Intelligent analysis and diagnosis method and system for optical fiber operation data

Abstract

The invention discloses an optical fiber operation data intelligent analysis and diagnosis method and system, and relates to the field of intelligent feedback control, and the method comprises the steps: receiving an optical fiber state data set, and obtaining an optical fiber fault data set through combining a preset optical fiber state fault contrast database; analyzing the set through a preset historical association database to obtain an association data judgment result; a judgment verification data set is obtained and analyzed, and a verification judgment result is obtained; optical fiber fault time and environment fault type sets are respectively obtained by combining optical fiber fault time characteristics and positions; constructing an alarm analysis model to analyze the set, and obtaining alarm grading judgment data; and associating the alarm judgment data with the fault position and the time characteristic, obtaining a repair scheme and effect feedback, and correcting the historical association database according to the repair scheme and effect feedback. According to the invention, the historical association database is utilized to accurately position the fault and repair the fault, and the fiber fault repair rate is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent feedback control, specifically to an intelligent analysis and diagnosis method and system for fiber optic operation data. Background Technology

[0002] The intelligent analysis and diagnosis system for fiber optic operation data is a comprehensive operation and maintenance tool that integrates sensing technology, data processing, and artificial intelligence. Its core function is to collect multi-dimensional data reflecting the physical state, transmission performance, and environmental impact of the fiber optic network in real time or periodically through various monitoring devices deployed in the fiber optic network. After preprocessing, relying on a preset fault comparison database and intelligent algorithms, the system performs feature extraction and multi-dimensional analysis on the data to accurately identify the type of fiber optic fault. Combining the fault location and time characteristics of BeiDou positioning, the system determines the level of fault impact and automatically generates targeted repair solutions. Through feedback data, the system continuously iterates the database and algorithm model, ultimately realizing the intelligent transformation of the fiber optic network from "passive fault response" to "proactive early warning and prevention." This significantly improves network operation and maintenance efficiency, reduces the risk of service interruption, and provides full lifecycle technical support for the stable operation of the fiber optic communication network.

[0003] The current process and algorithm revolve around a data closed loop. Monitoring equipment collects status parameters such as fiber optic attenuation coefficient, connector insertion loss, optical power, signal-to-noise ratio, and bit error rate, along with environmental location and time data. After cleaning, standardization, and time-series alignment, a dataset is formed. The system calls a fault comparison database, combines rule-based reasoning to determine the trend of parameter threshold changes and multi-parameter correlations to output a preliminary fault type. For complex scenarios, machine learning is introduced to extract features, train a model, and output a fault probability distribution. After fusion, the final type is determined. Then, combining time features and environmental type, a weighted score is used to determine the alarm level. A repair strategy is generated based on historical data matching. After repair, feedback data on the effect is collected and stored. Reinforcement learning is used to optimize database rules and model weights to improve diagnostic effectiveness.

[0004] Data acquisition relies on a single device, resulting in significant variations in format and accuracy, and incomplete collection of key parameters; static thresholds in preprocessing are ill-suited to complex environments, easily filtering out normal fluctuations or missing latent faults; rule-based reasoning in fault diagnosis is insufficient for identifying novel and multi-factor faults, and machine learning suffers from weak generalization ability due to imbalanced samples, leading to misclassification; alarm level weights rely on experience-based settings, failing to consider real-time business and topology, resulting in resource mismatch; remediation solutions overly rely on historical matching, lacking dynamic environmental adaptation; manual feedback of effects suffers from poor real-time performance and a lack of quantitative standards, and long model iteration cycles make it difficult to respond to network changes. Summary of the Invention

[0005] This application provides an intelligent analysis and diagnosis method and system for fiber optic operation data, which addresses the technical problems in the prior art, such as data acquisition relying on a single device leading to large differences in format accuracy, preprocessing static thresholds being difficult to adapt to complex environments, rule reasoning in fault diagnosis being insufficient for identifying new and multi-factor faults, poor real-time performance of manual input of effect feedback and lack of quantitative standards, and long model iteration cycles making it difficult to respond to network changes.

[0006] In view of the above problems, this application provides an intelligent analysis and diagnosis system for optical fiber operation data.

[0007] The first aspect of this application provides an intelligent analysis and diagnosis method for optical fiber operation data. The method includes: receiving an optical fiber status data set; pre-setting a historical association database and a fault classification database, analyzing the optical fiber status data set to obtain an optical fiber fault data set; receiving an optical fiber time data set, analyzing the optical fiber time data set based on the optical fiber fault data set to obtain an optical fiber time fault type set; receiving an optical fiber fault location, analyzing the received optical fiber environment data set based on the optical fiber fault data set to obtain an optical fiber environment fault type set; constructing an alarm analysis model, analyzing the optical fiber time fault type set and the optical fiber environment fault type set to obtain alarm classification judgment data; associating the alarm system judgment data with the optical fiber fault location and optical fiber time data to obtain a repair plan and effect feedback; and correcting the historical association database based on the obtained repair plan and effect feedback, combined with the association between the alarm system judgment data and the optical fiber fault location and optical fiber time data.

