CN118174788B - Fault detection method, device and equipment of optical fiber wiring cabinet and storage medium - Google Patents
Fault detection method, device and equipment of optical fiber wiring cabinet and storage medium Download PDFInfo
- Publication number
- CN118174788B CN118174788B CN202410586099.1A CN202410586099A CN118174788B CN 118174788 B CN118174788 B CN 118174788B CN 202410586099 A CN202410586099 A CN 202410586099A CN 118174788 B CN118174788 B CN 118174788B
- Authority
- CN
- China
- Prior art keywords
- fault
- signal
- data
- optical fiber
- signal transmission
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000013307 optical fiber Substances 0.000 title claims abstract description 108
- 238000001514 detection method Methods 0.000 title claims abstract description 77
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 238000012544 monitoring process Methods 0.000 claims abstract description 170
- 230000008054 signal transmission Effects 0.000 claims abstract description 127
- 238000009826 distribution Methods 0.000 claims abstract description 107
- 238000000034 method Methods 0.000 claims abstract description 44
- 230000011218 segmentation Effects 0.000 claims abstract description 16
- 239000013598 vector Substances 0.000 claims description 65
- 230000006870 function Effects 0.000 claims description 38
- 230000002159 abnormal effect Effects 0.000 claims description 37
- 238000012545 processing Methods 0.000 claims description 27
- 230000005540 biological transmission Effects 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000002457 bidirectional effect Effects 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 13
- 239000000835 fiber Substances 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 11
- 238000013499 data model Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 230000004931 aggregating effect Effects 0.000 claims description 4
- 238000013211 curve analysis Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 4
- 230000003213 activating effect Effects 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000004807 localization Effects 0.000 claims 1
- 230000008569 process Effects 0.000 description 17
- 230000008859 change Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012896 Statistical algorithm Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
- H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
- H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
- H04B10/0791—Fault location on the transmission path
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
- H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
- H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
- H04B10/0795—Performance monitoring; Measurement of transmission parameters
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Optical Communication System (AREA)
Abstract
The application relates to the technical field of fault detection and discloses a fault detection method, device and equipment of an optical fiber wiring cabinet and a storage medium. The method comprises the following steps: acquiring a basic signal transmission data set of an optical fiber distribution cabinet and constructing an hypersphere model; monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber wiring cabinet to obtain a target signal transmission data set; performing data segmentation and preliminary fault signal positioning on the target signal transmission data set to obtain initial fault signal data; performing key fault signal identification on the initial fault signal data according to the hypersphere model to obtain target fault signal data; the target fault signal data and the plurality of signal monitoring points are input into a preset signal transmission fault detection model to perform fault node positioning, so that at least one optical fiber connection fault node is obtained.
Description
Technical Field
The present application relates to the field of fault detection technologies, and in particular, to a fault detection method, apparatus, device, and storage medium for an optical fiber distribution cabinet.
Background
The optical fiber distribution cabinet is used as a key component of an optical fiber network, and the operation state and the stability of the optical fiber distribution cabinet are particularly important. However, the conventional fault detection method has the problems of insufficient adaptability to a large-scale network, low fault detection precision and difficult fault positioning.
The current optical fiber communication network has huge scale and increased complexity, and the traditional fault detection method often depends on manual inspection or simple fault alarm, so that the requirement of a large-scale network is difficult to meet. In addition, due to the complexity of the optical fiber communication system, the traditional method has certain limitation on the precision and efficiency of fault detection, and cannot effectively locate and identify complex faults, so that the network operation and maintenance efficiency is low.
Disclosure of Invention
The application provides a fault detection method, device and equipment of an optical fiber wiring cabinet and a storage medium, which are used for improving the fault detection accuracy of the optical fiber wiring cabinet.
In a first aspect, the present application provides a fault detection method for an optical fiber distribution cabinet, where the fault detection method for an optical fiber distribution cabinet includes:
acquiring a basic signal transmission data set of an optical fiber distribution cabinet and constructing an hypersphere model;
monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber wiring cabinet to obtain a target signal transmission data set;
Performing data segmentation and preliminary fault signal positioning on the target signal transmission data set to obtain initial fault signal data;
Performing key fault signal identification on the initial fault signal data according to the hypersphere model to obtain target fault signal data;
and inputting the target fault signal data and the plurality of signal monitoring points into a preset signal transmission fault detection model to locate fault nodes, so as to obtain at least one optical fiber connection fault node.
In a second aspect, the present application provides a fault detection device for an optical fiber distribution cabinet, the fault detection device for an optical fiber distribution cabinet comprising:
the acquisition module is used for acquiring a basic signal transmission data set of the optical fiber distribution cabinet and constructing an hypersphere model;
The monitoring module is used for monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber distribution cabinet to obtain a target signal transmission data set;
the segmentation module is used for carrying out data segmentation and preliminary fault signal positioning on the target signal transmission data set to obtain initial fault signal data;
the identification module is used for carrying out key fault signal identification on the initial fault signal data according to the hypersphere model to obtain target fault signal data;
And the positioning module is used for inputting the target fault signal data and the plurality of signal monitoring points into a preset signal transmission fault detection model to position the fault node so as to obtain at least one optical fiber connection fault node.
A third aspect of the present application provides a fault detection apparatus of an optical fiber distribution cabinet, comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the fault detection device of the fiber optic enclosure to perform the fault detection method of the fiber optic enclosure described above.
A fourth aspect of the present application provides a computer-readable storage medium having instructions stored therein that, when run on a computer, cause the computer to perform the above-described fault detection method of an optical fiber cabinet.
According to the technical scheme provided by the application, the signal transmission characteristics of the optical fiber distribution cabinet can be more accurately described by constructing the hyperspherical model, the fault signal identification and positioning capability is improved, the change and complexity of different optical fiber distribution cabinets can be better adapted, and the universality and expandability of the method are enhanced. By calculating the multidimensional data distribution characteristics, comprehensive understanding of data transmitted by the optical fiber distribution cabinet is increased, and the detection rate and accuracy of abnormal signals are improved. The application of the support vector data description algorithm enables the fault detection method to better identify abnormal modes in signal data, and improves the accuracy and stability of fault detection. The bidirectional gating cycle network can effectively extract hidden features of signal data, so that the fault detection method can better identify and analyze complex fault modes. The application comprehensively monitors and analyzes the signal monitoring points in the optical fiber distribution cabinet, and improves the comprehensiveness of fault detection. The dependence on manual experience is reduced, the detection efficiency and accuracy are improved, and the automation of fault detection is realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an embodiment of a fault detection method of an optical fiber distribution cabinet according to an embodiment of the present application;
fig. 2 is a schematic diagram of an embodiment of a fault detection device of an optical fiber distribution cabinet according to an embodiment of the present application.
