A communication integrated control system and method based on artificial intelligence

Abstract

The invention discloses an artificial intelligence-based communication integrated management and control system and method, which relate to the technical field of intelligent monitoring of pipe corridors. A layout map collection module collects layout topology sub-maps, a sample collection module collects fault feature vectors and fault state labels, a model training module trains a fault judgment model for each type of pipe corridor equipment, a fault probability calculation module obtains the fault probability of each pipe corridor equipment in real time, a cluster generation module obtains the position coordinates of all abnormal equipment in the equipment layout topology map, uses a DBSCACN algorithm to cluster the position coordinates of all abnormal equipment, and a sequential statistics module obtains a regional coverage set, calculates the maintenance weight value of each equipment cluster, sorts the maintenance weight values ​​from large to small, and obtains a maintenance order; the number and range of manual inspections are reduced, and the efficiency of manual inspections is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring of pipe corridors, and in particular to an integrated communication management and control system and method based on artificial intelligence. Background Art

[0002] With the continuous expansion of urban scale and economic development, problems such as insufficient scale of integrated pipeline construction and low management level have gradually become prominent. The traditional management method of the pipeline corridor can no longer meet the needs of modern urban management. Therefore, the communication integrated management and control system based on artificial intelligence has come into being. The system can realize the intelligent management of various facilities and equipment in the integrated pipeline corridor, improve management efficiency, and reduce the accident rate. It is of great significance to ensure urban safety and promote economic development.

[0003] The utility corridor is a kind of urban infrastructure, which is mainly used to centrally house various public pipelines, such as electricity, telecommunications, gas, water supply and drainage pipelines. At present, the management and monitoring of facilities and equipment in the utility corridor mainly face the following major technical problems:

[0004] Distributed monitoring of equipment:

[0005] The existing monitoring equipment is independent of each other and lacks centralized management, making it difficult to achieve comprehensive monitoring.

[0006] Failure prediction is difficult:

[0007] Traditional monitoring systems lack intelligent analysis capabilities and cannot effectively predict potential equipment failures.

[0008] High maintenance cost:

[0009] Due to the lack of intelligent means, the maintenance of the integrated pipeline corridor mainly relies on manual inspection, which is costly and inefficient.

[0010] The Chinese patent with publication number CN113467317A discloses a comprehensive monitoring system for pipe corridors based on industrial interconnection. The monitoring perception module is used to present the status of the pipe corridor through the monitoring screen, the environmental detection module is used to detect the environment in each partition of the pipe corridor, the lighting module is used to control the lighting facilities in the pipe corridor, the personnel positioning module is used to provide wireless coverage for the pipe corridor and locate and track the trajectory of public personnel, the fire protection module is used to monitor the fire alarm of the pipe corridor, the power monitoring module is used to monitor the power of the pipe corridor, and the alarm module is used to manage the alarms occurring in the pipe corridor for timely response and processing. However, the system failed to achieve intelligent fault prediction of pipe corridor equipment;

[0011] To this end, the present invention proposes an integrated communication management and control system and method based on artificial intelligence. Summary of the invention

[0012] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an integrated communication control system and method based on artificial intelligence, which can accurately locate the area where the fault occurs when a potential fault occurs, thereby reducing the number and scope of manual inspections and improving the efficiency of manual inspections.

[0013] To achieve the above-mentioned purpose, the present invention proposes an integrated communication management and control system based on artificial intelligence, including a layout diagram collection module, a sample collection module, a model training module, a fault probability calculation module, a cluster generation module and a sequential statistics module; wherein each module is electrically connected;

[0014] The layout map collection module collects the equipment layout topology map and divides the equipment layout topology map into several layout topology sub-maps, each layout topology sub-map corresponds to a piece of equipment area; and sends the layout topology sub-map to the sequential statistics module;

[0015] The sample collection module collects the historical operation data of each type of pipe gallery equipment, and builds several groups of fault feature vectors and fault status labels for each type of pipe gallery equipment based on the historical operation data, and sends the fault feature vectors and fault status labels to the model training module;

[0016] The model training module takes the fault feature vector as input and the fault state label as output, trains the fault judgment model for each type of pipe gallery equipment, and sends the fault judgment model to the fault probability calculation module;

[0017] The fault probability calculation module collects the actual fault characteristic data of each pipe gallery equipment in real time; based on the actual fault characteristic data and the corresponding fault judgment model, the fault probability of each pipe gallery equipment is obtained and sent to the cluster generation module;

[0018] The cluster generation module regards the pipe gallery equipment with a failure probability greater than a preset probability threshold as abnormal equipment; obtains the location coordinates of all abnormal equipment in the equipment layout topology map, clusters the location coordinates of all abnormal equipment using the DBSCACN algorithm, and obtains K equipment clusters; and sends the equipment clusters to the sequential statistics module;

[0019] The sequential statistics module obtains the regional coverage set based on the spatial position mapping relationship between each equipment cluster and the equipment area, calculates the maintenance weight value of each equipment cluster based on the failure probability of each abnormal equipment, sorts the maintenance weight values ​​from large to small, and obtains the maintenance order.

