A fiber-optic passive voltage and current fault diagnosis method, device, equipment and medium

Abstract

The present invention discloses a fiber-optic passive voltage and current fault diagnosis method, device, equipment and medium, relating to the technical field of fault diagnosis, and comprising the following steps: generating and sending a test signal at the sending end of a fiber-optic communication system, converting the test signal into an electrical signal, and performing time-domain and frequency-domain analysis on the electrical signal; extracting features from the electrical signal based on a multi-dimensional analysis algorithm to obtain a multi-dimensional feature data set of the electrical signal; obtaining monitoring data and fault history data corresponding to the multi-dimensional feature data set and constructing a fault evaluation model to generate a fault evaluation coefficient; obtaining the fault evaluation coefficients of each part of the transmission line at the same time interval within a period of time, and if the difference between the maximum and minimum fault evaluation coefficients of each part is greater than a preset threshold, monitoring is required. The present invention not only realizes early warning of faults, but also can quickly locate the faults when they occur, effectively improving the reliability and stability of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and more specifically, to a fiber optic passive voltage and current fault diagnosis method, device, equipment, and medium. Background Art

[0002] With the rapid development of modern communication technologies, fiber optic communication has become the main means of information transmission due to its advantages such as high speed, large capacity, long-distance transmission, and low loss. In fiber optic communication networks, passive optical networks (PONs) are widely used in access networks, metropolitan area networks, etc. due to their simple structure, easy expansion, and maintenance. However, with the continuous expansion of the scale of PON networks, the number of access nodes has increased sharply, posing an unprecedented challenge to the stability and reliability of the network.

[0003] Fiber optic passive voltage and current faults are one of the common problems in PON networks. Since voltage and current signals in fiber optic networks may be affected by various factors during transmission, such as fiber attenuation, connector loss, environmental temperature changes, etc., the signal quality deteriorates, and even faults may be triggered. These faults not only affect the normal communication of the network but may also damage equipment and increase maintenance costs.

[0004] Traditional fiber optic fault diagnosis methods mainly rely on manual inspections and offline tests. Manual inspections require maintenance personnel to regularly visit the site to check the equipment status. This method is not only time-consuming and laborious but also difficult to detect potential fault hazards. Offline testing is a detection method performed after a device fails. Although it can locate the fault point, it cannot monitor and give early warnings in real time, and cannot take timely measures to avoid the occurrence of faults.

[0005] For example, the invention with the publication number CN106646114A, a method for diagnosing faults in a distribution network, includes the following steps: obtaining device status information and switch action information; collecting the current on the primary side of the transmission line, including the alternating current and zero-sequence current on the primary side; comparing the collected alternating current value and zero-sequence current value with the preset short-circuit fault current threshold and ground fault current threshold respectively to determine the current fault information; comparing the processed device status information with the preset threshold information of the distribution terminal. If it exceeds the set range, it is determined that there is a device fault; obtaining the switch action information and comparing the switch action information with the current power supply information of the distribution network. If the switch action information is inconsistent with the power supply information of the grid at this time, it is determined that there is a switch fault. The present invention can accurately locate the cause and location of the fault, reduce the time of maintenance power outages, improve the utilization rate and operation reliability of equipment, and at the same time reduce maintenance costs.

[0006] For example, the invention with the announcement number: CN108414896A discloses a power grid fault diagnosis method, belonging to the technical field of power grid fault diagnosis. The power outage area is obtained from the protection and circuit breaker information, and the lines therein are taken as suspected fault lines, and their comprehensive current is constructed as a sample sequence; using the DTW algorithm, the DTW distance between the sample sequence of each suspected fault line and the reference sequence of the reference line outside the power outage area is obtained and constructed as the horizontal difference degree; at the same time, using the DTW algorithm, the DTW distance between the sample sequences before and after the fault of each suspected fault line is obtained and constructed as the vertical difference degree. The horizontal and vertical difference degrees of N suspected fault lines form a 2×N difference degree matrix, and the artificial fish swarm clustering algorithm is used to cluster this matrix into a fault class and a normal class, and the class with a larger clustering center value is diagnosed as the fault class, and the lines in the fault class are judged as fault lines. The present invention can accurately diagnose fault lines, is not affected by factors such as fault location, transition resistance, and fault type, and has good performance.

