Communication equipment fault prediction method and system based on big data and optical fiber sensors

Abstract

The present application provides a method and system for predicting communication equipment faults based on big data and optical fiber sensors. The method includes: generating a dynamic physical topology model based on a multi-source heterogeneous data matrix, combining the optical attenuation characteristic values ​​and signal transmission direction identifiers of distributed optical fiber sensors; separating abnormal fluctuation components associated with at least one of the abnormal temperature data and vibration data of the communication equipment from the optical signal data based on the spatial constraints of the dynamic physical topology model to generate a cross-modal abnormality correlation feature set; and generating communication equipment fault prediction results for early warning by analyzing the cross-modal abnormality correlation feature set through collaborative constraint rules. The technical solution provided by the present application constructs a multi-source heterogeneous matrix and a dynamic topology model based on distributed optical fiber sensor data, and provides an early and accurate warning of the fault location and type of the communication equipment through cross-modal abnormality correlation feature analysis.

Description

Technical Field

[0001] The present application relates to the technical field of communication equipment fault prediction, and in particular to a communication equipment fault prediction method and system based on big data and optical fiber sensors. Background Art

[0002] In the operation and maintenance of modern large-scale communication equipment, real-time, accurate, and spatially localized monitoring of the health status of communication equipment and early fault warning are crucial to ensuring high reliability and security of communications.

[0003] Existing solutions rely on single-modality or simple fusion machine learning models for analysis. These solutions utilize fiber optic sensor data, possibly combined with static network topology information, to train models to identify anomalies or patterns exceeding preset thresholds, thereby providing fault warnings.

[0004] However, the main drawbacks of existing solutions are that they are difficult to effectively model and do not consider the spatiotemporal correlation between multimodal anomalies. At the same time, the static topology model cannot fully reflect the dynamic changes in the physical state of the optical fiber network, resulting in insufficient spatial positioning accuracy of the anomaly source in complex environments, and weak ability to separate and associate cross-modal anomaly features, which is prone to false alarms or missed alarms, making it difficult to meet the needs of accurately locating fault points and types. Summary of the Invention

[0005] The present application provides a communication equipment fault prediction method and system based on big data and optical fiber sensors, which is used to solve the problem of low accuracy in communication equipment fault prediction in the prior art.

[0006] In a first aspect, the present application provides a communication equipment fault prediction method based on big data and optical fiber sensors, comprising:

[0007] The temperature data, vibration data and optical signal data of communication equipment collected by distributed optical fiber sensors are integrated into a multi-source heterogeneous data matrix according to time series and spatial position;

[0008] Generate a dynamic physical topology model based on the multi-source heterogeneous data matrix and the optical attenuation characteristic value and signal transmission direction identifier of the distributed optical fiber sensor;

[0009] Based on the spatial constraints of the dynamic physical topology model, separating an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device from the optical signal data to generate a cross-modal abnormality correlation feature set;

[0010] Through collaborative constraint rules, the cross-modal abnormal correlation feature set is analyzed to generate communication equipment fault prediction results. The collaborative constraint rules are jointly generated by support vector machines and extreme learning machines. The communication equipment fault prediction results include potential fault locations and fault types in the communication equipment, which are used for early warning.

[0011] Optionally, based on the spatial constraint condition of the dynamic physical topology model, separating an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device from the optical signal data to generate a cross-modal abnormality association feature set includes:

[0012] Decomposing optical signal data into independent spatial signal components corresponding to communication devices or connection segments based on spatial constraints of the dynamic physical topology model;

[0013] Performing time-frequency domain analysis on each of the independent spatial signal components, extracting abnormal fluctuation components that match the fluctuation morphology of historical abnormal temperature or vibration data, and obtaining abnormal temperature data and abnormal vibration data of the communication device;

[0014] Calculating the correlation weight between the abnormal temperature data and the abnormal vibration data of the communication device through multi-scale correlation analysis;

[0015] Performing weighted superposition on the abnormal fluctuation components based on the association weights to generate a weighted abnormal fluctuation component set;

[0016] Components that conflict with the device position or connection path are pruned from the weighted abnormal fluctuation component set to generate a cross-modal correlation feature set.

[0017] Optionally, performing weighted superposition on the abnormal fluctuation components based on the association weights to generate a weighted abnormal fluctuation component set includes:

[0018] Using the associated weight as an amplitude scaling factor, performing an element-by-element multiplication operation on each of the abnormal fluctuation components to generate a weighted abnormal fluctuation component;

[0019] Based on the spatial constraints of the dynamic physical topology model, assigning a unique spatial identifier to each of the weighted abnormal fluctuation components;

[0020] storing the weighted abnormal fluctuation component and the corresponding space identifier as a key-value pair in a hash map data structure;

[0021] Based on the complete storage state of the hash map data structure, a weighted abnormal fluctuation component set is output.

[0022] Optionally, the generating of a communication equipment fault prediction result by analyzing a cross-modal abnormality correlation feature set through collaborative constraint rules includes:

[0023] Based on the cross-modal anomaly correlation feature set, extracting a spatial identifier and a corresponding time-frequency domain feature vector of each cross-modal anomaly correlation feature;

[0024] Based on the time-frequency domain feature vector, a first feature selector is used to perform a feature compression operation to generate a dimension-reduced feature vector, and a second feature selector is used to perform a space mapping operation to generate a projected feature vector;

[0025] Performing a feature fusion operation on the reduced-dimensional feature vector and the projected feature vector to generate a fusion decision vector;

[0026] Retrieving the physical attributes of the corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier, and generating a fault type label in combination with the fusion decision vector;

[0027] The communication equipment fault prediction result is generated by taking the spatial identifier as the fault location index and the fault type label as an element.

[0028] Optionally, the retrieving physical attributes of a corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier and generating a fault type label in combination with the fusion decision vector includes:

[0029] Converting the physical attributes into a structured attribute vector, and generating a physical constraint rule library for the fusion decision vector based on the device material type, the connection segment signal transmission threshold, and the environmental tolerance threshold in the structured attribute vector;

[0030] Under the boundary conditions of the physical constraint rule library, performing a fault mode matching operation on the fusion decision vector to generate a candidate fault mode set;

[0031] When there are fault modes with conflicting physical properties in the candidate fault mode set, conflict resolution is performed based on the priority weights in the structured attribute vector, and a valid fault mode set is output;

[0032] The valid fault mode set is converted into a fault type label that is compatible with the structured attribute vector.

[0033] Optionally, the step of performing a feature compression operation based on the time-frequency domain feature vector using a first feature selector to generate a dimension-reduced feature vector, and performing a space mapping operation based on the second feature selector to generate a projected feature vector includes:

[0034] Based on the frequency band priority in the coordination constraint rule, assigning importance weights to each frequency band component in the time-frequency domain feature vector to generate a weighted time-frequency domain feature vector;

[0035] Inputting the weighted time-frequency domain feature vector into a first feature selector, removing redundant frequency band components and retaining key frequency band features according to a preset communication feature retention rule, and generating a reduced-dimensionality feature vector;

[0036] Constructing a spatial correlation index for the weighted time-frequency domain feature vector according to a mapping relationship between device ports and transmission paths in the dynamic physical topology model;

[0037] The weighted time-frequency domain feature vector and the spatial correlation index are synchronously input into a second feature selector, and feature components across ports are aggregated according to a physical connection topology to generate a projected feature vector.

