Multi-layer sensor data fusion cable bridge fault positioning method and system

Abstract

The application provides a multi-layer sensor data fusion cable bridge fault positioning method and system, relates to the technical field of data processing, and establishes a fault evaluation list for fault monitoring feature compensation by analyzing the fault characteristics of the cable bridge, and constructs a multi-layer sensor network; a digital twin space is constructed according to the physical structure and operation data of the cable bridge, a multi-layer induction configuration module is established based on the multi-layer sensor network and embedded in the digital twin space; and fault identification and positioning are performed through the digital twin space according to the sensing configuration parameters of the multi-layer induction configuration module. The application solves the technical problems of low fault identification accuracy and large positioning error caused by the monitoring blind area and data loss of the cable bridge fault monitoring in the prior art, and achieves the technical effects of improving fault positioning accuracy and reliability, and enhancing fault monitoring comprehensiveness and real-time performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a cable bridge fault positioning method and system based on multi-layer sensor data fusion. BACKGROUND

[0002] In modern power transmission and information wiring systems, cable bridges serve as key support structures, and their operating status is directly related to the stability of power supply and the safety of information transmission. Currently, cable bridge fault positioning mainly relies on single-type sensors such as vibration sensors, temperature sensors, or current sensors installed at bridge nodes or key positions for monitoring, and preliminary fault identification and positioning are performed in combination with abnormal threshold judgment. However, this method has a single sensor layout mode and cannot comprehensively perceive the fault characteristics of cable bridges under the combined action of factors such as structural vibration, temperature change, and current anomaly, which easily leads to monitoring blind spots and makes it difficult to detect some faults in a timely manner. In particular, for some complex fault types, single sensors cannot capture their key features, which easily leads to missed or false reports, affecting the stability of fault identification and positioning accuracy. SUMMARY

[0003] The present application provides a cable bridge fault positioning method and system based on multi-layer sensor data fusion, which solves the technical problems of low fault identification accuracy and large positioning error caused by the single sensor layout and the inability to configure different fault types hierarchically in the prior art, resulting in monitoring blind spots and data missing in cable bridge fault monitoring. The technical effects of improving fault positioning accuracy and reliability and enhancing comprehensive fault monitoring and real-time performance are achieved through multi-layer sensor fusion and digital twin space embedding.

[0004] In view of the above problems, on the one hand, the present application provides a cable bridge fault positioning method based on multi-layer sensor data fusion, which comprises: analyzing the fault characteristics of the cable bridge, establishing a fault evaluation list, wherein the fault evaluation list includes fault types, monitoring sensors, success probabilities, and abnormal missing features; performing fault monitoring feature compensation according to the fault evaluation list, and constructing a multi-layer sensor network, wherein the multi-layer sensor network has fault type labels and hierarchical granularity; constructing a digital twin space based on the physical structure and operating data of the cable bridge, embedding a multi-layer sensing configuration module based on the multi-layer sensor network into the digital twin space, and the multi-layer sensing configuration module is used for multi-level sensor data identification configuration; performing fault identification and positioning through the digital twin space according to the sensing configuration parameters of the multi-layer sensing configuration module, and obtaining fault positioning information.

[0005] In another aspect, the application also provides a multi-layer sensor data fusion cable bridge fault positioning system, comprising: a fault feature analysis module for analyzing the fault features of the cable bridge, establishing a fault evaluation list, wherein the fault evaluation list includes fault types, monitoring sensors, success probabilities, and abnormal missing features; a sensor construction module for compensating fault monitoring features according to the fault evaluation list, and constructing a multi-layer sensor network, wherein the multi-layer sensor network has fault type labels and hierarchical granularity; a digital twin module for constructing a digital twin space according to the physical structure and operation data of the cable bridge, embedding a multi-layer sensing configuration module in the digital twin space based on the multi-layer sensor network, and the multi-layer sensing configuration module is used for multi-level sensing data identification configuration; a fault positioning module for fault identification and positioning through the digital twin space according to the sensing configuration parameters of the multi-layer sensing configuration module, and obtaining fault positioning information.

[0006] The one or more technical solutions provided in the application have at least the following beneficial effects:

[0007] By analyzing the fault features of the cable bridge and establishing a fault evaluation list, a structured reference standard is provided for subsequent monitoring and identification. The introduction of success probability and abnormal missing features provides intelligent compensation capability under incomplete data and uncertain conditions. According to the fault evaluation list, fault monitoring features are compensated, and a multi-layer sensor network is constructed, which realizes fine perception of different fault types and different hierarchical data, improves the adaptability and perception ability of the sensor network to complex cable bridge faults, and provides a more comprehensive and accurate data basis for subsequent fault identification and positioning. By converting the physical structure and operation data of the cable bridge into a digital twin space, a virtual mapping environment highly matched with the physical entity is provided for fault identification and positioning. The embedded multi-layer sensing configuration module can perform multi-level sensing data identification configuration in the digital twin space, fully utilize the high fidelity and dynamic characteristics of digital twin technology, further improve the accuracy and reliability of fault identification, realize effective fusion and cooperation between physical entities and virtual spaces, and make subsequent fault identification efficient simulation, verification and reasoning in the virtual space, and improve fault dynamic identification capability. According to the sensing configuration parameters of the multi-layer sensing configuration module, fault identification and positioning are performed through the digital twin space, so that more accurate and stable fault positioning results are obtained.

