal data point.

2. The intelligent analysis method for connector data according to claim 1, wherein Before processing the parameter items of each data point in the connector dataset through the mean shift clustering algorithm, it further includes: Forming a data point with the value of the parameter item corresponding to each connector, and obtaining the connector dataset after preprocessing.

3. A method for intelligent analysis of connector data according to claim 2, characterized in that, Processing the parameter items of each data point in the connector dataset through the mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, including: Calculating the weighted average value within the preset bandwidth of the data point through the Gaussian kernel function, obtaining the mean shift vector of the data point according to the difference between the weighted average value and the data point; moving the data point along the direction of the corresponding mean shift vector; repeating the position movement step until the preset iteration termination condition is reached, and obtaining the clustering center corresponding to each data point.

4. The intelligent analysis method for connector data according to claim 3, wherein Processing the parameter items of each data point in the connector dataset through the mean shift clustering algorithm to obtain the clustering center of the clustering cluster where each data point is located, and further includes: Eliminating the data points without corresponding clustering clusters as abnormal data points.

5. A method for intelligent analysis of connector data according to claim 1, characterized in that, Obtaining the neighborhood of the data point with the data point as the center, including: Presetting the neighborhood radius of the data point; obtaining other data points within the neighborhood radius of the data point as the neighborhood of the data point.

6. The method for intelligent analysis of connector data according to claim 1, wherein The method for obtaining the weight of the parameter item includes: Obtaining the weight of each parameter item through the analytic hierarchy process.

7. A method for intelligent analysis of connector data according to claim 1, characterized in that Obtaining the analysis result of the connector data through the comparison result between the abnormal degree of the data point and the preset threshold, including: If the abnormal degree of the data point is greater than the preset threshold, the analysis result of the connector corresponding to the data point is abnormal; otherwise, the analysis result of the connector corresponding to the data point is normal.

8. A method for intelligent analysis of connector data according to claim 7, characterized in that After obtaining the analysis result of the connector data, it further includes: In response to the analysis result of the connector corresponding to the data point being abnormal, marking the abnormal connector and its corresponding parameter item values, and storing them in the abnormal database.

9. An intelligent connector data analysis system, characterized in that Including: A processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, it realizes a method for intelligent analysis of connector data according to any one of claims 1-8.

Images