[0007] In the above disclosed technical solution, there are at least the following technical problems: The voltage and current signals in the optical fiber network may be affected by various factors during transmission, and although the existing optical fiber-based voltage and current monitoring methods have a fast processing speed, they may not be able to capture some subtle optical fiber faults or losses, resulting in incorrect diagnostic results or delays in fault repair time, and it is difficult to meet the requirements of modern communication networks for fast and accurate fault diagnosis. In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0008] In order to overcome the above defects of the prior art, an embodiment of the present invention provides an optical fiber passive voltage and current fault diagnosis method, device, equipment and medium, which combines historical data with real-time monitoring of fault diagnosis to solve the problems such as insufficient accuracy, slow response and inaccurate diagnostic accuracy positioning existing in the optical fiber-based voltage and current monitoring method.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A fiber-optic passive voltage and current fault diagnosis method includes the following steps: generating and transmitting a test signal at the transmitting end of a fiber-optic communication system, converting the test signal into an electrical signal, and performing time-domain and frequency-domain analysis on the electrical signal; extracting features from the electrical signal based on a multi-dimensional analysis algorithm to obtain a multi-dimensional feature data set of the electrical signal; obtaining monitoring data and fault history data corresponding to the multi-dimensional feature data set and constructing a fault evaluation model to generate a fault evaluation coefficient; dividing the power transmission line of the fiber-optic communication system into several parts, respectively obtaining the fault evaluation coefficients of each part of the power transmission line at the same time interval within a period of time, and if the difference between the maximum and minimum fault evaluation coefficients of each part is greater than a preset threshold, monitoring is required; obtaining several fault evaluation coefficients of the power transmission line during the monitoring process, constructing a first data set, and re-judging the fault according to the mean and standard deviation of the first data set.

[0011] In a preferred embodiment, the step of generating and transmitting a test signal at the transmitting end of the fiber-optic communication system, converting the test signal into an electrical signal, and performing time-domain and frequency-domain analysis on the electrical signal is specifically as follows: at the transmitting end of the fiber-optic communication system, according to the test requirements, generating a test signal containing specific voltage and current identifiers; modulating the generated test signal onto a light source through a modulator, converting it into an optical signal and transmitting it through the optical fiber to the receiving end; the photodetector converts the received optical signal into an electrical signal;

[0012] Sampling the electrical signal based on the Nyquist sampling theorem, and using analog-to-digital conversion to convert the continuous electrical signal into a discrete digital signal; based on an oscilloscope, observing the amplitude, period, and phase information of the digital signal within t time, and identifying the instantaneous changes, periodic characteristics, and emergencies of the digital signal; converting the time signal into amplitude and phase information of different frequency components through fast Fourier transform; by calculating the FFT of the time signal, obtaining the frequency components of the time signal and the amplitudes corresponding to the frequency components, and obtaining the energy distribution of the time signal at different frequencies; using band-pass filters, low-pass filters, and high-pass filters to suppress interfering frequency components; initially identifying faults in fiber-optic passive voltage and current by combining the time-domain analysis and frequency-domain analysis results.

[0013] In a preferred embodiment, the multi-dimensional analysis algorithm is used to extract features from the electrical signal to obtain a multi-dimensional feature dataset of the electrical signal, specifically: removing noise from the electrical signal based on a digital filter and adjusting the electrical signal to the same scale through min-max normalization; decomposing the electrical signal into two dimensions of time domain and frequency domain, decomposing the electrical signal at different scales, extracting the local frequency information of the electrical signal and performing time-frequency spectrum decomposition on the electrical signal to obtain the time-local features of the electrical signal at different times and frequencies; converting the electrical signal from the time domain to the frequency domain and extracting the frequency components of the electrical signal; performing spatial and frequency decomposition on the electrical signal, extracting the local spatial frequency features of the electrical signal and filtering the electrical signal in the spatial and frequency domains to obtain the spatial-local features of the electrical signal; converting the electrical signal from the spatio-temporal domain to the time-frequency domain, simultaneously performing spatio-temporal image decomposition on the signal, and extracting the features in time and space through a feature extraction algorithm to obtain spatio-temporal local features; integrating the time-local features, spatial-local features and spatio-temporal local features to form a multi-dimensional feature dataset.

[0014] In a preferred embodiment, the monitoring data includes a voltage deviation index and a current fluctuation index; the fault history data includes an abnormal influence index; the specific method for obtaining the voltage deviation index is as follows: obtaining the time-domain data of the voltage signal to be analyzed in real time and displaying the obtained voltage data in the time domain; obtaining the voltage time-domain waveform and filtering out the part with the lowest fundamental frequency in the voltage time-domain waveform based on a band-stop filter; performing a fast Fourier transform on the filtered voltage time-domain waveform to obtain spectrum information; extracting the amplitudes and periods of the respective harmonics from the spectrum information; obtaining the amplitudes, frequencies and periods of the respective harmonics in the voltage signal through Fourier transform analysis; calculating the effective values of the respective harmonics in the voltage signal in combination with the fundamental frequency, the amplitudes of the respective harmonics, the harmonic periods and the phases, and calculating the voltage deviation index based on a preset voltage deviation index formula.

[0015] In a preferred embodiment, the specific method for obtaining the abnormal influence index is as follows: Perform time series processing on the collected fault historical data. For each abnormal scenario, record the instantaneous data of the fault occurrence in chronological order to obtain a number of time series data; According to the occurrence time of the fault event, cut the time series data at equal time intervals, and take the fault data within each period as a basic unit; Calibrate the time series data to distinguish the normal state, fault state, and recovery state; Construct a fault tree by logical gates for the basic events and map the time series data corresponding to multiple abnormal scenarios to the basic events of the fault tree; Take the abnormal events or system failures encountered by the entire system as the top event of the fault tree; For each basic event, assign a probability of occurrence through historical data; According to the time series data, calculate the influence coefficient of each event in the fault tree on the top event; Based on the Bayesian network, perform weighted calculation on the influence coefficients of each basic event to obtain the influence coefficient of the top event of the fault tree; Calculate the abnormal influence index through comprehensive calculation based on the weighted influence coefficients of each top event in multiple fault scenarios.

