Private network management method, device and equipment based on multi-source heterogeneous data fusion analysis

Abstract

The present application discloses a private network management method, device and equipment based on multi-source heterogeneous data fusion analysis, the method comprising: when it is detected that multiple base stations of a multi-mode wireless private network fail in the same period, the base station data of each failed base station is obtained; according to the base station data of each failed base station, the difficulty of network fault repair in each frequency band is preset, and the maintenance proficiency of each maintenance personnel is preset, a multi-source heterogeneous data array is constructed; the multi-source heterogeneous data array and the preset fault repair priority rules are input into a pre-trained fault repair scheduling model, and the maintenance personnel information corresponding to each failed base station is output; based on the maintenance personnel information corresponding to each failed base station, each maintenance personnel is dispatched. Therefore, the present application can improve the fault response speed of the multi-mode wireless private network, realize the optimal allocation of operation and maintenance resources when large-scale base station failures occur, reduce the operation and maintenance costs, further ensure the stability and reliability of the entire network, and thus improve the fault repair efficiency.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a private network management method, device and equipment based on multi-source heterogeneous data fusion analysis. Background Art

[0002] In modern cities, the stability and reliability of wireless communication networks are crucial to ensuring public safety, traffic management, emergency services, etc. Multimode wireless private network is an intelligent network system that integrates edge computing and cloud computing technologies, that is, a communication network under the edge-cloud unified heterogeneous technology architecture, which can provide efficient and secure data transmission services. However, in large-scale multimode network systems, such as 350MHz, 370MHz, 800MHz, 1.4GHz, 1.8GHz, 5.8GHz, and 6GHz, when multiple base stations fail at the same time, it is particularly important to quickly repair the entire network failure within the time limit with a limited number of maintenance personnel.

[0003] In the related art, when multiple base stations fail at the same time, in order to quickly repair the failures, as many maintenance personnel as possible are generally dispatched. The maintenance personnel are dispatched to repair the failures based on a single dimension such as the distance of the maintenance personnel from the failures and the urgency of the failures. As a result, when the maintenance personnel are limited, the failures cannot be quickly repaired in the shortest time, which prolongs the failure repair time and reduces the failure repair efficiency. Summary of the invention

[0004] The embodiments of the present application provide a method, device and equipment for managing a private network based on multi-source heterogeneous data fusion analysis. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0005] In a first aspect, an embodiment of the present application provides a private network management method based on multi-source heterogeneous data fusion analysis, the method comprising:

[0006] When multiple base stations of a multi-mode wireless private network are detected to have faults in the same period of time, base station data of each faulty base station is obtained;

[0007] A multi-source heterogeneous data array is constructed based on the base station data of each faulty base station, the difficulty of repairing network faults in each frequency band, and the maintenance proficiency of each maintenance personnel;

[0008] Input the multi-source heterogeneous data array and the preset fault repair priority rules into the pre-trained fault repair scheduling model, and output the maintenance personnel information corresponding to each faulty base station; the preset fault repair priority rules are used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array;

[0009] Based on the maintenance personnel information corresponding to each faulty base station, each maintenance personnel is dispatched.

[0010] Optionally, a multi-source heterogeneous data array is constructed according to the base station information of each faulty base station, the difficulty of repairing the network fault in each frequency band, and the maintenance proficiency of each maintenance personnel, including:

[0011] Perform data cleaning and data format standardization on the base station information of each faulty base station to obtain pre-processed data of each faulty base station;

[0012] Based on the preset matrix parameters and the number of maintenance personnel, an original matrix is ​​constructed;

[0013] According to the pre-processed data of each faulty base station and the difficulty of repairing the network faults in each frequency band, the maintenance proficiency of each maintenance personnel is fused to obtain the comprehensive embedded vector representation of each maintenance personnel;

[0014] The comprehensive embedding vector representation of each maintenance personnel is embedded into each element position in the original matrix to obtain a multi-source heterogeneous data array.

[0015] Optionally, the pre-processed data of each faulty base station includes base station location information, base station life cycle status information, base station impact information on users, and road congestion information within a preset range near the base station;

[0016] According to the pre-processed data of each faulty base station and the difficulty of repairing the network faults in each frequency band, the maintenance proficiency of each maintenance personnel is fused to obtain the comprehensive embedded vector representation of each maintenance personnel, including:

[0017] Get the employee information of each maintenance personnel;

[0018] Mapping the employee information of each maintenance personnel and the preset maintenance proficiency of each maintenance personnel into the same vector space to obtain the basic vector of each maintenance personnel;

[0019] The base station location information, base station life cycle status information, base station impact information on users, road congestion information within a preset range near the base station, and the difficulty of repairing network faults in each preset frequency band of each faulty base station are mapped into the same vector space to obtain the basic vector of each faulty base station;

[0020] Taking the basis vector of each maintenance personnel as the central vector, the basis vector of each faulty base station is mapped as a sub-vector to each central vector one by one to generate a comprehensive embedded vector representation of each maintenance personnel.

