Dual-active flow management and control method for centers of two places

By introducing a penalty term, a Gaussian kernel function, and a median filtering method, the problems of poor traffic monitoring accuracy and unstable clustering in dual-site active-active traffic control were solved, achieving higher clustering accuracy and faster convergence, and improving the precision and stability of traffic control.

CN118981650APending Publication Date: 2024-11-19STATE GRID INFORMATION & TELECOMM BRANCH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411119957.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

Existing dual-center active-active traffic control methods suffer from poor traffic monitoring accuracy, excessive noise impact, inappropriate distance measurement, unstable clustering, insufficient data consideration, and inability to converge quickly during the clustering process.

Method used

To enhance robustness, a penalty term is introduced, a Gaussian kernel function is used to measure distance, representativeness values ​​and multiple constraints are introduced, and the membership matrix is ​​processed by median filtering. The clustering strictness and data point influence are dynamically adjusted to achieve fast convergence.

Benefits of technology

This improves the accuracy and robustness of clustering, reduces noise interference, and achieves precision and stability in traffic control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118981650A_ABST
    Figure CN118981650A_ABST
Patent Text Reader

Abstract

The invention discloses a two-place center active-active flow management and control method. The method comprises the steps of data acquisition, data preprocessing, presetting, clustering processing and two-place center active-active flow management and control. The invention belongs to the technical field of flow management and control, and particularly relates to a two-place center active-active flow management and control method, which introduces a clustering process into flow management and control and introduces a penalty term into an objective function, thereby enhancing the robustness to noise, ensuring the stability of a clustering center and improving the accuracy and robustness of clustering. Introducing a Gaussian kernel function to obtain a distance between a data point and a clustering center; data characteristics are comprehensively considered, and a representative value and multiple reduction are introduced, so that the algorithm has better adaptive ability; dynamically considering the distance between the data point and the clustering center and the representativeness of the data point; in the clustering judgment process, the membership matrix is processed through a median filtering method, noise interference is reduced, the stability and accuracy of a clustering result are improved, and then the accuracy of flow management and control is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and in particular to a dual-center active-active traffic control method. Background Art

[0002] The dual-active traffic control method of the two centers aims to achieve effective control of data traffic by analyzing the historical monitoring data and real-time monitoring data of the two data centers. However, the general traffic control method has problems such as poor traffic monitoring accuracy, excessive noise influence when using clustering, improper distance measurement, and unstable clustering; the general traffic control method has problems such as insufficient data consideration, which makes the clustering process unable to fully reflect the data characteristics and the update cannot converge quickly. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a two-center active-active traffic control method. In view of the problems of poor traffic monitoring accuracy, excessive noise influence when using clustering processing, improper distance measurement, and unstable clustering in general traffic control methods, this solution introduces the clustering process into traffic control, introduces a penalty term into the objective function, enhances the robustness to noise, ensures the stability of the cluster center, and improves the accuracy and robustness of clustering; introduces a Gaussian kernel function to obtain the distance between the data point and the cluster center, which can better measure the distance between the data point and the cluster center, and improves the accuracy of clustering; in view of the fact that general traffic control methods have insufficient data consideration, In order to solve the problem that the clustering process cannot fully reflect the data characteristics and the update cannot converge quickly, this solution comprehensively considers the data characteristics, introduces representative values ​​and multiple constraints, and can flexibly adjust the clustering strictness of the algorithm and the influence of data points on the cluster center, so that the algorithm has better adaptability; when updating the membership matrix, cluster center and representative value, the distance between the data point and the cluster center and the representativeness of the data point are dynamically considered, which can achieve convergence more quickly and effectively; in the clustering determination process, the membership matrix is ​​processed by the median filtering method, which can effectively smooth the data, reduce noise interference, and improve the stability and accuracy of the clustering results, thereby achieving the accuracy of traffic control.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides a dual-center active-active traffic control method, which includes the following steps:

[0005] Step S1: data collection;

[0006] Step S2: data preprocessing;

[0007] Step S3: pre-setting;

[0008] Step S4: clustering processing;

[0009] Step S5: Active-active traffic control in the two centers.

[0010] Furthermore, in step S1, the data collection is to collect historical monitoring data and real-time monitoring data of the two locations; the historical monitoring data and real-time monitoring data of the two locations include network traffic data, equipment health data, application traffic data, user behavior and business data, link quality monitoring data, geographic location information of the two locations and time data; the historical monitoring data of the two locations also includes traffic status; the traffic status includes normal status and abnormal status; the traffic status is used as a data label, and the data label does not participate in clustering operations.