[0016] In a preferred embodiment, obtain a number of fault evaluation coefficients of the transmission line during monitoring, construct a first data set, and re-judge the fault according to the mean and standard deviation of the first data set. Specifically: If the average value of the first data set is greater than the average reference threshold of the first data set, there is a fault in the monitoring part and an alarm is issued; If the average value of the first data set is less than the average reference threshold of the first data set and the standard deviation of the first data set is greater than the preset standard deviation reference threshold, continuous monitoring is performed; If the average value of the first data set is less than the average reference threshold of the first data set and the standard deviation of the first data set is less than the preset standard deviation reference threshold, there is no fault risk.

[0017] An apparatus for a fiber-optic passive voltage and current fault diagnosis method, comprising: a signal generation module, a feature extraction module, a fault assessment model construction module, a fault assessment and monitoring trigger module, and a fault re-judgment module; the signal generation module: configured to generate and transmit a test signal at the transmitting end of the fiber-optic communication system, convert the test signal into an electrical signal, and perform time-domain and frequency-domain analysis on the electrical signal; the feature extraction module: configured to extract features from the electrical signal based on a multi-dimensional analysis algorithm to obtain a multi-dimensional feature data set of the electrical signal; the fault assessment model construction module: configured to obtain monitoring data and fault history data corresponding to the multi-dimensional feature data set and construct a fault assessment model to generate a fault assessment coefficient; the fault assessment and monitoring trigger module: configured to divide the power transmission line of the fiber-optic communication system into several parts, respectively obtain the fault assessment coefficients of each part of the power transmission line at the same time interval within a period of time, and if the difference between the maximum and minimum fault assessment coefficients of each part is greater than a preset threshold, monitoring is required; the fault re-judgment module: configured to obtain several fault assessment coefficients of the power transmission line during the monitoring process, construct a first data set, and re-judge the fault according to the mean and standard deviation of the first data set.

[0018] An electronic device includes:

[0019] At least one processor; and,

[0020] A memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the described fiber-optic passive voltage and current fault diagnosis method.

[0022] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, a fiber-optic passive voltage and current fault diagnosis method is implemented.

[0023] The technical effects and advantages of the fiber-optic passive voltage and current fault diagnosis method, apparatus, device and medium of the present invention:

[0024] 1. The present invention extracts features from the electrical signal through a multi-dimensional analysis algorithm, forming a multi-dimensional feature data set including time local features, space local features, and spatio-temporal local features. This multi-dimensional feature extraction method can comprehensively capture the features of the electrical signal at different scales and in different domains, thereby more accurately identifying and locating faults in fiber-optic passive voltage and current. Compared with single-dimensional analysis methods, multi-dimensional analysis can more carefully capture various characteristics of the electrical signal, avoid missing possible key signal features, and significantly improve the accuracy and reliability of fault diagnosis.

[0025] 2. The present invention adopts a real-time monitoring method to regularly obtain the fault evaluation coefficient of the power transmission line in the optical fiber communication system, and determines whether further monitoring is required according to the fluctuation of the fault evaluation coefficient. When the difference in the fault evaluation coefficient of a certain part exceeds the preset threshold, the system will automatically trigger the monitoring process to continuously monitor potential faults. At the same time, the system will also construct a data set based on the fault evaluation coefficients collected during the monitoring process, and re-judge the faults based on the mean and standard deviation of the data set. This real-time monitoring and intelligent early warning mechanism can timely detect and handle faults, avoid losses caused by the expansion of faults, improve the work efficiency of operation and maintenance personnel, and reduce the operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic flow chart of a fiber-optic passive voltage and current fault diagnosis method of the present invention;

[0027] Figure 2 is a schematic structural diagram of a fiber-optic passive voltage and current fault diagnosis device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0029] Embodiment 1, Figure 1 A fiber-optic passive voltage and current fault diagnosis method of the present invention is given, including the following steps:

[0030] S1, generate and send a test signal at the sending end of the optical fiber communication system, convert the test signal into an electrical signal, and perform time-domain and frequency-domain analysis on the electrical signal;

[0031] The generation and sending of the test signal at the sending end of the optical fiber communication system, the conversion of the test signal into an electrical signal, and the time-domain and frequency-domain analysis of the electrical signal are specifically as follows:

[0032] At the sending end of the optical fiber communication system, according to the test requirements, generate a test signal containing specific voltage and current identifiers;

[0033] Modulate the generated test signal onto the light source through a modulator, convert it into an optical signal and transmit it to the receiving end through the optical fiber;

[0034] The photodetector converts the received optical signal into an electrical signal;

[0035] Sample the electrical signal based on the Nyquist sampling theorem, and use analog-to-digital conversion (ADC) to convert the continuous electrical signal into a discrete digital signal;

[0036] Based on the oscilloscope to observe the amplitude, period, and phase information of the digital signal within time t, identify the instantaneous changes, periodic characteristics, and emergencies of the digital signal;

[0037] Convert the time signal into amplitude and phase information of different frequency components through fast Fourier transform;

[0038] By calculating the FFT of the time signal, obtain the frequency components of the time signal and the amplitudes corresponding to the frequency components, and get the energy distribution of the time signal at different frequencies;

[0039] Use band-pass filters, low-pass filters, and high-pass filters to suppress interfering frequency components;

[0040] Combine the time-domain analysis and frequency-domain analysis results to initially identify the faults of fiber-optic passive voltage and current.