[0008] A second aspect of this application provides an intelligent analysis and diagnosis system for optical fiber operation data. The system is equipped with the aforementioned method and includes: a first receiving unit for receiving an optical fiber status data set; a first analysis and judgment unit for pre-setting a historical association database and a fault classification database, analyzing the optical fiber status data set to obtain an optical fiber fault data set; a first analysis and output unit for receiving an optical fiber time data set, analyzing the optical fiber time data set based on the optical fiber fault data set to obtain an optical fiber time fault type set; a second analysis and output unit for receiving an optical fiber fault location, analyzing the received optical fiber environment data set based on the optical fiber fault data set to obtain an optical fiber environment fault type set; a second analysis and judgment unit for constructing an alarm analysis model, analyzing the optical fiber time fault type set and the optical fiber environment fault type set to obtain alarm classification judgment data; and a first processing and output unit for associating the alarm system judgment data with the optical fiber fault location and optical fiber time data to obtain a repair plan and effect feedback, and correcting the historical association database based on the obtained repair plan and effect feedback combined with the alarm system judgment data and the optical fiber fault location and optical fiber time data association.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages: The embodiments of this application receive a set of optical fiber status data; upon receiving the set of optical fiber status data, combine it with a preset optical fiber status fault comparison database to obtain a set of optical fiber fault data; analyze this set through a preset historical association database to obtain association data judgment results; obtain judgment verification datasets and merge them for analysis to obtain verification judgment results; combine optical fiber time data and location to obtain sets of optical fiber fault time and environmental fault types respectively; construct an alarm analysis model to analyze the above sets to obtain alarm classification judgment data; associate the alarm judgment data with fault location and time characteristics to obtain repair schemes and effect feedback, and thereby correct the historical association database. This application utilizes a historical association database to accurately locate faults and implement repairs, effectively improving the optical fiber fault repair rate.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] like Figure 1 This is a flowchart of an intelligent analysis and diagnosis method for optical fiber operation data according to this application. Detailed Implementation

[0013] This application provides an intelligent analysis and diagnosis method and system for fiber optic operation data, which addresses the technical problems in existing technologies, such as data acquisition relying on a single device leading to large differences in format accuracy, preprocessing static thresholds being difficult to adapt to complex environments, rule-based reasoning in fault diagnosis being insufficient for identifying novel and multi-factor faults, poor real-time performance of manual input of effect feedback and lack of quantitative standards, and long model iteration cycles making it difficult to respond to network changes.

[0014] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0015] Example 1 like Figure 1 As shown, this application provides an intelligent analysis and diagnosis method for optical fiber operation data, the method comprising: S100: Receive fiber optic status data set; Fiber optic status data refers to data indicators reflecting the status of fiber optic cables from different dimensions. These parameters include, but are not limited to, data collected using methods such as deploying optical time-domain reflectometers, optical power meters, spectrometers, and distributed sensing technologies. The acquired data is then directly obtained through methods such as correlating experimental equipment.

[0016] Step S100 in the method provided in this application embodiment includes: S110: Receive the fiber attenuation coefficient as the first fiber status data; S120: Receive connector insertion loss as second fiber status data; S130: Receive optical power as third fiber status data; S140: Receive signal-to-noise ratio as fourth fiber status data; S150: Received bit error rate as fifth fiber status data; S160: Receive optical power fluctuations as sixth fiber status data; S170: The first fiber status data, the second fiber status data, the third fiber status data, the fourth fiber status data, the fifth fiber status data, and the sixth fiber status data are used as a fiber status data set.

[0017] To acquire fiber optic status data, a test optical signal is first injected into the fiber using an optical time-domain reflectometer (OTDR) to detect signal attenuation during transmission and calculate attenuation values ​​at different wavelengths. This attenuation coefficient is then used as the first fiber optic status data. Simultaneously, the same device is used to detect power abrupt changes at the fiber optic connector, or a light source and optical power meter are used to measure the power before and after the connector and calculate the difference, yielding the connector insertion loss as the second fiber optic status data. Next, a portable optical power meter is used to measure the actual optical power at both the transmitting and receiving ends of the fiber optic link, providing the third fiber optic status data. An optical signal-to-noise ratio (SNR) analyzer is used to collect the useful optical signal power and noise power within the fiber, calculating their ratio to obtain the SNR as the fourth fiber optic status data. Then, a bit error rate (BER) analyzer is connected to the fiber optic link to simulate actual data transmission and statistically analyze the ratio of erroneous symbols to total symbols, obtaining the bit error rate as the fifth fiber optic status data. Finally, an optical power meter or network management system with real-time monitoring capabilities continuously collects optical power data over a set time period, analyzing the amplitude and frequency of changes to determine the optical power fluctuation as the sixth fiber optic status data. The fiber attenuation coefficient reflects the energy loss of the optical signal per unit length during transmission in the optical fiber. An abnormally high value will lead to a continuous weakening of the signal strength, affecting the quality of long-distance transmission. The connector insertion loss reflects the energy loss of the optical signal when passing through the fiber connector, reflecting the quality of the connector connection. An excessively high value will become a bottleneck in the link transmission, exacerbating signal attenuation. Optical power refers to the intensity of the optical signal acquired by the receiving equipment, directly determining the signal demodulation effect. Too low a value will cause the receiver to be unable to properly identify the signal, while too high a value may damage the receiving device. The signal-to-noise ratio (SNR) is the ratio of optical signal power to noise power, reflecting the signal's anti-interference capability. An insufficient value will lead to increased bit errors during high-speed transmission, affecting data integrity. The bit error rate (BER) represents the ratio of erroneous symbols to total symbols during transmission, directly reflecting the accuracy of data transmission. Excessive BER can lead to lost or erroneous service data, and in severe cases, communication interruption. Optical power fluctuation refers to the magnitude of changes in received optical power in the short term, reflecting link stability. Excessive fluctuation will cause instability in the receiver's operating point, increasing the risk of bit errors and affecting service continuity. For example, these data are obtained by periodically emitting test light pulses using an optical time-domain reflectometer to analyze backscattered waveforms, collect attenuation coefficients and connector insertion losses, collect optical power in real time using an optical power meter or a device-built-in monitoring unit, obtain signal-to-noise ratio by analyzing the optical signal spectrum using a spectrometer, statistically analyze bit error rate in real time using a built-in module of the transmission device, and calculate the standard deviation based on continuously sampled optical power data to obtain optical power fluctuations.

