Wiring monitoring method and system based on digital twinning

Abstract

The invention discloses a digital twinning-based wiring monitoring method and system, and relates to the technical field of wiring monitoring, and the method comprises the steps: building a digital twinning model of wiring based on the operation data of a wiring unit; dynamically updating the digital twin model in combination with real-time operation data of the wiring unit; the method comprises the following steps: acquiring multi-modal data of a wiring unit through a multi-modal sensor cluster, aligning a timestamp of the multi-modal data, and integrating the multi-modal data into original monitoring data; extracting original monitoring data features and generating state parameters by using multi-modal fusion based on deep learning; according to the fault condition report, a standby path is generated in combination with a digital twinborn model of wiring and a logic topological structure of a wiring unit; by constructing and dynamically updating the digital twinborn model of the wiring, accurate mapping and real-time synchronization of the structure and the state are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of wiring monitoring, and in particular to a wiring monitoring method and system based on digital twins. Background Art

[0002] With the rapid development of information technology, modern wiring has become increasingly complex and large-scale, which has put forward higher requirements on the reliability and stability of wiring. Traditionally, wiring monitoring mainly relies on management systems (NMS) based on standard protocols such as SNMP (Simple Management Protocol) and LLDP (Link Layer Discovery Protocol). These systems can provide basic topology discovery, equipment status monitoring and simple fault alarm functions.

[0003] In recent years, digital twin technology has gradually attracted attention as an emerging intelligent monitoring method. In the field, digital twin technology can be used to create virtual copies, allowing managers to simulate, optimize and maintain wiring in digital space. However, most existing technologies rely on static topological structures and cannot reflect the dynamic changes of wiring in real time. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a wiring monitoring method based on digital twins to solve the dynamic adaptability problem of existing wiring monitoring technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a wiring monitoring method based on digital twins, comprising:

[0008] Build a digital twin model of wiring based on the operation data of the wiring unit;

[0009] Dynamically update the digital twin model by combining the real-time operation data of the wiring unit;

[0010] Through the multimodal sensor cluster, the multimodal data of the wiring unit is collected, and the multimodal data timestamps are aligned to integrate into the original monitoring data;

[0011] Utilize multimodal fusion based on deep learning to extract features of original monitoring data and generate state parameters;

[0012] Based on the status parameters, the wiring unit is monitored in real time, abnormal events are detected and identified, and fault reports are generated;

[0013] Based on the fault report, an alternative path is generated by combining the digital twin model of the wiring and the logical topology of the wiring unit.

[0014] 11. As a preferred solution of the wiring monitoring method based on digital twins of the present invention, wherein: the digital twin model of wiring is constructed based on the operation data of the wiring unit, and the specific steps are as follows:

[0015] Collect the operation data of the wiring unit through monitoring sensors deployed at the nodes;

[0016] Use noise filtering algorithm to process the operation data;

[0017] Use SNMP and LLDP-based topology scanning tools to automatically scan the connection relationship of wiring devices;

[0018] Use the SNMP interface of switches and routers to obtain the MAC address, port status and connection information of the device and build the physical topology structure;

[0019] Matching the connection relationship data in the scan results with the operation data;

[0020] Use graph-theoretic methods to represent physical topology;

[0021] Construct the adjacency matrix A of the wiring unit and generate a logic graph describing the signal transmission path between nodes;

[0022] The output includes the logical topology of node locations, connection relationships, and signal transmission paths;

[0023] Use 3D laser scanning equipment to scan the physical position of the wiring unit and obtain the spatial coordinates of the wiring;

[0024] Process the spatial coordinates into a standardized point cloud data format, and align and integrate the data of multiple scanning areas through the ICP algorithm;

[0025] Automatically generate the geometric 3D model of the wiring unit using 3D modeling tools;

[0026] Match the nodes in the geometric 3D model with the node IDs in the logical topology structure;

[0027] The lines in the geometric model correspond to the connection relationships defined in the logical topology structure;

[0028] The operating data of nodes and transmission paths are mapped to the geometric three-dimensional model to obtain a digital twin model of the wiring.

[0029] As a preferred solution of the wiring monitoring method based on digital twins of the present invention, the digital twin model is dynamically updated in combination with the real-time operation data of the wiring unit. The specific steps are as follows:

[0030] By collecting real-time operation data, updating the node and line status, and mapping them to the corresponding nodes and lines of the digital twin model of the wiring;

[0031] When the logical topology changes, the adjacency matrix A is automatically updated.