[0038] Optionally, generating a dynamic physical topology model based on the multi-source heterogeneous data matrix and combining optical attenuation characteristic values ​​and signal transmission direction identifiers of distributed optical fiber sensors includes:

[0039] Extracting spatial coordinates of optical signal event points from the multi-source heterogeneous data matrix, binding the spatial coordinates to corresponding communication device ports based on a preset device port mapping table, and generating a port spatial coordinate set;

[0040] Analyzing the optical transmission connection relationship between adjacent communication device ports based on the optical attenuation characteristic value and the signal transmission direction identifier of the distributed optical fiber sensor to generate a port connection relationship set;

[0041] Based on the port connection relationship set and the transmission stability index, constructing an optical transmission connection relationship graph between ports;

[0042] The port space coordinate set and the optical transmission connection relationship diagram are bound to a preset device physical property library. The device physical properties include port material type, optical fiber length and environmental interference immunity level to generate a weighted dynamic physical topology model.

[0043] In a second aspect, the present application provides a communication equipment fault prediction system based on big data and optical fiber sensors, comprising:

[0044] The acquisition module is used to integrate the temperature data, vibration data, and optical signal data of the communication equipment collected by the distributed optical fiber sensor into a multi-source heterogeneous data matrix according to time series and spatial position;

[0045] A generation module, configured to generate a dynamic physical topology model based on the multi-source heterogeneous data matrix and in combination with the optical attenuation characteristic values ​​and signal transmission direction identifiers of the distributed optical fiber sensors;

[0046] a separation module, configured to separate, from the optical signal data, an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device based on the spatial constraints of the dynamic physical topology model, so as to generate a cross-modal abnormality correlation feature set;

[0047] The analysis module is used to generate communication equipment fault prediction results by analyzing the cross-modal abnormal correlation feature set through collaborative constraint rules. The collaborative constraint rules are jointly generated by a support vector machine and an extreme learning machine. The communication equipment fault prediction results include potential fault locations and fault types in the communication equipment, which are used for early warning.

[0048] In a third aspect, the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a communication equipment fault prediction method based on big data and optical fiber sensors as described in any one of the first aspects.

[0049] In a fourth aspect, the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a communication equipment fault prediction method based on big data and optical fiber sensors as described in any one of the first aspects.

[0050] In the present application, a communication equipment fault prediction method based on big data and optical fiber sensors is provided, which includes: integrating the temperature data, vibration data and optical signal data of the communication equipment collected by the distributed optical fiber sensor into a multi-source heterogeneous data matrix according to the time series and spatial position; generating a dynamic physical topology model based on the multi-source heterogeneous data matrix and combining the optical attenuation characteristic value and signal transmission direction identifier of the distributed optical fiber sensor; based on the spatial constraint conditions of the dynamic physical topology model, separating the abnormal fluctuation component associated with at least one abnormal data of the temperature abnormality data and the vibration abnormality data of the communication equipment from the optical signal data to generate a cross-modal abnormality correlation feature set; through collaborative constraint rules, analyzing the cross-modal abnormality correlation feature set to generate a communication equipment fault prediction result, the collaborative constraint rules are jointly generated by a support vector machine and an extreme learning machine, and the communication equipment fault prediction result includes potential fault locations and fault types in the communication equipment for early warning.

[0051] This application integrates the temperature, vibration and optical signal data collected from distributed optical fiber sensors into a multi-source heterogeneous data matrix according to the time and space dimensions, realizing a unified structured representation of multimodal data; a dynamic physical topology model generated based on the multi-source heterogeneous data matrix combined with optical attenuation characteristics and signal direction identification can map the physical state changes of the optical fiber network in real time; the spatial constraints of the model are used to accurately separate abnormal fluctuation components associated with temperature or vibration anomalies from the optical signal, construct a cross-modal anomaly correlation feature set, and effectively suppress environmental noise interference; finally, the feature set is analyzed through collaborative constraint rules jointly driven by a support vector machine and an extreme learning machine to generate high-precision spatial positioning fault prediction results, thereby improving the accuracy and reliability of early fault warning of communication equipment.

[0052] Furthermore, based on the spatial constraints of the dynamic physical topology model, the present application first decomposes the optical signal data into independent spatial signal components corresponding to the device or connection segment; then performs time-frequency domain analysis on each component to extract abnormal fluctuation components that match the historical temperature and vibration abnormality morphology; then calculates the association weight between the temperature and vibration abnormality data through multi-scale correlation analysis; uses the weight as an amplitude scaling factor to perform element-by-element multiplication on each abnormal fluctuation component to generate a weighted component, and assigns it a unique spatial identifier, which is stored in a hash map data structure in the form of a key-value pair to form a set of weighted abnormal fluctuation components; finally, components that conflict with the device location or connection path are pruned from this set to generate a cross-modal association feature set. This method ensures the precise positioning of anomaly analysis through spatially constrained signal decomposition; time-frequency domain analysis combined with historical morphological matching effectively extracts abnormal fluctuations that are strongly correlated with equipment failures; weighted superposition based on multi-scale correlation calculations strengthens the correlation characteristics between cross-modal anomalies; hash map data structure is used to manage weighted components with spatial identification, improving feature storage and retrieval efficiency; finally, spatial conflict pruning is used to significantly filter out environmental noise and irrelevant interference components, making the generated cross-modal correlation feature set highly reliable, with strong spatial directivity and fault characterization capabilities, laying a solid foundation for subsequent fault prediction.

[0053] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0055] Figure 1A flowchart of a communication equipment fault prediction method based on big data and optical fiber sensors provided in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of the structure of a communication equipment fault prediction system based on big data and optical fiber sensors provided in an embodiment of the present application;

[0057] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0059] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0060] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0061] In order to solve the problem of low accuracy in communication equipment fault prediction in the prior art, an embodiment of the present application provides a communication equipment fault prediction method based on big data and optical fiber sensors. The method adopts the following concept: to solve the problem of collaborative analysis and early fault warning of multi-source heterogeneous sensor data of communication equipment, first, a structured data matrix is ​​constructed by integrating temperature, vibration and optical signals through spatiotemporal alignment, laying the foundation for multimodal fusion; then, the optical fiber physical topology is dynamically modeled by combining optical attenuation characteristics and signal direction identification, breaking through the bottleneck of static topology in characterizing state evolution; based on the spatial constraints of this model, the fluctuation components in the optical signal that are strongly correlated with temperature and vibration anomalies are accurately separated, and a cross-modal anomaly feature set is constructed to suppress environmental noise interference; finally, a collaborative constraint rule driven jointly by a support vector machine and an extreme learning machine is designed, and the spatiotemporal correlation of cross-modal features is deeply integrated to achieve accurate positioning and early warning of fault location and type, thereby improving system reliability.