Detailed Description
The embodiment of the application provides a fault detection method, device and equipment of an optical fiber wiring cabinet and a storage medium. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present application is described below, referring to fig. 1, and an embodiment of a fault detection method for an optical fiber distribution cabinet in an embodiment of the present application includes:
S101, acquiring a basic signal transmission data set of an optical fiber distribution cabinet and constructing an hypersphere model;
It is to be understood that the execution body of the present application may be a fault detection device of an optical fiber distribution cabinet, and may also be a terminal or a server, which is not limited herein. The embodiment of the application is described by taking a server as an execution main body as an example.
Specifically, basic data acquisition is carried out on signal transmission through a sensor system arranged in the optical fiber distribution cabinet, and the signal intensity, the transmission rate and other key indexes of the optical fiber in a normal working state are captured to obtain an original signal transmission data set. The original signal transmission data set is subjected to data formatting processing, which includes data cleaning, normalization and normalization, to convert the original data into a base signal transmission data set having a consistent format and scale, for facilitating subsequent data analysis and model construction. And (3) carrying out multidimensional data distribution characteristic calculation on the basic signal transmission data set, and extracting statistical characteristics capable of reflecting signal transmission characteristics, such as mean value, variance, skewness, kurtosis and the like, from the data set. The distribution of data in a multidimensional space is understood by multidimensional data features. And (5) creating an initial data model of the optical fiber distribution cabinet in a fault-free state by adopting a support vector data description algorithm. The support vector data description algorithm is an algorithm based on a support vector machine, is suitable for unsupervised anomaly detection, and is used for surrounding most normal data points by finding a minimum hypersphere. And (3) performing parameter optimization of the model, such as adjusting regularization parameters, kernel function types and the like, so as to ensure that the hypersphere model can reflect the signal transmission characteristics of the optical fiber distribution cabinet in a normal working state most accurately. The optimized hypersphere model is used as a reference for identifying fault signals in a monitoring system, and any data point which deviates from the hypersphere obviously can indicate potential faults.
Step S102, monitoring signal transmission characteristics of a plurality of signal monitoring points in an optical fiber distribution cabinet to obtain a target signal transmission data set;
Specifically, a monitoring function of a plurality of preset signal monitoring points in the wiring cabinet is activated, and signal transmission characteristics of the points are monitored in real time. Signal data is collected from each monitoring point, including key parameters such as signal strength, frequency, phase, etc., and a first signal transmission data set is obtained for each monitoring point. The first signal transmission data set of each monitoring point is subjected to wavelet filtering processing. Wavelet filtering is an effective signal processing technique that can remove unnecessary noise while maintaining the temporal characteristics of data. And selecting proper wavelet base and decomposition level to adapt to specific signal characteristics, so as to obtain a clearer and more accurate second signal transmission data set. In order to synchronize the data of the plurality of monitoring points and eliminate time deviation, the second signal transmission data set of each signal monitoring point is subjected to time synchronization processing, so that the data from different monitoring points are aligned on a time axis. The time synchronization can significantly improve the accuracy of fault diagnosis when analyzing signal transmission paths or locating fault sources. By adjusting the time stamp of the data and correcting the time offset, it is ensured that the data sets of all monitoring points are compared and analyzed within the same time frame. And integrating the third signal transmission data set of each monitoring point subjected to time synchronization processing to obtain a target signal transmission data set, wherein the data set comprises comprehensive signal characteristic information of the whole optical fiber distribution cabinet.
Step S103, carrying out data segmentation and preliminary fault signal positioning on a target signal transmission data set to obtain initial fault signal data;
Specifically, a polynomial function is used as a basis function for curve fitting, and the polynomial function has good adaptability and flexibility in processing signal data with complex trend. And (3) carrying out single-point fitting on each signal transmission data point in the target signal transmission data set to obtain the deviation of each data point relative to an ideal signal model, wherein the single-point fitting result reflects the real-time state of the signal at a specific transmission point. And aggregating the single-point fitting results to construct a target polynomial curve which represents the continuous change of the signal state in the whole signal transmission process. The aggregated polynomial curve can exhibit overall trends and anomalies in signal transmission. And (3) performing curve segmentation on the target polynomial curve, decomposing the continuous polynomial curve into a plurality of initial curve segments, and analyzing local characteristics and potential fault signals in the curve more finely. In order to improve the accuracy of fault location, boundary point optimization is carried out on the initial curve segments, and the starting point and the ending point of each curve segment are adjusted so as to ensure that the curve segments can better reflect the real transmission characteristics of signals and obtain a plurality of target curve segments. And respectively carrying out fault curve analysis on the plurality of target curve segments to obtain a point fault analysis result of each target curve segment. Abrupt points and discontinuities in the curve segment are detected, and these points are often indicative of the occurrence of a fault. By means of point fault analysis of each curve segment, detailed fault indexes, such as fault positions, fault sizes and influences of fault positions and fault sizes on overall signal transmission, are obtained. And according to the result of the point fault analysis, carrying out preliminary fault signal positioning and fault data fusion on the plurality of target curve segments. The failure analysis results of all target curve segments are integrated to obtain initial failure signal data, and the data provides detailed information about potential failures of the optical fiber distribution cabinet.
Step S104, carrying out key fault signal identification on the initial fault signal data according to the hypersphere model to obtain target fault signal data;
Specifically, the coordinates of the center point of the hypersphere model are obtained, and the coordinates are key parameters in model calculation and represent ideal states of optical fiber transmission data in a fault-free state. And calculating the Euclidean distance from each abnormal point of the transmission data to the central point in the preliminary fault signal data. The distance value reflects the degree of deviation of each data point from ideal, a larger distance value may indicate a more serious signal anomaly or fault. In order to analyze the deviation degree, a Gaussian kernel function is used for converting the initial distance value of each transmission data abnormal point. The gaussian kernel function is a common method for increasing the separability of data points in a feature space, and the feature distance value obtained after conversion can more obviously reflect the abnormal properties of abnormal points. And comparing the characteristic distance value of each abnormal point of the transmission data with a threshold value, wherein the threshold value is preset according to the operation characteristics of the hypersphere model and the optical fiber distribution cabinet. Data points where the feature distance value exceeds the target threshold are considered target data outliers, which represent potential faults or significant signal deviations. The target data outliers are ranked and classified, so that the severity degree of each outlier is helped to be understood, and subsequent fault processing and maintenance work are facilitated. The ranking may be based on the magnitude of the feature distance values and the categorization may then be based on the type of anomaly or the extent of the effect. And the target fault signal data are obtained through comprehensive processing, so that the stable operation and the service reliability of the optical fiber distribution cabinet are ensured.
