ASON biplane networking equivalent diagram dynamic modeling intelligent method and system based on geographic information fusion

Abstract

The invention discloses an ASON biplane networking equivalent diagram dynamic modeling intelligent method and system based on geographic information fusion, and relates to the technical field of electric power communication, and the method comprises the steps: obtaining an optical cable geographic path graph layer and a plant station node information graph layer of a first plane and a second plane in ASON biplane networking; based on the optical cable geographic path layer and the plant station node information layer, fusing the two planes by adopting a layer merging algorithm based on a graph neural network to form a unified equivalent graph; based on the unified equivalent diagram, establishing a bidirectional mapping model between an ASON logic optical path taking a plant station node as a starting and ending point and a physical path in the unified equivalent diagram; and according to a graph neural network-based layer merging algorithm and a bidirectional mapping model, in response to a network topology change or a fault event, triggering local re-fusion updating of the unified equivalent graph. According to the biplane equivalent graph construction, accurate fusion of biplane physical resources is realized.

Description

Intelligent Method and System for Dynamic Modeling of ASON Biplane Network Equivalent Map Based on Geographic Information Fusion Technical Field

[0001] This invention relates to the field of power communication technology, and in particular to an intelligent method and system for dynamic modeling of ASON dual-plane networking equivalent graphs based on geographic information fusion. Background Technology

[0002] Automatically Switched Optical Networks (ASON) dual-plane networking, with its physical redundancy design, provides high reliability for backbone transmission in critical sectors such as power and telecommunications, and has become the core architecture of intelligent optical transport networks.

[0003] Currently, in the management of Automatically Switched Optical Networks (ASON) dual-plane networking, the network topology equivalent map is usually manually constructed based on static data, lacking the integration of the actual geographical paths of physical optical cables. Existing methods struggle to effectively identify the physical resource sharing relationships between the two planes, resulting in the equivalent map failing to accurately reflect the network structure. Summary of the Invention

[0004] To address the problem that existing technologies struggle to effectively identify physical resource sharing relationships between two planes, resulting in equivalent maps failing to accurately reflect the network structure, this invention provides an intelligent method and system for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion. By constructing dual-plane equivalent maps using a layer fusion algorithm based on graph neural networks, accurate fusion of physical resources between the two planes can be achieved, improving the accuracy of the equivalent map's representation of physical resource sharing relationships. The specific technical solution is as follows: This invention provides an intelligent method for dynamic modeling of the equivalent graph of ASON dual-plane networking based on geographic information fusion, including: acquiring the optical cable geographic path layers and plant node information layers of the first and second planes in the ASON dual-plane networking; based on the optical cable geographic path layers and plant node information layers, using a layer merging algorithm based on graph neural networks to merge the two planes, identifying physical resources with shared paths or co-cable relationships, and merging the physical resources as shared edges to form a unified equivalent graph reflecting the shared relationship of physical resources in the two planes; based on the unified equivalent graph, establishing a bidirectional mapping model of traceable bidirectional mapping relationship between the ASON logical optical paths with plant nodes as the starting and ending points and the physical paths in the unified equivalent graph; according to the layer merging algorithm based on graph neural networks and the bidirectional mapping model, in response to network topology changes or fault events, triggering a local re-fusion update of the unified equivalent graph.

[0005] Preferably, the layer merging algorithm based on graph neural networks for fusing the two planes includes: modeling the optical cable networks of the first plane and the second plane as graph structures respectively; identifying candidate common edge pairs from the two planes based on the similarity of spatial location and direction according to the graph structures of the first plane and the second plane; extracting the spatial geometric features, topological connection features and attribute features of the candidate common edge pairs, and inputting them into the graph neural network for common relationship determination; and merging the edges confirmed as common or of the same cable into a single shared edge according to the determination result.

