Airport network operation and maintenance management method and system based on digital twinning

Abstract

The present application relates to the technical field of network operation and management, in particular to an airport network operation and management method and system based on digital twinning. The present application firstly calibrates the three-dimensional absolute coordinates of the offline network address set and the equipment, generates the undirected wire slot connected graph; further determines the connected direction and eliminates the isolated nodes and equidistant redundant edges, outputs the one-way wire slot flow transfer graph; further obtains the wiring branch distribution ratio, according to the initial inflow and the expected cumulative value of the fault cable transmitted by all in-path, combined with the wire slot branch containing cross-sectional area, obtains the wire slot cross-section fault cable density; finally, the geographic location of the damaged wire slot is located, and the order issuing instruction is issued. First, generate the one-way wire slot flow transfer graph after the floor and fire compartment constraints are removed, and then determine the branch distribution ratio, obtain the wire slot cross-section fault cable density along the topological sequence, locate the geographic location of the maximum density and issue the order issuing instruction, improve the network communication efficiency.

Description

Technical Field

[0001] This invention relates to the field of network operation and maintenance management technology, specifically to an airport network operation and maintenance management method and system based on digital twins. Background Technology

[0002] Large airports and other complex buildings often house multiple logically isolated network service systems, whose physical cabling typically shares the building's main metal cable trays. During building renovations, there is a risk of damaging these main cable trays. If the shared cable trays are damaged, it can trigger a surge of inter-network segment device disconnection alarms within a very short period.

[0003] When troubleshooting sudden, high-concurrency physical disconnection events, existing network monitoring platforms can only provide the logical network addresses of the disconnected terminals. Since detailed drawings covering all cable routing paths are often lacking, current technologies typically rely on simple graph theory shortest path algorithms to attempt to find intersection points. However, this conventional method disregards common sense in engineering installations, ignoring the objective capacity limitations and diversion effects of the actual cross-sectional diameter of the conduits on cable routing. This results in the derived paths failing to accurately reflect the common convergence bottleneck of the disconnected cables, forcing on-site troubleshooting personnel to manually troubleshoot blindly through each conduit, significantly delaying the opportunity to restore network communication. Summary of the Invention

[0004] To address the technical problem of failing to quickly and accurately locate the physical damage points of damaged pipelines in network outages due to the lack of detailed wiring diagrams, thus delaying emergency repairs, this invention aims to provide a digital twin-based airport network operation and maintenance management method and system. The specific technical solution adopted is as follows: A digital twin-based airport network operation and maintenance management method, the method comprising: Extract the event analysis window based on the changes in offline incremental data of the device, and calibrate the set of offline network addresses; obtain the three-dimensional absolute coordinates of the device corresponding to the offline network address, and calculate the perpendicular foot by projecting it onto the main trunk of the digital twin model under the consistency constraints of floor and fire compartment, and generate an undirected trunk connectivity graph; determine the connectivity direction and remove isolated nodes and equidistant redundant edges based on the shortest routing length from each node to the end point of the computer room trunk based on the undirected trunk connectivity graph, and output a unidirectional trunk flow graph; Based on the cross-sectional area of ​​the outgoing branch and the building length of the branch in the unidirectional cable tray flow diagram, the cable branch allocation ratio is obtained; along the topological sequence of the unidirectional cable tray flow diagram, the initial inflow flow of the node is determined; based on the initial inflow flow and the expected cumulative value of faulty cables transmitted through all incoming paths, combined with the cable branch allocation ratio and the cross-sectional area of ​​the branch, the faulty cable density of the cable tray cross section is obtained. Locate the geographical location of the damaged cable tray with the highest density of faulty cables at the cross-section of the cable tray, and issue a scheduling order.

[0005] Furthermore, the method for obtaining the offline network address set includes: Obtain the differential slope of the offline incremental data of the device. When the differential slope is greater than the preset sudden disconnection rate threshold, open the event analysis window. When the incremental value returns to zero or the opening time reaches the preset maximum collection duration, close the event analysis window and generate an offline network address set.

[0006] Furthermore, the method for obtaining the undirected slotted connectivity graph includes: Extract the set of main building trunk lines from the digital twin model, select each device as a target device, and select line segments in the set of main building trunk lines that are on the same floor and in the same fire compartment as the target device in terms of the three-dimensional absolute coordinates to form a target line segment set; Obtain the shortest vertical distance from the three-dimensional absolute coordinates of the target device to all target line segments. When the minimum shortest vertical distance is less than the preset maximum adsorption distance, obtain the orthogonal perpendicular foot point on the corresponding target line segment. Call the three-dimensional coordinates of the historical access node and combine them with the preset merging distance threshold to determine whether the orthogonal perpendicular foot point is to be used as a new perpendicular foot node. By summarizing the cut-off cable trough segments and the undivided original cable trough segments, and combining each of the new vertical nodes and the historical access nodes, the undirected cable trough connectivity graph is generated.