[0032] As a preferred solution of the wiring monitoring method based on digital twins described in the present invention, wherein: the multimodal data of the wiring unit is collected through a multimodal sensor cluster, and the multimodal data timestamps are aligned to be integrated into the original monitoring data. The specific steps are as follows:

[0033] Deploy multimodal sensor clusters at the nodes of the wiring units and the branch points of the trunk lines;

[0034] Deploy fiber optic sensors on fiber optic lines and use distributed fiber optic sensing technology to monitor line stress changes and temperature;

[0035] Deploy electromagnetic sensors on cable lines and nodes to monitor cable current fluctuations and signal strength changes;

[0036] Deploy vibration sensors at physical nodes and line support locations to monitor mechanical vibrations;

[0037] The multimodal sensor cluster collects multimodal data at a uniform frequency to generate time series data;

[0038] Providing unified timestamps for multimodal sensor clusters through time synchronization protocols;

[0039] Integrate time-synchronized multimodal data into raw monitoring data in a unified format;

[0040] The original monitoring data is denoised, missing values ​​are filled and standardized.

[0041] As a preferred solution of the wiring monitoring method based on digital twins of the present invention, wherein: the multimodal fusion based on deep learning is used to extract the original monitoring data features and generate state parameters. The specific steps are as follows:

[0042] Extract features from unimodal raw monitoring data through a separate neural network (e.g., convolutional neural network);

[0043] Using the attention mechanism, the extracted unimodal original monitoring data feature vectors are weighted fused to generate a unified hidden feature representation;

[0044] Through the fully connected layer, the fused features are classified and regressed, and the output includes status parameters such as node load status, signal transmission strength, and line health status.

[0045] As a preferred solution of the wiring monitoring method based on digital twins of the present invention, the wiring unit is monitored in real time based on the state parameters, and abnormal events are detected and identified, and a fault report is generated. The specific steps are as follows:

[0046] Mapping state parameters into the digital twin model of wiring;

[0047] When the operating data is updated, the state parameters of the digital twin model are updated in real time;

[0048] Obtain a visualized digital twin model that displays the status parameters of the wiring unit in real time, including node load status, signal transmission strength, and line health status;

[0049] According to the impact of state parameter failure on the wiring unit, abnormal events are defined as node failure, signal attenuation failure and line vibration failure;

[0050] Through experimental analysis, the node fault threshold k1, signal attenuation fault threshold k2 and line vibration fault threshold k3 are set;

[0051] When the value of the node load status exceeds the node fault threshold k1, it is determined to be a node fault. When the signal transmission strength is less than the signal attenuation fault threshold k2, it is determined to be a signal attenuation fault. When the value of the line health status is less than the line vibration fault threshold k3, it is determined to be a line vibration fault.

[0052] A time series classification model based on LSTM (Long Short-Term Memory Network) to capture the temporal variation pattern of the operating status;

[0053] Use historical state parameter samples that annotate specific abnormal events (labeled as 1) and normal states (labeled as 0) and divide them into training sets and validation sets;

[0054] Input the training set into the classification model, output the predicted fault classification probability of the sample, and compare it with the actual label of the sample. When the maximum number of iterations is reached, the training ends.

[0055] Input the validation set, calculate the loss function of the classification model, and evaluate the performance of the classification model;

[0056] When the F1 score (an indicator that comprehensively considers precision and recall) of the classification model reaches the optimal value, the predicted probability of an abnormal event is the predicted probability threshold;

[0057] Input the state parameters obtained based on real-time multimodal data into the classification model, and output abnormal events and their probabilities;

[0058] If the probability of a certain type of abnormal event exceeds the predicted probability threshold, the abnormal event is determined to be a fact;

[0059] For detected abnormal events, the specific location of the fault is determined by combining the topological structure of the digital twin model;

[0060] Based on the transmission paths of the logical topology structure, the adjacency matrix A is used to calculate and summarize the paths connected to the faulty nodes and lines, and analyze the impact of the fault on the wiring;

[0061] The abnormal event type, specific location and impact range are obtained and integrated into a fault situation report.