[0018] S200: Preset historical correlation database and fault classification database, analyze fiber optic status data set to obtain fiber optic fault data set; The construction of the historical correlation database is centered on the full life cycle data of the fault. It integrates the status data set when the fiber optic fault occurs, the confirmed fault type data, the BeiDou-based location information, the time feature data, and the corresponding repair schemes and effect feedback. By assigning a unique identifier to each fault, the abnormal patterns of fault feature parameters are bound to the long-term stability records of parameter recovery after the implementation of repair measures. This forms a structured storage system that includes the correlation between fault type and location, time and environment, as well as the correspondence between repair schemes and effects. At the same time, by continuously incorporating new fault cases and effect feedback data, the feature mapping relationship and scheme optimization basis in the database are continuously enriched, providing data support for the generation of subsequent fault diagnosis and repair schemes and system iteration.

[0019] Step S200 in the method provided in this application embodiment includes: S201: Preset historical association database; S202: Preset fault classification database; S203: Analyze the fiber optic status data set through the fault classification database to obtain the fiber optic anomaly data set; S204: Receive sensor error as the first judgment and verification data; S205: Preset a reasonable threshold range, receive and analyze the threshold parameters according to the reasonable threshold range, and obtain the reasonable threshold data as the second judgment verification data; S206: Missing feature database data is used as third-party verification data; S207: Use the first judgment verification data, the second judgment verification data, and the third judgment verification data as a judgment verification data set; S208: Analyze the fiber optic anomaly data set based on the judgment and verification data set; S209: If the fiber optic fault data set has corresponding verification data in the judgment verification data set, the fault type is determined to be a misjudgment, and a misjudgment repair catalog is generated and output. If the fiber optic fault data set does not have corresponding verification data in the judgment verification data set, the fault type is determined to be a non-misjudgment, and the fiber optic fault data set is combined into a fiber optic fault data set. S210: Input the fiber optic fault data set into the historical correlation database; S211: If the fiber optic fault data set has corresponding fault data in the historical association database, then call and output the corresponding repair plan; if the fiber optic fault data set does not have corresponding fault data in the historical association database, then generate and send a pause command, and output the fiber optic fault data set.

[0020] In this embodiment, a historical correlation database and a fault classification database are first established. The fault classification database is used to analyze fiber optic status data, filtering out parameters that deviate from normal standards to construct an abnormal fiber optic data set. Simultaneously, sensor error data, threshold rationality data, and feature library missing data are collected to form a judgment verification data set, used to analyze the matching with fiber optic fault data. Reasonable data is obtained by pre-setting a reasonable threshold range and analyzing the threshold parameters. This is then combined with information such as whether sensor errors exist and whether the feature library is missing for comprehensive judgment: if the fault anomaly characteristics are consistent with the parameter deviation caused by sensor errors, or the threshold parameters exceed a reasonable range, or the fault characteristics are not included in the feature library, it is determined to be a misjudgment; otherwise, it is not a misjudgment. The verification data set is compared with the abnormal data set. If a match is found, it is determined to be a misjudgment and a misjudgment repair directory is generated; if no match is found, the abnormal data set is determined to be the fiber optic fault data set. Subsequently, the fault data set is compared with the historical correlation database. If a corresponding record is found, a repair plan is retrieved and output; if no corresponding record is found, a pause command is sent and the fault data set is output for subsequent analysis. For example, if an anomaly is detected due to sensor error, it is determined to be a misjudgment and a repair catalog for sensor calibration is generated; if there is no matching item for the abnormal data and a similar fault record is found in the historical database, the corresponding repair solution is output; if there is no record, the fault data set is output for further processing.

[0021] S300: Receives fiber optic time data set, analyzes fiber optic fault data set to obtain fiber optic time fault type set; Fiber optic fault time characteristic analysis obtains a set of fiber optic fault type data. This refers to further classifying and refining the identified fiber optic fault type data by combining time characteristics such as the specific time period of the fault occurrence, duration, and periodicity. This is achieved by analyzing the typical manifestations of different fault types in the time dimension.