[0021] Optionally, the multi-source heterogeneous data array is composed of comprehensive embedding vector representations of each maintenance personnel, the comprehensive embedding vector representations of each maintenance personnel include the basis vectors of the maintenance personnel and the basis vectors of each faulty base station, the basis vectors of the maintenance personnel are generated based on the employee information of the maintenance personnel and the maintenance proficiency of the preset maintenance personnel, and the basis vectors of each faulty base station are generated based on the base station information of each faulty base station; the pre-trained fault maintenance scheduling model includes an optimization target determination layer, a master node determination layer, a slave node determination layer, a comprehensive similarity calculation layer, a final slave node determination layer, a corresponding relationship determination layer, and an information output layer;

[0022] The multi-source heterogeneous data array and the preset fault repair priority rules are input into the pre-trained fault repair scheduling model, and the maintenance personnel information corresponding to each faulty base station is output, including:

[0023] The optimization target determination layer determines the optimization target based on the preset fault repair priority rules. The optimization target is used to minimize the total repair time or maximize resource utilization or balance the workload, and is used to constrain the number of maintenance personnel or skill matching requirements or geographical location restrictions;

[0024] The master node determination layer uses the basic vector of the maintenance personnel included in the comprehensive embedding vector representation of each maintenance personnel as the master node;

[0025] The slave node determination layer uses the basis vectors of each faulty base station included in the comprehensive embedding vector representation of each maintenance personnel as multiple slave nodes of the master node;

[0026] The comprehensive similarity calculation layer calculates the comprehensive similarity between the master node and each slave node on the master node;

[0027] The final slave node determination layer determines the final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, so as to obtain the optimal matching relationship between the maintenance personnel and the faulty base station;

[0028] The correspondence determination layer determines the correspondence between the maintenance personnel and the faulty base station one by one according to the final slave node corresponding to the master node;

[0029] The information output layer outputs the maintenance personnel information corresponding to each faulty base station based on the corresponding relationship between the maintenance personnel and the faulty base stations.

[0030] Optionally, determining a final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node includes:

[0031] Construct an objective function for the optimization goal, which is to minimize the total maintenance time or maximize the maintenance efficiency;

[0032] According to the comprehensive similarity between the master node and each slave node on the master node and the objective function, the total maintenance time or maintenance efficiency between the master node and each slave node on the master node is calculated;

[0033] The slave node corresponding to the minimum total maintenance time or the maximum maintenance efficiency is determined as the final slave node corresponding to the master node.

[0034] Optionally, the objective function is:

[0035]

[0036] Where MinZ is the total maintenance time or maintenance efficiency, m is the number of maintenance personnel, n is the number of faulty base stations, c ij is the cost of maintenance personnel i to repair faulty base station j, and the cost is the comprehensive similarity between the master node and each slave node on the master node, x ij is a decision variable; among them,

[0037] The constraint of the decision variables is: each faulty base station can only be repaired by one maintenance personnel: Each maintenance personnel can only repair one faulty base station:

[0038]

[0039] Optionally, a pre-trained fault maintenance scheduling model is generated by following the steps below, including:

[0040] Collect historical base station data of historical faulty base stations that have experienced historical faults;

[0041] Use neural network to build fault maintenance scheduling model;

[0042] Construct an objective function that minimizes the total maintenance time or maximizes the maintenance efficiency;

[0043] Integrate the objective function into the fault maintenance scheduling model;

[0044] By presetting the fault repair priority rules and objective functions, the historical base station data is labeled with the optimal maintenance personnel information to obtain the model training data;

[0045] The model training data is input into the fault maintenance scheduling model that integrates the objective function for machine learning to generate a pre-trained fault maintenance scheduling model.

[0046] In a second aspect, an embodiment of the present application provides a private network management device based on multi-source heterogeneous data fusion analysis, the device comprising:

[0047] The base station data acquisition module is used to acquire the base station data of each failed base station when multiple base stations in the multi-mode wireless private network are detected to have failed in the same period of time;

[0048] A multi-source heterogeneous data array construction module is used to construct a multi-source heterogeneous data array according to the base station data of each faulty base station, the difficulty of repairing the network faults of each frequency band, and the maintenance proficiency of each maintenance personnel;

[0049] A maintenance personnel information output module is used to input a multi-source heterogeneous data array and a preset fault repair priority rule into a pre-trained fault repair scheduling model, and output maintenance personnel information corresponding to each faulty base station; the preset fault repair priority rule is used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array;

[0050] The maintenance personnel scheduling module is used to schedule each maintenance personnel based on the maintenance personnel information corresponding to each faulty base station.

[0051] In a third aspect, an embodiment of the present application provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0052] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0053] In an embodiment of the present application, a multi-source heterogeneous data array and a preset fault repair priority rule can be input into a pre-trained fault repair scheduling model for processing, and then the maintenance personnel information corresponding to each faulty base station can be output. On the one hand, the multi-source heterogeneous data array integrates base station information, repair difficulty, and maintenance personnel information. The data provided by different data sources have different information and perspectives, which increases the depth and breadth of data analysis, thereby improving the reliability of the analysis results; on the other hand, the pre-trained fault repair scheduling model can fit the multi-source heterogeneous data array into master-slave nodes to quickly determine the corresponding relationship between maintenance personnel and faulty base stations, thereby improving the efficiency of fault handling; on the other hand, the preset fault repair priority rule is used to remind the model to output the best result according to the set priority strategy, which ensures the flexibility of the model and enables the model to output different optimal solutions under different scenario requirements, thereby avoiding the singleness of the analysis dimension.

[0054] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0056] Figure 1 It is a flowchart of a private network management method based on multi-source heterogeneous data fusion analysis provided by an embodiment of the present application;

[0057] Figure 2 This is a schematic diagram of relevant data of a base station provided in an embodiment of the present application;

[0058] Figure 3 It is a schematic diagram of a comprehensive embedding vector of a maintenance personnel provided in an embodiment of the present application;

[0059] Figure 4 It is a schematic diagram of the composition structure of a model provided in an embodiment of the present application;

[0060] Figure 5 This is a maintenance personnel allocation diagram provided by an embodiment of the present application;

[0061] Figure 6 This is a schematic diagram of maintenance personnel scheduling implementation provided by an embodiment of the present application;

[0062] Figure 7 It is a process diagram of a model training process provided by an embodiment of the present application;

[0063] Figure 8 It is a structural schematic diagram of a private network management device based on multi-source heterogeneous data fusion analysis provided by an embodiment of the present application;

[0064] Fig. 9 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] The following description and the accompanying drawings sufficiently illustrate specific embodiments of the present application to enable those skilled in the art to practice them.