[0020] The method of collecting the device layout topology diagram is:

[0021] Collect the plan design drawings and equipment location layout drawings of the integrated pipe gallery, obtain the relative positions of each pipe gallery equipment and the connection relationship between the pipe gallery equipment according to the plan design drawings, and construct the equipment nodes and the connection relationship between the equipment nodes based on the relative positions and connection relationships, so as to draw a two-dimensional plan view of the integrated pipe gallery;

[0022] Based on the equipment location layout diagram, the two-dimensional plane diagram is pre-processed with annotations to obtain the equipment layout topology diagram;

[0023] The step of dividing the device layout topology graph into a plurality of layout topology subgraphs comprises the following steps:

[0024] Step 11: Determine the regional division principles;

[0025] Step 12: Divide the space in the integrated pipe gallery according to the regional division principle in the equipment layout topology diagram, and mark the regional division lines on the equipment layout topology diagram;

[0026] Step 13: extract the device nodes, connection relationships and annotations in the area surrounded by each area dividing line to form a layout topology subgraph;

[0027] The method of collecting the historical operation data of each type of pipe gallery equipment is as follows:

[0028] For each type of pipe gallery equipment, determine its operating characteristic set;

[0029] The operation feature set includes several physical feature quantities that can reflect the operation status of the pipe gallery equipment;

[0030] Collect the characteristic values ​​of various operation characteristics of each pipe gallery equipment per unit time during the past operation of the integrated pipe gallery, as well as the abnormal labels of each pipe gallery equipment per unit time; the abnormal labels are 0 or 1;

[0031] The method of constructing several groups of fault feature vectors and fault status labels for each type of pipe gallery equipment based on historical operation data is as follows:

[0032] All the pipe gallery equipment in the integrated pipe gallery are grouped according to type to obtain M equipment groups, where M is the number of types of pipe gallery equipment, and all pipe gallery equipment of the same type form one equipment group;

[0033] For each device group:

[0034] The characteristic values ​​of each operation characteristic of each pipe gallery equipment at each unit time are combined into a set of fault characteristic vectors, and the abnormal label at each unit time is used as the fault state label;

[0035] The method of training the fault judgment model for each type of pipe gallery equipment using the fault feature vector as input and the fault state label as output is as follows:

[0036] For each type of pipe gallery equipment:

[0037] Each group of fault feature vectors in the corresponding equipment group is used as the input of the fault judgment model. The fault judgment model uses the predicted value of the fault probability of each pipe gallery equipment at each unit time as the output. The predicted value of the fault probability is a non-negative number and has a value range of [0, 1]; the fault judgment model uses the fault state label corresponding to each group of fault feature vectors as the prediction target, the difference between the predicted value of the fault probability and the fault state label as the prediction error, and minimizes the sum of squares of the prediction errors of all unit times of all pipe gallery equipment as the training target; the fault judgment model is trained until the sum of squares of the prediction errors converges and the training is stopped; the fault judgment model is any one of the classification models;

[0038] The method of clustering the location coordinates of all abnormal devices using the DBSCACN algorithm to obtain K device clusters includes the following steps:

[0039] Step 21: pre-set the parameter values ​​of the radius eps and the minimum number of points minPts of the DBSCACN algorithm;

[0040] Step 22: Obtain the location coordinates of each abnormal device, mark the number of the abnormal device as i, mark the location coordinates of the i-th abnormal device as (xi, yi), and mark the node corresponding to the i-th abnormal device in the device layout topology map as Pi;

[0041] Step 23: For each node Pi, calculate the number of nodes in the neighborhood with a radius of eps and the node Pi as the center;

[0042] Step 24: If the number of points in the neighborhood is greater than or equal to minPts, node Pi is regarded as a core node, and node Pi and the nodes in the neighborhood are regarded as a cluster; if the number of points in the neighborhood is less than minPts, but node Pi is in the neighborhood of a core node, node Pi is regarded as a boundary point and added to the corresponding cluster; the remaining points are marked as noise points; each cluster is regarded as a device clustering cluster;