And step 105, inputting the target fault signal data and the plurality of signal monitoring points into a preset signal transmission fault detection model to locate the fault node, and obtaining at least one optical fiber connection fault node.
Specifically, node feature mapping is performed on the target fault signal data, and key fault feature values, such as abnormal amplitude, frequency offset, abnormal phase and the like of signals, are extracted from fault data collected by each signal monitoring point, wherein the fault signal feature values reflect the deviation of the current state and the normal state of the signal monitoring point. In order to facilitate data processing and model input, the fault signal feature values are converted into a mathematically processable format, vector conversion is performed, and normalization processing is performed, so that comparison and calculation between different features are ensured to be performed on the same scale. The normalization process helps the model to better understand and process the data and avoid computational bias due to eigenvalue scale differences. And inputting fault signal feature vectors of each signal monitoring point into a preset signal transmission fault detection model, wherein the model comprises a bidirectional gating circulation network, a full connection layer and a probability density function. The bidirectional gating circulation network is an efficient recurrent neural network, and can extract forward and reverse information of time sequence data through the forward and backward gating circulation network respectively, capture forward and backward dependency relationship of signals, so that dynamic changes of the signals are more comprehensively understood. Through the network structure, fault signal feature vectors of each signal monitoring point are processed through the forward and backward gating circulation networks respectively, and respective hidden feature vectors are extracted. These hidden feature vectors represent the signal features observed from the forward and reverse directions of the time series, respectively, and then vector-stitching the two hidden feature vectors to form a comprehensive hidden feature vector. And performing feature conversion on the spliced hidden feature vector through the full connection layer, and mapping the extracted high-dimensional feature to a new feature space so as to facilitate subsequent classification or regression tasks. And performing fault probability calculation on the target hidden characteristic data of each signal monitoring point through a probability density function. The probability density function calculates the probability of occurrence of faults of each monitoring point based on a statistical method, and the probability reflects the probability of occurrence of faults of each point. And finally, positioning fault nodes of the system for the plurality of signal monitoring points according to the fault probability distribution data, and identifying one or more optical fiber connection fault nodes.
And calculating the fault posterior probability of each monitoring point through a Bayesian classifier. The bayesian classifier updates the probability of occurrence of an event based on prior knowledge and actual observation data, and the posterior probability calculation updates the probability of possible occurrence of a fault according to the fault probability distribution data of each point. And acquiring specific position distribution information of the signal monitoring points in the optical fiber distribution cabinet, and acquiring the relative position and potential network connection of each monitoring point in the whole distribution cabinet system. And creating a corresponding initial node network according to the position distribution information, and simulating physical connection and signal transmission paths in the optical fiber distribution cabinet. And carrying out clustering calculation on the initial node network based on the fault posterior probability of each signal monitoring point. The purpose of cluster computation is to generalize the probability of failure similar monitoring points into a set, which helps identify areas or paths that may be commonly affected by a failure of some sort. And calculating an influence index of each signal monitoring point according to the clustering result, wherein the influence index quantifies the contribution and importance of each point to the fault state of the whole wiring closet, and the monitoring points with high influence indexes can be key nodes or high risk areas of fault propagation. And positioning fault nodes of the plurality of signal monitoring points according to the influence indexes. The posterior probability of the faults of the single monitoring point is considered, the position relation of the single monitoring point in the wiring cabinet and the role of the single monitoring point in the fault network are integrated, and at least one optical fiber connection fault node can be accurately positioned.
In the embodiment of the application, the signal transmission characteristic of the optical fiber wiring cabinet can be more accurately described by constructing the hypersphere model, the identification and positioning capability of fault signals is improved, the change and complexity of different optical fiber wiring cabinets can be better adapted, and the universality and expandability of the method are enhanced. By calculating the multidimensional data distribution characteristics, comprehensive understanding of data transmitted by the optical fiber distribution cabinet is increased, and the detection rate and accuracy of abnormal signals are improved. The application of the support vector data description algorithm enables the fault detection method to better identify abnormal modes in signal data, and improves the accuracy and stability of fault detection. The bidirectional gating cycle network can effectively extract hidden features of signal data, so that the fault detection method can better identify and analyze complex fault modes. The application comprehensively monitors and analyzes the signal monitoring points in the optical fiber distribution cabinet, and improves the comprehensiveness of fault detection. The dependence on manual experience is reduced, the detection efficiency and accuracy are improved, and the automation of fault detection is realized.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) The method comprises the steps that a sensor is used for carrying out basic signal transmission acquisition on an optical fiber distribution cabinet to obtain an original signal transmission data set;
(2) Carrying out data formatting processing on the original signal transmission data set to obtain a basic signal transmission data set;
(3) Performing multidimensional data distribution feature calculation on the basic signal transmission data set to obtain a multidimensional data distribution feature set;
(4) And establishing an initial data model of the optical fiber wiring cabinet in a fault-free state according to the multidimensional data distribution characteristic set by adopting a support vector data description algorithm, and carrying out parameter optimization on the initial data model to obtain an hyperspherical model.