[0066] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0067] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0068] In the description of the present application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0069] The present application provides a private network management method, device and equipment based on multi-source heterogeneous data fusion analysis to solve the problems existing in the above-mentioned related technical problems. In the technical solution provided by the present application, the multi-source heterogeneous data array and the preset fault repair priority rules can be input into the pre-trained fault maintenance scheduling model for processing and then output the maintenance personnel information corresponding to each faulty base station. On the one hand, the multi-source heterogeneous data array integrates base station information, repair difficulty and maintenance personnel information. The data provided by different data sources have different information and perspectives, which increases the depth and breadth of data analysis, thereby improving the reliability of the analysis results; on the other hand, the pre-trained fault maintenance scheduling model can fit the multi-source heterogeneous data array to the master-slave node to quickly determine the corresponding relationship between the maintenance personnel and the faulty base station, thereby improving the efficiency of fault handling; on the other hand, the preset fault repair priority rules are used to remind the model to output the best result according to the set priority strategy, which ensures the flexibility of the model, so that the model can output different optimal solutions under different scenario requirements, thereby avoiding the singleness of the analysis dimension. The following exemplary embodiments are used for detailed description.

[0070] The following will be combined with the attached Figure 1 -Attached Figure 7 , a private network management method based on multi-source heterogeneous data fusion analysis provided by an embodiment of the present application is introduced in detail. The method can be implemented by a computer program and can be run on a private network management device based on multi-source heterogeneous data fusion analysis based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application.

[0071] See also Figure 1, which is a flowchart of a private network management method based on multi-source heterogeneous data fusion analysis for an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:

[0072] S101, when it is detected that multiple base stations of the multi-mode wireless private network fail in the same period, base station data of each failed base station is obtained;

[0073] Among them, the multi-mode wireless private network is an intelligent network system that integrates edge computing and cloud computing technologies, that is, a communication network under the edge-cloud unified heterogeneous technology architecture, which can provide efficient and secure data transmission services. Base station refers to a fixed device used to transmit and receive signals in a multi-mode wireless private network, and is a core component of the multi-mode wireless private network. The same period refers to a period of time, which can be set based on the actual scenario, such as 6 minutes or 12 minutes. Base station data includes the original base station location information, the original base station life cycle status information, the original base station impact information on users, and the original road congestion information within a preset range near the base station.

[0074] In an embodiment of the present application, the original base station location information can be obtained by following the GPS or other positioning systems on the base station; the original base station life cycle status information can be collected through the maintenance log and status monitoring system of the base station; the original base station impact information on the user can be obtained through user feedback, service interruption reports, etc.; the original road congestion information within a preset range near the base station can be obtained through a traffic monitoring system or third-party traffic APIs.

[0075] Specifically, GPS or other positioning system users perform real-time positioning, and store the positioning data and base station numbers as a mapping relationship and send them to the server for storage; the maintenance log and status monitoring system are used to record the operating status of each base station, and generate relevant status information and send it to the server for storage; the traffic monitoring system or third-party traffic APIs are used to analyze the congestion information of each road in real time.

[0076] Furthermore, in this application, the server can obtain relevant data by accessing a positioning system, a status monitoring system, a traffic monitoring system or third-party traffic APIs, and the acquisition of relevant data complies with privacy regulations.

[0077] Among them, the relevant data of the base station is as follows: Figure 2As shown, the base station ID can uniquely identify the number of each base station; the location information identifies the geographic coordinates or address of the base station, which can be used for map positioning and navigation; the activation date can identify the date when the base station started operating; the last maintenance date can identify the date when the base station was last maintained or inspected; the status information can describe the current operating status of the base station, such as "normal operation", "fault", "pending maintenance", etc.; the user impact can describe the impact of base station failure or performance degradation on user services, which is usually divided into levels such as "high", "medium", and "low"; the road congestion information can identify the congestion of roads around the base station, which may affect the speed at which maintenance personnel can reach the base station, and can also be divided into levels such as "mild", "moderate", and "severe".

[0078] S102, constructing a multi-source heterogeneous data array according to base station data of each faulty base station, the difficulty of repairing the network fault in each frequency band, and the maintenance proficiency of each maintenance personnel;

[0079] Among them, the difficulty of repairing network faults in each frequency band can be defined based on historical data and expert knowledge; the maintenance proficiency of each maintenance personnel can be determined through skill assessment and historical maintenance records.

[0080] In an embodiment of the present application, in the specific process of constructing a multi-source heterogeneous data array based on the base station information of each faulty base station, the preset difficulty of repairing network faults in each frequency band, and the preset maintenance proficiency of each maintenance personnel, it includes: performing data cleaning and data format standardization processing on the base station information of each faulty base station to obtain pre-processed data of each faulty base station; constructing an original matrix based on preset matrix parameters and the number of maintenance personnel; performing data fusion on the preset maintenance proficiency of each maintenance personnel based on the pre-processed data of each faulty base station and the preset difficulty of repairing network faults in each frequency band to obtain a comprehensive embedded vector representation of each maintenance personnel; embedding the comprehensive embedded vector representation of each maintenance personnel into each element position in the original matrix to obtain a multi-source heterogeneous data array.

[0081] Among them, data cleaning and data format standardization are key steps in data preprocessing to ensure data quality and consistency. The preset matrix parameters can limit the basic information of the matrix, and the number of maintenance personnel can limit the number of elements in the matrix. The comprehensive embedded vector representation can normalize multi-source data and complete the final generation of elements in the original matrix.