[0008] In summary, this application achieves panoramic perception and high-precision positioning of cable tray operating status by constructing a fault feature-driven evaluation mechanism, a multi-source sensor network deployed in a hierarchical manner, and a digital twin space that integrates cable tray structure and real-time data. Through the dynamic configuration and simulation of the digital twin environment, the accuracy and real-time response of cable tray fault identification under complex operating conditions are significantly improved, thereby enhancing the overall comprehensiveness and reliability of cable tray fault monitoring.

[0009] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the cable tray fault location method based on multi-layer sensor data fusion provided in this application embodiment.

[0011] Figure 2 This is a schematic diagram illustrating the process of constructing a multi-layer sensor network in the cable tray fault location method based on multi-layer sensor data fusion provided in the embodiments of this application.

[0012] Figure 3 A schematic diagram of the structure of the cable tray fault location system based on multi-layer sensor data fusion provided in this application embodiment.

[0013] Explanation of reference numerals in the attached diagram: Fault Feature Analysis Module 10, Sensor Construction Module 20, Digital Twin Module 30, Fault Location Module 40. Detailed Implementation

[0014] This application provides a cable tray fault location method and system based on multi-layer sensor data fusion. It solves the technical problems of existing technologies, such as blind spots and missing data in cable tray fault monitoring due to the single sensor deployment and inability to configure hierarchically for different fault types. This results in low fault identification accuracy and large location errors. The method achieves the technical effect of improving fault location accuracy and reliability, and enhancing the comprehensiveness and real-time nature of fault monitoring through multi-layer sensor fusion and digital twin spatial embedding.

[0015] Example 1, as Figure 1 As shown in the embodiment of this application, a method for locating cable tray faults using multi-layer sensor data fusion is provided. The method includes:

[0016] Step S100: Analyze the fault characteristics of the cable tray and establish a fault evaluation list, which includes fault type, monitoring sensor, success probability, and abnormal missing features.

[0017] In particular, the fault feature refers to abnormal forms that may occur during the operation of the cable bridge, such as temperature abnormalities, current abnormalities, structural stress deviations, etc. The fault evaluation list is a database or table structure for systematically describing the relationship between fault features and monitoring resources, including fault types, monitoring sensors, success probabilities, and missing abnormal features. Among them, the fault types include short circuit, overload, fracture, corrosion and other different types; the monitoring sensor is a sensor that performs real-time data collection, including temperature sensors, current sensors, voltage sensors, vibration sensors, image acquisition modules, etc. The success probability is the historical or statistical probability value of a certain type of sensor successfully detecting a specific fault, reflecting the monitoring reliability of the sensor for a specific fault. The missing abnormal feature refers to the key fault feature information that cannot be obtained due to insufficient sensor layout, insufficient performance or equipment failure. The common fault types of the cable bridge are classified and summarized, and the physical or electrical characteristics that may be presented by each type of fault are analyzed, and a fault evaluation list is established. The list lists in detail the key monitoring sensors corresponding to each fault type, the monitoring success probability for this type of fault in history, and the abnormal feature items that may be missing in a specific scenario, providing basic data support for subsequent compensation and layout of multi-layer sensor networks. For example, "cable overheating" belongs to the temperature abnormality type of fault, and requires the configuration of an infrared thermal imager and a point temperature sensor; its success probability is 85%, but there may be a thermal imaging dead angle at the corner of the bridge, resulting in "missing thermal imaging features".

[0018] In actual operation, the key features of each type of fault can be determined by consulting a large number of historical fault records, existing cable bridge fault cases, and using computer simulation to simulate the fault occurrence process, such as vibration abnormalities caused by mechanical damage, temperature changes caused by overheating, sudden increases in current caused by short circuits, and specific gas releases caused by insulation aging. At the same time, identify the monitoring sensors already deployed in the current cable bridge. Then, based on historical monitoring data and experimental statistics, estimate the success probability of each sensor in monitoring the corresponding fault, quantifying the reliability of each sensor in actual application. At the same time, by comparing with the ideal monitoring model, find out the abnormal missing features that appear in actual monitoring. These missing features may be caused by unreasonable sensor layout position, inability to cover the fault-prone parts, or insufficient monitoring accuracy of the sensor, etc. Finally, the fault type, monitoring sensor, success probability, and abnormal missing feature information obtained through the above analysis are organized into a detailed fault evaluation list, providing comprehensive and accurate basis for subsequent fault monitoring feature compensation and the construction of multi-layer sensor networks.

[0019] Step S200: compensating for fault monitoring features according to the fault evaluation list, and constructing a multi-layer sensor network, wherein the multi-layer sensor network has a fault type label and a hierarchical granularity.