[0041] S2. Extract features from the electrical signal based on the multi-dimensional analysis algorithm to obtain a multi-dimensional feature dataset of the electrical signal;

[0042] The multi-dimensional analysis algorithm (Multidimensional Analysis Algorithm) generally refers to the technologies and methods for analyzing data in a multi-dimensional space, which are widely used in fields such as data science, machine learning, data mining, and statistics. The goal of these algorithms is to extract valuable information by analyzing data in different dimensions to help better understand the structure, patterns, and trends of the data. The core idea of multi-dimensional analysis is to project the data from a high-dimensional space to reveal potential relationships and laws.

[0043] The multi-dimensional analysis algorithm includes:

[0044] Principal Component Analysis (PCA): PCA is a dimensionality reduction technique that linearly transforms the data from a high-dimensional space to a new, uncorrelated low-dimensional space (i.e., the principal component space) such that each dimension in the new space is the direction with the largest variance.

[0045] Feature dimensionality reduction: Reduce the dimensionality of the data and retain the main information of the data, which is commonly used in image processing, financial data analysis, etc.

[0046] Data visualization: Map high-dimensional data to a 2D or 3D space for easy human understanding.

[0047] Factor Analysis: Factor analysis is a statistical method used to explore the underlying factors behind multiple observed variables. By reducing the dimensions, multiple observed variables are transformed into a few factors, thus reducing complexity.

[0048] Cluster Analysis: Cluster algorithms divide data into several groups (clusters) such that the data within the same group has the highest possible similarity, while the similarity between different groups is as low as possible. Common cluster algorithms include K-means, hierarchical clustering, etc.

[0049] Multidimensional Scaling (MDS): MDS is an algorithm used to analyze the similarity or distance between objects, aiming to represent the relative distances between high-dimensional data points in a low-dimensional space.

[0050] Feature extraction of the electrical signal based on the multi-dimensional analysis algorithm to obtain a multi-dimensional feature dataset of the electrical signal, specifically:

[0051] Remove the noise in the electrical signal based on a digital filter and adjust the electrical signal to the same scale through min-max normalization;

[0052] Decompose the electrical signal into two dimensions, time domain and frequency domain, decompose the electrical signal at different scales, extract the local frequency information of the electrical signal and perform time-frequency spectrum decomposition on the electrical signal to obtain the time-local features of the electrical signal at different times and frequencies;

[0053] Convert the electrical signal from the time domain to the frequency domain and extract the frequency components of the electrical signal;

[0054] Perform spatial and frequency decomposition on the electrical signal, extract the local spatial frequency features of the electrical signal and filter the electrical signal in the spatial and frequency domains to obtain the spatial local features of the electrical signal;

[0055] Convert the electrical signal from the spatio-temporal domain to the time-frequency domain, simultaneously perform spatio-temporal image decomposition on the signal, and extract the features in time and space through a feature extraction algorithm to obtain spatio-temporal local features;

[0056] Integrate the time-local features, spatial local features and spatio-temporal local features to form a multi-dimensional feature dataset.

[0057] The method of feature extraction of the electrical signal based on the multi-dimensional analysis algorithm to obtain a multi-dimensional feature dataset of the electrical signal has many benefits for fiber optic passive voltage and current fault diagnosis, specifically reflected in the following aspects:

[0058] Improve signal noise suppression and signal quality:

[0059] Digital filter denoising: By removing noise based on digital filters, interference and noise in electrical signals can be effectively removed. This is particularly important for the signal quality in fiber optic communication systems because noise and interference often affect the accuracy of fault diagnosis. A clear signal can provide a more accurate data basis for subsequent feature extraction. Min-max normalization: After normalizing the signal, it can ensure that the amplitudes of different signals are within the same scale, so that when comparing or analyzing multiple signals, errors will not occur due to different amplitudes, ensuring the effectiveness and comparability of feature extraction.

[0060] Multi-dimensional time-domain, frequency-domain and time-frequency-domain analysis:

[0061] Time-domain and frequency-domain decomposition: By decomposing an electrical signal into two dimensions, the time domain and the frequency domain, the characteristics of the signal can be analyzed from different perspectives. The time-domain signal reflects the process of current or voltage changing over time, while the frequency-domain signal reveals the distribution of different frequency components in the signal. Through this decomposition, the characteristics of the electrical signal can be understood more comprehensively, which helps to discover potential fault signs.