[0082] In some embodiments, the specific process of data cleaning and data format standardization of the base station information of each faulty base station includes: identifying missing values ​​in the base station information of each faulty base station and deciding how to handle them, such as filling missing values, deleting records, or using statistical methods to estimate missing values; checking whether there are repeated base station IDs or other key information in the base station information of each faulty base station, and deciding how to handle these duplicates, such as deleting or merging; identifying abnormal values ​​or outliers in the base station information of each faulty base station, and deciding how to handle them, such as correcting, deleting or marking; ensuring that the data type of the data column in the base station information of each faulty base station is correct, such as converting the date column to a date format and converting the digital ID to an integer or a string; unifying the data types of each faulty base station The format of the data in the base station information of each faulty base station, such as the date format (YYYY-MM-DD), time format (HH:MM:SS), etc., is standardized for text data of the base station information of each faulty base station, such as the base station location information, such as unifying the case and removing extra spaces or special characters; the status information of the base station information of each faulty base station, such as "normal operation", "fault", and "pending maintenance", is coded and converted into numerical values ​​or classification labels; the qualitative description of the user impact of the base station information of each faulty base station is converted into a quantitative indicator, such as using a scale of 1-5 or 1-10; the road congestion information of the base station information of each faulty base station is quantified, for example, using a metric from 0 (no congestion) to 1 (severe congestion).

[0083] The pre-processed data of each faulty base station includes base station location information, base station life cycle status information, base station impact information on users, and road congestion information within a preset range near the base station.

[0084] In an embodiment of the present application, data fusion is performed on the preset maintenance proficiency of each maintenance personnel based on the pre-processed data of each faulty base station and the preset difficulty of repairing the network faults in each frequency band, and the specific process of obtaining a comprehensive embedded vector representation of each maintenance personnel includes: obtaining employee information of each maintenance personnel; mapping the employee information of each maintenance personnel and the preset maintenance proficiency of each maintenance personnel to the same vector space to obtain a basic vector of each maintenance personnel; mapping the base station location information of each faulty base station, the base station life cycle status information, the influence information of the base station on the user, the road congestion information within a preset range near the base station, and the preset difficulty of repairing the network faults in each frequency band to the same vector space to obtain the basic vector of each faulty base station; taking the basic vector of each maintenance personnel as the central vector, mapping the basic vector of each faulty base station as a sub-vector one by one to each central vector to generate a comprehensive embedded vector representation of each maintenance personnel.

[0085] For example Figure 3As shown, the present application can obtain a comprehensive embedded vector representation of each maintenance personnel by associating the basic vector of each faulty base station with the basic vector of the maintenance personnel. The comprehensive embedded vector representation of each maintenance personnel serves as an element in the matrix, thereby providing a basis for data analysis for the model.

[0086] S103, inputting the multi-source heterogeneous data array and the preset fault repair priority rule into the pre-trained fault repair scheduling model, and outputting the maintenance personnel information corresponding to each faulty base station; the preset fault repair priority rule is used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array;

[0087] Among them, the multi-source heterogeneous data array is composed of the comprehensive embedded vector representation of each maintenance personnel. The comprehensive embedded vector representation of each maintenance personnel includes the basic vector of the maintenance personnel and the basic vector of each faulty base station. The basic vector of the maintenance personnel is generated based on the employee information of the maintenance personnel and the preset maintenance proficiency of the maintenance personnel. The basic vector of each faulty base station is generated based on the base station information of each faulty base station.

[0088] For example Figure 4 As shown, the pre-trained fault maintenance scheduling model includes an optimization target determination layer, a master node determination layer, a slave node determination layer, a comprehensive similarity calculation layer, a final slave node determination layer, a corresponding relationship determination layer, and an information output layer.

[0089] In an embodiment of the present application, a multi-source heterogeneous data array and a preset fault repair priority rule are input into a pre-trained fault repair scheduling model, and a specific process of outputting maintenance personnel information corresponding to each faulty base station includes: an optimization target determination layer is used to determine an optimization target based on a preset fault repair priority rule, and the optimization target is used to minimize the total repair time or maximize resource utilization or balance the workload, and to constrain the number of maintenance personnel or skill matching requirements or geographical location restrictions; a master node determination layer is used to represent the basic vector of the maintenance personnel contained in the comprehensive embedding vector representation of each maintenance personnel as the master node; a slave node determination layer is used to represent the comprehensive embedding vector representation of each maintenance personnel The basic vectors of each faulty base station are included as multiple slave nodes of the master node; the comprehensive similarity calculation layer is used to calculate the comprehensive similarity between the master node and each slave node on the master node; the final slave node determination layer is used to determine the final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, so as to obtain the optimal matching relationship between the maintenance personnel and the faulty base station; the correspondence determination layer is used to determine the correspondence between the maintenance personnel and the faulty base station one by one according to the final slave node corresponding to the master node; the information output layer is used to output the maintenance personnel information corresponding to each faulty base station based on the correspondence between the maintenance personnel and the faulty base station.

[0090] In an embodiment of the present application, the specific process of determining the final slave node corresponding to the master node according to the optimization objective and the comprehensive similarity between the master node and each slave node on the master node includes: constructing an objective function of the optimization objective, the objective function is to minimize the total maintenance time or maximize the maintenance efficiency; calculating the total maintenance time or maintenance efficiency between the master node and each slave node on the master node according to the comprehensive similarity between the master node and each slave node on the master node and the objective function; determining the slave node corresponding to the minimum total maintenance time or the maximum maintenance efficiency as the final slave node corresponding to the master node.