[0062] As a preferred solution of the wiring monitoring method based on digital twins of the present invention, wherein: according to the fault report, the digital twin model of the wiring and the logical topology of the wiring unit are combined to generate an alternative path. The specific steps are as follows:

[0063] Extract the identified abnormal event data based on the specific content of the fault report;

[0064] Mapping fault starting points (such as nodes and lines) to nodes and lines in the logical topology structure in the digital twin model;

[0065] Input fault data into the digital twin model of wiring to simulate fault propagation;

[0066] Starting from the fault starting point (such as the faulty node and line), recursively traverse the directly connected nodes of the adjacency matrix A;

[0067] For each downstream node, the signal strength is calculated according to the signal attenuation model and updated;

[0068] For each connection line, check whether there is interruption or vibration abnormality according to the line health score. If it is abnormal, terminate the transmission, otherwise continue the transmission;

[0069] During the propagation simulation, each affected node ID and line number is recorded, and the affected nodes and lines are displayed in the 3D visualization interface of the digital twin model;

[0070] Remove faulty nodes and lines from the affected topology and update the adjacency matrix A;

[0071] The Dijkstra shortest path algorithm is used to calculate the shortest path from the source node to the target node, and the calculated alternative path is mapped into the digital twin model.

[0072] In a second aspect, the present invention provides a wiring monitoring system based on digital twins, including a model building module, a model updating module, a data acquisition module, a feature extraction module, a fault monitoring module and a backup path module.

[0073] The model building module is used to build a digital twin model of wiring based on the operation data of the wiring unit;

[0074] The model updating module is used to dynamically update the digital twin model in combination with the real-time operation data of the wiring unit;

[0075] The data acquisition module is used to collect multimodal data of the wiring unit through a multimodal sensor cluster, align the timestamps of the multimodal data, and integrate them into original monitoring data;

[0076] The feature extraction module is used to extract the features of the original monitoring data and generate state parameters by using multimodal fusion based on deep learning;

[0077] The fault monitoring module is used to monitor the wiring unit in real time based on the status parameters, detect and identify abnormal events, and generate a fault condition report;

[0078] The backup path module is used to generate a backup path based on the fault report and in combination with the digital twin model of the wiring and the logical topology of the wiring unit.

[0079] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the wiring monitoring method based on digital twins as described in the first aspect of the present invention is implemented.

[0080] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the wiring monitoring method based on digital twins as described in the first aspect of the present invention is implemented.

[0081] The beneficial effects of the present invention are as follows: by constructing and dynamically updating the digital twin model of wiring, accurate mapping and real-time synchronization of structure and status are achieved, which not only provides an intuitive visual management tool, but also significantly improves the fault location accuracy and maintenance efficiency. By deploying a multimodal sensor cluster to collect and integrate operating data, the system's perception ability is enhanced, the comprehensiveness and accuracy of the data are ensured, and a solid foundation is laid for subsequent analysis. By utilizing deep learning technology to fuse multimodal data features and generate comprehensive status parameters, the system can intelligently detect and predict anomalies, speed up diagnosis and improve accuracy. By comprehensively evaluating the health status of the wiring unit and quickly generating alternative paths when a fault occurs, the continuity and reliability of the wiring are guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0083] Figure 1 This is a flow chart of the wiring monitoring method based on digital twins in Example 1.

[0084] Figure 2 This is a module diagram of the wiring monitoring system based on digital twin in Example 1. DETAILED DESCRIPTION

[0085] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0086] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0087] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0088] Example 1, reference Figure 1 and Figure 2, which is the first embodiment of the present invention, provides a wiring monitoring method based on digital twins, comprising the following steps:

[0089] Build a digital twin model of wiring based on the operation data of the wiring unit;

[0090] Dynamically update the digital twin model by combining the real-time operation data of the wiring unit;

[0091] Through the multimodal sensor cluster, the multimodal data of the wiring unit is collected, and the multimodal data timestamps are aligned to integrate into the original monitoring data;

[0092] Utilize multimodal fusion based on deep learning to extract features of original monitoring data and generate state parameters;

[0093] Based on the status parameters, the wiring unit is monitored in real time, abnormal events are detected and identified, and fault reports are generated;

[0094] Based on the fault report, an alternative path is generated by combining the digital twin model of the wiring and the logical topology of the wiring unit.

[0095] S1. Build a digital twin model of wiring based on the operation data of the wiring unit.

[0096] The monitoring sensors (such as voltage sensors, current sensors, and temperature sensors) deployed at nodes (such as switches and routers) collect the operation data of the wiring unit, including node signal strength, line electrical impedance, temperature, and humidity.