[0006] Preferably, modeling the optical fiber networks of the first plane and the second plane as graph structures includes: modeling the optical fiber network of the first plane as a first graph structure. The optical fiber network in the second plane is modeled as a second graph structure. ,in, and For the set of plant nodes, and It is a set of edges based on the geographical path of the optical cable, and each edge is accompanied by actual geographical coordinate sequence information.

[0007] Preferably, the step of identifying candidate common edge pairs from the two planes based on the graph structure of the first and second planes and the similarity of their spatial positions and orientations includes: analyzing the edge sets of the first and second planes through spatial buffer analysis and calculation of the spatial relationships of linear features. and Pairwise analysis is performed, and when the spatial distance between two edges is within a preset threshold and the similarity of their paths is higher than a specific threshold, they are identified as candidate common-path edge pairs.

[0008] Preferably, the step of establishing a traceable bidirectional mapping model between the ASON logical optical path with the plant node as the starting and ending point and the physical path in the unified equivalent diagram based on the unified equivalent diagram includes: using an association rule learning algorithm to automatically mine the association relationship between the starting point, ending point and bandwidth parameters of the logical optical path and the edge sequence of the physical path in the unified equivalent diagram, and forming a traceable bidirectional mapping relationship table.

[0009] Preferably, the event that triggers the local re-fusion update of the unified equivalent map also includes a change event for the plant node information layer, which includes adding plant nodes, modifying node attributes, and deactivating nodes.

[0010] Preferably, triggering the local re-fusion update of the unified equivalent graph includes: calling the layer merging algorithm based on graph neural network to perform local recalculation and re-fusion of the affected area, and synchronously updating the bidirectional mapping model.

[0011] Preferably, the spatial geometric features include at least one of average Hausdorff distance, overlap, and parallelism; the topological connectivity features include the similarity of the hierarchy and type of the nodes connected by the candidate common-path pairs in the network; and the attribute features include at least one of the optical cable type, capacity, and ownership information.

[0012] Preferably, the attribute features also include node reliability levels. When determining common paths, the graph neural network-based layer merging algorithm incorporates the difference in node reliability levels as a penalty term into the topology connectivity features and matches different common path determination thresholds for nodes with different reliability levels.

[0013] This invention also provides an intelligent system for dynamic modeling of the equivalent map of ASON dual-plane networking based on geographic information fusion. Applying the aforementioned method, the system includes: a data acquisition unit for acquiring optical cable geographic path layers and plant node information layers of the first and second planes in the ASON dual-plane networking; a layer fusion unit for fusing the two planes based on the optical cable geographic path layers and plant node information layers using a layer merging algorithm based on graph neural networks, identifying physical resources with shared paths or cables, and merging these physical resources as shared edges to form a unified equivalent map reflecting the shared relationship of physical resources in the two planes; a mapping unit for establishing a bidirectional mapping model based on the unified equivalent map, showing a traceable bidirectional mapping relationship between the ASON logical optical paths starting and ending at plant nodes and the physical paths in the unified equivalent map; and a dynamic update unit for triggering a local re-fusion update of the unified equivalent map in response to network topology changes or fault events, based on the layer merging algorithm and bidirectional mapping model based on graph neural networks.

[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: The intelligent method and system for dynamic modeling of ASON dual-plane network equivalent graphs based on geographic information fusion, by fusing the optical cable geographic path layer and the plant node information layer, and employing a graph neural network layer merging algorithm, accurately identifies physical resource sharing relationships such as shared paths and co-cables between the two planes, and presents them uniformly in the form of shared edges. This solves the problem that traditional static modeling lacks geographic information fusion and cannot truly reflect the network structure. Simultaneously, a traceable bidirectional mapping model between logical optical paths and physical paths is established, enabling resource relationship matching and querying without manual intervention. Furthermore, based on an event-driven mechanism, local re-fusion and updates of the equivalent graph are triggered, addressing the limitations of traditional static models in adapting to dynamic network changes. Attached Figure Description

[0015] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0016] Figure 1 is a flowchart of the intelligent method for dynamic modeling of equivalent maps of ASON dual-plane networking based on geographic information fusion according to the present invention.