Specifically, the sensor is used for carrying out basic signal transmission and collection on the optical fiber wiring cabinet, the sensor is deployed at a key position of the wiring cabinet, and data such as the intensity, the frequency, the spectral distribution and the like of optical fiber signals are monitored and recorded in real time to form an original signal transmission data set. And carrying out data formatting processing on the original signal transmission data set to obtain a basic signal transmission data set. The formatting process mainly comprises data cleaning, such as filtering abnormal values caused by sensor errors or environmental interference, and data standardization or normalization, so that all data are ensured to be in the same magnitude, and comparison and further analysis are facilitated. And carrying out multidimensional data distribution characteristic calculation on the basic signal transmission data set. Including computing statistical features for each data point, such as mean, variance, bias, kurtosis, etc., as well as more complex features, such as time series analysis results and frequency domain features of the signal. For example, a particular frequency component in the signal is detected by frequency domain analysis, which may be indicative of a typical failure mode in the communication, such as signal attenuation or frequency drift. And establishing an initial data model of the optical fiber distribution cabinet in a fault-free state according to the multidimensional data distribution characteristic set by adopting a support vector data description algorithm. The support vector data description algorithm is an algorithm based on a support vector machine, is suitable for pattern recognition and anomaly detection, and particularly shows good effects when the number of data points is relatively small or the data dimension is high. The algorithm excludes outliers by finding the smallest closed sphere (hypersphere) that encompasses as much as possible all normal data points. And (3) carrying out parameter optimization on the initial data model, and ensuring the accuracy and the robustness of the model. Parameter optimization may involve adjusting the selection of kernel functions (e.g., linear kernel, polynomial kernel, or radial basis function kernel), and adjusting penalty parameters C, which determine the sensitivity of the model to outliers. The optimization aims to find an optimal hypersphere, and the hypersphere should reflect the signal transmission characteristics of the optical fiber distribution cabinet in a fault-free state as accurately as possible, and have higher detection capability on new and unseen abnormal signals. For example, if the intensity of the fiber signal continuously monitored by the sensor suddenly drops in a large data center, and the change is confirmed as an abnormal point after formatting and feature calculation, the abnormal point is analyzed by a support vector data description algorithm model, and if the abnormal point is located outside the hypersphere, possible fault reasons such as fiber breakage or connector damage can be rapidly diagnosed.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Activating a signal monitoring function of the signal monitoring points, and monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber distribution cabinet to obtain a first signal transmission data set of each signal monitoring point;
(2) Performing wavelet filtering processing on the first signal transmission data set of each signal monitoring point to obtain a second signal transmission data set of each signal monitoring point;
(3) Respectively carrying out time synchronization processing on the second signal transmission data set of each signal monitoring point to obtain a third signal transmission data set of each signal monitoring point;
(4) And integrating the data of the third signal transmission data set of each signal monitoring point to obtain a target signal transmission data set.
Specifically, the signal monitoring function of the signal monitoring points is activated, the built-in sensors of the monitoring points and related monitoring equipment are started, and signals transmitted from the optical fibers are captured. These sensors are typically deployed at critical nodes in the wiring closet and are capable of monitoring such important transmission characteristics as signal strength, frequency, phase and others to obtain a first signal transmission data set for each signal monitoring point. The first signal transmission data set of each signal monitoring point is subjected to wavelet filtering processing. Wavelet filtering is an effective technique for removing noise while maintaining the temporal characteristics of data. Wavelet transforms enable decomposition of signals at multiple scales, removing short-lived interference or persistent background noise from the original signal by selecting appropriate wavelet functions and decomposition levels. For example, if the monitored signal is subjected to some periodic disturbance, the disturbance components may be selectively attenuated or removed by wavelet filtering to obtain a second signal transmission data set for each signal monitoring point. And respectively carrying out time synchronization processing on the second signal transmission data set of each signal monitoring point, so as to ensure that the signal data from different monitoring points can be compared in the same time frame during analysis. The unsynchronized data may lead to analysis errors, especially in the case of fault diagnosis of the signal transmission path, due to small deviations in the operating clock of the sensor or delays that may occur in the transmission of the signal. The time synchronization process typically involves adjusting the time stamp of the data, unifying the clocks of the different devices using techniques such as network time protocol, ensuring that all data points correspond to the exact acquisition time, forming a third signal transmission data set for each signal monitoring point. And integrating the data of the third signal transmission data set subjected to time synchronization to obtain a comprehensive target signal transmission data set. For example, by integrating the data, identifying the transmission path of the signal throughout the wiring closet, detecting any possible signal attenuation or loss anomalies, further analysis may also indicate the specific location and nature of the fault potential.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Selecting a polynomial function as a basis function of curve fitting, and performing single-point fitting on each signal transmission data point in a target signal transmission data set through the basis function to obtain a single-point fitting result of each signal transmission data point;
(2) Aggregating the single-point fitting result of each signal transmission data point to obtain a target polynomial curve;
(3) Performing curve segmentation on the target polynomial curve to obtain a plurality of initial curve segments;
(4) Performing boundary point optimization on the initial curve segments to obtain target curve segments;
(5) Performing fault curve analysis on the plurality of target curve segments respectively to obtain a point fault analysis result of each target curve segment;
(6) And according to the point fault analysis result, performing primary fault signal positioning and fault data fusion on the plurality of target curve segments to obtain initial fault signal data.
Specifically, a polynomial function is selected as a basis function of curve fitting, single-point fitting is performed on each signal transmission data point in the target signal transmission data set through the basis function, and key trends and modes in the signals are extracted while original characteristics of the data are reserved. Each data point in the target signaling dataset is taken as an input, which is fitted using a polynomial function. Fitting typically uses a least squares method to find the polynomial equation that best represents the trend of the data points. For example, if the data points exhibit a trend that increases linearly with time, the first order polynomial may simulate such trend sufficiently well; if the data points show more complex fluctuations, a polynomial of degree two or more may be required to accurately fit these changes. And aggregating the single-point fitting result of each signal transmission data point to obtain a target polynomial curve. And integrating each fitting result to ensure the continuity and smoothness of the curve. The target polynomial curve is a comprehensive representation of the signal change across the data set and reveals the general trend and potential anomaly regions in the signal transmission process. Based on slope change, inflection point or other remarkable characteristics of the curve, the target polynomial curve is subjected to curve segmentation, and the continuous curve is decomposed into a plurality of initial curve segments according to different characteristics and change trends. Each curve segment represents the behavior of the signal within a particular interval. And (3) optimizing boundary points of the initial curve segments, ensuring that the definition of each curve segment is more accurate, and reflecting the real signal change. Boundary point optimization generally involves adjusting the start and end points of curve segments so that each segment fits more over the actual signal variations, reducing the loss of information or errors due to segmentation. Fault curve analysis is performed on each target curve segment, and potential fault or abnormal points are searched in each curve segment, wherein the points can be represented as abrupt signal changes or discontinuous signal changes. Tomographic analysis can indicate which regions of the signal exhibit deviations from the expected or normal range. And (5) according to the fault analysis result, carrying out primary fault signal positioning and fault data fusion. The fault analysis results of all target curve segments are integrated to form initial fault signal data which reflect the specific location of the potential fault and also indicate the nature and severity of the fault.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
(1) Acquiring the center point coordinates of the hypersphere model;
(2) Respectively calculating Euclidean distances from a plurality of transmission data abnormal points in the preliminary fault signal data to the coordinates of the central point to obtain an initial distance value of each transmission data abnormal point;
(3) Converting the initial distance value of each transmission data abnormal point through a Gaussian kernel function to obtain a characteristic distance value of each transmission data abnormal point;
(4) Threshold value comparison is carried out on the characteristic distance value of each transmission data abnormal point, and the characteristic distance value exceeds a target threshold value to serve as a target data abnormal point;
(5) And sequencing and classifying the abnormal points of the target data to obtain the target fault signal data.