[0007] Furthermore, the method for obtaining the unidirectional groove flow diagram includes: Identify isolated nodes whose shortest path length is infinite or greater than a preset maximum scalar value, and remove the isolated nodes and their associated connected edges from the undirected slot connectivity graph; For the undirected connected edges in the undirected slotted connected graph after removing isolated nodes, the shortest path lengths of the nodes at both ends of the undirected connected edge are compared, and the connection direction of the undirected connected edge is determined to be from the node with the longest shortest path length to the node with the shortest path length; when the shortest path lengths of the nodes at both ends of the undirected connected edge are equal, the undirected connected edge is removed as an equidistant redundant edge. The unidirectional groove flow diagram is constructed based on each connected edge after the connection direction is determined.

[0008] Furthermore, the method for obtaining the routing branch allocation ratio includes: For each outgoing branch, the length of the cable tray branch building is substituted into the preset logarithmic compensation function to obtain the length compensation logarithm. The cross-sectional area of ​​the cable tray branch per unit length compensation logarithm is used as the basic allocation ratio. The routing branch allocation ratio is obtained based on the basic allocation ratio values ​​of all outgoing branches under the same node.

[0009] Furthermore, the method for obtaining the density of faulty cables in the cross-section of the cable tray includes: The sum of the initial inflow flow of each node and the expected cumulative value of faulty cables of all incoming connected edges input to the node is used as the total expected convergence of the node; the total expected convergence of the node is combined with the routing branch allocation ratio of the node to obtain the expected cumulative value of faulty cables flowing through the corresponding outgoing branch. Based on the expected cumulative value of the faulty cables and the corresponding cross-sectional area of ​​the cable tray branch, the density of faulty cables in the cable tray cross-section is obtained.

[0010] Furthermore, the method for obtaining the expected cumulative value of the faulty cable includes: By deducing step by step along the topological sequence of the unidirectional cable tray flow diagram, the expected cumulative value of the faulty cable for each connected edge in the unidirectional cable tray flow diagram is obtained.

[0011] Furthermore, the method for issuing the scheduling instruction includes: Extract the center coordinates of the damaged cable tray corresponding to the fault cable density with the largest value of the cable tray cross-section, call the spatial inclusion relationship interface to obtain the structured geographic location text containing the building area code and manhole number, and encapsulate it with the physical connection operation instructions to generate a troubleshooting work order data package; obtain the three-dimensional spatial coordinates of each maintenance terminal, and send the troubleshooting work order data package to the maintenance terminal with the smallest straight-line distance from the center coordinates.

[0012] Furthermore, the method for obtaining the initial inflow flow includes: In the unidirectional cable tray flow diagram, the initial inflow traffic of non-cable tray access nodes is set to 0; the total number of devices corresponding to the offline network address bound to each cable tray access node is used as the initial inflow traffic.

[0013] The present invention also proposes an airport network operation and maintenance management system based on digital twins. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the steps of the airport network operation and maintenance management method based on digital twins.

[0014] The present invention has the following beneficial effects: This invention first calibrates the offline network address set and the three-dimensional absolute coordinates of the devices, generating an undirected cable tray connectivity graph. This legally attaches discrete device points scattered throughout the building space to the cable tray network, constructing a digital topology base that conforms to building physics rules while preserving the original pipeline connectivity. It further determines the connectivity direction and eliminates isolated nodes and equidistant redundant edges, outputting a unidirectional cable tray flow graph. This creates an irreversible transmission path for subsequent probability and expectation to be transmitted downstream in the determined direction. It further obtains the branch allocation ratio of the cable routing, providing a basis for the reasonable distribution of subsequent expected loads at branch intersections. Based on the initial inflow and the expected cumulative value of faulty cables transmitted through all inflow paths, combined with the cable tray branch capacity cross-sectional area, it obtains the faulty cable density of the cable tray cross-section, comprehensively reflecting the spatial congestion level and load-bearing capacity of the cable tray. Finally, it locates the geographical location of the damaged cable tray and issues a scheduling instruction. First, a unidirectional cable tray flow diagram is generated after being constrained by floors and fire compartments and isolating and eliminating equidistant redundancies. Then, the branch allocation ratio is determined, the density of faulty cables in the cable tray cross-section is obtained along the topology sequence, the geographical location of the location with the highest density is located, and a scheduling instruction is issued to improve the efficiency of restoring network communication. Attached Figure Description

[0015] Figure 1 A flowchart illustrating an airport network operation and maintenance management method based on digital twins, provided as an embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for obtaining the routing branch allocation ratio according to an embodiment of the present invention. Detailed Implementation

[0016] The following description, in conjunction with the accompanying drawings, details a specific solution for an airport network operation and maintenance management method and system based on digital twins provided by this invention.