[0017] Figure 2 is a schematic diagram of the fusion process of the layer merging algorithm based on graph neural network of the present invention.

[0018] Figure 3 is a schematic diagram of the intelligent system for dynamic modeling of ASON dual-plane networking equivalent map based on geographic information fusion according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] It should be understood that, when used in this specification, the terms “comprising” and “including” indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0021] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0022] It should also be further understood that the term "and / or" as used in this specification refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes such combinations.

[0023] Please refer to Figures 1 to 3 for the following examples.

[0024] This invention provides an intelligent method for dynamic modeling of equivalent maps of ASON dual-plane networking based on geographic information fusion, including: Step S1, acquiring the optical cable geographic path layers and power plant node information layers of the first and second planes in the ASON dual-plane networking; acquiring physical resource data of the ASON dual-plane networking from the system, wherein the physical resource data includes the optical cable geographic path layer and power plant node information layer of the first plane, and the optical cable geographic path layer and power plant node information layer of the second plane; cleaning and calibrating the acquired optical cable geographic path layers to form a standardized geographic information data base; in specific implementation, the physical resource data of the ASON dual-plane networking is retrieved in batches from the physical resource database of the power grid intelligent system through the API interface, including: the optical cable geographic path layer of the first plane: containing the starting coordinates, ending coordinates, intermediate point coordinate sequence, optical cable type, laying method and other attributes of the optical cable segment; the optical cable geographic path layer of the second plane: the data format is consistent with that of the first plane; the power plant node information layer: containing the unique identifier, latitude and longitude coordinates (WGS84 coordinate system), network level, equipment type, commissioning time and other attributes of all power plant nodes.

[0025] Abnormal data is detected and processed in batches, including deleting duplicate fiber optic cable segment records, repairing broken paths, and removing invalid attribute values. The preprocessing of the plant node information layer standardizes the attributes of the plant node information by unifying the enumeration values ​​at the network level and associating the start / end points of the fiber optic cable paths in the first / second plane using unique identifiers.

[0026] By standardizing attributes and associating them with unique identifiers, it is ensured that the start / end point of the dual-plane optical cable path corresponds one-to-one with the physical plant, avoiding topological distortion of the equivalent graph caused by edge and node mismatches. Unifying the attribute fields and formats of the dual-plane layers makes the optical cable resources of the first and second planes comparable.

[0027] Step S2: Based on the optical cable geographic path layer and the plant node information layer, a layer merging algorithm based on graph neural networks is used to fuse the two planes, identify physical resources with shared paths or co-cable relationships, and merge these physical resources as shared edges to form a unified equivalent graph reflecting the shared physical resource relationships of the two planes; the optical cable geographic path layers of the first plane and the second plane processed in Step S1 are superimposed, and a layer merging algorithm based on graph neural networks (GNN) is used to intelligently fuse the superimposed layers; wherein, the merging process of the layer merging algorithm of graph neural networks (GNN) is as follows: the optical cable networks of the two planes are modeled as graph structures respectively, and the features of nodes and edges are learned through GNN to identify optical cable paths in different planes that are spatially close and have shared paths or co-cable relationships, and these paths are merged into a shared physical edge in the equivalent graph; specifically, the merging of the two planes using the layer merging algorithm based on graph neural networks includes: modeling the optical cable networks of the first plane and the second plane as graph structures respectively; modeling the optical cable network of the first plane as a graph structure ,in For the set of plant nodes, Given the set of edges based on the geographical paths of optical cables; the optical cable network in the second plane is modeled as follows: Each edge Each of them comes with its actual geographic coordinate sequence information.