[0062] Time-frequency spectrum decomposition: Time-frequency analysis can capture the changes of a signal in both time and frequency simultaneously, especially suitable for processing signals with time-varying and frequency characteristics, such as sudden faults or short-time pulses. This characteristic is particularly important when diagnosing instantaneous changes in fiber optic voltage and current.

[0063] Local frequency and time-frequency characteristics: By extracting local frequency information and time-frequency spectra, some weak but meaningful characteristics in the signal can be captured, such as mutation points or abnormal fluctuations in frequency when a fault occurs. This is crucial for timely identifying and locating faults.

[0064] Extraction of spatial and frequency characteristics:

[0065] Spatial and frequency decomposition: By decomposing an electrical signal in the spatial and frequency domains, local characteristics of the signal at different spatial positions and different frequency ranges can be extracted. This analysis method is significantly helpful for signal propagation and fault location in fiber optic communication systems, especially in cases where the signal distribution is uneven or there is multi-path propagation.

[0066] Spatial local characteristics: Feature extraction in the spatial domain can reveal the changes of an electrical signal at different spatial positions, further helping to identify the fault sources in the circuit, such as fiber breakage, poor contact or cable aging.

[0067] Comprehensive analysis of spatio-temporal characteristics:

[0068] Space-time image decomposition: By converting the electrical signal from the space-time domain to the time-frequency domain and performing space-time image decomposition, the complex characteristics of the signal in space and time can be revealed, which is especially suitable for fault diagnosis tasks that require simultaneous consideration of time variations and spatial distributions.

[0069] Space-time local features: The extraction of space-time features helps to identify complex space-time fault patterns, such as time delay changes or frequency offsets during signal propagation. Some faults in fiber optic communication systems may only occur within specific time and space ranges, and the combination of space-time features can improve the accuracy and sensitivity of fault diagnosis.

[0070] Integration and comprehensive analysis of multi-dimensional feature sets:

[0071] Multi-dimensional feature integration: Combining time, space, and space-time features to form a multi-dimensional feature data set enables more comprehensive fault diagnosis and allows for signal analysis from multiple dimensions. This method can not only detect conventional faults but also discover some hard-to-detect potential fault patterns or early fault signs.

[0072] Improving diagnostic accuracy: The multi-dimensional analysis method can more precisely capture various characteristics of the electrical signal, avoiding signal features that may be missed in single-dimensional analysis, thereby improving the accuracy and reliability of fault diagnosis.

[0073] By extracting features from the electrical signal through multi-dimensional analysis algorithms, the features of the electrical signal at different scales and in different domains can be comprehensively captured, thus better identifying and locating faults in fiber optic passive voltage and current. This method not only improves the accuracy of fault diagnosis but also enhances the system's adaptability to various complex signal changes, which is of great significance for ensuring the stable operation of fiber optic communication systems.

[0074] S3. Obtain the monitoring data and fault history data corresponding to the multi-dimensional feature data set and construct a fault evaluation model to generate a fault evaluation coefficient;

[0075] The monitoring data includes a voltage deviation index and a current fluctuation index; the fault history data includes an abnormal impact index;

[0076] The voltage deviation index is used to measure the voltage fluctuation degree of a power system or equipment to judge the severity of voltage deviation. It reflects the stability and reliability of the voltage and is crucial for the normal operation of the power system. When the voltage deviation index is high, it means there is a large difference between the actual voltage and the rated voltage, which may cause unstable operation or even damage to power equipment. Therefore, monitoring and controlling the voltage deviation index is of great significance for ensuring the stable operation of the power system. By taking timely measures to adjust the voltage, the safety and reliability of the power system can be guaranteed, and potential safety hazards and economic losses can be avoided.

[0077] The specific method for obtaining the voltage deviation index is as follows:

[0078] Obtain the time-domain data of the voltage signal to be analyzed in real time, and display the obtained voltage data in the time domain;

[0079] Obtain the voltage time-domain waveform, and filter out the part with the lowest fundamental frequency in the voltage time-domain waveform based on a band-stop filter;

[0080] Perform a fast Fourier transform on the filtered voltage time-domain waveform to obtain spectral information;

[0081] Extract the amplitude and period of each harmonic from the spectral information;

[0082] Through the Fourier transform analysis method, obtain the amplitude, frequency, and period of each harmonic in the voltage signal;

[0083] Calculate the effective value of each harmonic in the voltage signal, combine the fundamental frequency, the amplitude of each harmonic, the harmonic period, and the phase, and calculate the voltage deviation index based on the preset voltage deviation index formula.

[0084] The specific calculation formula for the effective value of the harmonic is:

[0085]

[0086] The specific calculation formula for the voltage deviation index is:

[0087]

[0088] Among them, is the effective value of the harmonic, is the harmonic period, is the time-domain waveform of the voltage signal, is the voltage deviation index, is the fundamental frequency, is the amplitude of the k-th harmonic, is the phase of the k-th harmonic, and q is the total number of harmonics.

[0089] The current fluctuation index is used to describe the degree of fluctuation or change of the current signal within a certain time range. It reflects the stability or volatility of the current and can be used to evaluate the change of the current in an electrical system. Calculated based on the change of the current within a certain time, this index can help evaluate whether there is excessive fluctuation or instability in electrical equipment or power systems.