[0022] Step S300 in the method provided in this application embodiment includes: S301: Preset peak period interval, receive the time of fiber optic fault occurrence as the first fiber optic time data; S302: Preset time interval, receive the duration of fiber optic fault as the second fiber optic time data; S303: Preset optical power fluctuation range, receive periodic fluctuations of optical fiber faults as third optical fiber time data. S304: Collect the first fiber time data, the second fiber time data, and the third fiber time data as a fiber time data set; S305: Manually preset peak time periods, duration ranges, and preset optical power fluctuation ranges; S306: Analyze and judge the fault time feature set based on the fiber optic fault data set; S307: If the first fiber time data is within the peak period, the fiber fault type data is determined to be a peak period fault. If the fault occurs outside the peak period, the fiber fault type data is determined to be an off-peak period fault, and the first fiber time fault type is output. S308: If the second fiber time data is within the preset duration range, the fiber fault type data is determined to be a short-term fault; if the second fiber time data is outside the preset duration range, the fiber fault type is determined to be a long-term fault, and the second fiber time fault type is output. S309: If the third fiber time data is within the optical power fluctuation range, it is determined to be a periodic fault; if the third fiber time data is outside the optical power fluctuation range, it is determined to be a non-periodic fault, and the third fiber time fault type is output. S310: The first fiber time fault type, the second fiber time fault type, and the third fiber time fault type are taken as a set of fiber time fault types.

[0023] In this embodiment, the timing and duration of a received fiber optic fault are considered as a set of time-related characteristics. Fault type data are then classified based on these characteristics. If a fault occurs during peak traffic periods, it is classified as a peak-period fault; if it occurs during periods of low traffic, it is classified as an off-peak fault. Faults with long durations are classified as long-term faults, while those with short durations and no recurrence are classified as short-term faults. Regularly recurring optical power fluctuations correspond to periodic faults, while irregular fluctuations are classified as non-periodic faults. For example, if a fault occurs during peak evening internet usage, lasts for two hours, recovers, and does not recur, with no obvious periodicity in the optical power fluctuations, the generated set of fiber optic time-related fault types includes peak-period faults, short-term faults, and non-periodic faults.

[0024] S400: Receive fiber optic fault location. Analyze the received fiber optic environment data set based on the fiber optic fault data set to obtain the fiber optic environment fault type set. Analyzing fiber optic fault location data to obtain a set of fiber optic environmental fault types involves classifying identified fiber optic fault types based on their specific geographical location and surrounding environmental characteristics. This is achieved by matching and analyzing environmental information corresponding to the fault location (e.g., underground pipelines, overhead lines, mountainous areas, damp regions, or equipment rooms) with the fault type data to clarify the correlation patterns between different environmental factors and fault types.

[0025] Step S400 in the method provided in this application embodiment includes: S410: Location of receiver fiber optic fault; S420: Receives a set of fiber optic environment data based on the location of the fiber optic fault; S430: The received optical fiber environmental data set includes: receiving environmental humidity data as first optical fiber environmental data, receiving geographical optical fiber environmental data as second optical fiber environmental data, receiving vibration pressure data as third optical fiber environmental data, and using the first optical fiber environmental data, the second optical fiber environmental data, and the third optical fiber environmental data as an optical fiber environmental data set. S440: Preset environmental humidity critical point, abnormal range of geographical environment data, and abnormal range of vibration pressure; S450: Analyze and judge the fiber optic environment data set based on the fiber optic fault data set; S460: If the first fiber optic environment data is greater than the preset humidity threshold, the fiber optic environment fault type is determined to be a humidity data fault. If the first fiber optic environment data is less than or equal to the preset humidity threshold, the fiber optic environment fault type is determined to be a non-humidity data fault, and the first fiber optic environment fault type is output. S470: If the second fiber optic environment data is within the abnormal range of the geographic environment data, the fiber optic environment fault type is determined to be a geographic fiber optic environment data fault. If the second fiber optic environment data is outside the abnormal range of the geographic fiber optic environment data, the fiber optic environment fault type is determined to be a non-geographic environment data fault, and the second fiber optic environment fault type is output. S480: If the third fiber optic environment data is within the vibration and pressure abnormal range, the fiber optic environment fault type is determined to be a vibration and pressure data fault. If the third fiber optic environment data is outside the vibration and pressure abnormal range, the fiber optic environment fault type is determined to be a non-vibration and pressure data fault, and the third fiber optic environment fault type is output. S490: The first fiber optic environment fault type, the second fiber optic environment fault type, and the third fiber optic environment fault type are used as a set of fiber optic environment fault types.

[0026] In this embodiment, after receiving the location of the optical fiber fault, the corresponding environmental data is analyzed in conjunction with that location. The specific type of environmental fault is determined by anomalies in data such as humidity, geographical environment, vibration, and pressure. If the environmental humidity data exceeds the normal range, it is determined to be a humidity data fault; otherwise, it is a non-humidity data fault. If geographical environment data, such as terrain and soil, shows anomalies, it is determined to be a geographical environment data fault; otherwise, it is a non-geographical environment data fault. If vibration pressure data shows abnormal fluctuations, it is determined to be a vibration pressure data fault; otherwise, it is a non-vibration pressure data fault. For example, if an optical fiber fault is located in a low-lying wetland, and the detection shows that the environmental humidity is far above the normal range, while the geographical environment is normal and the vibration pressure data is normal, then the type of optical fiber environmental fault is determined to be a humidity data fault, not a geographical environment data fault, and not a vibration pressure data fault.