[0017] Please see Figure 1 The diagram illustrates a flowchart of an airport network operation and maintenance management method based on digital twins, according to an embodiment of the present invention, which specifically includes: Step S1: Extract the event analysis window based on the changes in offline incremental data of the device, and calibrate the set of offline network addresses; obtain the three-dimensional absolute coordinates of the device corresponding to the offline network address, and calculate the perpendicular foot by projecting it onto the main trunk line of the digital twin model under the consistency constraints of floor and fire compartment, and generate an undirected trunk line connection graph; determine the connection direction and remove isolated nodes and equidistant redundant edges based on the shortest routing length from each node in the undirected trunk line connection graph to the end point of the computer room trunk line, and output a unidirectional trunk line flow graph.

[0018] In one embodiment of the present invention, the system continuously monitors the connection status of the network monitoring station device, extracts the device change action from online to offline, then extracts the event analysis window based on the change of the device offline increment, calibrates the offline network address set, and outputs a large-scale offline device network address set that can completely characterize a single abnormal network outage event. As an example, we set the unit time period to 1 second, get the number of newly added offline terminals per second as the device offline increment, and then get the differential slope of the device offline increment, that is, the current device offline increment minus the device offline increment of the previous second, as the current differential slope. Based on the airport management's historical average fault reports, a preset threshold for the sudden disconnection rate is set at 5 units / second. When the differential slope is greater than the preset threshold for the sudden disconnection rate, the system marks that moment as the starting point of a sudden abnormal disconnection event, opens the event analysis window, and continuously collects all network packets that fall into the window that change from online to offline status. When the incremental value returns to zero or the opening time reaches the preset maximum collection duration, the event analysis window is closed, and an offline network address set is generated.

[0019] When the incremental value returns to zero, it indicates that the current avalanche-like chain reaction triggered by a single physical damage action (such as a large shearing action) has ended. At the same time, to prevent random network fluctuations from causing the collection window to lock, the preset maximum collection time is set to 5 minutes.

[0020] The system extracts all offline terminal state transition logs recorded in the event analysis window. It then extracts the network identifiers from these logs and constructs a set of offline network addresses for the current complete analysis batch.

[0021] It should be noted that in other embodiments of the present invention, the implementer may adjust the unit time period, the preset sudden disconnection rate threshold, and the preset maximum collection duration according to the specific scenario.

[0022] Considering that only by transforming the abstract network address into a specific location on the physical conduit carrying the device can the logical disconnection phenomenon of the network layer be mapped to the spatial structure of the physical layer for deduction, the three-dimensional absolute coordinates of the device corresponding to the offline network address are obtained. Under the constraint of consistency between the floor and the fire compartment, the perpendicular foot is calculated by projecting it onto the main trunk line of the digital twin model to generate an undirected trunk line connectivity graph. The discrete device points scattered in the building space are legally attached to the conduit network, and a digital topology base that conforms to the building physics rules and retains the original pipeline connectivity relationship is constructed, so that all subsequent deductions are based on a real and reliable physical wiring structure.

[0023] Furthermore, considering the existence of loops and equidistant symmetrical paths in an undirected conduit network, directly deriving cable routing in such a network would lead to logical deadlocks or infinite loops due to the non-uniqueness of directions. Therefore, based on the shortest routing length from each node in the undirected conduit network to the end point of the conduit network in the computer room, the connection direction is determined and isolated nodes and equidistant redundant edges are eliminated, outputting a unidirectional conduit flow graph. This transforms the entire conduit network from a bidirectional maze into a convergent structure where each edge has a clear direction, creating an irreversible transmission path for subsequent probabilities and expectations to be passed downstream in a defined direction.

[0024] As a preferred approach, the system first reads the digital twin model of the airport building and extracts the 3D line segment groups representing the metal pipes and wiring of the building entity from the digital twin model. These groups serve as the main wiring duct set. To avoid physically infeasible cross-connections between scattered coordinates (such as network points in the first-floor lobby being directly attached to lines hidden in the server room on the second floor), the system applies the first layer of anti-crossing attribute verification. Specifically: Each device is selected as a target device. Then, the system collaboratively retrieves the floor attribute field and fire compartment field of the building room where the target device is located, and obtains the floor and compartment attributes of each line segment in the building trunk line set. The system performs logical comparison to determine whether the attribute field text is completely equal. In this way, the system selects the line segment in the building trunk line set that is on the same floor and in the same fire compartment as the target device in terms of three-dimensional absolute coordinates, and uses it as the target line segment to form the target line segment set. Then, obtain the shortest vertical distance from the three-dimensional absolute coordinates of the target device to all target line segments. When the minimum shortest vertical distance is less than the preset maximum adsorption distance, it indicates that the distance between the device and the target line segment is within a reasonable wiring overhang range and that physical connection is feasible. Obtain the orthogonal perpendicular foot point on the corresponding target line segment. In low-voltage wiring, cables for end devices (such as cameras) are typically routed perpendicularly into the main cable tray at a 90-degree angle (T-junction). Therefore, obtaining the shortest vertical distance is equivalent to finding the shortest distance from a point to a straight line in three-dimensional space, a well-known technique. In this example, the preset maximum adsorption distance is 5 meters, representing the maximum legal physical length allowed for the end device cable to be suspended or laid in an exposed conduit to the main cable tray. This is used to eliminate wandering coordinates that do not belong to the main cable tray. If all shortest vertical distances are greater than or equal to the preset maximum adsorption distance, the target device is considered wandering interference data and is eliminated.