[0097] Use noise filtering algorithms (such as denoising algorithms based on wavelet transform) to process operating data to eliminate noise caused by environmental factors or equipment interference;

[0098] Furthermore, a topology scanning tool based on SNMP (Simple Management Protocol) and LLDP (Link Layer Discovery Protocol) is used to automatically scan the connection relationship of wiring devices;

[0099] Use the SNMP interface of switches and routers to obtain the MAC address, port status and connection information of the device and build the physical topology structure;

[0100] Match the connection relationship data in the scan results (such as port mapping and transmission paths between devices) with the operation data to ensure that the topology information can reflect the actual transmission status of the wiring;

[0101] Use graph-theoretic methods (such as adjacency matrix representation) to represent physical topology;

[0102] Construct the adjacency matrix A of the wiring unit and generate a logic graph describing the signal transmission path between nodes;

[0103] Specifically, the adjacency matrix A is an N×N matrix (N is the total number of nodes). Each element in the matrix indicates whether two nodes are directly connected. If the two nodes are directly connected, the element is 1, and if not, the element is 0.

[0104] For two nodes that are not directly connected, the path is calculated through matrix power operation (matrix multiplication) to find a transmission path of any length;

[0105] The output includes the logical topology of node locations, connection relationships, and signal transmission paths;

[0106] Furthermore, a three-dimensional laser scanning device is used to scan the physical position of the wiring unit to obtain the spatial coordinates of the wiring;

[0107] Process the spatial coordinates into a standardized point cloud data format, and align and integrate the data of multiple scanning areas through the ICP algorithm;

[0108] Use 3D modeling tools (such as Autodesk Revit or Open 3D) to automatically generate a geometric 3D model of the wiring unit, including the physical location of the wiring, node location, and line structure;

[0109] Match the nodes in the geometric 3D model with the node IDs in the logical topology structure;

[0110] The lines in the geometric model (e.g., fiber connections) correspond to the connection relationships defined in the logical topology;

[0111] The operation data of nodes and transmission paths are mapped to the geometric three-dimensional model, so that the geometric model can dynamically display the operation of the wiring unit and obtain a digital twin model of the wiring.

[0112] S2. Dynamically update the digital twin model based on the real-time operation data of the wiring unit.

[0113] By collecting real-time operation data, updating the node and line status, and mapping them to the corresponding nodes and lines of the digital twin model of the wiring;

[0114] When the logical topology changes (e.g., adding a new node or disconnecting a connection), the adjacency matrix A is automatically updated;

[0115] For example, assign a unique ID to a newly added node and map its location and operating status to the digital twin model of the wiring;

[0116] If the direct connection between two nodes is disconnected, the value of the corresponding position in the adjacency matrix is ​​changed from 1 to 0.

[0117] S3. Collect multimodal data of the wiring unit through a multimodal sensor cluster, align the timestamps of the multimodal data, and integrate them into the original monitoring data.

[0118] Deploy multimodal sensor clusters at the nodes of the wiring unit and the branch points of the trunk line to ensure that the monitoring range covers the entire wiring unit;

[0119] Deploy fiber optic sensors on fiber optic lines and use distributed fiber optic sensing technology (such as OTDR based on optical time domain reflectometry) to monitor line stress changes and temperature;

[0120] Deploy electromagnetic sensors on cable lines and nodes (such as switch ports) to monitor cable current fluctuations and signal strength changes;

[0121] Deploy vibration sensors at physical nodes (such as terminal equipment) and line support locations to monitor mechanical vibrations (such as when the line is disturbed by external forces or environmental factors);

[0122] The multimodal sensor cluster collects multimodal data at a uniform frequency to generate time series data;

[0123] Providing a unified timestamp for the multimodal sensor cluster through a time synchronization protocol (based on the IEEE 1588 Precision Time Protocol (PTP) synchronization method) to ensure time alignment of different data sources;

[0124] Integrate time-synchronized multimodal data into raw monitoring data in a unified format;

[0125] De-noising, filling missing values ​​and standardizing the original monitoring data;

[0126] Specifically, the wavelet denoising algorithm is used to denoise the original monitoring data, the missing time points of the original monitoring data are filled using linear interpolation, and the data of each mode are normalized to eliminate dimensional differences.

[0127] S4. Utilize multimodal fusion based on deep learning to extract the features of original monitoring data and generate state parameters.