[0043] The method of obtaining the regional coverage set based on the spatial position mapping relationship between each device cluster and the device area is as follows:

[0044] Count the overlap between the pipe gallery equipment in each layout topology subgraph and the pipe gallery equipment in each equipment cluster, and generate the area coverage set of each equipment cluster;

[0045] The statistical method of the area coverage set is:

[0046] Obtain the equipment area where each pipe gallery equipment in the equipment cluster is located, add it to the area coverage set, and remove duplicates from the area coverage set;

[0047] The maintenance weight value of each equipment cluster is calculated based on the failure probability of each abnormal equipment, and the maintenance weight values ​​are sorted from large to small to obtain the maintenance order in the following manner:

[0048] The number of the area coverage set is marked as j;

[0049] For the jth region covering set:

[0050] Count the number Sj of all abnormal devices in the area coverage set, as well as the maximum value Mj, average value Aj and median Zj of the predicted value of the failure probability of all abnormal devices;

[0051] The maintenance weight value is marked as Qj, and the calculation method of Qj is Qj=u1×Sj+u2×Mj+u3×Aj+u4×Zj; wherein u1, u2, u3 and u4 are all preset proportional coefficients.

[0052] The present invention also proposes a communication integrated management and control method based on artificial intelligence, comprising the following steps:

[0053] Step 1: Collect the device layout topology map, and divide the device layout topology map into several layout topology sub-maps, each layout topology sub-map corresponds to a device area;

[0054] Step 2: Collect historical operation data of each type of pipe gallery equipment, and build several groups of fault feature vectors and fault status labels for each type of pipe gallery equipment based on the historical operation data;

[0055] Step 3: Using the fault feature vector as input and the fault state label as output, train a fault judgment model for each type of tunnel equipment;

[0056] Step 4: Collect the actual fault characteristic data of each pipe gallery equipment in real time; based on the actual fault characteristic data and the corresponding fault judgment model, obtain the failure probability of each pipe gallery equipment;

[0057] Step 5: The pipe gallery equipment with a failure probability greater than a preset probability threshold is regarded as an abnormal device; the position coordinates of all abnormal devices in the equipment layout topology map are obtained;

[0058] Step 6: Use the DBSCACN algorithm to cluster the location coordinates of all abnormal devices to obtain K device clusters; K is the number of device clusters after clustering by the DBSCACN algorithm;

[0059] Step 7: Based on the spatial location mapping relationship between each device cluster and the device area, obtain the area coverage set;

[0060] Step 8: Calculate the maintenance weight value of each equipment cluster based on the failure probability of each abnormal equipment, sort the maintenance weight values ​​from large to small, and obtain the maintenance order.

[0061] The present invention further provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0062] The processor executes the above-mentioned artificial intelligence-based communication integrated management and control method by calling the computer program stored in the memory.

[0063] The present invention also provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0064] When the computer program runs on a computer device, the computer device executes the above-mentioned communication integrated management and control method based on artificial intelligence.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] The present invention collects an equipment layout topology map, and divides the equipment layout topology map into several layout topology sub-maps, each layout topology sub-map corresponds to an equipment area, collects historical operation data of each type of pipe gallery equipment, and constructs several groups of fault feature vectors and fault state labels for each type of pipe gallery equipment based on the historical operation data. With the fault feature vector as input and the fault state label as output, a fault judgment model is trained for each type of pipe gallery equipment, and the actual fault feature data of each pipe gallery equipment is collected in real time; based on the actual fault feature data and the corresponding fault judgment model, the failure probability of each pipe gallery equipment is obtained, and the pipe gallery equipment with a failure probability greater than a preset probability threshold is regarded as an abnormal device; the probability of all abnormal devices in the equipment layout topology is obtained. The location coordinates in the map are clustered using the DBSCACN algorithm to cluster the location coordinates of all abnormal devices to obtain K device clusters; K is the number of device clusters after clustering by the DBSCACN algorithm. Based on the spatial position mapping relationship between each device cluster and the device area, the regional coverage set is obtained. The maintenance weight value of each device cluster is calculated based on the failure probability of each abnormal device, and the maintenance weight values ​​are sorted from large to small to obtain the maintenance order; by comprehensively collating the monitoring data of each dispersed device and intelligently judging the occurrence of the fault based on the monitoring data, when a potential fault occurs, the area where the fault occurs can be accurately located, thereby reducing the number and scope of manual inspections and improving the efficiency of manual inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is a module connection relationship diagram of an artificial intelligence-based integrated communication management and control system in Example 1 of the present invention;