[0025] It should be noted that when the calculated orthogonal perpendicular foot point is located outside the two physical endpoints of the target line segment, the endpoint of the target line segment closest to that orthogonal perpendicular foot point will be updated as the final orthogonal perpendicular foot point.

[0026] Because in densely populated office areas, the absolute spatial coordinates of several terminal devices on different network segments may approach and overlap infinitely. If each device is interrupted, countless redundant zero-length displacement pipe corridors will be generated, resulting in resource waste and overlapping of graphic elements. Therefore, before performing entity interruption and reassembly, the system applies a second line of defense for coordinate deduplication and aggregation. Specifically: The system retrieves the 3D coordinates of historical access nodes and combines them with a preset merging distance threshold to determine whether the orthogonal perpendicular point should be used as a new perpendicular point. For example, the preset merging distance threshold is 0.05 meters. When the 3D coordinates of a historical access node coincide with the orthogonal perpendicular point, the distance is 0. When the distance between the 3D coordinates of a historical access node and the orthogonal perpendicular point is less than the preset merging distance threshold, the orthogonal perpendicular point is determined to be an existing node, and the system directly adds and binds the target device's network identifier or offline network address to this existing historical node. Otherwise, the orthogonal perpendicular point is used as a new perpendicular point, a completely new suspension pull-in position. The system cuts the original line segment, creating a new interconnected boundary at this position. If the new perpendicular node coincides with the physical endpoint of the target line segment, the operation of cutting the original line segment is skipped, and the physical endpoint is registered as an access node for retention. Finally, traverse all target devices, summarize the cut-out cable tray segments and the undivided original cable tray segments, combine each new vertical node and historical access node, and encapsulate them into digital topology graph elements that retain the interconnection of line segments but have not yet been given directional relationships, thus generating an undirected cable tray connected graph.

[0027] The system processes offline devices sequentially (e.g., device offline timing or other sequential schemes). The coordinates of the first processed devices are used to generate perpendicular nodes on the pipeline and stored in the cable tray access node set. When processing subsequent coordinates, The nodes that already exist in the set are the historical access nodes; The encapsulation generation rules are as follows: all new perpendicular nodes, historical access nodes, and the endpoints of undivided original pipe segments are taken as the set of vertices (Nodes) of the graph, and the cut-out cable trough segments and undivided original cable trough segments are taken as the set of undirected edges (Edges) of the graph. The connection topology is constructed based on the coincidence of the absolute spatial coordinates of the vertices and segments to generate a standard undirected graph data structure.

[0028] It should be noted that 3D absolute coordinates refer to the actual physical (x, y, z) location of network terminal devices (such as security cameras, departure terminals, office computers, etc.) within the airport's 3D architectural space. This binding is achieved during the initial construction of the airport's digital twin system or BIM (Building Information Modeling) system by recording and permanently storing the device's IP address and its physical installation location coordinates in the asset ledger database. In engineering practice, the 3D geometric center point of the device in the BIM model, or the physical location of the network cable interface on the device, is typically extracted as the 3D absolute coordinates.

[0029] Furthermore, the method for obtaining the unidirectional cable tray flow diagram includes: The system performs a traversal operation on all nodes in the undirected cable tray connected graph: it locates the endpoints of the cable trays corresponding to the core data center switches. The absolute center endpoint is set as the pathfinding calculation point. Then, the well-known Dijkstra's shortest path search algorithm is called, using the 3D measured length of each pipeline entity as the edge weight, to calculate the reach of each independent node along the connected pipeline. The shortest distance required to travel.

[0030] Among them, for For itself, the system forcibly assigns a value of 0 to its calculation result. For other ordinary computing nodes, the system defines the calculated distance value as the shortest routing length from the corresponding node to the end of the data center cable tray. The smaller this value, the closer the node is to the convergence center in the physical pipeline network.

[0031] After completing the full topology check calculation, in order to prevent some pipe segments from becoming headless pipe network islands due to the omission of key connection points during the twin model input, the system first performs an island removal operation: identify isolated nodes with a shortest route length of infinity or greater than a preset maximum scalar value, and remove the isolated nodes and their associated connected edges from the undirected cable trench connected graph; in this example, the preset maximum scalar value is 99999 meters.

[0032] Then, for the undirected connected edges in the undirected wire slot connected graph after removing isolated nodes, the shortest path lengths of the nodes at both ends of the undirected connected edge are compared, and the connection direction of the undirected connected edge is determined to be from the node with the longest shortest path length to the node with the shortest path length. Because in the location of large, regular low-voltage vertical shafts or multi-branch symmetrically arranged pipe corridors in buildings, there are objectively multiple physical routes of completely equal length. This edge cannot reduce the gradient by providing an effective convergence distance within the topology calculation cycle. Any lateral wandering attempt will lead to a dead loop. Therefore, when the shortest path lengths of the nodes at both ends of an undirected connected edge are equal, the undirected connected edge is eliminated as an equidistant redundant edge. The system sequentially traverses all the original undirected connected edges in the graph and constructs a unidirectional slot flow graph with an absolutely single convergent direction based on each connected edge after determining the connection direction.