[0091] Specifically, the objective function is:

[0092]

[0093] Where MinZ is the total maintenance time or maintenance efficiency, m is the number of maintenance personnel, n is the number of faulty base stations, c ij is the cost of maintenance personnel i to repair faulty base station j, and the cost is the comprehensive similarity between the master node and each slave node on the master node, x ij is a decision variable; among them,

[0094] The constraint of the decision variables is: each faulty base station can only be repaired by one maintenance personnel: Each maintenance personnel can only repair one faulty base station:

[0095]

[0096] In an embodiment of the present application, a pre-trained fault repair scheduling model can be generated according to the following steps: collecting historical base station data of historical faulty base stations where historical faults occurred; using a neural network to build a fault repair scheduling model; building an objective function that minimizes the total repair time or maximizes the repair efficiency; integrating the objective function into the fault repair scheduling model; by pre-setting fault repair priority rules and objective functions, labeling the historical base station data with the optimal maintenance personnel information to obtain model training data; inputting the model training data into the fault repair scheduling model that integrates the objective function for machine learning to generate a pre-trained fault repair scheduling model.

[0097] S104: dispatching maintenance personnel based on the maintenance personnel information corresponding to each faulty base station.

[0098] Among them, the maintenance personnel information corresponding to each faulty base station is as follows: Figure 5 shown.

[0099] In some embodiments, based on the maintenance personnel information corresponding to each faulty base station, the process of scheduling each maintenance personnel specifically includes: generating scheduling instructions according to the maintenance personnel information corresponding to each faulty base station, and clarifying the dispatch task of each maintenance personnel; sending the scheduling instructions to the corresponding maintenance personnel in real time through mobile devices or communication systems; monitoring the dynamics and fault repair progress of the maintenance personnel, and making scheduling adjustments when necessary; recording the completion status of maintenance tasks, and collecting feedback from maintenance personnel for improving future scheduling strategies; conducting post-analysis of completed maintenance tasks to evaluate scheduling efficiency and resource utilization.

[0100] Among them, the dynamics of maintenance personnel and the progress of fault repair, for example Figure 6 shown.

[0101] In an embodiment of the present application, a multi-source heterogeneous data array and a preset fault repair priority rule can be input into a pre-trained fault repair scheduling model for processing, and then the maintenance personnel information corresponding to each faulty base station can be output. On the one hand, the multi-source heterogeneous data array integrates base station information, repair difficulty, and maintenance personnel information. The data provided by different data sources have different information and perspectives, which increases the depth and breadth of data analysis, thereby improving the reliability of the analysis results; on the other hand, the pre-trained fault repair scheduling model can fit the multi-source heterogeneous data array into master-slave nodes to quickly determine the corresponding relationship between maintenance personnel and faulty base stations, thereby improving the efficiency of fault handling; on the other hand, the preset fault repair priority rule is used to remind the model to output the best result according to the set priority strategy, which ensures the flexibility of the model and enables the model to output different optimal solutions under different scenario requirements, thereby avoiding the singleness of the analysis dimension.

[0102] See also Figure 7 , which is a flow chart of a method for training an event classification model according to an embodiment of the present application. Figure 7 As shown, the method of the embodiment of the present application may include the following steps:

[0103] S201, collecting historical base station data of historical faulty base stations that have historically failed;

[0104] S202, constructing a fault maintenance scheduling model using a neural network;

[0105] S203, constructing an objective function that minimizes the total maintenance time or maximizes the maintenance efficiency;

[0106] S204, integrating the objective function into the fault maintenance scheduling model;

[0107] S205, annotating the optimal maintenance personnel information on the historical base station data by presetting the fault repair priority rule and the objective function to obtain the model training data;

[0108] S206, inputting the model training data into the fault maintenance scheduling model integrated with the objective function for machine learning, and generating a pre-trained fault maintenance scheduling model.

[0109] In an embodiment of the present application, a multi-source heterogeneous data array and a preset fault repair priority rule can be input into a pre-trained fault repair scheduling model for processing, and then the maintenance personnel information corresponding to each faulty base station can be output. On the one hand, the multi-source heterogeneous data array integrates base station information, repair difficulty, and maintenance personnel information. The data provided by different data sources have different information and perspectives, which increases the depth and breadth of data analysis, thereby improving the reliability of the analysis results; on the other hand, the pre-trained fault repair scheduling model can fit the multi-source heterogeneous data array into master-slave nodes to quickly determine the corresponding relationship between maintenance personnel and faulty base stations, thereby improving the efficiency of fault handling; on the other hand, the preset fault repair priority rule is used to remind the model to output the best result according to the set priority strategy, which ensures the flexibility of the model and enables the model to output different optimal solutions under different scenario requirements, thereby avoiding the singleness of the analysis dimension.

[0110] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0111] See also Figure 8 , which shows a schematic diagram of the structure of a private network management device based on multi-source heterogeneous data fusion analysis provided by an exemplary embodiment of the present application. The private network management device based on multi-source heterogeneous data fusion analysis can be implemented as all or part of an electronic device through software, hardware or a combination of both. The device 1 includes a base station data acquisition module 10, a multi-source heterogeneous data array construction module 20, a maintenance personnel information output module 30, and a maintenance personnel scheduling module 40.

[0112] The base station data acquisition module 10 is used to acquire the base station data of each failed base station when multiple base stations in the multi-mode wireless private network are detected to have failed in the same period of time;

[0113] A multi-source heterogeneous data array construction module 20 is used to construct a multi-source heterogeneous data array according to the base station data of each faulty base station, the difficulty of repairing the network fault of each frequency band, and the maintenance proficiency of each maintenance personnel;

[0114] The maintenance personnel information output module 30 is used to input the multi-source heterogeneous data array and the preset fault repair priority rule into the pre-trained fault repair scheduling model, and output the maintenance personnel information corresponding to each faulty base station; the preset fault repair priority rule is used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array;

[0115] The maintenance personnel scheduling module 40 is used to schedule each maintenance personnel based on the maintenance personnel information corresponding to each faulty base station.