[0068] Figure 2 This is a flow chart of an integrated communication management and control method based on artificial intelligence in Example 2 of the present invention;

[0069] Figure 3 This is a schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention;

[0070] Figure 4 This is a schematic diagram of the computer-readable storage medium structure in Example 4 of the present invention. DETAILED DESCRIPTION

[0071] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] Example 1

[0073] like Figure 1 As shown, an integrated communication management and control system based on artificial intelligence includes a layout diagram collection module, a sample collection module, a model training module, a fault probability calculation module, a cluster generation module and a sequential statistics module; wherein each module is electrically connected;

[0074] The layout map collection module collects the equipment layout topology map and divides the equipment layout topology map into several layout topology sub-maps, each layout topology sub-map corresponds to a piece of equipment area; and sends the layout topology sub-map to the sequential statistics module;

[0075] The sample collection module collects the historical operation data of each type of pipe gallery equipment, and builds several groups of fault feature vectors and fault status labels for each type of pipe gallery equipment based on the historical operation data, and sends the fault feature vectors and fault status labels to the model training module;

[0076] The model training module takes the fault feature vector as input and the fault state label as output, trains the fault judgment model for each type of pipe gallery equipment, and sends the fault judgment model to the fault probability calculation module;

[0077] The fault probability calculation module collects the actual fault characteristic data of each pipe gallery equipment in real time; based on the actual fault characteristic data and the corresponding fault judgment model, the fault probability of each pipe gallery equipment is obtained and sent to the cluster generation module;

[0078] The cluster generation module regards the pipe gallery equipment with a failure probability greater than a preset probability threshold as abnormal equipment; obtains the location coordinates of all abnormal equipment in the equipment layout topology map, clusters the location coordinates of all abnormal equipment using the DBSCACN algorithm, and obtains K equipment clusters; and sends the equipment clusters to the sequential statistics module;

[0079] The sequential statistics module obtains the regional coverage set based on the spatial location mapping relationship between each equipment cluster and the equipment area, calculates the maintenance weight value of each equipment cluster based on the failure probability of each abnormal equipment, sorts the maintenance weight values ​​from large to small, and obtains the maintenance order. ;

[0080] The method of collecting the device layout topology diagram is as follows:

[0081] Collect the plan design drawings and equipment location layout drawings of the integrated pipe gallery, obtain the relative positions of each pipe gallery equipment and the connection relationship between the pipe gallery equipment according to the plan design drawings, and construct the equipment nodes and the connection relationship between the equipment nodes based on the relative positions and connection relationships, so as to draw a two-dimensional plan view of the integrated pipe gallery;

[0082] Based on the equipment location layout diagram, the two-dimensional plane diagram is pre-processed with annotations to obtain the equipment layout topology diagram;

[0083] The annotation preprocessing methods include but are not limited to:

[0084] Mark the location coordinates of all pipe gallery equipment on the two-dimensional plan; set a unique equipment number for each pipe gallery equipment in the two-dimensional plan, and mark the type of each pipe gallery equipment;

[0085] Furthermore, dividing the device layout topology graph into a plurality of layout topology subgraphs comprises the following steps:

[0086] Step 11: Determine the regional division principle; the regional division principle includes but is not limited to division by functional area (such as power supply area, ventilation area, pipeline area, etc.), division by management authority (areas under the jurisdiction of different maintenance units) or division by physical location (divided by floor, corridor, etc.);

[0087] Step 12: Divide the space in the integrated pipe gallery according to the regional division principle in the equipment layout topology diagram, and mark the regional division lines on the equipment layout topology diagram;

[0088] Step 13: extract the device nodes, connection relationships and annotations in the area surrounded by each area dividing line to form a layout topology subgraph;

[0089] It can be understood that each layout topology sub-graph corresponds to a piece of equipment area covered by the area division line in the equipment layout topology graph;

[0090] Specifically, the types of pipe gallery equipment include but are not limited to:

[0091] Environmental and equipment monitoring equipment: including data monitoring subsystem, monitoring equipment control subsystem and monitoring alarm subsystem, responsible for real-time monitoring of environmental parameters and equipment operating status in the integrated pipeline corridor.

[0092] Security equipment: including data monitoring subsystem, security equipment control subsystem, security alarm subsystem and smart line security subsystem, responsible for security protection and emergency response in the integrated pipeline corridor.

[0093] Communication equipment: including data monitoring subsystem, communication equipment control subsystem, smart line communication subsystem and communication alarm subsystem, responsible for realizing communication and data transmission within the system.

[0094] Automatic fire alarm equipment: includes a data monitoring subsystem, which is responsible for real-time monitoring of fire hazards and automatic alarm when a fire occurs.