[0033] Step S2: Based on the cross-sectional area of ​​the outgoing branch of the unidirectional cable tray flow diagram and the building length of the cable tray branch, obtain the cable branch allocation ratio; determine the initial inflow flow of the node along the topological sequence of the unidirectional cable tray flow diagram; and obtain the fault cable density of the cable tray cross section based on the initial inflow flow and the expected cumulative value of fault cables transmitted through all incoming paths, combined with the cable branch allocation ratio and the cross-sectional area of ​​the cable tray branch.

[0034] Without detailed wiring diagrams, it is impossible to know which conduit each unknown-direction disconnected network cable specifically chose. However, the actual laying behavior of cables is objectively constrained by the physical properties of the conduits. The larger the cross-section of the conduit, the more attractive it is to cable. Therefore, based on the cross-sectional area of ​​the outgoing branches and the building length of the branch in the unidirectional cable tray flow diagram, the distribution ratio of the cable branches is obtained. The objective physical parameters of the engineering conduits are transformed into quantified routing probabilities, so that each outgoing branch obtains a weight value that conforms to the prior common sense of the project. This provides a basis for the reasonable distribution of the expected load at the branch intersection.

[0035] When a large number of devices go offline, such as due to physical damage caused by a large shearing operation, the distribution of the broken network cables is not isolated. The event of a cable entering a branch after passing through a branch node has a cumulative characteristic of being passed down from the source to the downstream. The number of offline devices bound to each access node constitutes the deterministic source of the disconnection at that point. Therefore, the initial inflow flow of the node is determined along the topological sequence of the unidirectional cable tray flow diagram. Based on the initial inflow flow and the expected cumulative value of the faulty cables transmitted through all incoming paths, combined with the branch allocation ratio and the cross-sectional area of ​​the cable tray branch, the density of faulty cables in the cable tray cross section is obtained. This series of recursive operations integrates deterministic breakpoints with probabilistic trends, converging and branching along the unidirectional convergent cable tray network, and finally obtaining the expected density of broken network cables carried by each segment of the cable tray. This density comprehensively reflects the spatial congestion level and load-bearing capacity of the cable tray, representing the vulnerability of a segment of the cable tray to the point where a single external force can cut off the most cross-segment cables.

[0036] Preferably, in one embodiment of the present invention, the system is designed for a unidirectional trough flow diagram. Outgoing (distribution) nodes in Assuming it exists Each outgoing branch connects to various downstream receiving nodes in the surrounding area. The system denotes the set of these outgoing physical pipe gallery components as follows: subscript Indicates the first Branch destination. For newly demarcated outgoing branches generated after the severing operation, the system retains and inherits the entity component codes of their original parent pipelines. The system interacts with the digital twin model's backend database to extract the branch destinations. Its own physical building parameters, specifically including the total area of ​​the internal hollowed-out channels of the long strip of cable, denoted as the cross-sectional area for cable trough branch accommodation. Obtain and extract the calculated straight-line length of the building space corresponding to a segment of the cable tray, and denot it as the branch building length of the cable tray. .

[0037] Further, please refer to Figure 2 The flowchart illustrates a method for obtaining the routing branch allocation ratio according to an embodiment of the present invention, specifically including: Step S201: For each outgoing branch, substitute the branch building length into the preset logarithmic compensation function to obtain the length compensation logarithm, and use the cross-sectional area of ​​the branch accommodated per unit length compensation logarithm as the basic allocation ratio.

[0038] The analysis process for each outgoing branch is the same, and only one example is described here. Considering that in engineering practice, the larger the cross-sectional area of ​​the conduit, the more cables it can accommodate. When the cable route is unknown, large-section conduits tend to be selected. Although the length of the conduit has a suppressive effect on cable selection, the repulsive force caused by the difference in length of hundreds of meters in actual wiring should not reach the order of hundreds of times. Therefore, the building length of the cable tray branch is substituted into the logarithmic compensation function to obtain the length compensation logarithm. Then, the cross-sectional area of ​​the trunking branch with unit length compensation logarithm is obtained, that is, the cross-sectional area of ​​the trunking branch with unit length compensation logarithm is divided by this length compensation logarithm value to obtain the basic allocation ratio. This makes large cross-section trunking obtain a higher basic score, while long-distance trunking is only subject to a mild penalty after logarithmic smoothing.