[0128] The single-modal raw monitoring data is subjected to feature extraction through a separate neural network (such as a convolutional neural network), and the expression is as follows:

[0129] h i =f Conv (W i *X i +b i );

[0130] Among them, h i is the feature vector of the original monitoring data of the i-th mode, i is the index of the number of modes, and fConv is the activation function after the convolution operation (ReLU), W i is the convolution kernel weight of the original monitoring data of the i-th modality, X i is the original monitoring data of the ith mode, b i is the bias term of the original monitoring data of the i-th modality, * represents the convolution operation;

[0131] Using the attention mechanism, the extracted unimodal original monitoring data feature vectors are weighted fused to generate a unified hidden layer feature representation, which is expressed as follows:

[0132]

[0133] Among them, α i is the attention weight of the original monitoring data of the i-th modality, which is used to measure the importance of the modality, N is the total number of modalities; ω is the parameter weight of the attention mechanism, and H is the feature vector of the original monitoring data after fusion;

[0134] Through the fully connected layer, the fused features are classified and regressed, and the output includes status parameters such as node load status, signal transmission strength, and line health status.

[0135] S5. Based on the status parameters, the wiring unit is monitored in real time, abnormal events are detected and identified, and a fault condition report is generated.

[0136] Mapping state parameters into the digital twin model of wiring;

[0137] Specifically, the node load status is mapped to the nodes of the digital twin model, and the node color changes dynamically to indicate the load condition (e.g., green for normal and red for overload);

[0138] The signal transmission strength is mapped to the line, and the thickness of the line indicates the signal strength (e.g., a thicker line indicates a high signal strength, and a thinner line indicates a signal attenuation);

[0139] The line health status is mapped to the line health score, and the line color changes reflect the health level (for example, green indicates healthy, yellow indicates sub-healthy, and red indicates fault);

[0140] When the operating data is updated, the state parameters of the digital twin model are updated in real time to ensure that the model is consistent with the actual operating status;

[0141] Obtain a visualized digital twin model that displays the status parameters of the wiring unit in real time, including node load status, signal transmission strength, and line health status;

[0142] According to the impact of state parameter failure on the wiring unit, abnormal events are defined as node failure, signal attenuation failure and line vibration failure;

[0143] Through experimental analysis, the node fault threshold k1, signal attenuation fault threshold k2 and line vibration fault threshold k3 are set;

[0144] Specifically, the node load state value is equal to the ratio of the current node load to the maximum node load. The node operation state under different load conditions is simulated by experiments. The results show that when the node load is 0, the node will disconnect the connection line, resulting in failure. When the node load exceeds 90% of the maximum processing value, the node response time is too long, resulting in task accumulation and processing failure. Therefore, k1 is set to 0.9;

[0145] Signal strength is negatively correlated with path loss. Through experimental simulation, the signal reception quality under different signal transmission strengths is compared. The experiment shows that when the signal transmission strength is less than 0.3, the wiring will be interrupted, resulting in the signal reception quality being far below the working requirements. Therefore, k2 is set to 0.3;

[0146] The line health value can be defined as the ratio of the actual signal transmission quality to the theoretical maximum signal transmission quality. When the line vibration fault occurs, it means that the line may be physically disturbed (such as vibration, bending or breakage). According to the optical fiber transmission theory, when the line health value is less than 0.5, the signal attenuation caused by vibration can no longer meet the demand. Therefore, k3 is set to 0.5.

[0147] When the value of the node load status exceeds the node fault threshold k1, it is determined to be a node fault. When the signal transmission strength is less than the signal attenuation fault threshold k2, it is determined to be a signal attenuation fault. When the value of the line health status is less than the line vibration fault threshold k3, it is determined to be a line vibration fault.

[0148] A time series classification model based on LSTM (Long Short-Term Memory Network) to capture the temporal variation pattern of the operating status;

[0149] Use historical state parameter samples that annotate specific abnormal events (labeled as 1) and normal states (labeled as 0) and divide them into training sets and validation sets;

[0150] Input the training set into the classification model, output the predicted fault classification probability of the sample, and compare it with the actual label of the sample. When the maximum number of iterations is reached, the training ends.