[0095] Data management equipment: responsible for centralized management and analysis of data from each subsystem, providing data support for decision-making.

[0096] Furthermore, the method of collecting the historical operation data of each type of pipe gallery equipment is as follows:

[0097] For each type of pipe gallery equipment, determine its operating characteristic set;

[0098] The operation feature set includes several physical feature quantities that can reflect the operation status of the pipe gallery equipment; for example, the operation characteristics of the environment and equipment monitoring equipment include environmental parameters such as temperature, humidity, pressure, and operation parameters such as voltage and current; the operation characteristics of the security equipment include state data such as access control and video monitoring; the operation characteristics of the communication equipment include communication quality data such as signal strength and delay; the operation characteristics of the automatic fire alarm equipment include sensing data of smoke sensors or other fire sensors;

[0099] Collect the characteristic values ​​of various operation characteristics of each pipe gallery equipment per unit time during the past operation of the integrated pipe gallery, as well as the abnormal labels of each pipe gallery equipment per unit time; the abnormal label is 0 or 1, when the pipe gallery equipment operates abnormally, the abnormal label is set to 1, when the pipe gallery equipment operates normally, the abnormal label is set to 0;

[0100] It is understandable that the abnormal tag can be obtained from the historical repair, maintenance or replacement records of each pipe gallery equipment in the integrated pipe gallery;

[0101] Furthermore, the method of constructing several groups of fault feature vectors and fault status labels for each type of pipe gallery equipment based on historical operation data is as follows:

[0102] All the pipe gallery equipment in the integrated pipe gallery are grouped according to type to obtain M equipment groups, where M is the number of types of pipe gallery equipment, and all pipe gallery equipment of the same type form one equipment group;

[0103] For each device group:

[0104] The characteristic values ​​of each operation characteristic of each pipe gallery equipment at each unit time are combined into a set of fault characteristic vectors, and the abnormal label at each unit time is used as the fault state label;

[0105] Furthermore, the method of training the fault judgment model for each type of pipe gallery equipment by taking the fault feature vector as input and the fault state label as output is as follows:

[0106] For each type of pipe gallery equipment:

[0107] Each group of fault feature vectors in the corresponding equipment group is used as the input of the fault judgment model. The fault judgment model uses the predicted value of the fault probability of each pipe gallery equipment at each unit time as the output. The predicted value of the fault probability is a non-negative number and has a value range of [0, 1]; the fault judgment model uses the fault state label corresponding to each group of fault feature vectors as the prediction target, the difference between the predicted value of the fault probability and the fault state label as the prediction error, and minimizes the sum of squares of the prediction errors of all unit times of all pipe gallery equipment as the training target; the fault judgment model is trained until the sum of squares of the prediction errors reaches convergence and the training is stopped; the fault judgment model is any one of the classification models, and the classification model includes but is not limited to a logistic regression model or an SVM model, etc.;

[0108] Furthermore, the actual fault characteristic data of each pipe gallery equipment is collected in real time; based on the actual fault characteristic data and the corresponding fault judgment model, the method of obtaining the fault probability of each pipe gallery equipment is as follows:

[0109] After the fault judgment model training is completed, during the operation of the integrated pipe corridor, for each pipe corridor equipment in the integrated pipe corridor, each unit collects the characteristic values ​​corresponding to various operating characteristics in real time at all times to form the actual fault characteristic data of the pipe corridor equipment;

[0110] The actual fault characteristic data of each pipe gallery equipment is input into the fault judgment model of the corresponding pipe gallery equipment type to obtain the real-time prediction value of the failure probability of each pipe gallery equipment;

[0111] Furthermore, the method of clustering the location coordinates of all abnormal devices using the DBSCACN algorithm to obtain K device clusters includes the following steps:

[0112] Step 21: pre-set the parameter values ​​of the radius eps and the minimum number of points minPts of the DBSCACN algorithm;

[0113] Step 22: Obtain the location coordinates of each abnormal device, mark the number of the abnormal device as i, mark the location coordinates of the i-th abnormal device as (xi, yi), and mark the node corresponding to the i-th abnormal device in the device layout topology map as Pi;

[0114] Step 23: For each node Pi, calculate the number of nodes in the neighborhood with a radius of eps and the node Pi as the center;

[0115] Step 24: If the number of points in the neighborhood is greater than or equal to minPts, node Pi is regarded as a core node, and node Pi and the nodes in the neighborhood are regarded as a cluster; if the number of points in the neighborhood is less than minPts, but node Pi is in the neighborhood of a core node, node Pi is regarded as a boundary point and added to the corresponding cluster; the remaining points are marked as noise points; each cluster is regarded as a device clustering cluster;