Specifically, the center point coordinates of the hypersphere model are obtained. The center point coordinates represent ideal conditions in the data under normal operating conditions, and are the results obtained by using the normal operating data when training the model. The optimal center position and radius is determined by minimizing the volume of the hypersphere containing all the training data points. Euclidean distances from a plurality of transmission data abnormal points in the preliminary fault signal data to the coordinates of the central point are calculated respectively. The Euclidean distance can quantify the degree of deviation of each data point from the center point, with greater distances implying greater likelihood of anomalies. And converting the initial distance value of each transmission data abnormal point by adopting a Gaussian kernel function. Gaussian kernel functions are a common type of nonlinear mapping used to increase the separability of data in feature space. By this conversion, the initial Euclidean distance is mapped to a new metric space, where the distance value of the outlier is further amplified, making the distinction between normal and outliers more apparent. The characteristic distance value of each transmission data outlier is compared with a predetermined target threshold. The threshold is determined based on the system's specific needs and historical data analysis and is used to distinguish between normal and potentially faulty conditions. Points of the feature distance value exceeding this threshold are marked as target data outliers. And sorting and classifying the abnormal points of the target data. The ranking is typically based on the magnitude of the feature distance values and may reveal which faults are more severe and which may be only slight deviations. Classification is based on characteristics of the anomaly, such as frequency, intensity, or time of occurrence, which helps the system understand the nature and possible cause of the fault, and thus takes corresponding maintenance or repair actions. For example, assume that in a routine inspection, a signal at a certain monitoring point is detected to suddenly increase in distance from the center point, and after the gaussian kernel conversion, the distance at this point far exceeds a set threshold. The system classifies this as a serious failure and the technician then inspects the section, possibly finding that the fiber optic connection is loose or fiber optic damage causes signal attenuation. Through timely feedback and accurate positioning, the fault can be quickly repaired, the normal operation of the system is recovered, and possible communication interruption or data loss is effectively avoided.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
(1) Node characteristic mapping is carried out on the target fault signal data, and a plurality of fault signal characteristic values of each signal monitoring point are obtained;
(2) Vector conversion and normalization processing are respectively carried out on a plurality of fault signal characteristic values of each signal monitoring point, so that fault signal characteristic vectors of each signal monitoring point are obtained;
(3) Inputting fault signal feature vectors of each signal monitoring point into a preset signal transmission fault detection model, wherein the signal transmission fault detection model comprises: a bi-directional gated loop network, a full connection layer, and a probability density function;
(4) Carrying out hidden feature extraction on fault signal feature vectors of each signal monitoring point through a forward gating loop network in the bidirectional gating loop network to obtain forward hidden feature vectors of each signal monitoring point;
(5) Carrying out hidden feature extraction on fault signal feature vectors of each signal monitoring point through a backward gating loop network in the bidirectional gating loop network to obtain backward hidden feature vectors of each signal monitoring point;
(6) Vector stitching is carried out on the forward hidden feature vector and the backward hidden feature vector to obtain stitched hidden feature vectors of each signal monitoring point, feature conversion is carried out on the stitched hidden feature vectors through a full connection layer, and target hidden feature data of each signal monitoring point are obtained;
(7) Performing fault probability calculation on the target hidden characteristic data of each signal monitoring point through a probability density function to obtain fault probability distribution data of each signal monitoring point;
(8) And positioning fault nodes of the plurality of signal monitoring points according to the fault probability distribution data of each signal monitoring point to obtain at least one optical fiber connection fault node.
Specifically, node feature mapping is performed on the target fault signal data to obtain a plurality of fault signal feature values of each signal monitoring point, such as signal strength change, frequency offset, waveform deformation and the like. And carrying out vector conversion on fault signal characteristic values of each signal monitoring point, encoding the fault characteristic values into a series of numerical vectors which can be processed by an algorithm, and converting original possibly unstructured data into a structured form so as to facilitate subsequent calculation processing. And carrying out normalization processing on the feature vectors, and avoiding deviation caused by the magnitude difference of feature dimensions. The normalization process typically involves scaling the data to a fixed interval, such as between 0 and 1, or making the data have a 0 mean and unit variance. The normalized fault signal feature vector is input into a preset signal transmission fault detection model, and the model is a composite neural network model comprising a bidirectional gating loop network, a full connection layer and a probability density function. A two-way gated loop network is a neural network suitable for processing sequence data, capable of extracting time-dependent features from the data. In the model, fault signal feature vectors of each signal monitoring point are processed through a forward gating loop network of a bidirectional gating loop network, and time forward dependence of signal features is extracted to obtain forward hidden feature vectors. Meanwhile, the backward gating loop network of the bidirectional gating loop network captures the backward dependent characteristics of time and generates a backward hidden characteristic vector. The forward and backward hidden feature vectors are spliced to form a comprehensive hidden feature vector containing the omni-directional features of the signal data over time. And the feature conversion is carried out on the spliced hidden feature vector through the full connection layer, so that the nonlinear expression capacity of the model is enhanced, and faults can be more accurately classified or predicted. And respectively carrying out fault probability calculation on the target hidden characteristic data of each signal monitoring point through a probability density function. The data reflect the possibility of faults of each monitoring point and are key bases for locating fault nodes. According to the fault probability distribution data, determining which monitoring points have abnormal indexes exceeding a normal range through statistical analysis and algorithm processing, so as to position at least one optical fiber connection fault node. For example, assume that the probability of failure at a point in a set of monitoring points suddenly rises, indicating that the point may have failed. By comparing the fault probability and historical data of the point with other monitoring points, whether the point is an isolated event or the problem of the whole network is judged.