[0090] The specific method for obtaining the current fluctuation index is as follows:

[0091] Collect the real-time current data within time t in real time and record the real-time current data in the form of a time-series signal;

[0092] Calibrate the current data with a preset standard current value and remove abnormal current data;

[0093] Integrate the calibrated current values into a current set and extract the current mean value, current standard value, and peak current from the current set as current characteristic values;

[0094] Calculate the current fluctuation index based on the current characteristic values according to a preset current fluctuation index calculation formula.

[0095] The specific calculation formula for the peak current is:

[0096]

[0097] The specific calculation formula for the current fluctuation index is:

[0098]

[0099] Where, is the peak current, is the current fluctuation index, h is the number of current data sets collected within time t, is the i-th current data.

[0100] The abnormal influence index is a quantitative index used to measure the impact of the abnormal state of a device or system on the overall operation or surrounding environment when a failure occurs. This index comprehensively considers various factors during the occurrence of a failure, such as the duration of the failure, the severity of the failure, the scope of failure spread, and the actual impact of the failure on production, safety, the environment, etc.

[0101] The specific method for obtaining the abnormal influence index is as follows:

[0102] Perform time series processing on the collected failure history data. For each abnormal scenario, record the instantaneous data of the failure occurrence in chronological order to obtain several time series data.

[0103] According to the occurrence time of the failure event, cut the time series data at equal time intervals, and regard the failure data within each period as a basic unit;

[0104] Calibrate the time series data to distinguish the normal state, failure state, and recovery state;

[0105] Construct a fault tree by using logic gates for the basic events and map the time series data corresponding to multiple abnormal scenarios to the basic events of the fault tree;

[0106] Regard the abnormal events or system failures encountered by the entire system as the top event of the fault tree;

[0107] For each basic event, assign the probability of occurrence based on historical data;

[0108] According to the time series data, calculate the influence coefficient of each event in the fault tree on the top event.

[0109] Based on the Bayesian network, perform weighted calculation on the influence coefficients of each basic event to obtain the influence coefficient of the top event of the fault tree;

[0110] According to the weighted influence coefficients of the top events in multiple fault scenarios, perform comprehensive calculation to obtain the abnormal influence index.

[0111] The specific calculation formula of the abnormal influence index is:

[0112]

[0113] Among them, is the abnormal influence index, m is the total number of fault scenarios, n is the total number of basic events in the fault tree, is the i-th basic event in the fault tree, is the basic event The probability of occurrence, is the basic event The weight in the fault tree, is the fault duration of the basic event.

[0114] The fault evaluation model is specifically:

[0115]

[0116] Among them, is the fault evaluation coefficient, is the voltage deviation index, is the current fluctuation index, is the abnormal influence index, is the preset voltage deviation index proportionality coefficient, is the preset current fluctuation index proportionality coefficient, is the preset abnormal influence index proportionality coefficient.

[0117] S4. Divide the power transmission line of the optical fiber communication system into several parts, and respectively obtain the fault evaluation coefficients of each part of the power transmission line at the same time interval within a period of time. If the difference between the maximum and minimum fault evaluation coefficients of each part is greater than the preset threshold, monitoring is required;

[0118] According to the geographical layout, transmission distance, relay station location, and optical fiber type factors of the optical fiber communication system, divide the power transmission line into several parts;

[0119] Assign a unique identifier to each part;

[0120] Determine a suitable monitoring period, such as daily, weekly, or monthly, for regularly evaluating the fault conditions of each part.

[0121] During each monitoring period, collect the fault data of each part of the transmission line, including the number of faults, duration, and impact range.

[0122] According to the collected fault data, calculate the fault evaluation coefficient for each part. The fault evaluation coefficient can be a comprehensive index reflecting the severity and frequency of faults in that part of the transmission line.

[0123] For each part, calculate the difference between the maximum and minimum fault evaluation coefficients within a monitoring period.

[0124] According to the actual situation of the system and the operation and maintenance requirements, set a reasonable threshold for determining whether further monitoring of this part is required.

[0125] Compare the difference of each part with the set threshold. If the difference is greater than the threshold, it indicates that the fault situation of this part fluctuates greatly and may require further monitoring and maintenance.

[0126] S5. Obtain several fault evaluation coefficients of the transmission line during the monitoring process, construct a first data set, and re-judge the faults based on the mean and standard deviation of the first data set.

[0127] The obtaining several fault evaluation coefficients of the transmission line during the monitoring process, constructing a first data set, and re-judging the faults based on the mean and standard deviation of the first data set is specifically as follows:

[0128] If the average value of the first data set is greater than the average reference threshold of the first data set, there is a fault in the monitored part and an alarm is issued;

[0129] If the average value of the first data set is less than the average reference threshold of the first data set and the standard deviation of the first data set is greater than the pre-set standard deviation reference threshold, continuous monitoring is performed;

[0130] If the average value of the first data set is less than the average reference threshold of the first data set and the standard deviation of the first data set is less than the pre-set standard deviation reference threshold, there is no fault risk.