[0027] S500: Construct an alarm analysis model to analyze the set of fiber optic time fault types and the set of fiber optic environmental fault types, and obtain alarm classification judgment data; Constructing an alarm analysis model to analyze the optical fiber time fault type set and the optical fiber environmental fault type set to obtain alarm classification judgment data refers to establishing an analysis model that integrates time characteristics and environmental characteristics. This model comprehensively evaluates time dimension information such as the time period, duration, and periodicity of the fault occurrence, along with environmental influencing factors such as humidity, geography, and vibration at the fault location. The model is based on the severity of the correlation between different time-type faults and environmental-type faults.

[0028] Step S500 in the method provided in this application embodiment includes: S510: Construct an alarm analysis model, and classify the alarm analysis model into three levels; S520: Obtain time fault weight score based on the set of fiber optic time fault types, and obtain environmental fault weight score based on the set of fiber optic environmental fault types. S530: Set the alarm level judgment formula, and substitute the time fault weight score and the environmental fault weight score into the alarm level judgment formula. The formula is as follows: ; in The total score is weighted. For time-based fault weighting, Environmental fault weighting; S540: Obtain the total weight score, input the total weight score into the alarm analysis model, and obtain alarm classification judgment data.

[0029] In this embodiment, an alarm analysis model is constructed and divided into three levels: general, severe, and emergency. During analysis, the time fault weight score is first determined based on the fiber optic time fault type set. For example, peak-period faults or long-term faults correspond to higher scores, while off-peak-period faults or short-term faults correspond to lower scores. Simultaneously, the environmental fault weight score is determined based on the fiber optic environmental fault type set. Environmental problems such as abnormal humidity or vibration / pressure are assigned corresponding scores, while normal environments receive lower scores. Next, the total weight score is calculated using a predefined alarm level judgment formula; that is, the total weight score equals the sum of the time fault weight score and the environmental fault weight score. The calculated total weight score is then input into the alarm analysis model. The model outputs results according to the different levels corresponding to the total score range. If a general alarm remains unresolved for 4 hours and the parameters deteriorate, it is automatically escalated to a severe alarm; if a severe alarm remains unresolved for 8 hours, it is automatically escalated to an emergency alarm. For example, a certain fault is a long-term fault that occurs during peak periods. The time fault has a weight of 8 points. At the same time, it is in an environment with abnormal humidity, so the environmental fault has a weight of 4 points. The total weight is 12 points. The alarm classification data obtained after inputting it into the model is an important alarm.

[0030] S600: Associate alarm system judgment data with fiber optic fault location and fiber optic time data to obtain repair solutions and effect feedback. Based on the obtained repair solutions and effect feedback, and in conjunction with the association between alarm system judgment data and fiber optic fault location and fiber optic time data, correct the historical association database. The alarm system's judgment data is correlated with the fiber optic fault location and time data. Based on this correlation information, the corresponding repair solutions are retrieved, and feedback on the effects after implementation is collected. The specific results of the applicability and effectiveness feedback of the repair solutions are then combined with the alarm judgment data and the fiber optic fault location and time data. The corresponding parameters, such as abnormal patterns, of the fault types and repair solutions stored in the historical correlation database are adjusted and optimized to ensure that the historical correlation database can incorporate new fault handling experience and characteristic patterns in a timely manner, thereby improving the accuracy of subsequent fault diagnosis and solution generation.

[0031] Step S600 in the method provided in this application embodiment includes: S610: Associates the alarm system with fiber optic fault location and fiber optic time data; S620: Obtain corresponding fault repair solutions based on different fiber optic fault locations and fiber optic time data; S630: Awaiting manual feedback on the repair results; S640: Based on human feedback, optimize and correct the historical correlation database and fault classification database.

[0032] In this embodiment, the alarm system is associated with the location and time data of fiber optic faults. Corresponding repair schemes are matched based on different fault locations and time data. After repair, feedback on the effectiveness is received manually, and the historical association database and fault classification database are optimized and corrected based on the feedback. For example, a waterproof repair scheme is used for joint faults in damp underground locations, and backup links are prioritized for fiber breakage faults during peak periods. After repair, effectiveness feedback is recorded using quantifiable indicators such as optical power recovery value and fault recurrence interval. If the optical power of a repaired underground pipe joint fault recovers to the standard value and does not recur for three months, the priority of this type of scheme is increased, and the correlation data between location, environmental, and temporal characteristics and repair effectiveness is included in the historical association database, achieving dynamic iterative optimization of the database.

[0033] Example 2 Based on the same inventive concept as the intelligent analysis and diagnosis method for optical fiber operation data in the foregoing embodiments, this application provides an intelligent analysis and diagnosis system for optical fiber operation data. This system is equipped with the aforementioned method, wherein the system includes: The first receiving unit is used to receive the optical fiber status data set; The first analysis and judgment unit is used to preset the historical correlation database and the fault classification database, analyze the optical fiber status data set, and obtain the optical fiber fault data set. The first analysis output unit is used to receive the fiber optic time data set, analyze the fiber optic time data set based on the fiber optic fault data set, and obtain the fiber optic time fault type set. The second analysis output unit is used to receive the fiber optic fault location, analyze the received fiber optic environment data set based on the fiber optic fault data set, and obtain the fiber optic environment fault type set. The second analysis and judgment unit is used to construct an alarm analysis model, analyze the set of fiber optic time fault types and the set of fiber optic environmental fault types, and obtain alarm classification judgment data. The first processing output unit is used to associate the alarm system judgment data with the fiber optic fault location and fiber optic time data, obtain the repair plan and effect feedback, and correct the historical association database based on the obtained repair plan and effect feedback and the association between the alarm system judgment data and the fiber optic fault location and fiber optic time data.