[0116] Optionally, the maintenance personnel information output module 30 includes:

[0117] An optimization target determination layer, which is used to determine the optimization target based on preset fault repair priority rules. The optimization target is used to minimize the total repair time or maximize resource utilization or balance the workload, and is used to constrain the number of maintenance personnel or skill matching requirements or geographical location restrictions;

[0118] A master node determination layer, used for taking the basic vector of the maintenance personnel included in the comprehensive embedding vector representation of each maintenance personnel as the master node;

[0119] A slave node determination layer, for using the basis vectors of each faulty base station included in the comprehensive embedded vector representation of each maintenance personnel as a plurality of slave nodes of the master node;

[0120] The comprehensive similarity calculation layer is used to calculate the comprehensive similarity between the master node and each slave node on the master node;

[0121] The final slave node determination layer is used to determine the final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, so as to obtain the optimal matching relationship between the maintenance personnel and the faulty base station;

[0122] The corresponding relationship determination layer is used to determine the corresponding relationship between the maintenance personnel and the faulty base station one by one according to the final slave node corresponding to the master node;

[0123] The information output layer is used to output the maintenance personnel information corresponding to each faulty base station based on the corresponding relationship between the maintenance personnel and the faulty base station.

[0124] It should be noted that the private network management device based on multi-source heterogeneous data fusion analysis provided in the above embodiment only uses the division of the above-mentioned functional modules as an example when executing the private network management method based on multi-source heterogeneous data fusion analysis. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the private network management device based on multi-source heterogeneous data fusion analysis provided in the above embodiment and the private network management method based on multi-source heterogeneous data fusion analysis belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[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 an embodiment of the present application, a multi-source heterogeneous data array and a preset fault repair priority rule can be input into a pre-trained fault repair scheduling model for processing, and then the maintenance personnel information corresponding to each faulty base station can be output. On the one hand, the multi-source heterogeneous data array integrates base station information, repair difficulty, and maintenance personnel information. The data provided by different data sources have different information and perspectives, which increases the depth and breadth of data analysis, thereby improving the reliability of the analysis results; on the other hand, the pre-trained fault repair scheduling model can fit the multi-source heterogeneous data array into master-slave nodes to quickly determine the corresponding relationship between maintenance personnel and faulty base stations, thereby improving the efficiency of fault handling; on the other hand, the preset fault repair priority rule is used to remind the model to output the best result according to the set priority strategy, which ensures the flexibility of the model and enables the model to output different optimal solutions under different scenario requirements, thereby avoiding the singleness of the analysis dimension.

[0127] The present application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implements the private network management method based on multi-source heterogeneous data fusion analysis provided by the above-mentioned various method embodiments.

[0128] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the private network management method based on multi-source heterogeneous data fusion analysis of the above-mentioned various method embodiments.

[0129] See also Fig. 9 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Fig. 9 As shown, the electronic device 1000 may include: at least one processor 1001 , at least one network interface 1004 , a user interface 1003 , a memory 1005 , and at least one communication bus 1002 .

[0130] The communication bus 1002 is used to realize the connection and communication between these components.

[0131] The user interface 1003 may include a display screen (Display) and a camera (Camera), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.

[0132] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0133] Among them, the processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect various parts within the entire electronic device 1000, and executes various functions and processes data of the electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, and it can be implemented separately through a chip.

[0134] Among them, the memory 1005 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1005 may also be optionally at least one storage system located away from the aforementioned processor 1001. As Fig. 9 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a private network management application based on multi-source heterogeneous data fusion analysis.

[0135] exist Fig. 9 In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 1001 can be used to call the private network management application based on multi-source heterogeneous data fusion analysis stored in the memory 1005, and specifically perform the following operations:

[0136] When multiple base stations of a multi-mode wireless private network are detected to have faults in the same period of time, base station data of each faulty base station is obtained;

[0137] A multi-source heterogeneous data array is constructed based on the base station data of each faulty base station, the difficulty of repairing network faults in each frequency band, and the maintenance proficiency of each maintenance personnel;

[0138] Input the multi-source heterogeneous data array and the preset fault repair priority rules into the pre-trained fault repair scheduling model, and output the maintenance personnel information corresponding to each faulty base station; the preset fault repair priority rules are used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array;

[0139] Based on the maintenance personnel information corresponding to each faulty base station, each maintenance personnel is dispatched.

[0140] In one embodiment, when the processor 1001 constructs a multi-source heterogeneous data array according to the base station information of each faulty base station, the preset difficulty of repairing the network fault of each frequency band, and the preset maintenance proficiency of each maintenance personnel, the processor 1001 specifically performs the following operations:

[0141] Perform data cleaning and data format standardization on the base station information of each faulty base station to obtain pre-processed data of each faulty base station;

[0142] Based on the preset matrix parameters and the number of maintenance personnel, an original matrix is ​​constructed;

[0143] According to the pre-processed data of each faulty base station and the difficulty of repairing the network faults in each frequency band, the maintenance proficiency of each maintenance personnel is fused to obtain the comprehensive embedded vector representation of each maintenance personnel;

[0144] The comprehensive embedding vector representation of each maintenance personnel is embedded into each element position in the original matrix to obtain a multi-source heterogeneous data array.