[0116] Furthermore, the method of obtaining the area coverage set based on the spatial position mapping relationship between each device cluster and the device area is as follows:

[0117] Count the overlap between the pipe gallery equipment in each layout topology subgraph and the pipe gallery equipment in each equipment cluster, and generate the area coverage set of each equipment cluster;

[0118] The statistical method of the area coverage set is:

[0119] Obtain the equipment area where each pipe gallery equipment in the equipment cluster is located, add it to the area coverage set, and remove duplicates from the area coverage set;

[0120] A simple example of generating a region cover set is as follows:

[0121] The utility corridor is divided into 10 equipment areas, numbered R1-R10. Each area has multiple utility corridor equipment;

[0122] DBSCAN clustering obtained four equipment clusters, which were recorded as cluster C1, cluster C2, cluster C3, and cluster C4. Each equipment cluster contained several pipe gallery equipment.

[0123] According to statistics, cluster C1 contains abnormal devices located in R2 and R3 areas; cluster C2 is completely located in R5 area; cluster C3 contains abnormal devices located in R7 and R8 areas; cluster C4 is located in R10 area;

[0124] Then the regional coverage set is: C1: {R2, R3}, C2: {R5}, C3: {R7, R8}, C4: {R10}; that is, according to this regional division, it can be considered that the equipment failures occurring in R2 and R3 have a clustering effect, and thus may be caused by the same reason;

[0125] Furthermore, the maintenance weight value of each equipment cluster is calculated based on the failure probability of each abnormal equipment, and the maintenance weight values ​​are sorted from large to small to obtain the maintenance order in the following manner:

[0126] The number of the area coverage set is marked as j;

[0127] For the jth region covering set:

[0128] Count the number Sj of all abnormal devices in the area coverage set, as well as the maximum value Mj, average value Aj and median Zj of the predicted value of the failure probability of all abnormal devices;

[0129] The maintenance weight value is marked as Qj, and the calculation method of Qj is Qj=u1×Sj+u2×Mj+u3×Aj+u4×Zj; wherein u1, u2, u3 and u4 are all preset proportional coefficients.

[0130] Example 2

[0131] like Figure 2 As shown, a communication integrated management and control method based on artificial intelligence includes the following steps:

[0132] Step 1: Collect the device layout topology map, and divide the device layout topology map into several layout topology sub-maps, each layout topology sub-map corresponds to a device area;

[0133] Step 2: Collect historical operation data of each type of pipe gallery equipment, and build several groups of fault feature vectors and fault status labels for each type of pipe gallery equipment based on the historical operation data;

[0134] Step 3: Using the fault feature vector as input and the fault state label as output, train a fault judgment model for each type of tunnel equipment;

[0135] Step 4: Collect the actual fault characteristic data of each pipe gallery equipment in real time; based on the actual fault characteristic data and the corresponding fault judgment model, obtain the failure probability of each pipe gallery equipment;

[0136] Step 5: The pipe gallery equipment with a failure probability greater than a preset probability threshold is regarded as an abnormal device; the position coordinates of all abnormal devices in the equipment layout topology map are obtained;

[0137] Step 6: Use the DBSCACN algorithm to cluster the location coordinates of all abnormal devices to obtain K device clusters; K is the number of device clusters after clustering by the DBSCACN algorithm;

[0138] Step 7: Based on the spatial location mapping relationship between each device cluster and the device area, obtain the area coverage set;

[0139] Step 8: Calculate the maintenance weight value of each equipment cluster based on the failure probability of each abnormal equipment, sort the maintenance weight values ​​from large to small, and obtain the maintenance order.

[0140] Example 3

[0141] Figure 3 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3 As shown, according to another aspect of the present application, an electronic device 100 is also provided. The electronic device 100 may include one or more processors and one or more memories. The memory stores a computer readable code, and when the computer readable code is executed by one or more processors, it can execute the above-mentioned communication integrated control method based on artificial intelligence.

[0142] The method or device according to the embodiment of the present application can also be used by Figure 3 The electronic device architecture shown in FIG. Figure 3 As shown, the electronic device 100 may include a bus 101, one or more CPUs 102, ROM 103, RAM 104, a communication port 105 connected to a network, an input / output component 106, a hard disk 107, etc. The storage device in the electronic device 100, such as ROM 103 or hard disk 107, may store an artificial intelligence-based communication integrated management and control method provided in the present application.

[0143] Furthermore, the electronic device 100 may further include a user interface 108. Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.