[0034] Preferred options also include: The second receiving unit is used to receive the fiber attenuation coefficient as the first fiber status data. The third receiving unit is used to receive the connector insertion loss as the second fiber status data. The fourth receiving unit is used to receive optical power as the third fiber status data; The fifth receiving unit is used to receive the signal-to-noise ratio as the status data of the fourth optical fiber; The sixth receiving unit is used to receive the bit error rate as the status data of the fifth optical fiber; The seventh receiving unit is used to receive optical power fluctuations as the status data of the sixth optical fiber. The second processing output unit is used to collect the first fiber status data, the second fiber status data, the third fiber status data, the fourth fiber status data, the fifth fiber status data, and the sixth fiber status data as a fiber status data set.

[0035] Preferred options also include: The first preset unit is used to preset the historical association database and the fault classification database; The third analysis output unit is used to analyze the optical fiber status data set through the fault classification database to obtain the optical fiber anomaly data set; The eighth receiving unit is used to receive sensor error as the first judgment and verification data; The ninth receiving unit is used to preset a reasonable threshold range, receive and analyze the threshold parameters according to the reasonable threshold range, and obtain threshold reasonableness data as the second judgment verification data; The tenth receiving unit is used to receive the third judgment verification data for feature library missingness; The third processing output unit is used to take the first judgment verification data, the second judgment verification data, and the third judgment verification data as a judgment verification data set. The third analysis and judgment unit is used to analyze and judge the fiber optic anomaly data set based on the judgment and verification data set; The fourth analysis and judgment unit is used to determine the fault type as a false alarm if the fiber optic fault data set has corresponding verification data in the judgment and verification data set, and to generate and output a false alarm repair catalog. If the fiber optic fault data set does not have corresponding verification data in the judgment and verification data set, the fault type is determined as a non-false alarm and the fiber optic fault data set is combined into a fiber optic fault data set. The fifth analysis and judgment unit is used to input the fiber optic fault data set into the historical correlation database for judgment; The fourth analysis output unit is used to call and output the corresponding repair scheme if the fiber optic fault data set has corresponding fault data in the historical association database, and to generate and send a pause command and output the fiber optic fault data set if the fiber optic fault data set does not have corresponding fault data in the historical association database.

[0036] Preferred options also include: The eleventh receiving unit is used to preset the peak period interval and receive the time of fiber optic failure as the first fiber optic time data. The twelfth receiving unit is used to receive the duration of optical fiber faults as second optical fiber time data within a preset time interval. The thirteenth receiving unit is used to preset the optical power fluctuation range and receive the periodic fluctuations of optical fiber faults as the third optical fiber time data. The fourth processing output unit is used to take the first fiber time data, the second fiber time data, and the third fiber time data as a fiber time data set. The second preset unit is used to manually preset peak time periods, duration ranges, and optical power fluctuation ranges. The sixth analysis and judgment unit is used to analyze and judge the fault time feature set based on the fiber optic fault data set; The fifth analysis output unit is used to determine the fiber fault type data as a peak business period fault if the first fiber time data is within the peak period, and to determine the fiber fault type data as an off-peak period fault if the fault occurs outside the peak period, and output the first fiber time fault type. The sixth analysis output unit is used to determine the fiber fault type as a short-term fault if the second fiber time data is within the preset duration interval, and to determine the fiber fault type as a long-term fault if the second fiber time data is outside the preset duration interval, and output the second fiber time fault type. The seventh analysis output unit is used to determine if the third fiber time data is within the optical power fluctuation range, and to determine if the third fiber time data is outside the optical power fluctuation range, and to output the third fiber time fault type. The fifth processing output unit is used to set the first fiber time fault type, the second fiber time fault type, and the third fiber time fault type as a fiber time fault type set.

[0037] Preferred options also include: The fourteenth receiving unit is used to receive the location of fiber optic faults. The first analysis and receiving unit is used to receive a set of fiber optic environment data based on the location of the fiber optic fault. The fifteenth receiving unit is used to receive the optical fiber environmental data set including: receiving environmental humidity data as first optical fiber environmental data, receiving geographical optical fiber environmental data as second optical fiber environmental data, receiving vibration pressure data as third optical fiber environmental data, and using the first optical fiber environmental data, the second optical fiber environmental data, and the third optical fiber environmental data as an optical fiber environmental data set. The third preset unit is used to manually preset the critical point of environmental humidity, the abnormal range of geographical environmental data, and the abnormal range of vibration pressure. The seventh analysis and judgment unit is used to analyze and judge the fiber optic environment data set based on the fiber optic fault data set; The eighth analysis output unit is used to determine the fiber optic environment fault type as a humidity data fault if the first fiber optic environment data is greater than the preset humidity threshold, and to determine the fiber optic environment fault type as a non-humidity data fault if the first fiber optic environment data is less than or equal to the preset humidity threshold, and output the first fiber optic environment fault type. The ninth analysis output unit is used to determine the fiber optic environment fault type as a geographic fiber optic environment data fault if the second fiber optic environment data is within the geographic environment data anomaly range, and to determine the fiber optic environment fault type as a non-geographic environment data fault if the second fiber optic environment data is outside the geographic fiber optic environment data anomaly range, and output the second fiber optic environment fault type. The tenth analysis output unit is used to determine the fiber optic environment fault type as a vibration pressure data fault if the third fiber optic environment data is within the vibration pressure abnormal range, and to determine the fiber optic environment fault type as a non-vibration pressure data fault if the third fiber optic environment data is outside the vibration pressure abnormal range, and outputs the third fiber optic environment fault type. The sixth processing output unit is used to set the first fiber optic environment fault type, the second fiber optic environment fault type, and the third fiber optic environment fault type as a set of fiber optic environment fault types.