[0020] Specifically, the fault type label is a fault type identifier for marking the monitoring task to which the sensor belongs. The hierarchical granularity refers to the layout density and monitoring range of sensors at different levels in the sensor network. The larger the hierarchical granularity, the wider the range monitored by the sensor at this level or the more macro the monitoring feature, and vice versa. According to the fault evaluation list established in step S100, the missing features in each fault type are analyzed, and sensors of the corresponding type are supplemented. For example, if stress monitoring is lacking in a certain area, a stress gauge is added; if image information needs to be compensated, a high-definition camera is introduced. A sensor set is constructed in units of fault types, and hierarchical levels are divided according to dimensions such as monitoring space size and sensing accuracy requirements, to generate a multi-layer sensor network with a layered structure and labeled markers, to compensate for sensing coverage blind areas and realize adaptive sensor combination layout for fault targets, improving monitoring comprehensiveness.

[0021] Step S300: constructing a digital twin space according to the physical structure and operation data of the cable bridge, embedding a multi-layer sensing configuration module based on the multi-layer sensor network in the digital twin space, and the multi-layer sensing configuration module is used for multi-level sensing data identification configuration.

[0022] Specifically, the multi-layer sensing configuration module is an embedded configuration unit with hierarchical feature management and fault label matching capability, used to integrate the sensing node parameters and configuration strategies required for each fault type. Based on the three-dimensional structure, wiring method and actual operation state data of the cable bridge, a virtual digital twin space is constructed, and the multi-layer sensor network constructed in step S200 is mapped and embedded in the digital twin space, that is, sensor nodes with fault labels and hierarchical granularity are deployed in the digital twin model, and a sensing configuration module is established for each layer of sensor network to configure reading parameters, data analysis logic, trigger threshold, etc., realizing virtual deployment of the sensor network, real-time synchronization and dynamic adjustable configuration of the state, providing a three-dimensional visual support and a highly controllable monitoring platform for fault identification, so that different levels of monitoring tasks can be virtually run, predicted, verified and updated in the digital twin space.

[0023] Step S400: fault identification and positioning are performed through the digital twin space according to the sensing configuration parameters of the multi-layer sensing configuration module, and fault positioning information is obtained.

[0024] Specifically, the sensing configuration parameters refer to the arrangement positions, sampling frequencies, threshold parameters, and identification algorithms of each sensor. After the digital twin space is constructed and the sensing configuration is completed, a positioning request is received, including a specified fault type and a required positioning level. By matching the corresponding sensing configuration module, the sensor network data is called to analyze the current bridge state in real time. Combined with historical data and model algorithms, the state simulation is performed to identify the specific fault location and further simulate the possible development trend and range of diffusion of the fault, outputting comprehensive positioning information including the fault point, severity, and warning range, thereby achieving high-precision and predictable fault positioning.

[0025] Further, as shown in Figure 2 Step S200 includes:

[0026] Step S210: According to the abnormal missing features in the fault evaluation list, a compensation monitoring sensor is matched.

[0027] Step S220: According to the monitoring sensor and the compensation monitoring sensor, hierarchical division is performed according to the range size of the compensation space to obtain a hierarchical granularity.

[0028] Step S230: A success probability compensation threshold is set, the success probability compensation threshold is used as a constraint condition, a compensation strategy analysis is performed on the compensation monitoring sensor based on the hierarchical granularity, and a compensation sensor and a sensor parameter of each hierarchical granularity are determined.

[0029] Step S240: According to the hierarchical granularity and the corresponding compensation sensor and sensor parameter, a hierarchical network block of each fault type is established.

[0030] Step S250: The hierarchical network blocks of all fault types are integrated to construct the multi-layer sensor network.

[0031] Specifically, based on the abnormal missing features marked in the fault evaluation list, the sensor type matching the feature is found, a sensor type list to be compensated, i.e., a compensation monitoring sensor, is generated, the gap in sensor arrangement is closed, and the ability to capture key fault signs in advance is enhanced. This matching process can be intelligently recommended based on a pre-stored fault and sensor mapping rule library. For example, if the uneven heat diffusion feature of the connection part is missing in the current deployment, an infrared point measurement sensor is compensated; if the structure micro-deformation feature is missing, a micro-displacement meter or strain gauge is installed.

[0032] The monitoring sensor information and its coverage range currently deployed are acquired. In combination with the missing features listed in the fault evaluation list, the monitoring blind area or the incomplete data area is identified, and then the spatial range of compensation monitoring is determined. According to the distribution range of the compensation area, in combination with the fault type and the identification accuracy requirement, a plurality of types of compensation monitoring sensors are introduced, including electrical sensors, physical sensors, chemical sensors and image sensors. After evaluating the coverage ability and monitoring resolution of each type of compensation sensor, the following granularity level division standard is set according to the spatial scale difference: microscopic granularity level: suitable for the area with a compensation spatial range less than 0.1 meters, and thermocouple sensors, micro pressure sensors, etc. are selected; mesoscopic granularity level: suitable for the area with a compensation spatial range between 0.1 meters and 1 meter, and infrared image collectors, tension sensors, etc. are configured; macroscopic granularity level: suitable for the area with a compensation spatial range greater than 1 meter, and distributed fiber temperature monitors, image recognition sensors, etc. are deployed. In the granularity division process, the identification accuracy and coverage range of each sensor in the target compensation area are evaluated based on fusion analysis, and an identification diffusion evaluation map is constructed. According to the diffusion trend and identification ability gradient of the evaluation map, the compensation space is clustered and divided, and the monitoring areas of different granularity levels are formed. Finally, the monitoring granularity structure including the microscopic, mesoscopic and macroscopic levels is output as the basis for subsequent compensation sensor configuration and multi-layer sensor network construction.