[0144] Example 4

[0145] Figure 4 Schematic diagram of a computer-readable storage medium structure provided by an embodiment of the present application. Figure 4As shown, a computer-readable storage medium 200 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are executed by the processor, a communication integrated control method based on artificial intelligence according to an embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium 200 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0146] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present application, and when the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0147] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0148] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0149] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0150] The above preset parameters or preset thresholds are all set by technicians in this field according to actual conditions or obtained through large-scale data simulation.

[0151] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A communication integrated management and control system based on artificial intelligence, characterized in that: It includes a layout collection module, a sample collection module, a model training module, a fault probability calculation module, a cluster generation module and a sequential statistics module; wherein each module is electrically connected; The layout map collection module collects the equipment layout topology map and divides the equipment layout topology map into layout topology sub-maps, each layout topology sub-map corresponds to a piece of equipment area; and sends the layout topology sub-maps to the sequential statistics module; The sample collection module collects the historical operation data of each type of pipe gallery equipment, and constructs the fault feature vector and fault status label for each type of pipe gallery equipment based on the historical operation data, and sends the fault feature vector and fault status label to the model training module; The model training module takes the fault feature vector as input and the fault state label as output, trains the fault judgment model for each type of pipe gallery equipment, and sends the fault judgment model to the fault probability calculation module; The fault probability calculation module collects the actual fault characteristic data of each pipe gallery equipment in real time; based on the actual fault characteristic data and the corresponding fault judgment model, the fault probability of each pipe gallery equipment is obtained and sent to the cluster generation module; The cluster generation module regards the pipe gallery equipment with a failure probability greater than a preset probability threshold as abnormal equipment; obtains the location coordinates of all abnormal equipment in the equipment layout topology map, clusters the location coordinates of all abnormal equipment using the DBSCACN algorithm, and obtains K equipment clusters; and sends the equipment clusters to the sequential statistics module; The sequential statistics module obtains the regional coverage set based on the spatial location mapping relationship between each equipment cluster and the equipment area, calculates the maintenance weight value of each equipment cluster based on the failure probability of each abnormal equipment, sorts the maintenance weight values ​​from large to small, and obtains the maintenance order; The method of obtaining the regional coverage set based on the spatial position mapping relationship between each device cluster and the device area is as follows: Count the overlap between the pipe gallery equipment in each layout topology subgraph and the pipe gallery equipment in each equipment cluster, and generate the area coverage set of each equipment cluster; The statistical method of the area coverage set is: Obtain the equipment area where each pipe gallery equipment in the equipment cluster is located, add it to the area coverage set, and remove duplicates from the area coverage set; The maintenance weight value of each equipment cluster is calculated based on the failure probability of each abnormal equipment, and the maintenance weight values ​​are sorted from large to small to obtain the maintenance order in the following manner: The number of the area coverage set is marked as j; For the jth region covering set: Count the number Sj of all abnormal devices in the area coverage set, as well as the maximum value Mj, average value Aj and median Zj of the predicted value of the failure probability of all abnormal devices; The maintenance weight value is marked as Qj, and the calculation method of Qj is as follows: , where u1, u2, u3 and u4 are all preset proportional coefficients.

2. According to the artificial intelligence-based integrated communication management and control system of claim 1, it is characterized in that: The method of collecting the device layout topology diagram is: Collect the plan design drawings and equipment location layout drawings of the integrated pipe gallery, obtain the relative positions of each pipe gallery equipment and the connection relationship between the pipe gallery equipment according to the plan design drawings, and construct the equipment nodes and the connection relationship between the equipment nodes based on the relative positions and connection relationships, so as to draw a two-dimensional plan view of the integrated pipe gallery; Based on the equipment location layout diagram, the two-dimensional plane diagram is pre-processed with annotations to obtain the equipment layout topology diagram; The step of dividing the device layout topology graph into layout topology subgraphs comprises the following steps: Step 11: Determine the regional division principles; Step 12: Divide the space in the integrated pipe gallery according to the regional division principle in the equipment layout topology diagram, and mark the regional division lines on the equipment layout topology diagram; Step 13: Extract the device nodes, connection relationships and annotations in the area surrounded by each area dividing line to form a layout topology subgraph.