[0011] Furthermore, in step S2, the data preprocessing is to perform data cleaning, data conversion, feature engineering processing and normalization processing on the collected data; the data cleaning is to process missing values, duplicate values ​​and outliers; the data conversion is to convert the data into vector form; the normalization processing is to normalize the data based on the maximum and minimum normalization method.

[0012] Further, in step S3, the presetting specifically includes the following steps:

[0013] Step S31: Set the objective function and introduce a penalty term to enhance the robustness to noise and ensure the stability of the cluster center. The formula used is as follows:

[0014] ;

[0015] Where J is the objective function; l is the data point index; q is the total number of data points; k is the cluster index; c is the total number of clusters; γ l is the data weight; u kl is the degree of membership of data point l to cluster k; t kl is the representative value of the data point l corresponding to cluster k; m is the index of fuzzy membership; is the index of the representative value; K(·) is the Gaussian kernel function; is the feature vector of data point l; v k is the center of cluster k; It is punitive;

[0016] Step S32: Initialization: initializing the cluster center and fuzzy membership based on the KPFCM algorithm;

[0017] Step S33: define the distance, introduce the Gaussian kernel function to obtain the distance between the data point and the cluster center, and the formula used is as follows:

[0018] ;

[0019] Where D klis the distance between data point l and cluster center k;

[0020] Step S34: Define the penalty term, the formula used is as follows:

[0021] ;

[0022] In the formula, is a penalty item; (·) limits the data points in the brackets to belong only to the kth cluster.

[0023] Furthermore, in step S4, the clustering process is to cluster the historical monitoring data and the real-time monitoring data of the two locations as data sets, which specifically includes the following steps:

[0024] Step S41: Initialize the membership matrix based on the distance defined in step S33;

[0025] Step S42: Update the cluster center. The formula used is as follows:

[0026] ;

[0027] In the formula, is the updated k-th cluster center;

[0028] Step S43: Update the membership matrix. The formula used to update the membership is as follows:

[0029] ;

[0030] In the formula, is the updated membership degree; v j is the jth cluster center;

[0031] Step S44: Update the representative value, which is only used for clustering determination; the formula used is as follows:

[0032] ;

[0033] In the formula, is the updated representative value;

[0034] Step S45: Clustering determination, a target threshold is set in advance. If the target function is lower than the target threshold, clustering ends; if the membership matrix converges or reaches the maximum number of iterations, clustering is reinitialized; otherwise, median filtering is performed on the membership matrix and the process returns to step S42; the formula used for median filtering is as follows:

[0035] ;

[0036] In the formula, and are the membership matrices after and before median filtering respectively; med is the median filtering operation.

[0037] Furthermore, in step S5, the active-active traffic control of the two centers is to establish an active-active architecture in the two locations; when clustering is completed, the cluster label is determined based on the number of labels of historical data, and the cluster label to which the real-time data belongs is used as the traffic status monitoring result of the real-time data; when the traffic status detection result of the data center in one location is abnormal, the traffic is switched to the active data center in another location; so as to realize active-active traffic control of the two centers.

[0038] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0039] (1) In view of the problems of poor traffic monitoring accuracy, excessive noise influence when using clustering, improper distance measurement, and unstable clustering in general traffic control methods, this scheme introduces the clustering process into traffic control and introduces a penalty term into the objective function, which enhances the robustness to noise, ensures the stability of the cluster center, and improves the accuracy and robustness of clustering; the Gaussian kernel function is introduced to obtain the distance between the data point and the cluster center, which can better measure the distance between the data point and the cluster center and improve the accuracy of clustering.

[0040] (2) In view of the problem that general traffic control methods do not fully consider data, resulting in the clustering process being unable to fully reflect data characteristics and the update being unable to converge quickly, this solution comprehensively considers data characteristics, introduces representative values ​​and multiple constraints, and can flexibly adjust the algorithm's clustering strictness and the influence of data points on the cluster center, making the algorithm more adaptive. When updating the membership matrix, cluster center and representative value, the distance between the data point and the cluster center and the representativeness of the data point are dynamically considered, which can achieve convergence more quickly and effectively. In the clustering determination process, the membership matrix is ​​processed by the median filtering method, which can effectively smooth the data, reduce noise interference, and improve the stability and accuracy of the clustering results, thereby achieving the accuracy of traffic control. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a flow chart of a dual-center active-active traffic control method provided by the present invention;

[0042] Figure 2 is a schematic flow chart of step S3;

[0043] Figure 3 It is a schematic diagram of the process of step S4.