[0151] Input the validation set, calculate the loss function of the classification model, and evaluate the performance of the classification model. The expression is as follows:

[0152]

[0153] Among them, L is the cross entropy loss value of the classification model, M is the total number of samples in the validation set, m is the sample number index, C is the number of abnormal event categories, c is the abnormal event category number index, and y is the number of samples in the validation set. m is the true label of the mth sample, is the predicted probability of the mth sample;

[0154] When the F1 score (an indicator that comprehensively considers precision and recall) of the classification model reaches the optimal value, the predicted probability of an abnormal event is the predicted probability threshold;

[0155] Input the state parameters obtained based on real-time multimodal data into the classification model, and output abnormal events and their probabilities;

[0156] If the probability of a certain type of abnormal event exceeds the predicted probability threshold, the abnormal event is determined to be a fact;

[0157] For detected abnormal events, the specific location of the fault is determined by combining the topological structure of the digital twin model;

[0158] Specifically, the signal transmission path is traced in reverse to determine the starting node where the signal strength decreases, the faulty node is directly located, the other nodes connected to it are confirmed through the adjacency matrix A, the line of the vibration fault is located, and the fault point is confirmed in combination with the vibration data of the multimodal sensor;

[0159] Based on the transmission paths of the logical topology structure, the adjacency matrix A is used to calculate and summarize the paths connected to the faulty nodes and lines, and analyze the impact of the fault on the wiring;

[0160] The abnormal event type, specific location and impact range are obtained and integrated into a fault situation report.

[0161] S6. Based on the fault report, a backup path is generated by combining the digital twin model of the wiring and the logical topology of the wiring unit.

[0162] Extract the identified abnormal event data based on the specific content of the fault report, including the fault node ID, signal attenuation strength, and line health score;

[0163] Mapping fault starting points (such as nodes and lines) to nodes and lines in the logical topology structure in the digital twin model;

[0164] Input fault data into the digital twin model of wiring to simulate fault propagation;

[0165] Starting from the fault starting point (such as the faulty node and line), recursively traverse the directly connected nodes of the adjacency matrix A;

[0166] For each downstream node, the signal strength is calculated according to the signal attenuation model and updated;

[0167] For each connection line, check whether there is interruption or vibration abnormality according to the line health score. If it is abnormal, terminate the transmission, otherwise continue the transmission;

[0168] The core logic of fault propagation is based on the following three characteristics:

[0169] Node dependency, fault propagation follows the direct connection relationship of nodes. If a node fails, its directly connected downstream nodes will be affected;

[0170] Signal attenuation characteristics: During the fault propagation process, the signal strength will gradually decrease along the path;

[0171] Propagation delay: There is a time delay for a fault to propagate from one node to another, which is calculated based on the physical distance between nodes and the signal propagation speed;

[0172] During the propagation simulation, each affected node ID and line number is recorded, and the affected nodes and lines are displayed in the 3D visualization interface of the digital twin model;

[0173] Remove faulty nodes and lines from the affected topology and update the adjacency matrix A;

[0174] The shortest path from the source node to the target node is calculated using the Dijkstra shortest path algorithm, and the calculated alternative path is mapped into the digital twin model and highlighted in green.

[0175] The present embodiment also provides a wiring monitoring system based on digital twins, including: a model construction module, a model update module, a data acquisition module, a feature extraction module, a fault monitoring module and a backup path module, the model construction module is used to build a digital twin model of the wiring based on the operation data of the wiring unit; the model update module is used to dynamically update the digital twin model in combination with the real-time operation data of the wiring unit; the data acquisition module is used to collect multimodal data of the wiring unit through a multimodal sensor cluster, and align the timestamps of the multimodal data to integrate them into original monitoring data; the feature extraction module is used to extract the features of the original monitoring data and generate state parameters by using multimodal fusion based on deep learning; the fault monitoring module is used to monitor the wiring unit in real time based on the state parameters, detect and identify abnormal events, and generate a fault situation report; the backup path module is used to generate a backup path based on the fault situation report, in combination with the digital twin model of the wiring and the logical topology of the wiring unit.

[0176] This embodiment also provides a computer device, which is suitable for the monitoring method of wiring based on digital twins, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the monitoring method of wiring based on digital twins proposed in the above embodiment.