[0028] Based on the graph structure of the first and second planes, and considering the similarity in spatial location and orientation, candidate common edge pairs are identified from the two planes; spatial buffer analysis and spatial relationship calculation of linear features are then used to... and All edges are analyzed pairwise. When the spatial distance between two edges (from the first plane and the second plane, respectively) is within a preset threshold and their path similarity is higher than a certain threshold, they are determined to be a pair of candidate co-path edges. It is then fed into a subsequent GNN for accurate determination.

[0029] Using spatial analysis tools and All edges are matched pairwise: Spatial distance calculation: For any pair of edges, the average Hausdorff distance between them (reflecting the overall spatial deviation) is calculated. When the distance is less than or equal to the distance threshold (which can be dynamically adjusted according to the terrain), the next step of judgment is performed. Direction similarity calculation: The equation of a straight line is fitted by the coordinate sequence of each edge, and the angle between the two straight lines is calculated. When the angle is less than or equal to the angle threshold (the directions are basically parallel), they are judged as candidate common-path edge pairs and stored in the candidate set.

[0030] The spatial geometric features, topological connectivity features, and attribute features of the candidate co-path edge pairs are extracted and input into a graph neural network for co-path relationship determination. Through the co-path relationship determination based on a graph neural network (GNN), each candidate co-path edge pair is... Construct a rich feature vector as input to the GNN. Features include: spatial geometric features such as the average Hausdorff distance between two edges, overlap, and parallelism.

[0031] Topological connectivity features: such as the nodes connected by two edges ( and Are the levels and types in the network similar?

[0032] Attribute characteristics: such as whether the model, capacity, and ownership information of the optical cable are consistent or compatible.

[0033] Based on the judgment results, edges that are confirmed to be on the same path or cable will be merged into a single shared edge.

[0034] Based on the classification results of GNN, the graph structures are fused, and the entire biplane fused graph is obtained. The data is input into the GNN model. GNNs, through a message-passing mechanism, aggregate information from nodes and their neighborhoods to learn the complex structure and relationships in the graph. The ultimate task of GNNs is to perform binary classification on each candidate co-path edge pair, outputting the probability that it is a true co-path or a non-co-path. Edge pairs determined to be true co-paths indicate that they are actually optical cables laid in the same bundle or channel in physical terms.

[0035] It should be noted that for edge pairs determined to be true colpaths... In the equivalent graph, edges are not drawn repeatedly but merged into a single shared physical edge; edges not judged as shared paths, and edges unique to either plane, are retained as independent edges in the equivalent graph. Finally, all nodes... Connect the merged edge set to generate a unified equivalent graph that accurately reflects the shared physical resources relationship between the two planes.

[0036] The generated unified equivalent graph accurately reflects the network's physical structure. By integrating dual-plane nodes and fused edge sets, the equivalent graph is no longer an abstract model detached from geographical reality, but a tool that can intuitively present the actual laying trajectory and sharing relationships of optical cables.

[0037] Step S3: Based on the unified equivalent diagram, establish a bidirectional mapping model with traceable bidirectional mapping relationship between the ASON logical optical path with the plant node as the starting and ending point and the physical path in the unified equivalent diagram; use an association rule learning algorithm to automatically mine the association relationship between the starting point, ending point and bandwidth parameters of the logical optical path and the physical path edge sequence in the unified equivalent diagram to form a traceable bidirectional mapping relationship table.

[0038] Logical optical path information is obtained from the path calculation unit (PCE) of the ASON control plane, including: optical path ID, starting logical node (the unique identifier of the corresponding plant node), ending logical node, bandwidth (e.g., 10G / 100G), protection type (1+1 / 1:1), establishment time, etc.

[0039] Association rule learning and mapping table construction employ an improved FP-growth algorithm to mine association relationships: frequent itemsets are constructed using the start, end, and bandwidth of the logical optical path as the prefixes and the edge sequences of the physical paths in the unified equivalent graph as the posteriors. Association rules are generated by setting minimum support and minimum confidence. A bidirectional mapping table is established, containing fields such as logical optical path ID, physical path edge sequence, association confidence, establishment time, and update time. It supports logical-to-physical lookup (inputting the optical path ID returns the edge sequence) and physical-to-logical lookup (inputting the edge ID returns all associated optical paths).