[0044] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0046] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0047] Example 1, see Figure 1 The present invention provides a dual-center active-active traffic control method for two locations, the method comprising the following steps:

[0048] Step S1: Data collection, collecting historical monitoring data and real-time monitoring data of the two locations;

[0049] Step S2: Data preprocessing, data cleaning, data conversion, feature engineering and normalization of the collected data;

[0050] Step S3: presetting, introducing a penalty term to set the objective function, and introducing a Gaussian kernel function to obtain the distance between the data point and the cluster center;

[0051] Step S4: Clustering processing, clustering the historical monitoring data and real-time monitoring data of the two places as data sets, introducing representative values, and dynamically considering the distance between the data point and the cluster center and the representativeness of the data point when updating the membership matrix, cluster center and representative value;

[0052] Step S5: Active-active traffic control in the two centers.

[0053] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the historical monitoring data and real-time monitoring data of the two places include network traffic data, equipment health data, application traffic data, user behavior and business data, link quality monitoring data, geographic location information of the two places and time data; the historical monitoring data of the two places also include traffic status; the traffic status includes normal status and abnormal status; the traffic status is used as a data label, and the data label does not participate in the clustering operation.

[0054] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S2, data cleaning is to process missing values, duplicate values ​​and outliers; data conversion is to convert data into vector form; and normalization is to normalize data based on the maximum and minimum normalization method.

[0055] Example 4, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S3, the presetting specifically includes the following steps:

[0056] Step S31: Set the objective function and introduce a penalty term to enhance the robustness to noise and ensure the stability of the cluster center. The formula used is as follows:

[0057] ;

[0058] Where J is the objective function; l is the data point index; q is the total number of data points; k is the cluster index; c is the total number of clusters; γ l is the data weight; u kl is the degree of membership of data point l to cluster k; t kl is the representative value of the data point l corresponding to cluster k; m is the index of fuzzy membership; is the index of the representative value; K(·) is the Gaussian kernel function; is the feature vector of data point l; v k is the center of cluster k; It is punitive;

[0059] Step S32: Initialization: initializing the cluster center and fuzzy membership based on the KPFCM algorithm;

[0060] Step S33: define the distance, introduce the Gaussian kernel function to obtain the distance between the data point and the cluster center, and the formula used is as follows:

[0061] ;

[0062] Where D kl is the distance between data point l and cluster center k;

[0063] Step S34: Define the penalty term, the formula used is as follows:

[0064] ;

[0065] In the formula, is a penalty item; (·) limits the data points in the brackets to belong only to the kth cluster.

[0066] By performing the above operations, this solution introduces the clustering process into traffic control and the penalty term into the objective function to address the problems of poor traffic monitoring accuracy, excessive noise influence when using clustering, improper distance measurement, and unstable clustering in general traffic control methods. This enhances the robustness to noise, ensures the stability of the cluster center, and improves the accuracy and robustness of clustering. The Gaussian kernel function is introduced to obtain the distance between the data point and the cluster center, which can better measure the distance between the data point and the cluster center and improve the accuracy of clustering.

[0067] Example 5, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S4, the clustering process specifically includes the following steps:

[0068] Step S41: Initialize the membership matrix based on the distance defined in step S33;

[0069] Step S42: Update the cluster center. The formula used is as follows:

[0070] ;

[0071] In the formula, is the updated k-th cluster center;

[0072] Step S43: Update the membership matrix. The formula used to update the membership is as follows:

[0073] ;

[0074] In the formula, is the updated membership degree; v j is the jth cluster center;

[0075] Step S44: Update the representative value, which is only used for clustering determination; the formula used is as follows:

[0076] ;

[0077] In the formula, is the updated representative value;

[0078] Step S45: Clustering determination, a target threshold is set in advance. If the target function is lower than the target threshold, clustering ends; if the membership matrix converges or reaches the maximum number of iterations, clustering is reinitialized; otherwise, median filtering is performed on the membership matrix and the process returns to step S42; the formula used for median filtering is as follows:

[0079] ;

[0080] In the formula, and are the membership matrices after and before median filtering respectively; med is the median filtering operation.

[0081] By performing the above operations, in order to solve the problem that general traffic control methods do not take data into consideration sufficiently, resulting in the clustering process being unable to fully reflect data characteristics and the update being unable to converge quickly, this solution comprehensively considers data characteristics, introduces representative values ​​and multiple constraints, and can flexibly adjust the algorithm's clustering strictness and the influence of data points on the cluster center, making the algorithm more adaptive; when updating the membership matrix, cluster center and representative value, the distance between the data point and the cluster center and the representativeness of the data point are dynamically considered, which can achieve convergence more quickly and effectively; in the clustering determination process, the membership matrix is ​​processed by the median filtering method, which can effectively smooth the data, reduce noise interference, and improve the stability and accuracy of the clustering results, thereby achieving the accuracy of traffic control.