[0177] The computer device may be a terminal, and the computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to connect to an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field network) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0178] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by the processor, the monitoring method for wiring based on digital twins proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0179] In summary, the present invention achieves accurate mapping and real-time synchronization of structure and status by: constructing and dynamically updating a digital twin model of wiring, which not only provides an intuitive visual management tool, but also significantly improves fault location accuracy and maintenance efficiency. By deploying a multimodal sensor cluster to collect and integrate operating data, the system's perception capability is enhanced, the comprehensiveness and accuracy of the data are ensured, and a solid foundation is laid for subsequent analysis. By utilizing deep learning technology to fuse multimodal data features and generate comprehensive status parameters, the system can intelligently detect and predict anomalies, speed up diagnosis and improve accuracy. By comprehensively evaluating the health status of the wiring unit and quickly generating alternative paths when a fault occurs, continuity and reliability are guaranteed.

[0180] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A wiring monitoring method based on digital twins, characterized in that: include, Build a digital twin model of wiring based on the operation data of the wiring unit; Dynamically update the digital twin model by combining the real-time operation data of the wiring unit; Through the multimodal sensor cluster, the multimodal data of the wiring unit is collected, and the multimodal data timestamps are aligned to integrate into the original monitoring data; Utilize multimodal fusion based on deep learning to extract features of original monitoring data and generate state parameters; Based on the status parameters, the wiring unit is monitored in real time, abnormal events are detected and identified, and fault reports are generated; Based on the fault report, an alternative path is generated by combining the digital twin model of the wiring and the logical topology of the wiring unit.

2. The wiring monitoring method based on digital twin according to claim 1, characterized in that: The specific steps of building a digital twin model of wiring based on the operation data of the wiring unit are as follows: Collect the operation data of the wiring unit through monitoring sensors deployed at the nodes; Use noise filtering algorithm to process the operation data; Use SNMP and LLDP-based topology scanning tools to automatically scan the connection relationship of wiring devices; Use the SNMP interface of switches and routers to obtain the MAC address, port status and connection information of the device and build the physical topology structure; Matching the connection relationship data in the scan results with the operation data; Use graph-theoretic methods to represent physical topology; Construct the adjacency matrix A of the wiring unit and generate a logic graph describing the signal transmission path between nodes; The output includes the logical topology of node locations, connection relationships, and signal transmission paths; Use 3D laser scanning equipment to scan the physical position of the wiring unit and obtain the spatial coordinates of the wiring; Process the spatial coordinates into a standardized point cloud data format, and align and integrate the data of multiple scanning areas through the ICP algorithm; Automatically generate the geometric 3D model of the wiring unit using 3D modeling tools; Match the nodes in the geometric 3D model with the node IDs in the logical topology structure; The lines in the geometric model correspond to the connection relationships defined in the logical topology structure; The operating data of nodes and transmission paths are mapped to the geometric three-dimensional model to obtain a digital twin model of the wiring.

3. The wiring monitoring method based on digital twins according to claim 2, characterized in that: The digital twin model is dynamically updated by combining the real-time operation data of the wiring unit. The specific steps are as follows: By collecting real-time operation data, updating the node and line status, and mapping them to the corresponding nodes and lines of the digital twin model of the wiring; When the logical topology changes, the adjacency matrix A is automatically updated.

4. The wiring monitoring method based on digital twins according to claim 3 is characterized in that: The multimodal sensor cluster is used to collect multimodal data of the wiring unit, align the timestamps of the multimodal data, and integrate them into the original monitoring data. The specific steps are as follows: Deploy multimodal sensor clusters at the nodes of the wiring units and the branch points of the trunk lines; Deploy fiber optic sensors on fiber optic lines and use distributed fiber optic sensing technology to monitor line stress changes and temperature; Deploy electromagnetic sensors on cable lines and nodes to monitor cable current fluctuations and signal strength changes; Deploy vibration sensors at physical nodes and line support locations to monitor mechanical vibrations; The multimodal sensor cluster collects multimodal data at a uniform frequency to generate time series data; Providing unified timestamps for multimodal sensor clusters through time synchronization protocols; Integrate time-synchronized multimodal data into raw monitoring data in a unified format; The original monitoring data is denoised, missing values ​​are filled and standardized.

5. The wiring monitoring method based on digital twin according to claim 4, characterized in that: The method utilizes multimodal fusion based on deep learning to extract the original monitoring data features and generate state parameters. The specific steps are as follows: Extract features from unimodal raw monitoring data through a separate neural network (e.g., convolutional neural network); Using the attention mechanism, the extracted unimodal original monitoring data feature vectors are weighted fused to generate a unified hidden feature representation; Through the fully connected layer, the fused features are classified and regressed, and the output includes status parameters such as node load status, signal transmission strength, and line health status.