[0038] Preferred options also include: The seventh processing output unit is used to construct the alarm analysis model, which is divided into three levels. The eighth processing output unit is used to obtain time fault weight scores based on the set of fiber optic time fault types and to obtain environmental fault weight scores based on the set of fiber optic environmental fault types. The first calculation unit is used to set the alarm level judgment formula, and to input the time fault weight score and the environmental fault weight score into the alarm level judgment formula. The formula is as follows: ; in The total score is weighted. For time-based fault weighting, Environmental fault weighting; The eighth analysis and judgment unit is used to obtain the total weight score, which is then input into the alarm analysis model to obtain alarm classification judgment data.

[0039] Preferred options also include: The first association unit is used to associate the alarm system with the fiber optic fault location and fiber optic time data. The ninth processing output unit is used to obtain corresponding fault repair solutions based on different fiber optic fault locations and fiber optic time data. The eleventh analysis output unit is used to wait for manual feedback on the effect after the repair is completed; The tenth processing output unit is used to optimize and correct the historical correlation database and fault classification database based on the feedback from humans.

[0040] The steps of the methods or algorithms described in this application can be directly embedded in hardware, software units executed by a processor, or a combination of both. Exemplarily, a storage medium can be connected to a processor so that the processor can read information from and write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. Optionally, the processor and the storage medium can also be located in different components within a terminal. These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0041] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely illustrative examples of this application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for intelligent analysis and diagnosis of optical fiber operation data, characterized in that, The method includes: Receive fiber optic status data set; Pre-set historical correlation database and fault classification database, analyze fiber optic status data set to obtain fiber optic fault data set; Receive the fiber optic time data set, analyze the fiber optic fault data set to obtain the fiber optic time fault type set; Receive the location of fiber optic faults, and analyze the received fiber optic environment data set based on the fiber optic fault data set to obtain the set of fiber optic environment fault types. Construct an alarm analysis model to analyze the set of fiber optic time fault types and the set of fiber optic environmental fault types, and obtain alarm classification judgment data; The alarm system's judgment data is correlated with the fiber optic fault location and fiber optic time data to obtain repair solutions and effect feedback. Based on the obtained repair solutions and effect feedback, the historical correlation database is corrected by combining the alarm system's judgment data with the fiber optic fault location and fiber optic time data.

2. The method according to claim 1, characterized in that, The received fiber status data set includes: The fiber attenuation coefficient is received as the first fiber status data. The insertion loss of the receiving connector is used as the second fiber status data. Received optical power as third fiber status data; The received signal-to-noise ratio is used as the status data of the fourth fiber optic cable; The received bit error rate is used as the status data of the fifth fiber optic cable. Receive optical power fluctuations as sixth fiber status data; The first fiber status data, the second fiber status data, the third fiber status data, the fourth fiber status data, the fifth fiber status data, and the sixth fiber status data are used as the fiber status data set.

3. The method according to claim 2, characterized in that, The preset historical association database and fault classification database are used to analyze the fiber optic status data set to obtain the fiber optic fault data set, including: Preset historical relational database; Pre-defined fault classification database; By analyzing the fiber optic status data set through a fault classification database, a set of fiber optic anomaly data is obtained. The received sensor error is used as the first judgment and verification data; A reasonable threshold range is preset, and the threshold parameters are received and analyzed based on the reasonable threshold range to obtain threshold reasonableness data as the second judgment and verification data; The missing feature library data is used as the third judgment and verification data. The first judgment verification data, the second judgment verification data, and the third judgment verification data are used as the judgment verification data set; The fiber optic anomaly data set is analyzed and judged based on the judgment and verification data set; If the fiber optic anomaly data set has corresponding verification data in the judgment verification data set, the fault type is determined to be a false alarm, and a false alarm repair catalog is generated and output. If the fiber optic fault data set does not have corresponding verification data in the judgment verification data set, the fault type is determined to be a non-false alarm, and the fiber optic anomaly data set is combined into a fiber optic fault data set. The fiber optic fault data set is input into the historical correlation database for judgment; If the fiber optic fault data set has corresponding fault data in the historical association database, the corresponding repair solution is invoked and output. If the fiber optic fault data set does not have corresponding fault data in the historical association database, a pause command is generated and sent, and the fiber optic fault data set is output.