[0062] Figure 1 A flowchart of a communication equipment fault prediction method based on big data and optical fiber sensors provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0063] S11. Integrate the temperature data, vibration data, and optical signal data of the communication equipment collected by the distributed optical fiber sensor into a multi-source heterogeneous data matrix according to time series and spatial position.

[0064] Among them, the distributed fiber optic sensor may refer to a sensing device that is continuously distributed along the fiber optic laying path, and obtains the physical parameters of the monitored object by measuring the change in the intensity of Rayleigh scattered light. In an embodiment of the present application, the monitored object is a communication device, and the communication device includes network infrastructure such as an optical transmission cabinet and a base station radio frequency unit. Temperature data can be understood as a sequence of physical quantities used to reflect the thermal distribution state of the equipment surface. Vibration data can be a sequence of acceleration measurements that characterizes the mechanical vibration intensity of the equipment. Optical signal data refers to a sequence of optical power values ​​transmitted in the optical fiber that changes with time. A time series refers to a data set arranged at a fixed sampling interval. Spatial position refers to a physical coordinate point based on distance on the optical fiber path. A multi-source heterogeneous data matrix refers to a three-dimensional data structure that integrates temperature, vibration, and optical signals.

[0065] In an embodiment of the present application, first, the temperature data, vibration data, and optical signal data of the communication equipment are synchronously collected through distributed optical fiber sensors; secondly, the three types of data are aligned according to millisecond timestamps to form a time series, and the spatial position is mapped according to the distance coordinates of the sensor on the optical fiber path; then, the time series and spatial position information are fused through multi-dimensional array splicing technology, and finally a multi-source heterogeneous data matrix containing three dimensions of temperature, vibration, and optical signal is generated.

[0066] S12. Generate a dynamic physical topology model based on the multi-source heterogeneous data matrix and the optical attenuation characteristic values ​​and signal transmission direction identifiers of the distributed optical fiber sensors.

[0067] The optical attenuation characteristic value refers to the optical power loss coefficient per unit length of optical fiber, calculated by dividing the input optical power by the output optical power and taking the logarithm. The signal transmission direction identifier is a logical variable that identifies the direction of optical signal propagation in the optical fiber. The dynamic physical topology model can be used to establish a multi-dimensional mapping relationship between potential fault areas and sensor data collection points.

[0068] In an embodiment of the present application, first, the optical attenuation characteristic value is extracted from the optical signal dimension of the multi-source heterogeneous data matrix, which represents the optical power loss per unit length of optical fiber; secondly, the propagation path of the optical signal in the optical fiber is determined in combination with the signal transmission direction identifier; then, a graph theory algorithm is used to construct a network skeleton with communication equipment as nodes and optical fiber links as edges; finally, the real-time calculated optical attenuation characteristic value is injected into the network skeleton as a dynamic edge weight to generate a dynamic physical topology model that reflects the changes in physical state.

[0069] S13. Based on the spatial constraints of the dynamic physical topology model, separate the abnormal fluctuation component associated with at least one of the abnormal temperature data and the abnormal vibration data of the communication device from the optical signal data to generate a cross-modal abnormality association feature set.

[0070] Among them, the spatial constraint condition refers to the geometric rules that limit the signal analysis range, and the effective radius is set with the device location as the center. The implementation of this application does not specifically limit the specific content of the spatial constraint condition. Temperature anomaly data can be a set of continuous data points that exceed the normal range of historical temperature fluctuations. Vibration anomaly data refers to a set of frequency bands in which the vibration energy exceeds a preset threshold. The cross-modal anomaly correlation feature set refers to a set of feature vectors that include spatial position, temperature anomaly intensity, and vibration anomaly intensity. It should be noted that the embodiments of this application do not specifically limit the numerical values ​​of each type of threshold.

[0071] In an embodiment of the present application, first, a signal analysis area around the communication equipment is delineated based on the spatial constraints of a dynamic physical topology model; secondly, wavelet packet decomposition is performed on the optical signal data within the area; then, similarity matching is performed between the decomposed sub-band components and the spectral characteristics of historical temperature anomaly data and vibration anomaly data. The embodiment of the present application does not specifically limit the formula used for similarity matching; finally, the sub-band fluctuation components whose matching degree exceeds a preset threshold are extracted and aggregated to generate a cross-modal anomaly correlation feature set.

[0072] S14. Through collaborative constraint rules, the cross-modal abnormal correlation feature set is analyzed to generate communication equipment fault prediction results. The collaborative constraint rules are jointly generated by support vector machines and extreme learning machines. The communication equipment fault prediction results include potential fault locations and fault types in the communication equipment, which are used for early warning.

[0073] Collaborative constraint rules can be derived through machine learning training based on the correlation between historical multi-source heterogeneous data matrices and dynamic physical topology models on a big data analysis platform. Communication equipment fault prediction results refer to output data containing device location coordinates and fault type identifiers. A support vector machine (SVM) is a classification algorithm based on the principle of structural risk minimization. An extreme learning machine (ELM) is a single-hidden-layer feedforward neural network algorithm. The fault location refers to the spatial coordinates of the device where the anomaly occurred on the optical fiber path. The fault type refers to a predefined device anomaly category identifier.

[0074] In an embodiment of the present application, a collaborative decision-making framework of a support vector machine and an extreme learning machine is first constructed: the support vector machine is responsible for spatially classifying the cross-modal abnormality correlation feature set, and the extreme learning machine is responsible for calculating the probability distribution of the fault type; secondly, the output results of the two types of models are integrated through a weighted fusion algorithm; then, the location of the faulty equipment is mapped according to the spatial identifier of the feature set; finally, the prediction result containing the coordinates of the communication equipment fault location and the fault type code is output and the early warning protocol is triggered.

[0075] The following is a specific example: First, distributed fiber optic sensors are used to collect raw data on temperature, vibration, and optical signals in the communication equipment area in real time. Second, the three types of data are aligned according to a unified timestamp and mapped to optical fiber spatial coordinates to form a multi-source heterogeneous data matrix. Then, the optical attenuation eigenvalues ​​are extracted from the matrix and integrated with the signal direction identifier to construct a dynamic physical topology model. The optical signal data is then segmented based on the spatial constraints of the model, and components associated with temperature and vibration anomalies are extracted through spectrum matching to generate a cross-modal feature set. Finally, a support vector machine and an extreme learning machine are used to collaboratively analyze the feature set, output the fault location coordinates and fault type identifier of the communication equipment, and trigger the early warning process of the operation and maintenance system.

[0076] By executing S11 to S14, the embodiment of the present application builds a unified analysis foundation through the fusion of multi-source spatiotemporal data, uses dynamic topology modeling to accurately describe the changes in the physical state of the optical fiber, and effectively enhances the abnormal signal identification capability based on cross-modal feature separation based on spatial constraints. Combined with the collaborative decision-making mechanism of the dual machine learning model, it achieves high-precision prediction of the location and type of communication equipment faults, thereby improving the reliability of early warning.

[0077] In one possible embodiment, S13, based on the spatial constraints of the dynamic physical topology model, separating an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device from the optical signal data to generate a cross-modal abnormality correlation feature set, including:

[0078] Step 131: Based on the spatial constraints of the dynamic physical topology model, decompose the optical signal data into independent spatial signal components corresponding to the communication devices or connection segments.

[0079] The connection segment can refer to the physical section of optical fiber between adjacent communication devices, for example, with a length ranging from 10 to 500 meters. The independent spatial signal component refers to the discretized representation of the optical signal within the spatial constraints, including the starting point coordinates, the end point coordinates, and the signal curve.

[0080] In an embodiment of the present application, the location of the communication equipment and the range of the optical fiber connection segment are first delineated based on the spatial constraints of the dynamic physical topology model; secondly, the original optical signal data is divided into independent signal segments corresponding to the devices or connection segments according to the spatial coordinates; then, the crosstalk between segments is eliminated through a digital filter; finally, a set of independent spatial signal components for each device or connection segment is output.

[0081] Step 132: Perform time-frequency domain analysis on each independent spatial signal component, extract abnormal fluctuation components that match the fluctuation shape of historical abnormal temperature or vibration data, and obtain abnormal temperature data and abnormal vibration data of the communication device.

[0082] Historical abnormal temperatures refer to a collection of temperature anomaly data marked as fault-related during past monitoring periods, stored in a two-dimensional time-amplitude array. Vibration data fluctuations refer to the time-domain waveform envelope characteristics of the vibration acceleration signal, extracted using a Hilbert transform. Abnormal fluctuation components are signal segments in the time-frequency domain that match the historical abnormal morphology and include three-dimensional features: center frequency, duration, and energy intensity.

[0083] In an embodiment of the present application, a continuous wavelet transform is first performed on each independent spatial signal component; secondly, the time-frequency distribution characteristics of the wavelet coefficients are extracted; then, the characteristics are matched with the fluctuation morphology library of historical abnormal temperature data and the fluctuation morphology library of historical vibration data for similarity; finally, the component with the highest similarity is screened and marked as the abnormal fluctuation component, and a set of abnormal temperature data and abnormal vibration data is generated.

[0084] Step 133: Calculate the correlation weight between the abnormal temperature data and the abnormal vibration data of the communication device through multi-scale correlation analysis.

[0085] Among them, multi-scale correlation analysis calculates data correlation in three dimensions: time domain, frequency domain, and time-frequency domain.

[0086] In this embodiment, the temperature anomaly data and vibration anomaly data are first aligned according to the time window. The Pearson correlation coefficient is then calculated at the time, frequency, and time-frequency scales. The multi-scale correlation coefficients are then weighted averaged. Finally, the correlation weights are normalized to obtain the correlation weights between the temperature and vibration anomaly data. The correlation weights are quantified values ​​of the correlation between the temperature and vibration anomaly data, ranging from 0 to 1, and can be calculated as the arithmetic average of the correlation coefficients at each scale.

[0087] Step 134: Perform weighted superposition on the abnormal fluctuation components based on the associated weights to generate a weighted abnormal fluctuation component set.

[0088] Among them, the weighted abnormal fluctuation component set refers to the set of abnormal fluctuation components after amplitude scaling, and the data structure is a key-value pair consisting of a space identifier and a weighted signal vector.

[0089] In an embodiment of the present application, the associated weight is first converted into an amplitude scaling coefficient; secondly, a coefficient multiplication operation is performed on each abnormal fluctuation component; then a spatial identifier is assigned to the weighted component; finally, the device location index is stored in a hash table to form a set of weighted abnormal fluctuation components.

[0090] Step 135: prune the components that conflict with the device position or connection path from the weighted abnormal fluctuation component set to generate a cross-modal correlation feature set.

[0091] The component of connection path conflict refers to the component of the distance deviation between the spatial identifier and the actual connection path of the topology model.

[0092] In an embodiment of the present application, the device location and connection path data of the dynamic physical topology model are first loaded; secondly, the spatial identifiers of the weighted abnormal fluctuation component set are traversed; then, the geometric deviation between the identifier and the actual location of the device is detected; finally, the components with deviations exceeding the threshold are deleted to generate a cross-modal association feature set.

[0093] The following is a specific example: first, the optical signal is divided into independent spatial signal components associated with the device based on the spatial constraints of the dynamic topology model; second, the time-frequency characteristics of the components are extracted through wavelet transform and matched with the historical temperature vibration anomaly morphology to generate abnormal fluctuation components; then the multi-scale association weights of the temperature vibration anomaly data are calculated; then the abnormal fluctuation components are amplitude-weighted according to the weights and stored as a set; finally, the spatial offset components are pruned according to the actual position of the device to generate a cross-modal association feature set.

[0094] By executing steps 131 to 135, the embodiment of the present application ensures the accuracy of anomaly positioning through spatially constrained signal decomposition, enhances the fault feature extraction capability through time-frequency domain feature matching, strengthens cross-modal correlation through multi-scale correlation weighting, and effectively filters out interference components through spatial conflict pruning, ultimately generating a highly reliable cross-modal correlation feature set.

[0095] In a possible embodiment, step 134, performing weighted superposition on the abnormal fluctuation components based on the associated weights to generate a weighted abnormal fluctuation component set, includes:

[0096] Step a1: Use the associated weight as an amplitude scaling factor, perform element-by-element multiplication on each abnormal fluctuation component, and generate a weighted abnormal fluctuation component.

[0097] The amplitude scaling factor is a scaling factor used to adjust the signal amplitude, including values ​​calculated based on associated weights, used to amplify or reduce the strength of the abnormal fluctuation component. Element-by-element multiplication refers to the multiplication of elements at corresponding positions in two arrays or matrices, including element-by-element multiplication, to generate a new array. The weighted abnormal fluctuation component is the abnormal fluctuation component weighted by the amplitude scaling factor, including the resulting data reflecting the product of the original fluctuation and the weight.

[0098] Step a2: Based on the spatial constraints of the dynamic physical topology model, a unique spatial identifier is assigned to each weighted abnormal fluctuation component.

[0099] Among them, the spatial identifier refers to an identifier used to uniquely identify a physical space location, including a unique code generated based on the spatial constraints of a dynamic physical topology model.

[0100] Step a3: Store the weighted abnormal fluctuation component and the corresponding space identifier as a key-value pair in a hash mapping data structure.

[0101] A key-value pair is a data structure element consisting of a key and a value. The key uniquely identifies an entity, while the value stores the associated data. A hash map data structure is a hash table-based data storage structure that includes a key-value pair storage mechanism for efficient data lookup and access.

[0102] Step a4: Based on the complete storage state of the hash map data structure, output a set of weighted abnormal fluctuation components.

[0103] The complete storage state refers to a state in which all data in the hash map data structure has been successfully stored, including a confirmation state reflecting that all key-value pairs have been saved without omission.