[0033] The success probability compensation threshold is the minimum success monitoring probability lower limit set for judging whether the current configuration meets the fault monitoring reliability requirement. The minimum reliability (such as 85% success probability) of monitoring a certain target fault category is set, the compensation monitoring sensors at each level are simulated and analyzed, the sensor configuration scheme meeting or exceeding the threshold is selected, and its key configuration parameters including sampling rate, power consumption, layout mode, etc. are output to ensure that each fault type has stable sensing ability after compensation.

[0034] The hierarchical network block refers to a sub-network with a sensor layout hierarchy structure constructed for a single fault type. According to the hierarchical granularity and the sensor types and configuration parameters corresponding to each level, a sensing block structure for a single fault type is formed, i.e. a hierarchical network block is constructed. Each block has its self-consistent internal sensing hierarchy, layout scheme and feedback mechanism, realizing independent, clear and structured sensing network construction for each type of fault, which is beneficial to modular management and fault special diagnosis. For example, for the "short circuit" fault, two-layer blocks of temperature sensing and current monitoring are established; for the "deformation" fault, three-layer blocks of stress sensing, image sensing and displacement sensing are established.

[0035] The fusion operation is performed on the hierarchical network blocks of each fault type, the shared area, the shared sensor, and the adjacent block merging condition are identified, the topology optimization is performed, the unified sensor network structure across the fault dimension and multi-granularity is formed, and the sensor deployment graph and the network parameter configuration file are output to construct the multi-layer sensor network and realize the collaborative sensing and data multiplexing capability.

[0036] Further, step S250 includes:

[0037] Step S251: Identifying the homologous sensors between the hierarchical network blocks of each fault type.

[0038] Step S252: Connecting the hierarchical network blocks of each fault type based on the homologous sensors, combining the hierarchical network block connections of all fault types, and constructing the multi-layer sensor network.

[0039] Specifically, the homologous sensors refer to sensors with the same data source or shareable sensing information in multiple fault monitoring blocks, for example, the same temperature sensor is used for overload identification and contact failure monitoring. Cross-analysis is performed on all constructed fault type hierarchical network blocks to identify sensor nodes with overlapping data sources or overlapping sampling areas. This process can be matched by comparing the sensor ID, installation location code, sampling variable type, etc. information, and outputting a list of homologous sensors as the basis for subsequent integration.

[0040] Based on the identified homologous sensors, the originally independent fault identification sub-networks are connected through these common nodes to form an integrated structure network that can share data and decision resources. At the same time, conflict coordination (such as data access priority, bandwidth allocation), function mapping (matching of multi-task sensing functions), and path optimization (rearrangement of node access sequence) are performed to finally form an integrated multi-layer sensor network topology, realize the integration of monitoring resources across fault types, reduce redundant deployment, construct an efficient collaborative data acquisition and processing platform, and support complete data-driven input for subsequent digital twin models.

[0041] Further, step S220 includes:

[0042] Step S221: The compensation monitoring sensors include electrical, physical, chemical, and image monitoring types, and the abnormal missing features are compensated according to the electrical, physical, chemical, and image monitoring types and their sensing parameters to obtain compensation features.

[0043] Step S222: Based on the compensation features of the monitoring sensors and each monitoring type, the fault fusion recognition accuracy and recognition range evaluation are performed to obtain an identification diffusion evaluation graph.

[0044] Step S223: hierarchically segmenting the identification diffusion evaluation graph according to the positioning level range of each fault type to obtain the hierarchical granularity.

[0045] Specifically, the compensation monitoring sensor is a sensor device for functional compensation of missing or abnormal data in the original monitoring link. According to the detection object, it can be divided into electrical types (such as current, voltage, impedance), physical types (such as temperature, vibration, displacement), chemical types (such as gas concentration, corrosion products), and image monitoring types (such as visual recognition, thermal image analysis). Analyze the "abnormal missing feature" field marked in the fault evaluation list, select the corresponding type of compensation sensor according to the missing content type, and compensate for the missing features according to their sensing parameters (such as sampling frequency, sensitivity, sampling coverage radius, etc.) to obtain compensation features that can accurately reflect fault features.

[0046] The fault fusion recognition accuracy refers to the correctness index of the monitoring sensor network formed by combining multiple sensor types in actual fault detection. The recognition range refers to the area range that the monitoring sensor can effectively recognize the fault. The identification diffusion evaluation graph is a graph structure that represents the trend of fusion recognition ability changing with spatial or type dimensions, which can be used to evaluate the balance between coverage and accuracy. The feature data of existing monitoring sensors and compensation monitoring sensors are fused, and a fault recognition algorithm (such as support vector machine, neural network, etc.) is used to evaluate the accuracy and range of the fused data in fault recognition. Through simulation and actual test, the identification diffusion evaluation graph is generated, which intuitively shows the gain and "diffusion" boundary under each type of compensation configuration, providing objective data basis for subsequent hierarchical granularity division, making the sensor network more targeted in configuration.