[0040] The traceability guarantee of the mapping is achieved by adding a version control field to the mapping relationship table to record the timestamp and changes of each rule update; when the logical optical path is adjusted or the physical path is changed, a historical snapshot is automatically generated, which supports backtracking queries based on timestamps, such as querying the association status of the optical path and physical path at a certain point in time.

[0041] Step S4: Based on the graph neural network-based layer merging algorithm and bidirectional mapping model, in response to network topology changes or fault events, trigger a local re-fusion update of the unified equivalent graph.

[0042] Specifically, the events that trigger the local re-fusion update of the unified equivalent map also include the change event of the plant node information layer, which includes adding plant nodes, modifying node attributes, and deactivating nodes.

[0043] Specifically, trigger events are captured in real time through the following methods: Network topology changes: Monitor the topology announcement protocol of the ASON control plane, and trigger an update when the addition / deletion of optical cables or changes in node connection relationships are detected; Fault events: Receive alarm information from the Network Management System (NMS), such as optical cable interruption alarms and node failure alarms, and extract the fault location; Plant node changes: Monitor the database change log of the plant node information layer, such as the addition of nodes, attribute modification, and node deactivation, and trigger an update through triggers.

[0044] Specifically, triggering the local re-fusion update of the unified equivalent graph includes: calling the layer merging algorithm based on graph neural network to perform local recalculation and re-fusion of the affected area, and synchronously updating the bidirectional mapping model.

[0045] The affected area is defined based on the topology of the unified equivalent graph. Breadth-First Search (BFS) can be used to define the minimum affected area. Only the affected part is recalculated, rather than the entire graph is refused, which greatly shortens the update time and ensures a fast response even in scenarios with frequent network changes. The affected area is determined by starting from the faulty / changed node and searching for edges that are directly related within 1 hop and nodes that are indirectly related within 2 hops to form the minimum affected area. For changes in optical cable paths, only the changed path and its associated nodes are included in the affected area.

[0046] The GNN layer merging algorithm in step S2 is invoked, and the graph structure within the affected area is recalculated only: candidate edge pairs are re-selected, features are extracted, and common path relationships are determined to generate a local fusion result; the local result is then concatenated with the original equivalent graph, replacing the edge / node information in the affected area. For the physical path edge sequence in the affected area, the association rule learning in step S3 is re-executed to update the associated logical optical path records in the mapping table.

[0047] By monitoring events such as topology changes, faults, and node alterations in real time, updates are triggered immediately, preventing the equivalent graph from becoming disconnected from the actual network. Local re-fusion design improves update efficiency and reduces system computing power consumption.

[0048] Specifically, in a preferred embodiment of this application, the attribute features further include node reliability levels. When determining common paths, the graph neural network-based layer merging algorithm incorporates the difference in node reliability levels as a penalty term into the topology connection features and matches different common path determination thresholds for nodes with different reliability levels.

[0049] In practice, the specific implementation process for introducing node reliability levels into attribute features and applying them to GNN co-path determination is as follows: Based on the equipment redundancy, historical failure rate, and importance of the services carried by the plant nodes, the reliability levels of the plant nodes are divided into different levels. The level information is stored in the reliability level field of the plant node information layer and is represented numerically.

[0050] Among them, the fusion method of node reliability level in GNN co-path determination includes (i) calculating the penalty term as a topological connection feature for candidate co-path edge pairs. Extract the reliability level of its connected nodes, and assume... The connection node is (grade ), (grade ); The connection node is (grade ), (grade ).