[0145] In one embodiment, when the processor 1001 performs data fusion on the maintenance proficiency of each maintenance personnel according to the pre-processed data of each faulty base station and the difficulty of repairing the network fault of each frequency band, and obtains the comprehensive embedded vector representation of each maintenance personnel, the following operations are specifically performed:

[0146] Get the employee information of each maintenance personnel;

[0147] Mapping the employee information of each maintenance personnel and the preset maintenance proficiency of each maintenance personnel into the same vector space to obtain the basic vector of each maintenance personnel;

[0148] The base station location information, base station life cycle status information, base station impact information on users, road congestion information within a preset range near the base station, and the difficulty of repairing network faults in each preset frequency band of each faulty base station are mapped into the same vector space to obtain the basic vector of each faulty base station;

[0149] Taking the basis vector of each maintenance personnel as the central vector, the basis vector of each faulty base station is mapped as a sub-vector to each central vector one by one to generate a comprehensive embedded vector representation of each maintenance personnel.

[0150] In one embodiment, when the processor 1001 inputs the multi-source heterogeneous data array and the preset fault repair priority rule into the pre-trained fault repair scheduling model and outputs the maintenance personnel information corresponding to each faulty base station, the processor 1001 specifically performs the following operations:

[0151] The optimization target determination layer determines the optimization target based on the preset fault repair priority rules. The optimization target is used to minimize the total repair time or maximize resource utilization or balance the workload, and is used to constrain the number of maintenance personnel or skill matching requirements or geographical location restrictions;

[0152] The master node determination layer uses the basic vector of the maintenance personnel included in the comprehensive embedding vector representation of each maintenance personnel as the master node;

[0153] The slave node determination layer uses the basis vectors of each faulty base station included in the comprehensive embedding vector representation of each maintenance personnel as multiple slave nodes of the master node;

[0154] The comprehensive similarity calculation layer calculates the comprehensive similarity between the master node and each slave node on the master node;

[0155] The final slave node determination layer determines the final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, so as to obtain the optimal matching relationship between the maintenance personnel and the faulty base station;

[0156] The correspondence determination layer determines the correspondence between the maintenance personnel and the faulty base station one by one according to the final slave node corresponding to the master node;

[0157] The information output layer outputs the maintenance personnel information corresponding to each faulty base station based on the corresponding relationship between the maintenance personnel and the faulty base stations.

[0158] In one embodiment, when the processor 1001 determines the final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, the processor 1001 specifically performs the following operations:

[0159] Construct an objective function for the optimization goal, which is to minimize the total maintenance time or maximize the maintenance efficiency;

[0160] According to the comprehensive similarity between the master node and each slave node on the master node and the objective function, the total maintenance time or maintenance efficiency between the master node and each slave node on the master node is calculated;

[0161] The slave node corresponding to the minimum total maintenance time or the maximum maintenance efficiency is determined as the final slave node corresponding to the master node.

[0162] In one embodiment, when generating a pre-trained fault repair scheduling model, the processor 1001 specifically performs the following operations:

[0163] Collect historical base station data of historical faulty base stations that have experienced historical faults;

[0164] Use neural network to build fault maintenance scheduling model;

[0165] Construct an objective function that minimizes the total maintenance time or maximizes the maintenance efficiency;

[0166] Integrate the objective function into the fault maintenance scheduling model;

[0167] By presetting the fault repair priority rules and objective functions, the historical base station data is labeled with the optimal maintenance personnel information to obtain the model training data;

[0168] The model training data is input into the fault maintenance scheduling model that integrates the objective function for machine learning to generate a pre-trained fault maintenance scheduling model.

[0169] In an embodiment of the present application, a multi-source heterogeneous data array and a preset fault repair priority rule can be input into a pre-trained fault repair scheduling model for processing, and then the maintenance personnel information corresponding to each faulty base station can be output. On the one hand, the multi-source heterogeneous data array integrates base station information, repair difficulty, and maintenance personnel information. The data provided by different data sources have different information and perspectives, which increases the depth and breadth of data analysis, thereby improving the reliability of the analysis results; on the other hand, the pre-trained fault repair scheduling model can fit the multi-source heterogeneous data array into master-slave nodes to quickly determine the corresponding relationship between maintenance personnel and faulty base stations, thereby improving the efficiency of fault handling; on the other hand, the preset fault repair priority rule is used to remind the model to output the best result according to the set priority strategy, which ensures the flexibility of the model and enables the model to output different optimal solutions under different scenario requirements, thereby avoiding the singleness of the analysis dimension.

[0170] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program for private network management based on multi-source heterogeneous data fusion analysis can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium of the program for private network management based on multi-source heterogeneous data fusion analysis can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0171] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A private network management method based on multi-source heterogeneous data fusion analysis, characterized in that: The method comprises: When multiple base stations of a multi-mode wireless private network are detected to have faults in the same period of time, base station data of each faulty base station is obtained; A multi-source heterogeneous data array is constructed based on the base station data of each faulty base station, the difficulty of repairing the network faults in each frequency band, and the maintenance proficiency of each maintenance personnel; wherein, The multi-source heterogeneous data array is constructed according to the base station data of each faulty base station, the difficulty of repairing the network fault of each frequency band, and the maintenance proficiency of each maintenance personnel, including: Performing data cleaning and data format standardization processing on the base station data of each faulty base station to obtain pre-processed data of each faulty base station; Based on the preset matrix parameters and the number of the maintenance personnel, construct an original matrix; According to the pre-processed data of each faulty base station and the difficulty of repairing the network faults in each frequency band, data fusion is performed on the maintenance proficiency of each preset maintenance personnel to obtain a comprehensive embedded vector representation of each maintenance personnel; The comprehensive embedding vector representation of each maintenance personnel is embedded into each element position in the original matrix to obtain a multi-source heterogeneous data array; wherein, The pre-processed data of each faulty base station includes base station location information, base station life cycle status information, base station impact information on users, and road congestion information within a preset range near the base station; The method of performing data fusion on the maintenance proficiency of each maintenance personnel according to the pre-processed data of each faulty base station and the difficulty of repairing the network fault of each frequency band to obtain the comprehensive embedded vector representation of each maintenance personnel includes: Obtaining employee information of each maintenance personnel; Mapping the employee information of each maintenance personnel and the preset maintenance proficiency of each maintenance personnel into the same vector space to obtain a basic vector of each maintenance personnel; Mapping the base station location information, base station life cycle status information, base station impact information on users, road congestion information within a preset range near the base station, and the difficulty of repairing network faults in each preset frequency band of each faulty base station to the same vector space to obtain the basic vector of each faulty base station; Taking the basic vector of each maintenance personnel as the central vector, mapping the basic vector of each faulty base station as a sub-vector to each central vector one by one, and generating a comprehensive embedded vector representation of each maintenance personnel; The multi-source heterogeneous data array and the preset fault repair priority rule are input into a pre-trained fault repair scheduling model, and the maintenance personnel information corresponding to each faulty base station is output; the preset fault repair priority rule is used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array; The maintenance personnel are dispatched based on the maintenance personnel information corresponding to the faulty base stations.