[0039] As an example, the default logarithmic compensation function is ln(x+e), where x is the independent variable, ln() is the logarithmic function with base e, and e is the natural constant. The purpose of adding e to the function as compensation is as follows: If the original road segment is less than or equal to the value 1 (for example, the extreme length of the pipe opening is only 0.2 meters), the logarithmic operation will trigger the system to divide by zero or obtain an abnormal negative infinity / minimal number, thus disrupting the overall rule algorithm operation process; the area of ​​zero in the model usually indicates data omission. The actual physical cable trench must have space, so for the cable trench branch accommodating cross-sectional area in the numerator, if it is 0, a very small constant is assigned (0.01 square meters in this example) to maintain the minimum allocation weight of the branch, preventing the normal deduction path from being blocked due to missing model data. At the same time, in subsequent calculations, when the cable trench branch accommodating cross-sectional area is used as the denominator, the denominator is avoided to be 0.

[0040] In determining the direction, the shortest path lengths of the nodes at both ends of the undirected connected edge are compared, and the connection direction of the undirected connected edge is determined as the direction from the node with the longest shortest path length to the node with the shortest path length. The node with the longest shortest path length corresponds to the outgoing direction, and the node with the shortest path length corresponds to the incoming direction. This corresponds to an outgoing branch under the node with the longest shortest path length.

[0041] Step S202: Obtain the routing branch allocation ratio based on the basic allocation ratio of all outgoing branches under the same node.

[0042] In this example, the basic allocation ratio of all outgoing branches under the same node is normalized to output the routing branch allocation ratio. The basic allocation ratio of each outgoing branch under the same node is divided by the sum of the basic allocation ratios of all outgoing branches under the corresponding node, and the result is used as the routing branch allocation ratio of each outgoing branch under the corresponding node. This design transforms the absolute score of each outgoing branch into a set of probability weights with a sum of 1, so that the subsequent expected load splitting ratio at each branch intersection objectively quantifies the tendency of the duct's physical properties to guide the unknown cable route.

[0043] Furthermore, in order to simulate the state of cables converging towards the computer room along a predetermined physical pipe corridor in the blind area without drawings, the system does not cut the physical cables. Instead, it abstracts the disconnection event into the expected probability value that is passed down level by level along the pipe corridor. It calculates the expected cumulative value of faulty cables at each connected edge by starting from the outermost access node and working down level by level along the topology sequence.

[0044] Since the unidirectional cable channel flow graph has transformed the cable network into a unique unidirectional convergent structure where each edge points to a single point, but there are strict upstream and downstream dependencies between nodes, the convergent expectation of any node can only be calculated after the expectations of all its upstream nodes have been calculated. Therefore, the system first performs a computational ordering operation on the entire graph, calling a conventional topology sorting algorithm and sequentially marking all computational structure nodes within the graph according to a predetermined unidirectional gradient direction. This action constructs a recursive computation sequence that flows hierarchically from the outermost scattered nodes to the core data center terminal.

[0045] With any node For example: The system retrieves the set of cable tray access nodes. In the unidirectional cable tray flow diagram, the node identity is determined. belong If the location is determined to be a cable tray access node, then the coordinates do indeed show one or more reported physical peripheral communication terminal interface network address hosts. The total number of devices corresponding to the offline network addresses bound to each cable tray access node is used as the initial inflow traffic. .like Not belonging to If the node is identified as a non-groove access node, the initial inflow flow of the non-groove access node is set to 0.

[0046] After assigning the initial inflow flow value, the expected convergence of each node is calculated sequentially according to the recursive sequence determined by the topology sorting. Then, the density of faulty cables in the cable tray cross-section is calculated, specifically including: In a unidirectional cable tray flow diagram, the sum of the initial inflow flow of each node and the expected cumulative value of faulty cables on all incoming connected edges to the node is used as the node's total expected convergence. Each incoming connected edge represents an incoming path. The expected total convergence of the fusion node and the allocation ratio of the corresponding routing branch of the node are used to obtain the expected cumulative value of the faulty cable flowing through the corresponding outgoing branch; Based on the expected cumulative value of faulty cables and the corresponding cross-sectional area of ​​the cable tray branch, the density of faulty cables in the cable tray cross section is obtained.

[0047] As an example, With nodes The distribution ratio of routing branches along the j-th outgoing connected edge (outgoing branch) The product of the two nodes is used as a node. Expected cumulative value of faulty cables flowing out along the j-th outgoing connected edge ; Then Divide by the cross-sectional area of ​​the cable tray branch The quotient is used as the corresponding fault cable density of the cable tray cross-section. This ratio eliminates the dilution effect of large cross-sections on the expected total, making the vulnerable pipe sections with small cross-sections but high expected loads numerically prominent, accurately characterizing the relative congestion significance of each pipe tray as a common convergence throat.