[0082] Example 6, see Figure 1 This embodiment is based on the above embodiment. In step S5, the active-active traffic control of the two centers is to establish an active-active architecture in the two locations; when clustering is completed, the cluster label is determined based on the number of labels of historical data, and the cluster label to which the real-time data belongs is used as the traffic status monitoring result of the real-time data; when the traffic status detection result of a data center in one location is abnormal, the traffic is switched to the active data center in another location; so as to realize the active-active traffic control of the two centers.

[0083] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" 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 other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0084] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0085] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A dual-center active-active traffic control method, characterized in that: The method comprises the following steps: Step S1: data collection; Step S2: data preprocessing; Step S3: presetting, introducing penalty terms to set the objective function, and introducing Gaussian kernel function to obtain the distance between the data point and the cluster center; Step S4: Clustering processing, clustering the historical monitoring data and real-time monitoring data of the two places as data sets, introducing representative values, and dynamically considering the distance between the data point and the cluster center and the representativeness of the data point when updating the membership matrix, cluster center and representative value; Step S5: Active-active traffic control in the two centers.

2. According to claim 1, a dual-center active-active traffic control method is characterized by: In step S3, the presetting specifically includes the following steps: Step S31: Set the objective function and introduce the penalty term; the formula used is as follows: ; Where J is the objective function; l is the data point index; q is the total number of data points; k is the cluster index; c is the total number of clusters; γ l is the data weight; u kl is the degree of membership of data point l to cluster k; t kl is the representative value of the data point l corresponding to cluster k; m is the index of fuzzy membership; is the index of the representative value; K(·) is the Gaussian kernel function; is the feature vector of data point l; v k is the center of cluster k; It is punitive; Step S32: Initialization: initializing the cluster center and fuzzy membership based on the KPFCM algorithm; Step S33: define the distance, introduce the Gaussian kernel function to obtain the distance between the data point and the cluster center, and the formula used is as follows: ; Where D kl is the distance between data point l and cluster center k; Step S34: Define the penalty term, the formula used is as follows: ; In the formula, is a penalty item; (·) limits the data points in the brackets to belong only to the kth cluster.

3. According to claim 1, a dual-center active-active traffic control method is characterized by: In step S4, the clustering process specifically includes the following steps: Step S41: Initialize the membership matrix based on the distance defined in step S33; Step S42: Update the cluster center. The formula used is as follows: ; In the formula, is the updated k-th cluster center; Step S43: Update the membership matrix. The formula used to update the membership is as follows: ; In the formula, is the updated membership degree; v j is the jth cluster center; Step S44: Update the representative value, which is only used for clustering determination; the formula used is as follows: ; In the formula, is the updated representative value; Step S45: Clustering determination, a target threshold is set in advance. If the target function is lower than the target threshold, clustering ends; if the membership matrix converges or reaches the maximum number of iterations, clustering is reinitialized; otherwise, median filtering is performed on the membership matrix and the process returns to step S42; the formula used for median filtering is as follows: ; In the formula, and are the membership matrices after and before median filtering respectively; med is the median filtering operation.

4. According to claim 1, a dual-center active-active traffic control method is characterized by: In step S1, the data collection is to collect historical monitoring data and real-time monitoring data of the two locations; the historical monitoring data and real-time monitoring data of the two locations include network traffic data, equipment health data, application traffic data, user behavior and business data, link quality monitoring data, geographic location information of the two locations and time data; the historical monitoring data of the two locations also includes traffic status; the traffic status includes normal status and abnormal status; the traffic status is used as a data label, and the data label does not participate in clustering operations.

5. According to claim 1, a dual-center active-active traffic control method is characterized by: In step S5, the active-active traffic control of the two centers is to set up an active-active architecture in the two locations; when clustering is completed, the cluster label is determined based on the number of labels of the historical data, and the cluster label to which the real-time data belongs is used as the traffic status monitoring result of the real-time data; when the traffic status detection result of the data center in one location is abnormal, the traffic is switched to the active data center in another location; To achieve active-active traffic control in two centers.

6. A dual-center active-active traffic control method according to claim 1, characterized in that: In step S2, the data preprocessing is to perform data cleaning, data conversion, feature engineering processing and normalization processing on the collected data; the data cleaning is to process missing values, duplicate values ​​and outliers; The data conversion is to convert the data into a vector form; the normalization processing is to normalize the data based on the maximum and minimum normalization method.