2. The method according to claim 1, characterized in that The multi-source heterogeneous data array is composed of the comprehensive embedded vector representation of each maintenance personnel, the comprehensive embedded vector representation of each maintenance personnel includes the basic vector of the maintenance personnel and the basic vector of each faulty base station, the basic vector of the maintenance personnel is generated based on the employee information of the maintenance personnel and the maintenance proficiency of the preset maintenance personnel, and the basic vector of each faulty base station is generated according to the base station data of each faulty base station; the pre-trained fault maintenance scheduling model includes an optimization target determination layer, a master node determination layer, a slave node determination layer, a comprehensive similarity calculation layer, a final slave node determination layer, a corresponding relationship determination layer and an information output layer; The step of inputting the multi-source heterogeneous data array and the preset fault repair priority rule into a pre-trained fault repair scheduling model and outputting the maintenance personnel information corresponding to each faulty base station includes: The optimization target determination layer determines the optimization target based on the preset fault repair priority rule, wherein the optimization target is used to minimize the total repair time or maximize resource utilization or balance the workload, and to constrain the number of maintenance personnel or skill matching requirements or geographical location restrictions; The master node determination layer uses the basic vector of the maintenance personnel included in the comprehensive embedding vector representation of each maintenance personnel as the master node; The slave node determination layer uses the basis vectors of each faulty base station included in the comprehensive embedded vector representation of each maintenance personnel as multiple slave nodes of the master node; The comprehensive similarity calculation layer calculates the comprehensive similarity between the master node and each slave node on the master node; The final slave node determination layer determines the final slave node corresponding to the master node according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, so as to obtain the optimal matching relationship between the maintenance personnel and the faulty base station; The correspondence determination layer determines the correspondence between the maintenance personnel and the faulty base station one by one according to the final slave node corresponding to the master node; The information output layer outputs the maintenance personnel information corresponding to each faulty base station based on the corresponding relationship between the maintenance personnel and the faulty base station.

3. The method according to claim 2, characterized in that The determining, according to the optimization target and the comprehensive similarity between the master node and each slave node on the master node, a final slave node corresponding to the master node comprises: Constructing an objective function of the optimization target, wherein the objective function is to minimize the total maintenance time or maximize the maintenance efficiency; Calculating the total maintenance time or maintenance efficiency between the master node and each slave node on the master node according to the comprehensive similarity between the master node and each slave node on the master node and the objective function; The slave node corresponding to the minimum total maintenance time or the maximum maintenance efficiency is determined as the final slave node corresponding to the master node.

4. The method according to claim 3, characterized in that The objective function is: Where MinZ is the total maintenance time or maintenance efficiency, m is the number of maintenance personnel, n is the number of faulty base stations, c ij is the cost of maintenance personnel i to repair faulty base station j, the cost is the comprehensive similarity between the master node and each slave node on the master node, x ij is a decision variable; among them, The constraint condition of the decision variable is: each faulty base station can only be repaired by one maintenance personnel: Each maintenance personnel can only repair one faulty base station:

5. The method according to claim 1, characterized in that The following steps are used to generate a pre-trained fault maintenance scheduling model, including: Collect historical base station data of historical faulty base stations that have experienced historical faults; Use neural network to build fault maintenance scheduling model; Construct an objective function that minimizes the total maintenance time or maximizes the maintenance efficiency; Integrating the objective function into the fault maintenance scheduling model; By presetting the fault repair priority rule and the objective function, the historical base station data is labeled with the optimal maintenance personnel information to obtain model training data; The model training data is input into a fault maintenance scheduling model that integrates an objective function for machine learning to generate a pre-trained fault maintenance scheduling model.

6. A private network management device based on multi-source heterogeneous data fusion analysis implemented by the method described in any one of claims 1 to 5, characterized in that: The device comprises: The base station data acquisition module is used to acquire the base station data of each failed base station when multiple base stations in the multi-mode wireless private network are detected to have failed in the same period of time; A multi-source heterogeneous data array construction module, used to construct a multi-source heterogeneous data array according to the base station data of each faulty base station, the difficulty of repairing the network fault of each frequency band, and the preset maintenance proficiency of each maintenance personnel; A maintenance personnel information output module is used to input the multi-source heterogeneous data array and the preset fault repair priority rule into a pre-trained fault repair scheduling model, and output the maintenance personnel information corresponding to each faulty base station; the preset fault repair priority rule is used to guide the pre-trained fault repair scheduling model to generate the optimal maintenance personnel information corresponding to each faulty base station based on the multi-source heterogeneous data array; The maintenance personnel scheduling module is used to schedule the maintenance personnel based on the maintenance personnel information corresponding to the faulty base stations.

7. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method according to any one of claims 1 to 5.

Images