[0048] It should be noted that in the unidirectional cable tray flow diagram, the expected cumulative value of faulty cables carried by the incoming connected edge of each node is the result passed down from the upstream node along the corresponding outgoing connected edge. Once the total expected value of the upstream node has been calculated, this expected value needs to be distributed downwards according to the routing branch allocation ratio of each outgoing branch of the upstream node. The portion of the expected value allocated to a particular outgoing branch is defined as the expected cumulative value of faulty cables for that outgoing connected edge. This value serves two purposes: firstly, it indicates the load carried by the connected edge as an attribute; secondly, it is passed down along the topology sequence to the downstream node, participating in the summation as the expected value of the downstream node's incoming connected edge. For the access node at the very front of the topology sequence, since no upstream node passes on the expected value to it, its set of incoming connected edges is empty. Therefore, the total expected value of this node is directly equal to its initial inflow, which is the total number of offline devices bound to this node. This serves as the starting source of the entire recursive chain.

[0049] Step S3: Locate the geographical location of the damaged cable tray with the highest density of faulty cables across the cable tray cross-section, and issue a dispatch order.

[0050] Considering that the emergency repair personnel need to go directly to the only point of failure to carry out the work after arriving at the site, and that the density value directly reflects the urgency of each cable tray becoming a common convergence bottleneck, the geographical location of the damaged cable tray with the largest cross-sectional area of ​​faulty cables is located, and a dispatch order is issued. This allows the load information scattered throughout the pipeline network to be finally narrowed down to a clear physical target, and the massive number of network-level outage alarms are attributed to physical damage points on the ground that urgently need to be repaired, thereby improving the efficiency of restoring network communication.

[0051] In a preferred embodiment of the present invention, the center coordinates (in the digital twin model) of the damaged cable tray corresponding to the fault cable density with the largest numerical value are extracted. Then, the pre-defined spatial inclusion relationship interface of the digital twin model is called to... Input a query to obtain structured geographic location text containing building area codes (such as specific terminal floor and area codes) and well numbers, directly indicating the specific physical door or well number where the damage occurred, thus completing the substantive conversion from machine positioning to engineering scheduling instructions.

[0052] Then, it is encapsulated with the physical connection operation instructions (the operation instruction text that requires the restoration of physical connections between cross-network segment cables) to generate a troubleshooting work order data packet; Obtain the three-dimensional spatial coordinates of each maintenance terminal, and send the troubleshooting work order data packet to the center coordinates. The maintenance terminal with the smallest straight-line distance (Euclidean distance).

[0053] The maintenance terminal displays the text content of the troubleshooting work order data package through a visualization module such as a screen. After receiving the work order, the field engineer holding the maintenance terminal can directly read the structured geographical location text and go directly to the corresponding low-voltage pipe well to carry out the connection and repair work at the damaged cable tray.

[0054] It should be noted that when the density of faulty cables in multiple cable tray sections is the highest, the center coordinates of the damaged cable trays corresponding to the density of faulty cables in each cable tray section are extracted. The Euclidean distance from the center coordinates of each damaged cable tray to the end point of the cable tray in the computer room is calculated, and the coordinate with the smallest Euclidean distance is selected as the final extracted center coordinates of the damaged cable tray.

[0055] An embodiment of the present invention also provides an airport network operation and maintenance management system based on digital twins. The system includes a memory, a processor, and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and the computer program, when running in the processor, can implement the airport network operation and maintenance management method based on digital twins described in steps S1-S3.

[0056] In summary, to address the technical problem of failing to quickly and accurately locate the physical damage points of damaged pipelines in network outages due to the lack of detailed wiring diagrams, thus delaying repair opportunities, this invention aims to provide an airport network operation and maintenance management method and system based on digital twins. This invention first calibrates the offline network address set and the three-dimensional absolute coordinates of the devices to generate an undirected cable tray connectivity graph; further, it determines the connectivity direction and removes isolated nodes and equidistant redundant edges, outputting a unidirectional cable tray flow graph; it further obtains the routing branch allocation ratio, and based on the initial inflow traffic and the expected cumulative value of faulty cables transmitted through all inflow paths, combined with the cable tray branch capacity cross-sectional area, it obtains the faulty cable density of the cable tray cross-section; finally, it locates the geographical location of the damaged cable tray and issues a scheduling command. First, a unidirectional cable tray flow graph is generated after being constrained by floor and fire compartment, and isolated and equidistant redundancies are removed. Then, the branch allocation ratio is determined, the faulty cable density of the cable tray cross-section is obtained along the topology sequence, the geographical location of the point with the highest density is located, and a scheduling command is issued, improving the efficiency of restoring network communication.