4. The method according to claim 3, characterized in that, The received fiber optic time data set is analyzed based on the fiber optic fault data set to obtain a set of fiber optic time fault types, including: The peak period interval is preset, and the time of fiber optic failure is received as the first fiber optic time data. The preset time interval is used to receive the duration of fiber optic faults as the second fiber optic time data. The optical power fluctuation range is preset, and the periodic fluctuations of optical fiber faults are received as third optical fiber time data. The first fiber time data, the second fiber time data, and the third fiber time data are used as a fiber time data set. Manually preset peak time periods, duration ranges, and optical power fluctuation ranges; The fault time feature set is judged and analyzed based on the fiber optic fault data set; If the first fiber time data is within the peak period, the fiber fault type data is determined to be a peak period fault. If the fault occurs outside the peak period, the fiber fault type data is determined to be an off-peak period fault, and the first fiber time fault type is output. If the second fiber time data is within the preset duration range, the fiber fault type data is determined to be a short-term fault; if the second fiber time data is outside the preset duration range, the fiber fault type is determined to be a long-term fault, and the second fiber time fault type is output. If the third fiber time data is within the optical power fluctuation range, it is determined to be a periodic fault; if the third fiber time data is outside the optical power fluctuation range, it is determined to be a non-periodic fault, and the third fiber time fault type is output. The first fiber time fault type, the second fiber time fault type, and the third fiber time fault type are defined as a set of fiber time fault types.

5. The method according to claim 3, characterized in that, The location of the receiving optical fiber fault is determined by analyzing the receiving optical fiber environmental data set based on the optical fiber fault data set, to obtain a set of optical fiber environmental fault types, including: Location of the receiving fiber optic fault; Based on the location of the fiber optic fault, receive a set of fiber optic environmental data; The received optical fiber environmental data set includes: received environmental humidity data as first optical fiber environmental data, received geographical optical fiber environmental data as second optical fiber environmental data, received vibration pressure data as third optical fiber environmental data, and the first optical fiber environmental data, the second optical fiber environmental data, and the third optical fiber environmental data as an optical fiber environmental data set. Artificially preset environmental humidity critical points, abnormal geographical environment data ranges, and abnormal vibration pressure ranges; The fiber optic environment data set is analyzed and judged based on the fiber optic fault data set; If the first fiber optic environment data is greater than the preset humidity threshold, the fiber optic environment fault type is determined to be a humidity data fault. If the first fiber optic environment data is less than or equal to the preset humidity threshold, the fiber optic environment fault type is determined to be a non-humidity data fault, and the first fiber optic environment fault type is output. If the second fiber optic environment data is within the abnormal range of the geographic environment data, the fiber optic environment fault type is determined to be a geographic fiber optic environment data fault. If the second fiber optic environment data is outside the abnormal range of the geographic fiber optic environment data, the fiber optic environment fault type is determined to be a non-geographic environment data fault, and the second fiber optic environment fault type is output. If the third fiber optic environment data is within the vibration and pressure abnormal range, the fiber optic environment fault type is determined to be a vibration and pressure data fault. If the third fiber optic environment data is outside the vibration and pressure abnormal range, the fiber optic environment fault type is determined to be a non-vibration and pressure data fault, and the third fiber optic environment fault type is output. The first fiber optic environment fault type, the second fiber optic environment fault type, and the third fiber optic environment fault type are defined as a set of fiber optic environment fault types.

6. The method according to claim 5, characterized in that, The alarm analysis model is constructed to analyze the fiber optic time fault type set and the fiber optic environmental fault type set to obtain alarm classification judgment data, including: Build an alarm analysis model and classify the alarm analysis model into three levels; Based on the set of fiber optic time fault types, obtain the time fault weight score; based on the set of fiber optic environmental fault types, obtain the environmental fault weight score. A formula for determining alarm levels is established, incorporating the time-based fault weight score and the environmental fault weight score. The formula is as follows: ； in The total score is weighted. For time-based fault weighting, Environmental fault weighting; Obtain the total weighted score, input the total weighted score into the alarm analysis model, and obtain alarm classification judgment data.

7. The method according to claim 6, characterized in that, The process of associating alarm system judgment data with fiber optic fault location and fiber optic time data to obtain repair solutions and effect feedback, and then correcting the historical association database based on the obtained repair solutions and effect feedback in conjunction with the association between alarm system judgment data and fiber optic fault location and fiber optic time data, includes: Associate the alarm system with fiber optic fault location and fiber optic time data; Based on the location and time data of the fiber optic fault, a corresponding fault repair plan is obtained; Waiting for manual feedback on the repair results; Based on human feedback, optimize and correct the historical correlation database and fault classification database.

8. An intelligent analysis and diagnosis system for fiber optic operation data, the system being used to implement the above method, the system comprising: The first receiving unit is used to receive the optical fiber status data set; The first analysis and judgment unit is used to preset the historical correlation database and the fault classification database, analyze the optical fiber status data set, and obtain the optical fiber fault data set. The first analysis output unit is used to receive the fiber optic time data set, analyze the fiber optic time data set based on the fiber optic fault data set, and obtain the fiber optic time fault type set. The second analysis output unit is used to receive the fiber optic fault location, analyze the received fiber optic environment data set based on the fiber optic fault data set, and obtain the fiber optic environment fault type set. The second analysis and judgment unit is used to construct an alarm analysis model, analyze the set of fiber optic time fault types and the set of fiber optic environmental fault types, and obtain alarm classification judgment data. The first processing output unit is used to associate the alarm system judgment data with the fiber optic fault location and fiber optic time data, obtain the repair plan and effect feedback, and correct the historical association database based on the obtained repair plan and effect feedback and the association between the alarm system judgment data and the fiber optic fault location and fiber optic time data.

Images