3. According to the artificial intelligence-based integrated communication management and control system of claim 2, it is characterized in that: The method of collecting the historical operation data of each type of pipe gallery equipment is as follows: For each type of pipe gallery equipment, determine its operating characteristic set; The operation feature set includes physical feature quantities that can reflect the operation status of the pipe gallery equipment; Collect the characteristic values ​​of various operation characteristics of each pipe gallery equipment per unit time during the past operation of the integrated pipe gallery, as well as the abnormal labels of each pipe gallery equipment per unit time; the abnormal labels are 0 or 1; The method of constructing the fault feature vector and fault status label for each type of pipe gallery equipment based on historical operation data is as follows: All the pipe gallery equipment in the integrated pipe gallery are grouped according to type to obtain M equipment groups, where M is the number of types of pipe gallery equipment, and all pipe gallery equipment of the same type form one equipment group; For each device group: The characteristic values ​​of the operating characteristics of each pipe gallery equipment at each unit time are combined into a set of fault feature vectors, and the abnormal label at each unit time is used as the fault state label.

4. According to claim 3, the integrated communication management and control system based on artificial intelligence is characterized in that: The method of training the fault judgment model for each type of pipe gallery equipment using the fault feature vector as input and the fault state label as output is as follows: For each type of pipe gallery equipment: Each group of fault feature vectors in the corresponding equipment group is used as the input of the fault judgment model. The fault judgment model uses the predicted value of the failure probability of each pipe gallery equipment at each unit time as the output. The predicted value of the failure probability is a non-negative number and has a value range of [0,1]; The fault judgment model uses the fault state label corresponding to each group of fault feature vectors as the prediction target, the difference between the predicted value of the fault probability and the fault state label as the prediction error, and minimizes the sum of squares of the prediction errors of all pipe gallery equipment at all unit times as the training target; The fault judgment model is trained until the sum of squares of the prediction errors reaches convergence and the training is stopped; the fault judgment model is any one of the classification models.

5. According to claim 4, the integrated communication management and control system based on artificial intelligence is characterized in that: The method of clustering the location coordinates of all abnormal devices using the DBSCACN algorithm to obtain K device clusters includes the following steps: Step 21: pre-set the parameter values ​​of the radius eps and the minimum number of points minPts of the DBSCACN algorithm; Step 22: Obtain the location coordinates of each abnormal device, mark the number of the abnormal device as i, mark the location coordinates of the i-th abnormal device as (xi, yi), and mark the node corresponding to the i-th abnormal device in the device layout topology map as Pi; Step 23: For each node Pi, calculate the number of nodes in the neighborhood with the node Pi as the center and the radius eps; Step 24: If the number of points in the neighborhood is greater than or equal to minPts, node Pi is taken as a core node, and node Pi and the nodes in the neighborhood are regarded as a cluster; if the number of points in the neighborhood is less than minPts, but node Pi is in the neighborhood of a core node, node Pi is taken as a boundary point and added to the corresponding cluster; the remaining points are marked as noise points; each cluster is regarded as a device clustering cluster.

6. A communication integrated management and control method based on artificial intelligence, which is implemented based on the communication integrated management and control system based on artificial intelligence according to any one of claims 1 to 5, characterized in that: The following steps are involved: Step 1: Collect the device layout topology map, and divide the device layout topology map into layout topology sub-maps, each layout topology sub-map corresponds to a device area; Step 2: Collect historical operation data of each type of pipe gallery equipment, and build fault feature vectors and fault status labels for each type of pipe gallery equipment based on the historical operation data; Step 3: Using the fault feature vector as input and the fault state label as output, train a fault judgment model for each type of tunnel equipment; Step 4: Collect the actual fault characteristic data of each pipe gallery equipment in real time; Based on the actual fault feature data and the corresponding fault judgment model, the failure probability of each pipe gallery equipment is obtained; Step 5: The pipe gallery equipment with a failure probability greater than a preset probability threshold is regarded as an abnormal device; the position coordinates of all abnormal devices in the equipment layout topology map are obtained; Step 6: Use the DBSCACN algorithm to cluster the location coordinates of all abnormal devices to obtain K device clusters; K is the number of device clusters after clustering by the DBSCACN algorithm; Step 7: Based on the spatial location mapping relationship between each device cluster and the device area, obtain the area coverage set; Step 8: Calculate the maintenance weight value of each equipment cluster based on the failure probability of each abnormal equipment, sort the maintenance weight values ​​from large to small, and obtain the maintenance order.

7. An electronic device, characterized in that: include: processor and memory, wherein The memory stores a computer program that can be called by the processor; The processor executes the artificial intelligence-based communication integrated management and control method described in claim 6 in the background by calling the computer program stored in the memory.

8. A computer-readable storage medium, characterized in that: A rewritable computer program is stored thereon; When the computer program runs on a computer device, the computer device executes the artificial intelligence-based communication integrated management and control method according to claim 6 in the background.

Images