Claims

1. A method for airport network operation and maintenance management based on digital twins, characterized in that, The method includes: Extract the event analysis window based on the changes in offline incremental data of the device, and calibrate the set of offline network addresses; obtain the three-dimensional absolute coordinates of the device corresponding to the offline network address, and calculate the perpendicular foot by projecting it onto the main trunk of the digital twin model under the consistency constraints of floor and fire compartment, and generate an undirected trunk connectivity graph; determine the connectivity direction and remove isolated nodes and equidistant redundant edges based on the shortest routing length from each node to the end point of the computer room trunk based on the undirected trunk connectivity graph, and output a unidirectional trunk flow graph; Based on the cross-sectional area of ​​the outgoing branch and the building length of the branch in the unidirectional cable tray flow diagram, the cable branch allocation ratio is obtained; along the topological sequence of the unidirectional cable tray flow diagram, the initial inflow flow of the node is determined; based on the initial inflow flow and the expected cumulative value of faulty cables transmitted through all incoming paths, combined with the cable branch allocation ratio and the cross-sectional area of ​​the branch, the faulty cable density of the cable tray cross section is obtained. Locate the geographical location of the damaged cable tray with the highest density of faulty cables at the cross-section of the cable tray, and issue a scheduling order; The method for obtaining the undirected cable tray connectivity graph includes: extracting the building trunk cable tray set from the digital twin model; selecting each device as a target device; selecting line segments in the building trunk cable tray set that are on the same floor and in the same fire compartment as the target device's three-dimensional absolute coordinates to form a target line segment set; obtaining the shortest vertical distance from the target device's three-dimensional absolute coordinates to all target line segments; when the minimum shortest vertical distance is less than a preset maximum adsorption distance, obtaining an orthogonal perpendicular point on the corresponding target line segment; calling the three-dimensional coordinates of historical access nodes and combining them with a preset merging distance threshold to determine whether the orthogonal perpendicular point is a new perpendicular point node; summarizing the cable tray segments formed by cutting and the undivided original cable tray segments, and combining each new perpendicular point node and the historical access node to generate the undirected cable tray connectivity graph. The method for obtaining the expected cumulative value of the faulty cable includes: deducing step by step along the topological sequence of the unidirectional cable tray flow diagram to obtain the expected cumulative value of the faulty cable for each connected edge in the unidirectional cable tray flow diagram; The method for obtaining the initial inflow traffic includes: in the unidirectional cable tray flow diagram, setting the initial inflow traffic of non-cable tray access nodes to 0; and using the total number of devices corresponding to the offline network address bound to each cable tray access node as the initial inflow traffic.

2. The airport network operation and maintenance management method based on digital twins according to claim 1, characterized in that, The method for obtaining the offline network address set includes: Obtain the differential slope of the offline incremental data of the device. When the differential slope is greater than the preset sudden disconnection rate threshold, open the event analysis window. When the incremental value returns to zero or the opening time reaches the preset maximum collection duration, close the event analysis window and generate an offline network address set.

3. The airport network operation and maintenance management method based on digital twins according to claim 1, characterized in that, The method for obtaining the unidirectional wire groove flow diagram includes: Identify isolated nodes whose shortest path length is infinite or greater than a preset maximum scalar value, and remove the isolated nodes and their associated connected edges from the undirected slot connectivity graph; For the undirected connected edges in the undirected slotted connected graph after removing isolated nodes, the shortest path lengths of the nodes at both ends of the undirected connected edge are compared, and the connection direction of the undirected connected edge is determined to be from the node with the longest shortest path length to the node with the shortest path length; when the shortest path lengths of the nodes at both ends of the undirected connected edge are equal, the undirected connected edge is removed as an equidistant redundant edge. The unidirectional groove flow diagram is constructed based on each connected edge after the connection direction is determined.

4. The airport network operation and maintenance management method based on digital twins according to claim 1, characterized in that, The method for obtaining the routing branch allocation ratio includes: For each outgoing branch, the length of the cable tray branch building is substituted into the preset logarithmic compensation function to obtain the length compensation logarithm. The cross-sectional area of ​​the cable tray branch per unit length compensation logarithm is used as the basic allocation ratio. The routing branch allocation ratio is obtained based on the basic allocation ratio values ​​of all outgoing branches under the same node.

5. The airport network operation and maintenance management method based on digital twins according to claim 1, characterized in that, The method for obtaining the density of faulty cables in the cross-section of the cable tray includes: The sum of the initial inflow flow of each node and the expected cumulative value of faulty cables of all incoming connected edges input to the node is used as the total expected convergence of the node; the total expected convergence of the node is combined with the routing branch allocation ratio of the node to obtain the expected cumulative value of faulty cables flowing through the corresponding outgoing branch. Based on the expected cumulative value of the faulty cables and the corresponding cross-sectional area of ​​the cable tray branch, the density of faulty cables in the cable tray cross-section is obtained.

6. The airport network operation and maintenance management method based on digital twin according to claim 1, characterized in that, The method for issuing scheduling instructions includes: Extract the center coordinates of the damaged cable tray corresponding to the fault cable density with the largest value of the cable tray cross-section, call the spatial inclusion relationship interface to obtain the structured geographic location text containing the building area code and manhole number, and encapsulate it with the physical connection operation instructions to generate a troubleshooting work order data package; obtain the three-dimensional spatial coordinates of each maintenance terminal, and send the troubleshooting work order data package to the maintenance terminal with the smallest straight-line distance from the center coordinates.

7. A digital twin-based airport network operation and maintenance management system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the airport network operation and maintenance management method based on digital twins as described in any one of claims 1 to 6.

Images