[0131] Embodiment 2 Figure 2 A fiber-optic passive voltage and current fault diagnosis device according to the present invention is provided, including: a signal generation module, a feature extraction module, a fault evaluation model construction module, a fault evaluation and monitoring trigger module, and a fault re-judgment module;

[0132] Signal generation module 201: It is used to generate and send test signals at the transmitting end of the optical fiber communication system, convert the test signals into electrical signals, and perform time-domain and frequency-domain analysis on the electrical signals;

[0133] Feature extraction module 202: It is used to extract features from electrical signals based on multi-dimensional analysis algorithms to obtain a multi-dimensional feature data set of electrical signals;

[0134] Fault assessment model construction module 203: It is used to obtain monitoring data and fault history data corresponding to the multi-dimensional feature data set and construct a fault assessment model to generate a fault assessment coefficient;

[0135] Fault assessment and monitoring trigger module 204: It is used to divide the power transmission line of the optical fiber communication system into several parts, respectively obtain the fault assessment coefficients of each part of the power transmission line at the same time interval within a period of time, and if the difference between the maximum and minimum fault assessment coefficients of each part is greater than a preset threshold, monitoring is required;

[0136] Fault re-judgment module 205: It is used to obtain several fault assessment coefficients of the power transmission line during the monitoring process, construct a first data set, and re-judge the fault according to the mean and standard deviation of the first data set.

[0137] The present invention further includes an electronic device, and the electronic device includes:

[0138] At least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute a method for diagnosing optical fiber passive voltage and current faults.

[0139] The present invention includes a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, it implements a method for diagnosing optical fiber passive voltage and current faults.

[0140] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0141] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0142] Those of ordinary skill in the art will realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0143] In addition, the functional modules in various embodiments of the present invention can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0144] As mentioned above, the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0145] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A fiber optic passive voltage and current fault diagnosis method, characterized in that The following steps are involved: Generate and send a test signal at a transmitting end of an optical fiber communication system, convert the test signal into an electrical signal, and perform time domain and frequency domain analysis on the electrical signal; Extract features of electrical signals based on a multi-dimensional analysis algorithm to obtain a multi-dimensional feature data set of electrical signals; Obtain monitoring data and fault history data corresponding to the multidimensional feature data set and build a fault assessment model to generate a fault assessment coefficient; Among them, is the fault evaluation coefficient, is the voltage deviation index, is the current fluctuation index, is the abnormal influence index, is the preset voltage deviation index proportionality coefficient, is the preset current fluctuation index proportionality coefficient, is the preset abnormal influence index proportionality coefficient; The transmission line of the optical fiber communication system is divided into several parts, and the fault assessment coefficients of each part of the transmission line at the same time interval within a period of time are obtained respectively. If the difference between the maximum and minimum fault assessment coefficients of each part is greater than a preset threshold, monitoring is required; A plurality of fault assessment coefficients of the transmission line during the monitoring process are obtained, a first data set is constructed, and the fault is re-judged according to the mean and standard deviation of the first data set.

2. The fiber optic passive voltage and current fault diagnosis method according to claim 1, wherein The generating and sending of a test signal at the transmitting end of the optical fiber communication system, converting the test signal into an electrical signal, and performing time domain and frequency domain analysis on the electrical signal are specifically as follows: At the transmitting end of the optical fiber communication system, a test signal including specific voltage and current identifiers is generated according to the test requirements; The generated test signal is modulated onto the light source through a modulator, converted into an optical signal and transmitted to the receiving end through an optical fiber; The photodetector converts the received light signal into an electrical signal; The electrical signal is sampled based on the Nyquist sampling theorem, and the continuous electrical signal is converted into a discrete digital signal using analog-to-digital conversion; Based on the oscilloscope observation of the amplitude, period, and phase information of the digital signal within time t, the instantaneous changes, periodic characteristics, and sudden events of the digital signal can be identified; The time signal is converted into the amplitude and phase information of different frequency components through fast Fourier transform; By calculating the FFT of the time signal, the frequency components of the time signal and the amplitudes corresponding to the frequency components are obtained, and the energy distribution of the time signal at different frequencies is obtained; Use bandpass filters, lowpass filters, and highpass filters to suppress interfering frequency components; Combining the time domain analysis and frequency domain analysis results, the passive voltage and current faults of the optical fiber can be preliminarily identified.

3. The fiber-optic passive voltage and current fault diagnosis method according to claim 2, characterized in that The feature extraction of the electrical signal based on the multi-dimensional analysis algorithm is performed to obtain a multi-dimensional feature data set of the electrical signal, specifically: The noise in the electrical signal is removed based on a digital filter, and the electrical signal is adjusted to the same scale through minimum-maximum normalization; Decompose the electrical signal into two dimensions: time domain and frequency domain, decompose the electrical signal at different scales, extract the local frequency information of the electrical signal, and perform time-frequency spectrum decomposition on the electrical signal to obtain the temporal local characteristics of the electrical signal at different times and frequencies; Convert the electrical signal from the time domain to the frequency domain and extract the frequency component of the electrical signal; Decompose the electrical signal in space and frequency, extract the local spatial frequency characteristics of the electrical signal, and filter the electrical signal in space and frequency domains to obtain the spatial local characteristics of the electrical signal; The electrical signal is converted from the time-space domain to the time-frequency domain, and the signal is decomposed into a time-space image, and the time and space features are extracted through a feature extraction algorithm to obtain the time-space local features; The temporal local features, spatial local features and spatiotemporal local features are integrated to form a multidimensional feature data set.