In a specific embodiment, the performing step locates the fault node of the plurality of signal monitoring points according to the fault probability distribution data of each signal monitoring point, and the process of obtaining at least one optical fiber connection fault node may specifically include the following steps:
(1) Performing posterior probability calculation on the fault probability distribution data of each signal monitoring point through a Bayesian classifier to obtain the fault posterior probability of each signal monitoring point;
(2) Acquiring position distribution information of a plurality of signal monitoring points in an optical fiber distribution cabinet, and creating a corresponding initial node network according to the position distribution information;
(3) Clustering calculation is carried out on the initial node network according to the fault posterior probability of each signal monitoring point to obtain a clustering result, and the influence index of each signal monitoring point is calculated;
(4) And positioning the fault nodes of the plurality of signal monitoring points according to the influence indexes to obtain at least one optical fiber connection fault node.
Specifically, the posterior probability calculation is carried out on the fault probability distribution data of each signal monitoring point through a Bayesian classifier, and the Bayesian classifier updates the posterior probability of each monitoring point in fault according to priori knowledge and actually collected data. The posterior probability is a conditional probability reflecting the occurrence probability of an event based on specific input data. For example, in a fiber optic network, if a monitoring point frequently fails in the past, in the event of a similar signal anomaly, the posterior probability of the point failing again increases accordingly. And acquiring position distribution information of a plurality of signal monitoring points in the optical fiber distribution cabinet. The location information is one of the key factors for effective fault detection and helps determine potential fault propagation paths and impact ranges. For example, the physical layout of the monitoring points may exhibit a particular topology, such as a star, ring, or grid, with such structural information helping to understand how the fault propagates in the network. An initial node network is constructed by the location information, which reflects the relative locations of all monitoring points in space and the possible connection relationships between them. And clustering calculation is carried out on the initial node network by utilizing the fault posterior probability of each signal monitoring point. The monitoring points with similar fault probabilities are grouped into a group, so that the area with concentrated faults can be identified, and possible fault sources can be revealed. Cluster analysis is derived by various algorithms such as K-means or hierarchical clustering. Each cluster group represents a collection of monitoring points with similar fault behavior. And calculating an influence index of each monitoring point, wherein the index quantifies the importance of each point in the overall fault diagnosis based on the posterior probability of the fault of the monitoring point and the position relation of the fault of the monitoring point in the network. And positioning fault nodes of the plurality of signal monitoring points according to the influence indexes. The probability of failure of each monitoring point and their positional relationship in the network are analyzed to locate at least one or more fiber optic connection failure nodes. For example, if the posterior probability of failure of multiple monitoring points within a particular area suddenly increases and they are closely connected in the network, it can be inferred that a major source of failure is likely to exist in that area.
The method for detecting a fault in an optical fiber distribution cabinet according to the embodiment of the present application is described above, and the device for detecting a fault in an optical fiber distribution cabinet according to the embodiment of the present application is described below, referring to fig. 2, where an embodiment of the device for detecting a fault in an optical fiber distribution cabinet according to the embodiment of the present application includes:
an acquisition module 201, configured to acquire a basic signal transmission data set of the optical fiber distribution cabinet and construct an hypersphere model;
the monitoring module 202 is configured to monitor signal transmission characteristics of a plurality of signal monitoring points in the optical fiber distribution cabinet, so as to obtain a target signal transmission data set;
The segmentation module 203 is configured to perform data segmentation and preliminary fault signal positioning on the target signal transmission data set to obtain initial fault signal data;
the identification module 204 is configured to identify key fault signals of the initial fault signal data according to the hypersphere model, so as to obtain target fault signal data;
And the positioning module 205 is configured to input the target fault signal data and the plurality of signal monitoring points into a preset signal transmission fault detection model to perform fault node positioning, so as to obtain at least one optical fiber connection fault node.
Through the cooperation of the components, the signal transmission characteristics of the optical fiber distribution cabinet can be more accurately described by constructing an hyperspherical model, the fault signal identification and positioning capability is improved, the method can be better adapted to the changes and the complexity of different optical fiber distribution cabinets, and the universality and the expandability of the method are enhanced. By calculating the multidimensional data distribution characteristics, comprehensive understanding of data transmitted by the optical fiber distribution cabinet is increased, and the detection rate and accuracy of abnormal signals are improved. The application of the support vector data description algorithm enables the fault detection method to better identify abnormal modes in signal data, and improves the accuracy and stability of fault detection. The bidirectional gating cycle network can effectively extract hidden features of signal data, so that the fault detection method can better identify and analyze complex fault modes. The application comprehensively monitors and analyzes the signal monitoring points in the optical fiber distribution cabinet, and improves the comprehensiveness of fault detection. The dependence on manual experience is reduced, the detection efficiency and accuracy are improved, and the automation of fault detection is realized.
The application also provides fault detection equipment of the optical fiber wiring cabinet, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the processor executes the steps of the fault detection method of the optical fiber wiring cabinet in the above embodiments.