[0051] Compute node reliability level difference: Starting level difference: Δ =| - |;Difference in final grade: ΔR t =| - |; Take the maximum of the two values ​​as the basis for the penalty coefficient: ΔR = max(Δ ΔR t ), penalty term P=ΔR×0.1 (if ΔR=2, P=0.2), used to adjust the weight of topological connection features: adjusted topological feature value = original topological feature value × (1-P) (II) Dynamically adjust the common path judgment threshold According to the lowest reliability level of the edge pair connection nodes, set different GNN output probability thresholds: if there are high reliability level nodes in the edge pair connection nodes, the judgment threshold is increased to 0.9 (assuming the default threshold is 0.85); if the edge pair connection nodes are all medium reliability level, the judgment threshold remains at 0.85; if the edge pair connection nodes contain low reliability level nodes, the judgment threshold is reduced to 0.8.

[0052] After the GNN outputs the common path probability, it performs a second verification by combining the dynamic threshold and the adjusted topological feature value. When the probability is greater than or equal to the dynamic threshold and the adjusted topological feature value is ≥0.5, it is finally determined to be a true common path; otherwise, it is determined to be a non-common path.

[0053] By differentiating node reliability levels, co-path identification becomes more aligned with service requirements. High-reliability node connections require higher redundancy, and strict thresholds prevent double failures caused by misidentification as co-paths. Conversely, low-reliability node connections have lower redundancy requirements, and relaxed thresholds allow for more flexible identification of actual co-path resources. Using the difference in node reliability levels as a penalty reduces the probability of high-reliability and low-reliability node connections being identified as co-paths.

[0054] This invention also provides an intelligent system for dynamic modeling of the equivalent map of ASON dual-plane networking based on geographic information fusion. Applying the aforementioned method, the system includes: a data acquisition unit for acquiring optical cable geographic path layers and plant node information layers of the first and second planes in the ASON dual-plane networking; a layer fusion unit for fusing the two planes based on the optical cable geographic path layers and plant node information layers using a layer merging algorithm based on graph neural networks, identifying physical resources with shared paths or cables, and merging these physical resources as shared edges to form a unified equivalent map reflecting the shared relationship of physical resources in the two planes; a mapping unit for establishing a bidirectional mapping model based on the unified equivalent map, showing a traceable bidirectional mapping relationship between the ASON logical optical paths starting and ending at plant nodes and the physical paths in the unified equivalent map; and a dynamic update unit for triggering a local re-fusion update of the unified equivalent map in response to network topology changes or fault events, based on the layer merging algorithm and bidirectional mapping model based on graph neural networks.

[0055] The functional explanation of each unit in this embodiment is the same as that of the ASON dual-plane networking equivalent map dynamic modeling intelligent method based on geographic information fusion, and the technical effects are the same, so it will not be repeated here.

[0056] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0057] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical functional division. In actual implementation, there may be other division methods, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored.

[0058] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0059] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the specification of the present invention.

Claims

1. An intelligent method for dynamic modeling of equivalent maps of ASON dual-plane networking based on geographic information fusion, characterized in that, include: Obtain the optical cable geographic path layers and plant node information layers in the first and second planes of the ASON dual-plane network; Based on the optical cable geographic path layer and the plant node information layer, a layer merging algorithm based on graph neural networks is used to fuse the two planes, identify physical resources with shared paths or cables, and represent these physical resources as shared edges to form a unified equivalent graph reflecting the shared relationship of physical resources in both planes. Based on the unified equivalent graph, a bidirectional mapping model is established to establish a traceable bidirectional mapping relationship between the ASON logical optical path with plant nodes as the starting and ending points and the physical path in the unified equivalent graph. According to the layer merging algorithm based on graph neural networks and the bidirectional mapping model, in response to network topology changes or fault events, a local re-fusion update of the unified equivalent graph is triggered.

2. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 1, characterized in that, The layer merging algorithm based on graph neural networks for fusing the two planes includes: modeling the optical cable networks of the first and second planes as graph structures respectively; identifying candidate common edge pairs from the two planes based on the similarity of spatial location and direction according to the graph structures of the first and second planes; extracting the spatial geometric features, topological connection features, and attribute features of the candidate common edge pairs, and inputting them into the graph neural network for common relationship determination; and merging the edges confirmed as common or on the same cable into a single shared edge according to the determination result.

3. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 2, characterized in that, The step of modeling the optical fiber networks of the first plane and the second plane as graph structures includes: modeling the optical fiber network of the first plane as a first graph structure. The optical fiber network in the second plane is modeled as a second graph structure. ,in, and For the set of plant nodes, and It is a set of edges based on the geographical path of the optical cable, and each edge is accompanied by actual geographical coordinate sequence information.

4. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 3, characterized in that, The step of identifying candidate common edge pairs from the two planes based on the graph structure of the first and second planes and the similarity of their spatial positions and orientations includes: analyzing the edge sets of the first and second planes through spatial buffer analysis and calculation of the spatial relationships of linear features. and Pairwise analysis is performed, and when the spatial distance between two edges is within a preset threshold and the similarity of their paths is higher than a specific threshold, they are identified as candidate common-path edge pairs.

5. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 1, characterized in that, The step of establishing a traceable bidirectional mapping model based on the unified equivalent graph, which establishes a traceable bidirectional mapping relationship between the ASON logical optical path with the plant node as the starting and ending point and the physical path in the unified equivalent graph, includes: using an association rule learning algorithm to automatically mine the association relationship between the starting point, ending point, and bandwidth parameters of the logical optical path and the edge sequence of the physical path in the unified equivalent graph, and forming a traceable bidirectional mapping relationship table.

6. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 1, characterized in that, The events that trigger the local re-fusion update of the unified equivalent map also include the change event of the plant node information layer, which includes adding plant nodes, modifying node attributes, and deactivating nodes.

7. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 1, characterized in that, The triggering of the local re-fusion update of the unified equivalent graph includes: calling the layer merging algorithm based on graph neural network to perform local recalculation and re-fusion of the affected area, and synchronously updating the bidirectional mapping model.

8. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 2, characterized in that, The spatial geometric features include at least one of average Hausdorff distance, overlap, and parallelism; the topological connectivity features include the similarity of the hierarchy and type of the nodes connected by the candidate common-path pairs in the network; and the attribute features include at least one of the optical cable type, capacity, and ownership information.

9. The intelligent method for dynamic modeling of ASON dual-plane network equivalent maps based on geographic information fusion according to claim 8, characterized in that, The attribute features also include node reliability levels. When determining common paths, the graph neural network-based layer merging algorithm incorporates the difference in node reliability levels as a penalty term into the topology connection features and matches different common path determination thresholds for nodes with different reliability levels.

10. An intelligent system for dynamic modeling of equivalent maps of ASON dual-plane networking based on geographic information fusion, characterized in that, The method according to any one of claims 1 to 9 comprises: a data acquisition unit, configured to acquire optical cable geographic path layers and plant node information layers of the first and second planes in an ASON dual-plane network; a layer fusion unit, configured to fuse the two planes based on the optical cable geographic path layers and plant node information layers using a layer merging algorithm based on a graph neural network, identify physical resources with shared paths or co-cable relationships, and merge the physical resources as shared edges to form a unified equivalent graph reflecting the shared relationship of physical resources in the two planes; a mapping unit, configured to establish a bidirectional mapping model based on the unified equivalent graph, establishing a traceable bidirectional mapping relationship between the ASON logical optical paths with plant nodes as the starting and ending points and the physical paths in the unified equivalent graph; and a dynamic update unit, configured to trigger a local re-fusion update of the unified equivalent graph in response to network topology changes or fault events, according to the layer merging algorithm based on a graph neural network and the bidirectional mapping model.

Images