6. The wiring monitoring method based on digital twin according to claim 5, characterized in that: Based on the state parameters, the wiring unit is monitored in real time, abnormal events are detected and identified, and a fault report is generated. The specific steps are as follows: Mapping state parameters into the digital twin model of wiring; When the operating data is updated, the state parameters of the digital twin model are updated in real time; Obtain a visualized digital twin model that displays the status parameters of the wiring unit in real time, including node load status, signal transmission strength, and line health status; According to the impact of state parameter failure on the wiring unit, abnormal events are defined as node failure, signal attenuation failure and line vibration failure; Through experimental analysis, the node fault threshold k1, signal attenuation fault threshold k2 and line vibration fault threshold k3 are set; When the value of the node load status exceeds the node fault threshold k1, it is determined to be a node fault. When the signal transmission strength is less than the signal attenuation fault threshold k2, it is determined to be a signal attenuation fault. When the value of the line health status is less than the line vibration fault threshold k3, it is determined to be a line vibration fault. A time series classification model based on LSTM (Long Short-Term Memory Network) to capture the temporal variation pattern of the operating status; Use historical state parameter samples that annotate specific abnormal events (labeled as 1) and normal states (labeled as 0) and divide them into training sets and validation sets; Input the training set into the classification model, output the predicted fault classification probability of the sample, and compare it with the actual label of the sample. When the maximum number of iterations is reached, the training ends. Input the validation set, calculate the loss function of the classification model, and evaluate the performance of the classification model; When the F1 score (an indicator that comprehensively considers precision and recall) of the classification model reaches the optimal value, the predicted probability of an abnormal event is the predicted probability threshold; Input the state parameters obtained based on real-time multimodal data into the classification model, and output abnormal events and their probabilities; If the probability of a certain type of abnormal event exceeds the predicted probability threshold, the abnormal event is determined to be a fact; For detected abnormal events, the specific location of the fault is determined by combining the topological structure of the digital twin model; Based on the transmission paths of the logical topology structure, the adjacency matrix A is used to calculate and summarize the paths connected to the faulty nodes and lines, and analyze the impact of the fault on the wiring; The abnormal event type, specific location and impact range are obtained and integrated into a fault situation report.

7. The wiring monitoring method based on digital twin according to claim 6, characterized in that: According to the fault report, the digital twin model of the wiring and the logical topology of the wiring unit are combined to generate an alternative path. The specific steps are as follows: Extract the identified abnormal event data based on the specific content of the fault report; Mapping fault starting points (such as nodes and lines) to nodes and lines in the logical topology structure in the digital twin model; Input fault data into the digital twin model of wiring to simulate fault propagation; Starting from the fault starting point (such as the faulty node and line), recursively traverse the directly connected nodes of the adjacency matrix A; For each downstream node, the signal strength is calculated according to the signal attenuation model and updated; For each connection line, check whether there is interruption or vibration abnormality according to the line health score. If it is abnormal, terminate the transmission, otherwise continue the transmission; During the propagation simulation, each affected node ID and line number is recorded, and the affected nodes and lines are displayed in the 3D visualization interface of the digital twin model; Remove faulty nodes and lines from the affected topology and update the adjacency matrix A; The Dijkstra shortest path algorithm is used to calculate the shortest path from the source node to the target node, and the calculated alternative path is mapped into the digital twin model.

8. A wiring monitoring system based on digital twins, based on the wiring monitoring method based on digital twins according to any one of claims 1 to 7, characterized in that: It includes model building module, model updating module, data acquisition module, feature extraction module, fault monitoring module and backup path module. The model building module is used to build a digital twin model of wiring based on the operation data of the wiring unit; The model updating module is used to dynamically update the digital twin model in combination with the real-time operation data of the wiring unit; The data acquisition module is used to collect multimodal data of the wiring unit through a multimodal sensor cluster, align the timestamps of the multimodal data, and integrate them into original monitoring data; The feature extraction module is used to extract the features of the original monitoring data and generate state parameters by using multimodal fusion based on deep learning; The fault monitoring module is used to monitor the wiring unit in real time based on the status parameters, detect and identify abnormal events, and generate a fault condition report; The backup path module is used to generate a backup path based on the fault report and in combination with the digital twin model of the wiring and the logical topology of the wiring unit.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the wiring monitoring method based on digital twins according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the wiring monitoring method based on digital twins according to any one of claims 1 to 7 are implemented.

Images