4. A fiber optic passive voltage and current fault diagnosis method according to claim 3, characterized in that, The monitored data includes a voltage deviation index and a current fluctuation index; the fault history data includes an abnormal influence index; The specific method for obtaining the voltage deviation index is as follows: Obtain the time-domain data of the voltage signal to be analyzed in real time, and display the obtained voltage data in the time domain; Obtain the voltage time-domain waveform, and filter out the part with the lowest fundamental frequency in the voltage time-domain waveform based on a band-stop filter; Perform a fast Fourier transform on the filtered voltage time-domain waveform to obtain spectral information; Extract the amplitude and period of each harmonic from the spectral information; Through the Fourier transform analysis method, obtain the amplitude, frequency, and period of each harmonic in the voltage signal; Calculate the effective value of each harmonic in the voltage signal, combine the fundamental frequency, the amplitude of each harmonic, the harmonic period, and the phase, and calculate the voltage deviation index based on a preset voltage deviation index formula.

5. The fiber optic passive voltage and current fault diagnosis method according to claim 4, characterized in that, The specific method for obtaining the abnormal influence index is as follows: Perform time series processing on the collected fault history data. For each abnormal scenario, record the instantaneous data of the fault occurrence in chronological order to obtain several time series data; According to the occurrence time of the fault event, cut the time series data at equal time intervals, and use the fault data within each period as a basic unit; Calibrate the time series data to distinguish the normal state, the fault state, and the recovery state; Construct a fault tree by using basic events through logic gates, and map the time series data corresponding to multiple abnormal scenarios to the basic events of the fault tree; Regard the abnormal events or system failures encountered by the entire system as the top event of the fault tree; For each basic event, assign a probability of occurrence through historical data; According to the time series data, calculate the influence coefficient of each event in the fault tree on the top event; Based on the Bayesian network, perform weighted calculation on the influence coefficients of each basic event to obtain the influence coefficient of the top event of the fault tree; Perform comprehensive calculation based on the weighted influence coefficients of each top event in multiple fault scenarios to obtain the abnormal influence index.

6. The fiber-optic passive voltage and current fault diagnosis method according to claim 5, characterized in that Obtain several fault evaluation coefficients of the transmission line during the monitoring process, construct a first data set, and re-judge the fault according to the mean and standard deviation of the first data set. Specifically: If the average value of the first data set is greater than the average reference threshold of the first data set, there is a fault in the monitoring part, and an alarm is issued; If the average value of the first data set is less than the average reference threshold of the first data set, and the standard deviation of the first data set is greater than the preset standard deviation reference threshold, continuous monitoring is performed; If the average value of the first data set is less than the average reference threshold of the first data set, and the standard deviation of the first data set is less than the preset standard deviation reference threshold, there is no fault risk.

7. A fiber optic passive voltage and current fault diagnosis method according to claim 6, characterized in that The specific calculation formula of the abnormal influence index is: Among them, is the abnormal influence index, m is the total number of failure scenarios, n is the total number of basic events in the fault tree, is the i-th basic event in the fault tree, is the basic event the probability of occurrence, is the basic event the weight in the fault tree, is the fault duration of the basic event.

8. An apparatus using a fiber optic passive voltage and current fault diagnosis method according to any one of claims 1-7, characterized in that, including: a signal generation module, a feature extraction module, a fault evaluation model construction module, a fault evaluation and monitoring trigger module, and a fault re-judgment module; The signal generation module: used to generate and send a test signal at the sending end of the optical fiber communication system, convert the test signal into an electrical signal, and perform time-domain and frequency-domain analysis on the electrical signal; Feature extraction module: used to extract features from electrical signals based on a multi-dimensional analysis algorithm to obtain a multi-dimensional feature dataset of electrical signals; Fault assessment model construction module: used to obtain the monitoring data and fault history data corresponding to the multi-dimensional feature dataset and construct a fault assessment model to generate a fault assessment coefficient; Fault assessment and monitoring trigger module: used to divide the power transmission line of the optical fiber communication system into several parts, respectively obtain the fault assessment coefficients of each part of the power transmission line at the same time interval within a period of time, and if the difference between the maximum and minimum fault assessment coefficients of each part is greater than a preset threshold, monitoring is required; Fault re-judgment module: used to obtain several fault assessment coefficients of the power transmission line during the monitoring process, construct a first dataset, and re-judge the fault based on the mean and standard deviation of the first dataset.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute a fiber optic passive voltage and current fault diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a fiber optic passive voltage and current fault diagnosis method according to any one of claims 1 to 7.

Images