The present application also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, where the instructions when executed on a computer cause the computer to perform the steps of the fault detection method of the optical fiber distribution cabinet.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (8)
1. The fault detection method of the optical fiber wiring cabinet is characterized by comprising the following steps of:
Acquiring a basic signal transmission data set of an optical fiber distribution cabinet and constructing an hypersphere model; the method specifically comprises the following steps: the method comprises the steps that a sensor is used for carrying out basic signal transmission acquisition on an optical fiber distribution cabinet to obtain an original signal transmission data set; carrying out data formatting processing on the original signal transmission data set to obtain a basic signal transmission data set; performing multidimensional data distribution feature calculation on the basic signal transmission data set to obtain a multidimensional data distribution feature set; adopting a support vector data description algorithm to create an initial data model of the optical fiber distribution cabinet in a fault-free state according to the multidimensional data distribution feature set, and carrying out parameter optimization on the initial data model to obtain a hyperspherical model;
monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber wiring cabinet to obtain a target signal transmission data set;
Performing data segmentation and preliminary fault signal positioning on the target signal transmission data set to obtain initial fault signal data;
Performing key fault signal identification on the initial fault signal data according to the hypersphere model to obtain target fault signal data;
Inputting the target fault signal data and the plurality of signal monitoring points into a preset signal transmission fault detection model to locate fault nodes, so as to obtain at least one optical fiber connection fault node; the method specifically comprises the following steps: performing node characteristic mapping on the target fault signal data to obtain a plurality of fault signal characteristic values of each signal monitoring point; vector conversion and normalization processing are respectively carried out on a plurality of fault signal characteristic values of each signal monitoring point, so that fault signal characteristic vectors of each signal monitoring point are obtained; inputting fault signal feature vectors of each signal monitoring point into a preset signal transmission fault detection model, wherein the signal transmission fault detection model comprises the following components: a bi-directional gated loop network, a full connection layer, and a probability density function; carrying out hidden feature extraction on fault signal feature vectors of each signal monitoring point through a forward gating loop network in the bidirectional gating loop network to obtain forward hidden feature vectors of each signal monitoring point; carrying out hidden feature extraction on fault signal feature vectors of each signal monitoring point through a backward gating loop network in the bidirectional gating loop network to obtain backward hidden feature vectors of each signal monitoring point; vector stitching is carried out on the forward hidden feature vector and the backward hidden feature vector to obtain stitched hidden feature vectors of each signal monitoring point, and feature conversion is carried out on the stitched hidden feature vectors through the full connection layer to obtain target hidden feature data of each signal monitoring point; performing fault probability calculation on the target hidden characteristic data of each signal monitoring point through the probability density function to obtain fault probability distribution data of each signal monitoring point; and positioning the fault nodes of the plurality of signal monitoring points according to the fault probability distribution data of each signal monitoring point to obtain at least one optical fiber connection fault node.
2. The method for detecting a fault in an optical fiber distribution cabinet according to claim 1, wherein the monitoring signal transmission characteristics of the plurality of signal monitoring points in the optical fiber distribution cabinet to obtain a target signal transmission data set comprises:
Activating a signal monitoring function of the signal monitoring points, and monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber distribution cabinet to obtain a first signal transmission data set of each signal monitoring point;
performing wavelet filtering processing on the first signal transmission data set of each signal monitoring point to obtain a second signal transmission data set of each signal monitoring point;
Respectively carrying out time synchronization processing on the second signal transmission data set of each signal monitoring point to obtain a third signal transmission data set of each signal monitoring point;
and integrating the data of the third signal transmission data set of each signal monitoring point to obtain a target signal transmission data set.
3. The method for detecting a fault in an optical fiber distribution cabinet according to claim 1, wherein the performing data segmentation and preliminary fault signal localization on the target signal transmission data set to obtain initial fault signal data includes:
selecting a polynomial function as a basis function of curve fitting, and performing single-point fitting on each signal transmission data point in the target signal transmission data set through the basis function to obtain a single-point fitting result of each signal transmission data point;
Aggregating the single-point fitting result of each signal transmission data point to obtain a target polynomial curve;
performing curve segmentation on the target polynomial curve to obtain a plurality of initial curve segments;
performing boundary point optimization on the initial curve segments to obtain target curve segments;
Performing fault curve analysis on the plurality of target curve segments respectively to obtain a point fault analysis result of each target curve segment;
and according to the point fault analysis result, performing preliminary fault signal positioning and fault data fusion on the plurality of target curve segments to obtain initial fault signal data.
4. The fault detection method of an optical fiber distribution cabinet according to claim 1, wherein the performing key fault signal recognition on the initial fault signal data according to the hypersphere model to obtain target fault signal data includes:
acquiring the center point coordinates of the hypersphere model;
respectively calculating Euclidean distances from a plurality of transmission data abnormal points in the initial fault signal data to the coordinates of the central point to obtain an initial distance value of each transmission data abnormal point;
converting the initial distance value of each transmission data abnormal point through a Gaussian kernel function to obtain a characteristic distance value of each transmission data abnormal point;
Threshold value comparison is carried out on the characteristic distance value of each transmission data abnormal point, and the characteristic distance value exceeds a target threshold value to serve as a target data abnormal point;
And sequencing and classifying the target data abnormal points to obtain target fault signal data.
5. The method for detecting a fault in an optical fiber distribution cabinet according to claim 1, wherein the positioning the fault node of the plurality of signal monitoring points according to the fault probability distribution data of each signal monitoring point to obtain at least one optical fiber connection fault node comprises:
Performing posterior probability calculation on the fault probability distribution data of each signal monitoring point through a Bayesian classifier to obtain the fault posterior probability of each signal monitoring point;
acquiring position distribution information of the plurality of signal monitoring points in the optical fiber distribution cabinet, and creating a corresponding initial node network according to the position distribution information;
Clustering calculation is carried out on the initial node network according to the fault posterior probability of each signal monitoring point to obtain a clustering result, and the influence index of each signal monitoring point is calculated;
and positioning the fault node of the plurality of signal monitoring points according to the influence indexes to obtain at least one optical fiber connection fault node.