[0104] The following is a specific example: First, the sensor network acquires data on abnormal fluctuation components at each node in the power grid. The associated weights are extracted as amplitude scaling factors, and each component is subjected to element-by-element multiplication to generate weighted abnormal fluctuation components. Second, spatial constraints are queried based on the dynamic physical topology model of the power grid, and a unique spatial identifier is assigned to each weighted component. The weighted component and spatial identifier are then combined into a key-value pair and stored in a hash map data structure. Finally, the hash map's complete storage state is verified, and the set of weighted abnormal fluctuation components is output.

[0105] By executing steps a1 to a4, the embodiment of the present application enhances the identifiability of abnormal fluctuations through weighted processing, combines spatial constraints to achieve accurate association between components and physical locations, uses hash mapping data structures to improve data storage and access efficiency, and ultimately outputs a complete weighted set, thereby improving the system's processing accuracy and response speed to abnormal fluctuations.

[0106] In a possible embodiment, S14, analyzing the cross-modal abnormality correlation feature set through collaborative constraint rules to generate a communication device fault prediction result, including:

[0107] Step 141: Based on the cross-modal anomaly correlation feature set, extract the spatial identifier and the corresponding time-frequency domain feature vector of each cross-modal anomaly correlation feature.

[0108] The time-frequency domain feature vector refers to a feature vector that contains both time variation and frequency distribution, including the time-frequency characteristics of the signal extracted by short-time Fourier transform or wavelet transform.

[0109] In an embodiment of the present application, first, based on the cross-modal anomaly association feature set, the spatial identifier and the corresponding time-frequency domain feature vector of each cross-modal anomaly association feature are extracted through a feature extraction algorithm, where the spatial identifier is used to locate the physical location, and the time-frequency domain feature vector contains feature information in the time and frequency dimensions.

[0110] Step 142: Based on the time-frequency domain feature vector, use the first feature selector to perform a feature compression operation to generate a dimension-reduced feature vector, and use the second feature selector to perform a space mapping operation to generate a projected feature vector.

[0111] Among them, the first feature selector refers to a feature processing module for dimensionality reduction, including algorithms based on principal component analysis or autoencoders. Feature compression operation refers to the process of reducing feature dimensions, including removing redundant information through linear transformation or neural network. The reduced dimensionality feature vector refers to a low-dimensional feature vector after the compression operation, including a simplified feature representation that retains key information. The second feature selector refers to a feature processing module for spatial mapping. Spatial mapping operation refers to the transformation process of mapping features to a new spatial coordinate system, including feature reconstruction through matrix multiplication. The projected feature vector refers to the new feature vector generated after spatial mapping, including feature data reflecting the distribution of low-dimensional space.

[0112] In an embodiment of the present application, first, based on the time-frequency domain feature vector, a first feature selector is used to perform a feature compression operation through principal component analysis or an autoencoder algorithm to remove redundant information and generate a reduced-dimensional feature vector; secondly, a second feature selector is used to perform a spatial mapping operation through a spatial transformation algorithm to project the feature into a low-dimensional space to generate a projected feature vector.

[0113] Step 143: Perform a feature fusion operation on the reduced-dimensional feature vector and the projected feature vector to generate a fusion decision vector.

[0114] Among them, the fused decision vector refers to a composite vector after integrating multiple features, including decision input data generated by splicing or weighted fusion.

[0115] In an embodiment of the present application, a feature fusion operation is first performed on the reduced-dimensional feature vector and the projected feature vector, and the two types of features are integrated through a weighted splicing or attention mechanism algorithm, and finally a fusion decision vector is generated as the input basis for fault judgment.

[0116] Step 144: Retrieve the physical attributes of the corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier, and generate a fault type label in combination with the fusion decision vector.

[0117] Among them, the fault type label refers to the label that identifies the equipment fault category, including the classification results based on physical attributes and fusion decision vectors.

[0118] In an embodiment of the present application, the specific implementation process of step 144 is as follows: first, a retrieval operation is performed in the dynamic physical topology model according to the spatial identifier to locate the communication device entity or network connection segment corresponding to the identifier; second, the physical attribute data of the target device or connection segment is extracted, including the device model, material parameters, topological connection relationship and historical operation threshold; then, the fused decision vector and the physical attribute data are input into the pre-trained fault classification model, and the fault mode probability distribution is calculated by matching the vector features with the physical constraint rules; finally, the dominant fault type is determined based on the probability distribution peak, and the corresponding fault type label is generated.

[0119] Step 145: Generate a communication equipment fault prediction result using the spatial identifier as the fault location index and the fault type label as an element.

[0120] Among them, in the embodiment of the present application, the spatial identifier is first used as the fault location index and the fault type label is used as the element, and the spatial and type information are integrated through the data encapsulation algorithm to finally generate the communication equipment fault prediction result.

[0121] The following is a specific example: First, based on the cross-modal anomaly correlation feature set, the spatial identifiers and time-frequency domain feature vectors of the vibration and temperature anomaly characteristics of the communication base station are extracted. Secondly, the principal component analysis algorithm is used to perform feature compression on the time-frequency domain feature vectors to generate a reduced-dimensional feature vector. At the same time, the manifold learning algorithm is used to perform spatial mapping to generate a projected feature vector. Then, a weighted splicing fusion is performed on the two types of feature vectors to generate a fused decision vector. The heat dissipation properties and connection relationships of the base station equipment in the dynamic physical topology model are then retrieved based on the spatial identifier. The overload or short circuit fault type label is output in combination with the fused decision vector. Finally, the spatial identifier is used to locate the location of the faulty base station and the fault type label is used as the factor to generate the base station fault prediction result.

[0122] By executing steps 141 to 145, the embodiment of the present application achieves deep compression and fusion of fault features through spatial-temporal-frequency joint analysis of cross-modal features, accurately associates device attributes with the physical topology model, generates integrated prediction results of spatial location and fault type, and improves the accuracy and positioning efficiency of communication network fault diagnosis.

[0123] In one possible embodiment, step 144, retrieving the physical attributes of the corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier and generating a fault type label in combination with the fusion decision vector, includes:

[0124] Step b1: Convert the physical attributes into a structured attribute vector, and generate a physical constraint rule library for the fusion decision vector based on the device material type, connection segment signal transmission threshold, and environmental tolerance threshold in the structured attribute vector.

[0125] Physical attributes refer to the inherent characteristics of a communication device or connection segment, including conductivity determined by the device material type, maximum load capacity specified by the signal transmission threshold for the connection segment, and temperature and humidity adaptability defined by the environmental tolerance threshold. A structured attribute vector is a mathematical vector that encodes discrete physical attributes in a dimensionally coded manner, including standardized feature representations generated through one-hot encoding or normalization. Device material type refers to the material category of the communication device, including core material properties that affect electrical performance, such as metal conductors, optical fiber, or semiconductor components. The signal transmission threshold for the connection segment refers to the maximum signal strength threshold allowed for transmission in the network connection segment, including a theoretical safety margin calculated based on the signal-to-noise ratio and attenuation coefficient. The environmental tolerance threshold refers to the extreme environmental parameter limits that the device can withstand, including maximum operating temperature, maximum humidity range, and electromagnetic interference resistance level. A physical constraint rule base is a database that stores the correlations between device physical characteristics and faults, including mapping rules between material thermal expansion coefficients and temperature faults, and causal logic between transmission thresholds and overload faults.

[0126] Step b2: Under the boundary conditions of the physical constraint rule base, perform a fault pattern matching operation on the fused decision vector to generate a candidate fault pattern set.

[0127] Fault pattern matching is the process of comparing physical properties with pre-set fault rules, including identifying potential fault types through decision tree traversal or graph neural network reasoning. The candidate fault pattern set is a list of fault types generated by preliminary matching, including fault hypotheses that satisfy some physical constraints but may conflict.

[0128] Step b3: When there are fault modes with conflicting physical properties in the candidate fault mode set, conflict resolution is performed based on the priority weights in the structured attribute vector, and a valid fault mode set is output.

[0129] Physical property conflicts occur when candidate failure modes conflict with the device's actual characteristics, including the logical error of requiring a "fiber break" failure in copper cable equipment. Conflict resolution is the process of resolving conflicts through priority rules, including sorting failure possibilities based on material tolerance thresholds and eliminating low-probability options.

[0130] Step b4: Convert the valid fault mode set into a fault type label that is compatible with the structured attribute vector.

[0131] The following is a specific example: under the boundary conditions of the physical constraint rule base, a fault pattern matching operation is performed on the fused decision vector to generate a set of candidate fault patterns.

[0132] Fault pattern matching is the process of comparing physical properties with pre-set fault rules, including identifying potential fault types through decision tree traversal or graph neural network reasoning. The candidate fault pattern set is a list of fault types generated by preliminary matching, including fault hypotheses that satisfy some physical constraints but may conflict.

[0133] Step b3: When there are fault modes with conflicting physical properties in the candidate fault mode set, conflict resolution is performed based on the priority weights in the structured attribute vector, and a valid fault mode set is output.

[0134] Physical property conflicts occur when candidate failure modes conflict with the device's actual characteristics, including the logical error of requiring a "fiber break" failure in copper cable equipment. Conflict resolution is the process of resolving conflicts through priority rules, including sorting failure possibilities based on material tolerance thresholds and eliminating low-probability options.

[0135] Step b4: Convert the valid fault mode set into a fault type label that is compatible with the structured attribute vector.

[0136] The following is a specific example: First, the dynamic physical topology data of a Fifth Generation Mobile Communication Systems (5G) base station is retrieved based on the spatial identifier. The aluminum alloy chassis material type, the transmission threshold of the optical fiber connection segment, and the environmental tolerance threshold are extracted and integrated into a structured attribute vector. This vector is then input into the physical constraint rule library to match the candidate fault mode sets of "overheating and frequency reduction" and "optical failure interruption." A conflict between "optical failure interruption" and the aluminum alloy material is detected. Based on the material's thermal conductivity, the conflict is resolved, retaining the valid "overheating and frequency reduction" fault mode. Finally, a similarity calculation is performed between the fused decision vector and the valid fault mode to generate a fault type label of "overheating and frequency reduction due to heat dissipation failure."

[0137] By executing steps b1 to b4, the embodiment of the present application generates fault diagnosis results that conform to the actual characteristics of the equipment under the protection of the conflict resolution mechanism through deep binding of physical properties and fault rules, avoiding misjudgments that violate physical laws, and improving the engineering applicability and reliability of communication network fault location.

[0138] In a possible embodiment, step 142, based on the time-frequency domain feature vector, using a first feature selector to perform a feature compression operation to generate a dimension-reduced feature vector, and using a second feature selector to perform a spatial mapping operation to generate a projected feature vector, includes:

[0139] Step c1: Based on the frequency band priority in the coordination constraint rule, assign importance weights to the frequency band components in the time-frequency domain feature vector to generate a weighted time-frequency domain feature vector.

[0140] Frequency band priority refers to the importance of frequency bands as determined by the communication protocol, including weight coefficients that are dynamically adjusted based on service type or channel quality. Weighted time-frequency feature vectors refer to time-frequency feature data scaled by frequency band priority, including numerical representations reflecting the relative importance of each frequency band.

[0141] Step c2: input the weighted time-frequency domain feature vector into the first feature selector, remove redundant frequency band components according to the preset communication feature retention rule, retain the key frequency band features, and generate a reduced-dimensionality feature vector.

[0142] The preset communication feature retention rules refer to predefined frequency band screening criteria, including logical conditions such as requiring the signal-to-noise ratio to be above a set threshold or the energy ratio to exceed a minimum retention ratio. Frequency band components refer to subvectors within a time-frequency domain feature vector that represent specific frequency intervals, including discrete spectral units derived through Fourier transform decomposition. Key frequency band features refer to the core frequency band components that have been screened and retained, including the effective spectral components that carry key fault information.

[0143] Step c3: construct a spatial correlation index for the weighted time-frequency domain feature vector according to the mapping relationship between the device ports and the transmission paths in the dynamic physical topology model.

[0144] Among them, the spatial association index refers to the index structure that identifies the relationship between features and physical locations, including the encoding data of device port numbers, transmission path identifiers, and topological connection directions.

[0145] Step c4: synchronously input the weighted time-frequency domain feature vector and the spatial correlation index into the second feature selector, aggregate the feature components across ports according to the physical connection topology, and generate a projected feature vector.

[0146] Among them, the physical connection topology refers to the actual connection architecture between communication devices, including the logical model of port cascade relationship, transmission path direction and signal flow direction.

[0147] The following is a specific example: First, based on the frequency band priority of the 5G base station, a high weight is assigned to the high-frequency band components in the time-frequency domain feature vector to generate a weighted time-frequency domain feature vector. Secondly, the first feature selector is used to eliminate redundant frequency band components with a signal-to-noise ratio below the threshold value, retaining the key features of the high-frequency band to generate a reduced-dimensional feature vector. Subsequently, based on the mapping relationship between the device port of the remote radio frequency unit of the base station and the antenna transmission path, a spatial association index containing the port identifier is constructed for the weighted feature. Finally, the weighted features and index are input into the second feature selector, and the cross-port features are aggregated according to the physical connection topology of the antenna cluster to generate a projected feature vector that characterizes the signal coverage quality.

[0148] By executing steps c1 to c4, the embodiment of the present application strengthens key fault features through frequency band priority weighting, combines physical topology constraints to achieve feature dimensionality reduction and spatial aggregation, and generates projection features that retain core frequency band information and conform to device connection relationships, thereby improving the recognition and spatial interpretability of communication system fault features.

[0149] In a possible embodiment, S12, generating a dynamic physical topology model based on the multi-source heterogeneous data matrix and combining the optical attenuation characteristic values ​​and signal transmission direction identifiers of the distributed optical fiber sensor, includes:

[0150] Step 121: extract the spatial coordinates of optical signal event points from the multi-source heterogeneous data matrix, bind the spatial coordinates to corresponding communication device ports based on a preset device port mapping table, and generate a port spatial coordinate set.

[0151] An optical signal event point refers to the location of an abnormal optical signal caused by bending, breaking, or connector loss in an optical fiber, including spatial coordinate data detected through sudden changes in scattered signal intensity. A preset device port mapping table is a configuration table that stores the correspondence between the logical port numbers of communication equipment and their actual physical locations, including mapping rules for room numbers, rack locations, and port sequences. A port spatial coordinate set refers to a three-dimensional location dataset for bound device ports, including a set of port geographic coordinates consisting of longitude, latitude, and elevation information.

[0152] Step 122: Analyze the optical transmission connection relationship between adjacent communication device ports based on the optical attenuation characteristic value and the signal transmission direction identifier of the distributed optical fiber sensor, and generate a port connection relationship set.

[0153] The optical attenuation characteristic value refers to the quantified value of the signal attenuation intensity along the optical fiber transmission path, including the loss coefficient per unit distance calculated based on the backscattered optical power. The port connection relationship set refers to the data set that describes the physical connection between the ports of communication devices, including the connection records of the source port, destination port, and transmission direction.

[0154] Step 123: Based on the port connection relationship set and the transmission stability index, construct an optical transmission connection relationship graph between ports.

[0155] The transmission stability index measures the reliability of optical transmission links, including a comprehensive score calculated based on historical bit error rates, jitter rates, and outage frequencies. The optical transmission connection graph is a topological graph composed of ports as nodes and connections as edges, including a graph structure whose edge weights reflect transmission stability. The transmission stability index is calculated based on a comprehensive score calculated from historical bit error rates, jitter rates, and outage frequencies.

[0156] Step 124 : Bind the port space coordinate set and the optical transmission connection relationship diagram with the preset device physical property library. The device physical properties include port material type, optical fiber length, and environmental interference immunity level, and generate a weighted dynamic physical topology model.

[0157] Among them, the port material type refers to the physical material category of the optical fiber port interface, including characteristics such as ceramic ferrule, metal shell or plastic connector that affect signal loss. The optical fiber length refers to the physical length value of the optical fiber between the ports of two communication devices, including the precise distance data obtained based on the optical time domain reflectometer measurement. The environmental immunity level refers to the port's tolerance rating in a complex environment, including the electromagnetic interference resistance strength, waterproof and dustproof level and temperature adaptability range. The weight calculation in the weighted dynamic physical topology model is obtained by integrating the transmission stability index with the key physical property factors. The calculation formula of the weight W is: ,in, is the transmission stability index, is the fiber length factor, is the environmental interference immunity factor, is the port material factor.

[0158] The following is a specific example: First, the spatial coordinates of the optical cable bend event points are extracted from the optical fiber vibration monitoring data matrix. These coordinates are then bound to the optical line terminal ports according to the core equipment room port mapping table to generate a set of port spatial coordinates. Next, based on the optical attenuation characteristic values ​​and transmission directions of the distributed sensors, the optical transmission connection relationship set between adjacent optical lines and optical network unit ports is parsed. A weighted trunk optical cable connection relationship diagram is then constructed, combining transmission stability indicators. Finally, a dynamic physical topology model is generated by linking the port spatial coordinate set, the connection relationship diagram, and the ceramic material type, optical fiber length, and interference immunity rating from the device physical property library.

[0159] By executing steps 121 to 124, the embodiment of the present application constructs a dynamic topology model that reflects the real-time connection status and reliability by integrating the spatial positioning of optical signals with the physical properties of the equipment, thereby providing an accurate physical layer basic framework for optical fiber network fault diagnosis.

[0160] Figure 2 A schematic diagram of the structure of a communication equipment fault prediction system based on big data and optical fiber sensors provided in an embodiment of the present application is shown as follows: Figure 2 As shown, the system includes:

[0161] The acquisition module 21 is used to integrate the temperature data, vibration data and optical signal data of the communication equipment collected by the distributed optical fiber sensor into a multi-source heterogeneous data matrix according to time series and spatial position.

[0162] The generation module 22 is used to generate a dynamic physical topology model based on the multi-source heterogeneous data matrix and in combination with the optical attenuation characteristic value and signal transmission direction identifier of the distributed optical fiber sensor.

[0163] The separation module 23 is used to separate the abnormal fluctuation component associated with at least one of the abnormal temperature data and the abnormal vibration data of the communication device from the optical signal data based on the spatial constraints of the dynamic physical topology model, so as to generate a cross-modal abnormality association feature set.

[0164] The analysis module 24 is used to generate communication equipment fault prediction results by analyzing the cross-modal abnormal correlation feature set through collaborative constraint rules. The collaborative constraint rules are jointly generated by a support vector machine and an extreme learning machine. The communication equipment fault prediction results include potential fault locations and fault types in the communication equipment, which are used for early warning.

[0165] Figure 2 The communication equipment fault prediction system based on big data and optical fiber sensors can be performed Figure 1 The implementation principles and technical effects of the communication equipment fault prediction method based on big data and fiber optic sensors described in the illustrated embodiment are not further elaborated. The specific manner in which each module and unit performs operations in the communication equipment fault prediction system based on big data and fiber optic sensors described in the above embodiment has been described in detail in the relevant embodiments of the method and will not be elaborated on here.

[0166] In one possible design, Figure 2 The communication equipment fault prediction system based on big data and optical fiber sensors of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32 .

[0167] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0168] The processing component 32 is used to execute the following process: integrating the temperature data, vibration data and optical signal data of the communication equipment collected by the distributed optical fiber sensor into a multi-source heterogeneous data matrix according to time series and spatial position; generating a dynamic physical topology model based on the multi-source heterogeneous data matrix and combining the optical attenuation characteristic value and signal transmission direction identifier of the distributed optical fiber sensor; based on the spatial constraint conditions of the dynamic physical topology model, separating the abnormal fluctuation component associated with at least one abnormal data of the temperature abnormality data and the vibration abnormality data of the communication equipment from the optical signal data to generate a cross-modal abnormality correlation feature set; through collaborative constraint rules, analyzing the cross-modal abnormality correlation feature set to generate a communication equipment fault prediction result, the collaborative constraint rules are jointly generated by a support vector machine and an extreme learning machine, and the communication equipment fault prediction result includes potential fault locations and fault types in the communication equipment for early warning.

[0169] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0170] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0171] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0172] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0173] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0174] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0175] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment provides a communication equipment fault prediction method based on big data and optical fiber sensors.

[0176] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0177] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0178] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A communication equipment fault prediction method based on big data and optical fiber sensors, characterized in that: include: The temperature data, vibration data and optical signal data of communication equipment collected by distributed optical fiber sensors are integrated into a multi-source heterogeneous data matrix according to time series and spatial position; Generate a dynamic physical topology model based on the multi-source heterogeneous data matrix and the optical attenuation characteristic value and signal transmission direction identifier of the distributed optical fiber sensor; Based on the spatial constraints of the dynamic physical topology model, separating an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device from the optical signal data to generate a cross-modal abnormality correlation feature set; By analyzing cross-modal anomaly correlation feature sets through collaborative constraint rules, communication equipment fault prediction results are generated. The collaborative constraint rules are jointly generated using a support vector machine and an extreme learning machine. The communication equipment fault prediction results include potential fault locations and fault types in the communication equipment, which are used for early warning. The method of generating a communication equipment fault prediction result by analyzing a cross-modal abnormal correlation feature set through collaborative constraint rules includes: Based on the cross-modal anomaly correlation feature set, extracting a spatial identifier and a corresponding time-frequency domain feature vector of each cross-modal anomaly correlation feature; Based on the time-frequency domain feature vector, a first feature selector is used to perform a feature compression operation to generate a dimension-reduced feature vector, and a second feature selector is used to perform a space mapping operation to generate a projected feature vector; Performing a feature fusion operation on the reduced-dimensional feature vector and the projected feature vector to generate a fusion decision vector; Retrieving the physical attributes of the corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier, and generating a fault type label in combination with the fusion decision vector; The communication equipment fault prediction result is generated by taking the spatial identifier as the fault location index and the fault type label as an element.

2. The method according to claim 1, characterized in that The method of separating, based on the spatial constraint of the dynamic physical topology model, an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device from the optical signal data to generate a cross-modal abnormality correlation feature set includes: Decomposing optical signal data into independent spatial signal components corresponding to communication devices or connection segments based on spatial constraints of the dynamic physical topology model; Performing time-frequency domain analysis on each of the independent spatial signal components, extracting abnormal fluctuation components that match the fluctuation morphology of historical abnormal temperature or vibration data, and obtaining abnormal temperature data and abnormal vibration data of the communication device; Calculating the correlation weight between the abnormal temperature data and the abnormal vibration data of the communication device through multi-scale correlation analysis; Performing weighted superposition on the abnormal fluctuation components based on the association weights to generate a weighted abnormal fluctuation component set; Components that conflict with the device position or connection path are pruned from the weighted abnormal fluctuation component set to generate a cross-modal correlation feature set.

3. The method according to claim 2, characterized in that The step of weighting and superimposing the abnormal fluctuation components based on the association weights to generate a weighted abnormal fluctuation component set includes: Using the associated weight as an amplitude scaling factor, performing an element-by-element multiplication operation on each of the abnormal fluctuation components to generate a weighted abnormal fluctuation component; Based on the spatial constraints of the dynamic physical topology model, assigning a unique spatial identifier to each of the weighted abnormal fluctuation components; storing the weighted abnormal fluctuation component and the corresponding space identifier as a key-value pair in a hash map data structure; Based on the complete storage state of the hash map data structure, a weighted abnormal fluctuation component set is output.

4. The method according to claim 1, wherein The retrieving the physical attributes of the corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier and generating a fault type label in combination with the fusion decision vector includes: Converting the physical attributes into a structured attribute vector, and generating a physical constraint rule library for the fusion decision vector based on the device material type, the connection segment signal transmission threshold, and the environmental tolerance threshold in the structured attribute vector; Under the boundary conditions of the physical constraint rule library, performing a fault mode matching operation on the fusion decision vector to generate a candidate fault mode set; When there are fault modes with conflicting physical properties in the candidate fault mode set, conflict resolution is performed based on the priority weights in the structured attribute vector, and a valid fault mode set is output; The valid fault mode set is converted into a fault type label that is compatible with the structured attribute vector.

5. The method according to claim 1, wherein The method of performing a feature compression operation based on the time-frequency domain feature vector using a first feature selector to generate a dimension-reduced feature vector, and performing a space mapping operation using a second feature selector to generate a projected feature vector includes: Based on the frequency band priority in the coordination constraint rule, assigning importance weights to each frequency band component in the time-frequency domain feature vector to generate a weighted time-frequency domain feature vector; Inputting the weighted time-frequency domain feature vector into a first feature selector, removing redundant frequency band components and retaining key frequency band features according to a preset communication feature retention rule, and generating a reduced-dimensionality feature vector; Constructing a spatial correlation index for the weighted time-frequency domain feature vector according to a mapping relationship between device ports and transmission paths in the dynamic physical topology model; The weighted time-frequency domain feature vector and the spatial correlation index are synchronously input into a second feature selector, and feature components across ports are aggregated according to a physical connection topology to generate a projected feature vector.

6. The method according to claim 1, characterized in that The generating of a dynamic physical topology model based on the multi-source heterogeneous data matrix and in combination with the optical attenuation characteristic value and the signal transmission direction identifier of the distributed optical fiber sensor includes: Extracting the spatial coordinates of optical signal event points from the multi-source heterogeneous data matrix, binding the spatial coordinates to corresponding communication device ports based on a preset device port mapping table, and generating a port spatial coordinate set; Analyzing the optical transmission connection relationship between adjacent communication device ports based on the optical attenuation characteristic value and the signal transmission direction identifier of the distributed optical fiber sensor to generate a port connection relationship set; Constructing an optical transmission connection relationship graph between ports based on the port connection relationship set and the transmission stability index; The port space coordinate set and the optical transmission connection relationship diagram are bound to a preset device physical property library. The device physical properties include port material type, optical fiber length and environmental interference immunity level to generate a weighted dynamic physical topology model.

7. A communication equipment fault prediction system based on big data and optical fiber sensors, characterized in that: include: The acquisition module is used to integrate the temperature data, vibration data, and optical signal data of the communication equipment collected by the distributed optical fiber sensor into a multi-source heterogeneous data matrix according to time series and spatial position; A generation module, configured to generate a dynamic physical topology model based on the multi-source heterogeneous data matrix and in combination with the optical attenuation characteristic values ​​and signal transmission direction identifiers of the distributed optical fiber sensors; a separation module, configured to separate, from the optical signal data, an abnormal fluctuation component associated with at least one of abnormal temperature data and abnormal vibration data of the communication device based on the spatial constraints of the dynamic physical topology model, so as to generate a cross-modal abnormality correlation feature set; An analysis module is configured to generate a communication equipment fault prediction result by analyzing a cross-modal anomaly correlation feature set using collaborative constraint rules. The collaborative constraint rules are jointly generated using a support vector machine and an extreme learning machine. The communication equipment fault prediction result includes potential fault locations and fault types in the communication equipment, which are used for early warning. The method of generating a communication equipment fault prediction result by analyzing a cross-modal abnormal correlation feature set through collaborative constraint rules includes: Based on the cross-modal anomaly correlation feature set, extracting a spatial identifier and a corresponding time-frequency domain feature vector of each cross-modal anomaly correlation feature; Based on the time-frequency domain feature vector, a first feature selector is used to perform a feature compression operation to generate a dimension-reduced feature vector, and a second feature selector is used to perform a space mapping operation to generate a projected feature vector; Performing a feature fusion operation on the reduced-dimensional feature vector and the projected feature vector to generate a fusion decision vector; Retrieving the physical attributes of the corresponding communication device or connection segment in the dynamic physical topology model based on the spatial identifier, and generating a fault type label in combination with the fusion decision vector; The communication equipment fault prediction result is generated by taking the spatial identifier as the fault location index and the fault type label as an element.

8. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a communication equipment fault prediction method based on big data and optical fiber sensors as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, a communication equipment fault prediction method based on big data and optical fiber sensors as described in any one of claims 1 to 6 is implemented.

Images