[0047] The positioning level range pointer refers to the positioning accuracy range required for a certain type of fault. According to the possible propagation range or safety response requirement of different fault types in the physical structure, set the target positioning resolution level, for example, require accurate to node level for overload fault, accurate to branch level for corrosion fault. Then the identification diffusion evaluation graph is hierarchically partitioned according to these requirements, and the corresponding hierarchical granularity level of each layer is output, realizing accurate and hierarchical monitoring design for different fault types, improving resource configuration efficiency and dynamic adaptation ability of identification response.

[0048] Further, step S230 includes:

[0049] With the success probability compensation threshold as a constraint, according to the identification diffusion evaluation graph, search for sensor layout distance and sensing control parameters to determine the compensation sensor and sensing parameters of each hierarchical granularity, which are the compensation sensor types and their layout strategies and control parameters that meet the success probability compensation threshold and the identification range.

[0050] Specifically, the sensor deployment distance refers to compensating for the deployment distance between the sensor and the target monitoring point or other sensors in space, affecting data synchronization and signal interference. The sensor control parameter refers to the parameter affecting the working performance of the sensor, such as sampling frequency, sensitivity, detection threshold, communication frequency band, etc. Based on the identification diffusion evaluation map, the fusion identification performance at each level is extracted. Under the premise of meeting the success probability compensation threshold, the sensor deployment distance optimization and control parameter matching search are carried out: the distance between the sensors deployed at the same layer or adjacent layers is adjusted to evaluate whether the number of redundant deployment can be reduced without reducing the accuracy; search for parameter combinations that meet the identification range requirements and success probability threshold in the existing sensor type candidate set; consider the identification coverage, sensor available resources, deployment space, energy consumption and other constraints in the search process, and select the optimal configuration scheme in a multi-objective optimization manner. Finally, the selected compensation sensor type, deployment position distance and its sensor parameters at each level granularity are output.

[0051] Further, step S300 comprises:

[0052] Step S310: constructing a three-dimensional geometric model according to the physical entity structure data of the cable bridge.

[0053] Step S320: collecting multi-physical field data of the cable bridge and coupling the multi-physical field data to the three-dimensional geometric model.

[0054] Step S330: combining the three-dimensional geometric model and the actual operation data of the cable bridge to create a digital twin base and construct the digital twin space.

[0055] Further, the multi-physical field data includes thermal field, stress field, electromagnetic field.

[0056] Specifically, the physical entity structure data refers to the size, shape, material properties, support structure, interface and other information of the bridge structure. The BIM, CAD or 3D modeling tool is used to import the bridge structure drawing to identify geometric components (such as trays, support arms, connecting plates) and establish a three-dimensional geometric model of the cable bridge, providing a spatial carrier for digital simulation and mapping.

[0057] Multi-physical field data includes thermal field (temperature distribution), stress field (deformation and load), electromagnetic field (inductive interference) and other field information that can affect the health status of the bridge. Using a multi-physical field simulation tool, import the three-dimensional model, add boundary conditions, material parameters, and input measured or simulated data to obtain the spatial distribution of each physical field. Superimpose and map in the three-dimensional geometric model to realize the synchronous visualization and prediction of the stress, electrical, thermal and electromagnetic effects of the bridge under complex working conditions. Among them, the thermal field is collected by thermocouple sensors or infrared thermometers to reflect the heating of the cable; the stress field is collected by strain gauges or pressure sensors to reflect the stress of the structure; the electromagnetic field is collected by electromagnetic sensors to reflect electromagnetic interference and electromagnetic compatibility.

[0058] From the data acquisition system, edge sensor terminal, and historical database, obtain the bridge operation data (such as temperature, current, vibration, cracking signal, etc.), and perform time-space mapping binding with the three-dimensional geometric model to create a digital twin base as the starting state of the virtual twin. On this basis, further build a digital twin space to simulate various state changes of the cable bridge in actual operation, such as temperature changes, stress changes, electromagnetic field changes, etc.

[0059] Further, step S300 further includes:

[0060] Step S340: Obtain the hierarchical sensing configuration parameters of each fault type according to the fault type label and hierarchical granularity of the multi-layer sensor network.

[0061] Step S350: Establish the multi-layer sensing configuration module according to the hierarchical sensing configuration parameters of each fault type.

[0062] Step S360: Add the multi-layer sensing configuration module as a built-in sensing parameter configuration tool of the digital twin space to the digital twin space.

[0063] Specifically, after the multi-layer sensor network is constructed, the relevant configuration information of each fault type is extracted from the calibrated hierarchical network blocks, including hierarchical labels, data requirements, compensation strategies, etc. The hierarchical sensing configuration parameter table for each fault is sorted out to realize the accurate mapping between fault types and multi-level sensing requirements and provide basic data support for the structure of the configuration module.

[0064] According to the configuration parameters of each fault type at different granularity levels, a modular configuration unit structure is generated, including sensor selection logic, parameter setting template and data access specification, and packaged into a functional module that can be deployed in the twin space to form a standardized deployable configuration module unit, enhancing the automatic adaptation capability of sensing strategies in different scenarios.

[0065] The multi-layer sensing configuration module is mounted in the digital twin platform in the form of a service interface or a dynamic component, supports calling the corresponding level of sensing configuration in different fault evolution simulation stages, realizes autonomous completion of parameter activation and deployment by the system, and provides real-time adaptation capability for virtual-real mapping of the cable bridge operation state.

[0066] Further, step S400 includes:

[0067] Step S410: obtaining a fault positioning request, including a fault type and a positioning level granularity.

[0068] Step S420: the multi-layer sensing configuration module performs sensing configuration matching according to the fault type and the positioning level granularity, obtains monitoring data of a corresponding sensor network based on the matched sensing configuration parameters, performs state simulation through the digital twin space, identifies fault information and performs fault positioning, simultaneously performs fault evolution simulation based on the fault information to obtain a fault prediction range, and outputs the fault positioning information.

[0069] Specifically, the fault positioning request is a positioning task request proposed by a system user or a monitoring system, and the content includes a fault type to be analyzed and an expected monitoring accuracy level. The request data including the fault type (such as overload, breakdown, corrosion, etc.) and the positioning level granularity are received as input parameters for subsequent sensing configuration and identification simulation. When no specific level granularity parameter is provided in the positioning request, the granularity classification is set based on the identification diffusion evaluation map and the current monitoring coverage range (for example, the microscopic layer is used for detail confirmation, and the macroscopic layer is used for approximate range identification).

[0070] The multi-layer sensing configuration module calls a matching mechanism, filters sensing units corresponding to the fault type and the level granularity from a database, inputs cable bridge monitoring data (temperature, current, image, etc.) collected in real time according to sensing parameters into the digital twin space, calls a twin simulation model to perform multi-physical field simulation on the current state of the cable bridge, identifies the current fault point position and the specific fault type by comparison with the normal state model, simultaneously performs fault evolution simulation in the digital twin space based on the fault information, predicts the possible development range and trend of the fault, and finally outputs detailed fault positioning information, including the fault type, the specific position, the granularity level, and the risk diffusion map.

[0071] In summary, the multi-layer sensing data fusion cable bridge fault positioning method provided by the embodiments of the present application has the following beneficial effects:

[0072] By analyzing the fault characteristics of the cable bridge, a fault evaluation list is established to provide a structured reference standard for subsequent monitoring and identification. The introduction of success probability and abnormal missing features provides intelligent compensation capability for handling incomplete data and uncertain conditions. According to the fault evaluation list, the fault monitoring feature compensation is performed, and a multi-layer sensor network is constructed to realize fine perception of different fault types and different level data, improve the adaptability and perception ability of the sensor network to complex cable bridge faults, and provide a more comprehensive and accurate data basis for subsequent fault identification positioning. By converting the physical structure and operation data of the cable bridge into a digital twin space, a virtual mapping environment highly matched with the physical entity is provided for fault identification positioning. The embedded multi-level sensing configuration module can perform multi-level sensing data identification configuration in the digital twin space, fully utilize the high fidelity and dynamic characteristics of digital twin technology, and further improve the accuracy and reliability of fault identification, realize effective fusion and cooperation between physical entity and virtual space, and make subsequent fault identification efficient simulation, verification and reasoning in virtual space, improve fault dynamic identification capability. According to the sensing configuration parameters of the multi-level sensing configuration module, the fault identification positioning is performed through the digital twin space, so as to obtain more accurate and stable fault positioning results.

[0073] Overall, the embodiments of the present application realize panoramic perception and high-precision positioning of the operation state of the cable bridge by constructing an evaluation mechanism based on fault feature driving, a multi-source sensor network deployed in layers, and a digital twin space that integrates cable bridge structure and real-time data. Through dynamic configuration and deduction in the digital twin environment, the accuracy and real-time response of cable bridge fault identification under complex working conditions are greatly improved, and the comprehensiveness and reliability of cable bridge fault monitoring are overall enhanced.

[0074] Embodiment two, as Figure 3 shown, based on the same inventive concept as the preceding embodiment one, the embodiments of the present application provide a multi-layer sensing data fusion cable bridge fault positioning system, which comprises:

[0075] A fault feature analysis module 10 is configured to analyze the fault characteristics of the cable bridge and establish a fault evaluation list, which includes fault type, monitoring sensor, success probability, and abnormal missing feature.

[0076] A sensing construction module 20 is configured to perform fault monitoring feature compensation according to the fault evaluation list and construct a multi-layer sensor network, wherein the multi-layer sensor network has fault type label and hierarchical granularity.

[0077] The digital twin module 30 is configured to construct a digital twin space according to a physical structure and operation data of the cable bridge, embed a multi-layer sensing configuration module based on the multi-layer sensor network into the digital twin space, and configure the multi-layer sensing configuration module to perform multi-layer sensing data identification.

[0078] The fault positioning module 40 is configured to identify and locate a fault in the digital twin space according to a sensing configuration parameter of the multi-layer sensing configuration module, and obtain fault positioning information.

[0079] Further, the sensing construction module 20 is further configured to perform the following steps:

[0080] According to the abnormal missing features in the fault evaluation list, a compensation monitoring sensor is matched; according to the monitoring sensor and the compensation monitoring sensor, a hierarchical granularity is obtained by performing hierarchical division according to a range size of a compensation space; a success probability compensation threshold is set, the success probability compensation threshold is taken as a constraint condition, a compensation strategy of the compensation monitoring sensor is analyzed based on the hierarchical granularity, and a compensation sensor and a sensing parameter of each hierarchical granularity are determined; a hierarchical network block of each fault type is established according to the hierarchical granularity and the corresponding compensation sensor and sensing parameter; and the hierarchical network blocks of all fault types are integrated to construct the multi-layer sensor network.

[0081] Further, the sensing construction module 20 is further configured to perform the following steps:

[0082] Homologous sensors between the hierarchical network blocks of each fault type are identified; the hierarchical network blocks of each fault type are connected based on the homologous sensors, and the multi-layer sensor network is constructed by combining the connections of the hierarchical network blocks of all fault types.

[0083] Further, the sensing construction module 20 is further configured to perform the following steps:

[0084] The compensation monitoring sensor includes electrical, physical, chemical, and image monitoring types, the abnormal missing features are compensated according to the electrical, physical, chemical, and image monitoring types and sensing parameters thereof to obtain compensation features, fault fusion identification accuracy and identification range evaluation are performed based on the monitoring sensor and the compensation features of each monitoring type to obtain an identification diffusion evaluation map, and the identification diffusion evaluation map is hierarchically segmented according to a positioning hierarchical range of each fault type to obtain the hierarchical granularity.

[0085] Further, the sensing construction module 20 is further configured to perform the following steps:

[0086] Constrained by the success probability compensation threshold, according to the identified diffusion evaluation map, sensor layout distance and sensor control parameter search are performed to determine compensation sensors and sensor parameters at each hierarchical granularity, the compensation sensors and sensor parameters at each hierarchical granularity being compensation sensor types and their layout strategies and control parameters that meet the success probability compensation threshold and the identification range.

[0087] Further, the digital twin module 30 of the embodiment of the application is further used to perform the following steps:

[0088] According to the physical entity structure data of the cable bridge, a three-dimensional geometric model is constructed; multi-physical field data of the cable bridge is collected and coupled to the three-dimensional geometric model; the three-dimensional geometric model and actual operation data of the cable bridge are combined to create a digital twin base, and the digital twin space is constructed.

[0089] Further, the multi-physical field data includes thermal field, stress field, and electromagnetic field.

[0090] Further, the digital twin module 30 of the embodiment of the application is further used to perform the following steps:

[0091] According to the fault type label and hierarchical granularity of the multi-layer sensor network, hierarchical sensor configuration parameters of each fault type are obtained; the multi-layer sensor configuration module is established according to the hierarchical sensor configuration parameters of each fault type; and the multi-layer sensor configuration module is added to the digital twin space as a built-in sensor parameter configuration tool of the digital twin space.

[0092] Further, the fault positioning module 40 of the embodiment of the application is further used to perform the following steps:

[0093] A fault positioning request is obtained, including a fault type and a positioning hierarchical granularity; the multi-layer sensor configuration module performs sensor configuration matching according to the fault type and the positioning hierarchical granularity, obtains monitoring data of a corresponding sensor network based on the matched sensor configuration parameters, performs state simulation through the digital twin space, identifies fault information and performs fault positioning, simultaneously performs fault evolution simulation based on the fault information to obtain a fault prediction range, and outputs the fault positioning information.

[0094] Through the foregoing detailed description of the method for cable bridge fault positioning based on multi-layer sensor data fusion, those skilled in the art can clearly know the system for cable bridge fault positioning based on multi-layer sensor data fusion in the embodiment. For the system disclosed in the second embodiment, since it corresponds to the method disclosed in the first embodiment, it has corresponding functional modules and beneficial effects, and the related parts are described in the method part.

[0095] The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Modifications of these embodiments will occur to persons of skill in the art, and that the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for locating a fault of a cable bridge with multi-layer sensing data fusion, characterized in that, The method comprises the following steps: analyzing the fault characteristics of the cable bridge, establishing a fault evaluation list, which includes fault types, monitoring sensors, success probabilities, and abnormal missing features; compensating for fault monitoring features according to the fault evaluation list, and constructing a multi-layer sensor network, wherein the multi-layer sensor network has fault type labels and hierarchical granularity; constructing a digital twin space based on the physical structure and operation data of the cable bridge, embedding a multi-layer sensing configuration module based on the multi-layer sensor network into the digital twin space, and using the multi-layer sensing configuration module for multi-level sensing data identification configuration; performing fault identification and positioning through the digital twin space according to the sensing configuration parameters of the multi-layer sensing configuration module, and obtaining fault positioning information; wherein, according to the fault evaluation list, the fault monitoring feature compensation is performed to construct a multi-layer sensor network, which comprises: matching and compensating for monitoring sensors according to the abnormal missing features in the fault evaluation list; dividing the levels according to the range of the compensation space based on the monitoring sensors and the compensation monitoring sensors, and obtaining the hierarchical granularity; setting a success probability compensation threshold, using the success probability compensation threshold as a constraint condition, and performing compensation strategy analysis on the compensation monitoring sensors based on the hierarchical granularity to determine the compensation sensors and sensing parameters of each hierarchical granularity; establishing hierarchical network blocks of each fault type according to the hierarchical granularity and the corresponding compensation sensors and sensing parameters; integrating the hierarchical network blocks of all fault types to construct the multi-layer sensor network; wherein, according to the monitoring sensors and the compensation monitoring sensors, the hierarchical granularity is obtained by dividing the levels according to the range of the compensation space, which comprises: the compensation monitoring sensors include electrical, physical, chemical, and image monitoring types, the abnormal missing features are compensated according to the electrical, physical, chemical, and image monitoring types and their sensing parameters to obtain compensation features; based on the monitoring sensors and the compensation features of each monitoring type, the fault fusion identification accuracy and identification range evaluation are performed to obtain an identification diffusion evaluation map; according to the positioning level range of each fault type, the hierarchical network blocks of all fault types are integrated to construct the multi-layer sensor network. wherein, based on the hierarchical granularity, the compensation strategy analysis is performed on the compensation monitoring sensors to determine the compensation sensors and sensing parameters of each hierarchical granularity, which comprises: taking the success probability compensation threshold as a constraint, searching for sensor layout distance and sensing control parameters according to the identification diffusion evaluation map, and determining the compensation sensors and sensing parameters of each hierarchical granularity, which are the compensation sensor types and their layout strategies and control parameters that meet the success probability compensation threshold and the identification range.

2. The multi-tiered sensor data fusion cable tray fault location method of claim 1, wherein, identifying homologous sensors between the hierarchical network blocks of each fault type; based on the homologous sensors, connecting the hierarchical network blocks of each fault type, combining the hierarchical network blocks of all fault types, and constructing the multi-layer sensor network. ​ 3. The multi-tiered sensor data fusion cable tray fault location method of claim 1, wherein, According to the physical structure and operation data of the cable bridge, a digital twin space is constructed, including: According to the physical entity structure data of the cable bridge, a three-dimensional geometric model is constructed; Collecting multi-physical field data of the cable bridge, coupling multi-physical field data to the three-dimensional geometric model; The three-dimensional geometric model and the actual operation data of the cable bridge are combined to create a digital twin base, and the digital twin space is constructed.

4. The multi-tiered sensor data fusion cable tray fault location method of claim 3, wherein, The multi-physical field data includes: thermal field, stress field, electromagnetic field.

5. The multi-tiered sensor data fusion cable tray fault location method of claim 1, wherein, Based on the multi-layer sensor network, a multi-layer sensing configuration module is established and embedded in the digital twin space, including: According to the fault type label and hierarchical granularity of the multi-layer sensor network, the hierarchical sensing configuration parameters of each fault type are obtained; According to the hierarchical sensing configuration parameters of each fault type, the multi-layer sensing configuration module is established; The multi-layer sensing configuration module is used as a built-in sensing parameter configuration tool of the digital twin space and is added to the digital twin space.

6. The multi-tiered sensor data fusion cable tray fault location method of claim 5, wherein, According to the sensing configuration parameters of the multi-layer sensing configuration module, fault identification and positioning are performed through the digital twin space to obtain fault positioning information, including: Obtain fault positioning request, including fault type and positioning hierarchical granularity; The multi-layer sensing configuration module performs sensing configuration matching according to the fault type and positioning hierarchical granularity, obtains the monitoring data of the corresponding sensor network based on the matched sensing configuration parameters, performs state simulation through the digital twin space, identifies fault information and performs fault positioning, and simultaneously performs fault evolution simulation based on the fault information to obtain fault prediction range, and outputs the fault positioning information.

7. A multi-layer sensing data fusion cable tray fault location system characterized by, The system is used to perform the multi-layer sensing data fusion cable bridge fault positioning method of any one of claims 1-6, including: A fault feature analysis module is used to analyze the fault features of the cable bridge and establish a fault evaluation list, including fault type, monitoring sensor, success probability, and missing features; A sensing construction module is used to compensate for fault monitoring features according to the fault evaluation list and construct a multi-layer sensor network, wherein the multi-layer sensor network has fault type labels and hierarchical granularity; A digital twin module is used to construct a digital twin space according to the physical structure and operation data of the cable bridge, and a multi-layer sensing configuration module is established based on the multi-layer sensor network and embedded in the digital twin space, which is used for multi-level sensing data identification configuration; A fault positioning module is used to perform fault identification and positioning through the digital twin space according to the sensing configuration parameters of the multi-layer sensing configuration module to obtain fault positioning information.

Images