6. A fault detection device for an optical fiber distribution cabinet, characterized in that the fault detection device for an optical fiber distribution cabinet comprises:
the acquisition module is used for acquiring a basic signal transmission data set of the optical fiber distribution cabinet and constructing an hypersphere model; the method specifically comprises the following steps: the method comprises the steps that a sensor is used for carrying out basic signal transmission acquisition on an optical fiber distribution cabinet to obtain an original signal transmission data set; carrying out data formatting processing on the original signal transmission data set to obtain a basic signal transmission data set; performing multidimensional data distribution feature calculation on the basic signal transmission data set to obtain a multidimensional data distribution feature set; adopting a support vector data description algorithm to create an initial data model of the optical fiber distribution cabinet in a fault-free state according to the multidimensional data distribution feature set, and carrying out parameter optimization on the initial data model to obtain a hyperspherical model;
The monitoring module is used for monitoring signal transmission characteristics of a plurality of signal monitoring points in the optical fiber distribution cabinet to obtain a target signal transmission data set;
the segmentation module is used for carrying out data segmentation and preliminary fault signal positioning on the target signal transmission data set to obtain initial fault signal data;
the identification module is used for carrying out key fault signal identification on the initial fault signal data according to the hypersphere model to obtain target fault signal data;
The positioning module is used for inputting the target fault signal data and the plurality of signal monitoring points into a preset signal transmission fault detection model to position fault nodes so as to obtain at least one optical fiber connection fault node; the method specifically comprises the following steps: performing node characteristic mapping on the target fault signal data to obtain a plurality of fault signal characteristic values of each signal monitoring point; vector conversion and normalization processing are respectively carried out on a plurality of fault signal characteristic values of each signal monitoring point, so that fault signal characteristic vectors of each signal monitoring point are obtained; inputting fault signal feature vectors of each signal monitoring point into a preset signal transmission fault detection model, wherein the signal transmission fault detection model comprises the following components: a bi-directional gated loop network, a full connection layer, and a probability density function; carrying out hidden feature extraction on fault signal feature vectors of each signal monitoring point through a forward gating loop network in the bidirectional gating loop network to obtain forward hidden feature vectors of each signal monitoring point; carrying out hidden feature extraction on fault signal feature vectors of each signal monitoring point through a backward gating loop network in the bidirectional gating loop network to obtain backward hidden feature vectors of each signal monitoring point; vector stitching is carried out on the forward hidden feature vector and the backward hidden feature vector to obtain stitched hidden feature vectors of each signal monitoring point, and feature conversion is carried out on the stitched hidden feature vectors through the full connection layer to obtain target hidden feature data of each signal monitoring point; performing fault probability calculation on the target hidden characteristic data of each signal monitoring point through the probability density function to obtain fault probability distribution data of each signal monitoring point; and positioning the fault nodes of the plurality of signal monitoring points according to the fault probability distribution data of each signal monitoring point to obtain at least one optical fiber connection fault node.
7.A fault detection device for an optical fiber distribution cabinet, characterized in that the fault detection device for an optical fiber distribution cabinet comprises: a memory and at least one processor, the memory having instructions stored therein;
The at least one processor invokes the instructions in the memory to cause the fault detection device of the fiber optic enclosure to perform the fault detection method of the fiber optic enclosure of any of claims 1-5.
8. A computer readable storage medium having instructions stored thereon, wherein the instructions when executed by a processor implement the method of fault detection for a fiber optic distribution cabinet of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410586099.1A CN118174788B (en) | 2024-05-13 | 2024-05-13 | Fault detection method, device and equipment of optical fiber wiring cabinet and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410586099.1A CN118174788B (en) | 2024-05-13 | 2024-05-13 | Fault detection method, device and equipment of optical fiber wiring cabinet and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118174788A CN118174788A (en) | 2024-06-11 |
| CN118174788B true CN118174788B (en) | 2024-07-26 |
Family
ID=91360638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410586099.1A Active CN118174788B (en) | 2024-05-13 | 2024-05-13 | Fault detection method, device and equipment of optical fiber wiring cabinet and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118174788B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118590141A (en) * | 2024-08-05 | 2024-09-03 | 杭州奥克光电设备有限公司 | Internet of things operation and maintenance method and system based on intelligent fiber melting disc and intelligent fiber melting disc |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103822793A (en) * | 2014-01-20 | 2014-05-28 | 北京邮电大学 | Complicated equipment acoustic fault recognition and location method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113342476B (en) * | 2021-07-05 | 2023-04-07 | 中山大学 | Equipment cluster cross-domain fault prediction method and system based on transfer learning |
| CN115526234A (en) * | 2022-08-30 | 2022-12-27 | 长沙理工大学 | Cross-domain model training and log anomaly detection method and device based on transfer learning |
| CN115981280A (en) * | 2022-12-21 | 2023-04-18 | 浙江大学 | Complex equipment fault diagnosis system based on deep learning |
| CN117493841A (en) * | 2023-09-15 | 2024-02-02 | 国网湖北省电力有限公司信息通信公司 | Power distribution network line fault prediction method based on BiGRU-Attention model |
| CN117763432A (en) * | 2023-11-15 | 2024-03-26 | 成都态坦测试科技有限公司 | Communication fault prediction method, device, equipment and storage medium |
-
2024
- 2024-05-13 CN CN202410586099.1A patent/CN118174788B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103822793A (en) * | 2014-01-20 | 2014-05-28 | 北京邮电大学 | Complicated equipment acoustic fault recognition and location method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118174788A (en) | 2024-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220283576A1 (en) | Automatic diagnosis method, system and storage medium for equipment | |
| CN111902781B (en) | Apparatus and method for controlling a system | |
| US11243524B2 (en) | System and method for unsupervised root cause analysis of machine failures | |
| US20210397501A1 (en) | System and method for unsupervised prediction of machine failures | |
| CN111459700B (en) | Equipment fault diagnosis method, diagnosis device, diagnosis equipment and storage medium | |
| CN118174788B (en) | Fault detection method, device and equipment of optical fiber wiring cabinet and storage medium | |
| US10373065B2 (en) | Generating database cluster health alerts using machine learning | |
| CN111242323B (en) | Active automatic system and method for repairing suboptimal operation of a machine | |
| US20150219530A1 (en) | Systems and methods for event detection and diagnosis | |
| US20120041575A1 (en) | Anomaly Detection Method and Anomaly Detection System | |
| US11657121B2 (en) | Abnormality detection device, abnormality detection method and computer readable medium | |
| US20070239629A1 (en) | Cluster Trending Method for Abnormal Events Detection | |
| Lim et al. | Identifying recurrent and unknown performance issues | |
| US11733688B2 (en) | System and method for recognizing and forecasting anomalous sensory behavioral patterns of a machine | |
| Michau et al. | Fleet PHM for critical systems: bi-level deep learning approach for fault detection | |
| CN111027615A (en) | Middleware fault early warning method and system based on machine learning | |
| CN115454778A (en) | Intelligent monitoring system for abnormal time sequence indexes in large-scale cloud network environment | |
| US20190265088A1 (en) | System analysis method, system analysis apparatus, and program | |
| CN117131110B (en) | Method and system for monitoring dielectric loss of capacitive equipment based on correlation analysis | |
| US11928565B2 (en) | Automated model building and updating environment | |
| Atzmueller et al. | Anomaly detection and structural analysis in industrial production environments | |
| Hubauer et al. | Analysis of data quality issues in real-world industrial data | |
| WO2017127260A1 (en) | System and method for allocating machine behavioral models | |
| Duan et al. | Dynamic causal modeling for nonstationary industrial process performance degradation analysis and fault prognosis | |
| Hu et al. | TS-InvarNet: Anomaly detection and localization based on tempo-spatial KPI invariants in distributed services |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |