Intelligent Operation and Maintenance Method and System for Optical Cables in Port Areas Based on AI Technology

Abstract

This invention discloses an intelligent operation and maintenance method and system for optical fiber cables in port areas based on AI technology. The method includes: real-time acquisition of optical fiber cable acoustic vibration intensity parameters, environmental parameters, and positioning signal parameters; quantification and preprocessing of each parameter to generate vibration energy entropy coefficient, environmental stress coefficient, and positioning signal coefficient; calculation of a health index based on an optical fiber cable health assessment model that integrates these coefficients; triggering corresponding response measures based on the health index; generating a maintenance priority list and performing path planning for multiple operation and maintenance nodes based on the Dijkstra algorithm; and updating the health assessment model. The system includes a data acquisition module, a database storage module, an intelligent operation and maintenance hub, and an operation and maintenance module. This method improves the ability to perceive hidden damage and potential risks to optical fibers by collecting multi-dimensional sensing data and quantifying it into corresponding coefficients. The established assessment model can predict and accurately determine the operating status of optical fibers, guiding the implementation of proactive preventative and restorative maintenance, and improving the overall operation and maintenance capabilities of the system.

Description

Technical Field

[0001] This invention relates to the field of optical cable operation and maintenance algorithm technology, specifically to an intelligent operation and maintenance method and system for port communication optical cables based on AI technology. Background Technology

[0002] With the accelerating pace of intelligent and automated port development globally, port communication optical cables, as the core infrastructure for data transmission, directly determine the efficiency of port logistics scheduling, equipment coordination, and safety monitoring based on their operational stability. However, traditional operation and maintenance models rely on manual inspections and passive fault handling, making it difficult to detect hidden damage and environmental stress in optical cables in real time. This results in delayed fault response, high maintenance costs, and severely restricts the efficient operation of ports.

[0003] Current optical cable operation and maintenance technologies suffer from weak comprehensiveness, insufficient predictive and disaster prevention capabilities, and an inability to make accurate and efficient maintenance decisions. Specifically: First, there is insufficient integration capability for multi-source heterogeneous data. For example, the integration and collaborative analysis capabilities of DAS acoustic temperature and humidity monitoring are weak, and the coupled impact of environmental stress and vibration intensity is not fully quantified, making it difficult to construct a comprehensive optical cable condition assessment model. Second, the fault analysis capability of SDH alarm information is insufficient. SDH alarm information can only provide a vague judgment of fiber optic faults, failing to integrate multi-dimensional data and historical operation and maintenance data to improve the accuracy of fault diagnosis. Third, there is a lack of dynamic self-optimization mechanisms, making it impossible to dynamically adjust the assessment model based on historical operation and maintenance data to improve operation and maintenance efficiency. Fourth, maintenance strategies rely on manual experience and lack intelligent priority decisions based on multi-dimensional factors such as the scope of fault impact, service interruption level, environmental risks, and maintenance costs, resulting in low resource scheduling efficiency and difficulty in meeting the high real-time and high reliability operation and maintenance needs of ports.

[0004] Therefore, there is an urgent need for an AI-based intelligent operation and maintenance method and system for port communication optical cables to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide an intelligent operation and maintenance method and system for port communication optical cables based on AI technology, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] A method for intelligent operation and maintenance of optical fiber cables in port areas based on AI technology includes the following contents and steps:

[0008] S1. Real-time acquisition of acoustic vibration intensity parameters, environmental parameters, and positioning signal parameters of the optical cable;

[0009] S2. Perform quantitative preprocessing on the acoustic vibration intensity parameters, environmental parameters, and positioning signal parameters to generate vibration energy entropy coefficient, environmental stress coefficient, and positioning signal coefficient, respectively.

[0010] S3. Establish a health assessment model for optical cables that integrates vibration energy entropy coefficient, environmental stress coefficient, and location information coefficient, and generate a health index.

[0011] S4. Preset health index thresholds and response measures, determine the operation and maintenance nodes based on the health index values, and trigger the response measures;

[0012] S5. If there are multiple maintenance nodes, a maintenance priority list is generated based on the fault impact range, service interruption level, environmental risk level and equipment availability status. Maintenance path planning is performed on the maintenance nodes. Node edge weights are generated by combining the maintenance priority, distance cost, time cost and risk cost of the access nodes. The maintenance path planning is implemented based on the Dijkstra algorithm.

[0013] S6. Perform repairs on each of the maintenance nodes according to the maintenance path, and update the optical cable health assessment model after the processing is completed.

[0014] Furthermore, the acoustic vibration intensity parameters include signal amplitude. Statistics are performed on the M frames of vibration signals for each set of vibration data, and the vibration energy entropy coefficient is used as the parameter. Characterizing the energy distribution of vibration signals:

[0015] ;

[0016]

[0017]

[0018] in, Let k be the energy probability of the m-th frame, and k be the energy sensitivity factor. Let L be the energy of the m-th frame, L be the number of signal sampling points per frame, and s(n) be the signal amplitude of the n-th sampling point.

[0019] Furthermore, environmental parameters include temperature and humidity. The impact of environmental parameters on optical cables manifests as cumulative stress caused by changes in temperature and humidity, expressed as the environmental stress coefficient. Perform comprehensive characterization:

[0020] ;

[0021] ;

[0022] ;

[0023] in, and These are the weighting coefficients for temperature and humidity, respectively. + =1; For temperature-accumulated stress, This is the temperature influence coefficient. for Instantaneous temperature value at a given time point This refers to the standard operating temperature of the optical cable. For the current time, This is the initial time for the statistics; For humidity-accumulated stress, Humidity influence coefficient for Instantaneous humidity value at a given time.

[0024] Furthermore, considering the inherent errors of the positioning terminal equipment, and calculating the positioning accuracy coefficient C based on signal strength and signal fluctuations:

[0025] ;

[0026] Signal strength influence factor ;

[0027] Signal fluctuation influencing factors ;

[0028] Equipment error impact factor ;

[0029] in, , and These are the weight coefficients of each influencing factor, and + + =1; The signal strength of the current acquisition frame. This represents the maximum signal strength. The signal conditioning coefficient, This represents the standard deviation of the signal fluctuation. The adjustment coefficient for the positioning terminal equipment. This is due to the inherent error of the positioning terminal equipment.

[0030] Furthermore, the optical cable health assessment model is as follows:

[0031] ;

[0032] in, For health index, It is a non-linear adjustment factor. For health threshold, , and These are the dynamic weighting coefficients for the vibration energy entropy coefficient, environmental stress coefficient, and positional information coefficient, respectively.

[0033] Furthermore, when the health index If the health index threshold is ≤, a level 3 compensation measure will be triggered, specifically:

[0034] The first level is: activate the backup fiber optic link to switch data routes;

[0035] The second level involves using thermal imaging technology to inspect along the line and search for fiber optic cable fault points.

[0036] The third level involves performing local fiber optic fusion splicing to repair fiber optic cable fault points.

[0037] Furthermore, it also includes fault analysis, collecting SDH alarm information, system error information, and optical power information data, and performing fault analysis by integrating the vibration energy entropy coefficient, the environmental stress coefficient, and the positioning signal coefficient. The analysis methods include:

[0038] 1) If an R-Los alarm occurs, and the vibration energy entropy coefficient changes abruptly and the positioning information coefficient is abnormally high, and there is a breakage in the historical fault data of the fault reporting point, then the fault of the fault reporting point is determined to be a fiber optic cable breakage.

[0039] 2) For system bit error alarms, calculate the bit error rate. If the bit error rate fluctuates periodically and is related to the change period of the environmental parameters, combine this with the environmental stress coefficient within that time period. When the environmental stress coefficient exceeds the environmental stress threshold, the bit error fault is determined to be a fault caused by environmental influence.

[0040] 3) For the system error alarm, if the bit error rate suddenly increases without obvious periodicity and no R-Los alarm occurs, but the optical power fluctuates abnormally, then the fault point is determined to be optical cable aging or optical cable fault.

[0041] Furthermore, the maintenance priority list is created based on the Mamdani fuzzy inference method, which assigns fuzzy sets to the scope of fault impact, service interruption level, environmental risk level, and equipment availability status, and establishes a rule base for maintenance priorities by combining the various fuzzy sets, and matches maintenance priorities to each operation and maintenance node.

[0042] Different maintenance priorities are matched with different cost influencing factors. We establish a multi-objective cost function, using the cost value as the node edge weight from the current location to the access node, and perform operation and maintenance path planning based on Dijkstra's algorithm.

[0043] ;

[0044] Where a, b, and c are the weighting coefficients for distance cost, time cost, and risk cost, respectively. The physical distance from the current location to the access node. The time taken to reach the access node from the current location. This is the comprehensive risk assessment value during the process of reaching the access node from the current location.

[0045] Furthermore, after completing the repair of the maintenance nodes, the true health index of the optical cable is reassessed, a loss function is established, and the nonlinear adjustment factor, health threshold, and dynamic weight coefficient are gradually adjusted according to the direction of the error gradient. The prediction error is then minimized using the gradient descent algorithm. Optimize and update the optical cable health assessment model; among which,

[0046] Loss function: ;

[0047] Where n represents the training samples, which come from the repaired node data and historical node data; This is the predicted health index value based on the data of the i-th sample. This represents the true value of the health index based on the i-th sample data.

[0048] A smart operation and maintenance system for optical fiber cables in port areas includes a data acquisition module, an AI analysis terminal, a database storage module, a smart operation and maintenance hub, and an operation and maintenance module; wherein,

[0049] The data acquisition module includes: acoustic wave sensing DNS equipment, SDH transmission equipment, environmental monitoring instruments, and positioning terminals;

[0050] The AI ​​analysis terminal is used to calculate the vibration energy entropy coefficient, environmental stress coefficient, and positioning information coefficient based on real-time monitoring data, and then send them to the intelligent operation and maintenance center.

[0051] The database storage module includes: a time-series database for receiving and storing real-time monitoring data uploaded by the data acquisition module; a relational database for storing operation and maintenance data uploaded by the operation and maintenance module, as well as parameters and definitions of various algorithm models; and an optical cable status characteristic database, which includes optical cable abnormal event types, fault location mapping relationships, and a maintenance strategy knowledge base.

[0052] The intelligent operation and maintenance center is used to receive and process the real-time monitoring data from the data acquisition module, calculate the health index based on the vibration energy entropy coefficient, the environmental stress coefficient, and the positioning information coefficient, and generate a maintenance priority list; the intelligent operation and maintenance center analyzes the causes of failures based on the coefficients and the information data from the acoustic wave sensor DNS device.

[0053] The operation and maintenance module includes: a scheduling terminal, used to receive the maintenance priority list and the positioning information of the positioning terminal, and generate scheduling instructions according to the built-in cost analysis algorithm and operation and maintenance strategy; a drone swarm, used to perform thermal imaging inspection of optical cables according to the scheduling instructions; an intelligent robot, used to perform fiber optic splice repair work according to the scheduling instructions; and maintenance personnel, used to perform in-depth repair work according to the scheduling instructions.

[0054] Compared with existing technologies, the intelligent operation and maintenance method and system for port communication optical cables based on AI technology of the present invention has the following beneficial effects:

[0055] 1. This operation and maintenance method enhances the perception of potential risks caused by hidden damage to optical cables and environmental stress by introducing vibration energy entropy coefficient, environmental stress coefficient, and location information coefficient. It also establishes an assessment model based on these three coefficients to calculate a health index, enabling the prediction and accurate assessment of the optical cable's operating status. Different operation and maintenance strategies are implemented based on the health index, guiding the appropriate implementation of proactive preventative and restorative maintenance. Fault response is rapid and appropriate, reducing potential fault risks, lowering operation and maintenance costs, improving the overall operation and maintenance capabilities of optical cables, and enhancing the overall resilience of the port optical cable network. Furthermore, it optimizes path scheduling to rationally allocate maintenance resources and automatically updates and optimizes the assessment model after maintenance, improving the model's adaptability.

[0056] 2. The system collects and integrates heterogeneous data from multiple sources, calls different operation and maintenance strategies according to the health status of optical cables, comprehensively analyzes the causes of faults based on historical data and collected data, generates instructions for comprehensive operation and maintenance and dispatches maintenance equipment and personnel through the scheduling terminal, optimizes the paths of multiple fault points, rationally allocates maintenance resources, and reduces the impact of faults. Attached Figure Description

[0057] Figure 1 This is a flowchart of the intelligent operation and maintenance method for optical fiber cables in port areas disclosed in this invention.

[0058] Figure 2 This is a schematic diagram of the composition of the intelligent operation and maintenance system for optical fiber cables in port areas disclosed in this invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely the best embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] The term "embodiment" as used herein means that a particular method, step, or content described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0061] This embodiment provides an intelligent operation and maintenance method for port area communication optical cables based on AI technology, such as... Figure 1 As shown, based on the port area communication optical cable intelligent operation and maintenance system, such as Figure 2 As shown, the operation and maintenance system includes a data acquisition module, a database storage module, an AI analysis terminal, an intelligent operation and maintenance hub, and an operation and maintenance module. The following describes the implementation method of this operation and maintenance method in conjunction with the operation and maintenance system. Specifically, it includes the following contents and steps:

[0062] S1. First, a database storage module is constructed. The database storage module adopts a cloud database, which includes a time-series database, a relational database, and an optical cable status feature database. The time-series database InfluxDB receives and stores high-frequency real-time monitoring data uploaded by the data acquisition module. The relational database MySQL records and stores the operation and maintenance data uploaded by the operation and maintenance module, the calculation formulas of various coefficients and the parameters of the health index model, the definition of fuzzy sets, etc., so that the intelligent operation and maintenance center can call them. The optical cable status feature database records the types of historical abnormal events of optical cables, fault location mapping relationships, and maintenance strategy knowledge base.

[0063] Secondly, the data acquisition module collects the acoustic vibration intensity parameters, environmental parameters, and positioning signal parameters of the optical cable in real time and uploads them to the time-series database InfluxDB. In this embodiment, the data acquisition module specifically includes a distributed fiber optic acoustic sensing DNS device, an SDH transmission device, environmental monitoring instruments, and a positioning terminal.

[0064] Distributed fiber optic acoustic wave sensing DNS devices are based on the principle of scattering. When an external acoustic wave or vibration acts on the optical cable, it will cause local vibration of the optical fiber, thereby changing the phase of the Rayleigh scattered light. The signal processing device demodulates and extracts the phase change of the vibration signal, thereby obtaining the acoustic wave vibration intensity parameter, i.e., the signal amplitude.

[0065] SDH transmission equipment receives optical signals from optical fibers and has a built-in optical power monitoring function that monitors the optical signal strength in real time. When the received optical signal strength falls below a preset minimum received optical power threshold, an R-LOS alarm is triggered on the optical board. Simultaneously, the SDH transmission equipment uses a bit error rate (BER) detection algorithm (such as BIP-8) to detect bit errors in the transmitted signal in real time and calculate the BER. , Number of error symbols The total number of transmitted symbols is the bit error rate. When the bit error rate exceeds the preset threshold, a system bit error alarm is triggered.

[0066] The environmental monitoring instruments are temperature and humidity sensors, which are installed at various key nodes along the optical cable line to collect temperature and humidity data at each node in real time.

[0067] The positioning terminal consists of a satellite positioning system and a signal strength sensor. It is used to acquire the location coordinate data of key nodes or fault points, monitor the signal strength in real time, and calculate the positioning signal coefficient of the coordinate point based on the signal strength.

[0068] It should be noted that distributed fiber optic acoustic sensing DNS devices, environmental monitoring instruments, positioning terminals, and AI analysis terminals are deployed along the port area's communication optical cables to collect real-time optical cable status signal data at various key locations.

[0069] S2, the AI ​​analysis terminal performs quantitative preprocessing on the acoustic vibration intensity parameters, environmental parameters and positioning signal parameters of the area of ​​interest, and generates vibration energy entropy coefficient, environmental stress coefficient and positioning signal coefficient respectively, and analyzes the fault causes of SDH alarm information and system error information.

[0070] Quantitative preprocessing of acoustic vibration intensity parameters: Statistical analysis is performed on the M frames of each vibration data set, and the vibration energy entropy coefficient is used as the parameter. Characterizing the energy distribution of vibration signals:

[0071] ;

[0072]

[0073]

[0074] in, Let k be the energy probability of the m-th frame, and k be the energy sensitivity factor, with a value range of 1. The default value is 1.5. Let be the energy of the m-th frame, L be the number of sampling points in each frame, and s(n) be the signal amplitude of the n-th sampling point.

[0075] Vibration energy entropy coefficient The value of the vibration energy entropy coefficient reflects the distribution pattern of the vibration signal. A larger value indicates a more dispersed energy distribution of the optical cable's vibration signal, suggesting multi-source interference or abnormal impact, and a greater risk of cable breakage. The smaller the value, the more balanced the energy distribution of the optical cable vibration signal, indicating that the optical cable is in good working condition;

[0076] Quantitative preprocessing of environmental parameters: The impact of environmental parameters on optical cables manifests as cumulative stress caused by changes in temperature and humidity, expressed as the environmental stress coefficient. Perform comprehensive characterization:

[0077] ;

[0078] ;

[0079] ;

[0080] in, and These are the weighting coefficients for temperature and humidity, respectively. + =1; For temperature-accumulated stress, This is the temperature influence coefficient. for Instantaneous temperature value at a given time point , A represents the standard operating temperature of the optical cable, and A represents the amplitude of temperature fluctuation. For the current time, This is the initial time for the statistics; For humidity-accumulated stress, Humidity influence coefficient for Instantaneous humidity value at a given time , B represents the standard operating humidity for the optical cable, and B represents the amplitude of humidity fluctuation.

[0081] Environmental stress coefficient The larger the value, the greater the cumulative stress of environmental changes, indicating that the environment has a more serious impact on the optical cable and the greater the risk of optical cable failure or breakage.

[0082] Quantization preprocessing of positioning signal parameters: The accuracy of the positioning signal is characterized by the positioning accuracy coefficient C, taking into account the inherent error of the positioning terminal device, signal strength, and the impact of signal fluctuation on positioning.

[0083] ;

[0084] Signal strength influence factor ;

[0085] Signal fluctuation influencing factors ;

[0086] Equipment error impact factor ;

[0087] in, , and These are the weight coefficients of each influencing factor, and + + =1; The signal strength of the current acquisition frame. This represents the maximum signal strength. The signal conditioning coefficient, The standard deviation of the signal fluctuation; This refers to the adjustment coefficient for the positioning terminal equipment;

[0088] Due to the inherent error of the positioning terminal equipment, , The mean error is given for a known location ( Perform n measurements and calculate the coordinate values ​​for each measurement. ) error , ,but , The standard deviation of the error is... ;

[0089] Signal strength influence factor The higher the value, the closer the current signal strength is to its maximum value, indicating a more stable signal transmission and higher positioning accuracy; signal fluctuation influence factor This represents signal fluctuations over a period of time; a higher value indicates a more stable signal and higher positioning reliability. It also reflects the inherent error of the positioning terminal device. The larger the value, the smaller the equipment error; the positioning reliability coefficient C quantifies the influence of signal strength, signal fluctuation and equipment error. The larger the value, the higher the positioning reliability, the more accurate and stable the positioning results, and the higher the credibility of the positioning data.

[0090] The intelligent operation and maintenance center integrates vibration energy entropy coefficient, environmental stress coefficient, and location information coefficient to analyze the fault causes of SDH alarm information and system error information of SDH transmission equipment. The analysis method is as follows:

[0091] 1) If an R-Los alarm occurs, and the vibration energy entropy coefficient changes abruptly and the location accuracy coefficient is abnormally high, and the historical fault data of the fault location shows a breakage, then the fault of the node is determined to be a fiber optic cable breakage. The location coordinates of the fault point are sent to the operation and maintenance module, and the operation and maintenance node path planning is carried out in combination with the operation and maintenance priority list. The abrupt change refers to a significant abnormal change in the vibration energy entropy coefficient in a very short period of time.

[0092] 2) For system bit error alarms, calculate the bit error rate (BER). If the BER fluctuates periodically and is related to the temperature or humidity changes in environmental parameters, consider the environmental stress coefficient during that time period. The change in environmental stress coefficient When the environmental stress threshold is exceeded, the bit error fault is determined to be caused by environmental factors.

[0093] 3) For system error alarms, if the bit error rate suddenly increases without obvious periodicity and no R-Los alarm occurs, but the optical power monitoring of the SDH transmission equipment shows abnormal fluctuations in optical power, then the fault is determined to be optical cable aging or optical cable fault.

[0094] S3. By integrating the vibration energy entropy coefficient, environmental stress coefficient, and location confidence coefficient, a health assessment model for optical cables is established, and the health index of the optical cables is calculated. ;

[0095] ;

[0096] in, It is a non-linear adjustment factor. The default value is 0.5; For health threshold, , Based on the adjustment of port area importance, its default value is 0.5, and the core area should be appropriately increased. value; , and These are the dynamic weighting coefficients for the vibration energy entropy coefficient, environmental stress coefficient, and positional information coefficient, respectively, with an initial default value of [value missing]. =0.65, =0.25, =0.1;

[0097] Health Index The smaller the value, the worse the health of the optical cable; conversely, the larger the value, the healthier the cable.

[0098] S4. The maintenance strategy knowledge base records the operation and maintenance strategies that match different optical cable health states, sets health index thresholds and operation and maintenance thresholds for each level of optical cable, and different health indices trigger different response measures, as shown in the example in Table 1.

[0099] Table 1. Response Measures for the Health Status Classification of Optical Cables

[0100]

[0101] For maintenance strategies employing a three-tiered compensation response, the intelligent maintenance hub generates a maintenance priority list and sends it to the corresponding maintenance modules. Multiple maintenance modules are set up at key nodes along the port area's communication optical cables, including dispatch terminals, maintenance equipment, and maintenance personnel. The maintenance equipment includes drone swarms and intelligent robots. Based on maintenance instructions, the dispatch terminal initiates the three-tiered compensation measures, specifically including the following:

[0102] The first level is: the dispatch terminal starts the backup fiber optic link to switch data routes. The backup fiber optic link adopts dynamic load balancing technology, with a switching time of ≤50ms. It can support multi-link aggregation and has an automatic recovery mechanism. The fault recovery time is ≤2 minutes and the total bandwidth is ≥10Gbps.

[0103] The second level is: the dispatch terminal dispatches a swarm of drones to conduct thermal imaging inspections along the fault area, search for and determine the fault point;

[0104] The third level involves: deploying intelligent robots at the dispatch terminal to perform local fiber optic splicing at the fault point to repair the fault; recording the repair data; and updating the cloud database.

[0105] It is important to know that if the compensation measures do not meet expectations, the dispatch terminal will automatically dispatch maintenance personnel to perform in-depth repairs. After the repairs are completed, the maintenance personnel will update the cloud database.

[0106] S5. There are multiple maintenance nodes that require compensation measures. The intelligent maintenance center generates a maintenance priority list and assigns corresponding priority cost impact values. The scheduling terminal comprehensively accesses the priority cost impact value, distance cost, time cost and risk cost of the nodes to calculate the node edge weights and performs maintenance path planning based on the Dijkstra algorithm.

[0107] The maintenance priority list is determined based on the Mamdani inference method. Its impact items include the scope of fault impact, service interruption level, environmental risk level and equipment availability status. In a specific embodiment of this example, as shown in Tables 2-5, the membership function formula of each impact item is established to determine its fuzzy set.

[0108] Table 2 Fuzzy Set of Fault Impact Range

[0109]

[0110] Table 3 Fuzzy Set Table of Service Interruption Levels

[0111]

[0112] Table 4 Fuzzy Set Table of Environmental Risk Levels

[0113]

[0114] Table 5. Fuzzy Set of Equipment Availability Status

[0115]

[0116] Tables 2-5 not only give the definition of the fuzzy set of the influencing items, but also the membership function formula. By calculating the membership degree of the influencing items, it is convenient to arrange the priority queue of multiple operation and maintenance nodes within the same triggering rule according to the membership degree. The preset rule base is used for reasoning and outputs the maintenance priority. In a specific embodiment of this example, the preset rule base is shown in Table 6.

[0117] Table 6 Rule Base Table

[0118]

[0119] Based on the maintenance priority list output by the intelligent operation and maintenance center, the scheduling terminal of the operation and maintenance module retrieves the location information of the optical cable fault point, including the location information of the fault point where the fault is determined to be a break by fault analysis, determines the target location of each maintenance task, and plans the best path; at the same time, by communicating with the satellite positioning system, it obtains the location information of the equipment or personnel (such as intelligent robots, drone swarms and maintenance personnel) performing the maintenance task, and uses them as the starting point for path planning.

[0120] Quantify the maintenance priority of operation and maintenance nodes into cost influencing factors. Among them, maintenance priority is an urgent cost factor. =0.3, a cost impact factor with high maintenance priority. =0.6, a cost impact factor with low maintenance priority. =0.9, establish the multi-objective cost function for path planning:

[0121]

[0122] Where a, b, and c are the weighting coefficients of each cost impact item, a + b + c = 1, with default values ​​of a = 0.7, b = 0.2, and c = 0.1. The physical distance from the current location to the access node. The time taken to reach the access node from the current location. The comprehensive risk assessment value during the execution of this path, The default value is 0.5;

[0123] The dispatch terminal uses the cost value as the edge weight for reaching the access node from the current location. It integrates the port area's geographical information, road distribution, optical cable laying lines, and access rules for each area to construct a feasible path network graph that includes all operation and maintenance nodes. It uses Dijkstra's algorithm to calculate the lowest cost path from the starting point to each operation and maintenance node, generates a detailed path planning scheme, and dispatches a swarm of drones to perform thermal imaging inspections and intelligent robots to perform fiber optic splicing operations in real time. If the compensation measures do not meet expectations, it dispatches maintenance personnel to perform in-depth maintenance. After the task is completed, it uploads the maintenance path, time, fault handling details, and equipment status analysis results to the database storage module.

[0124] S6. After completing the repair of the operation and maintenance nodes, update and optimize the health assessment model of the intelligent operation and maintenance center based on the repair result data and historical data.

[0125] Specifically, a loss function is established, and after the maintenance nodes are repaired, the true health index is reassessed. The repaired collected data is used as a partial sample, combined with historical sample data, and the nonlinear adjustment factor, health threshold, and various dynamic weight coefficients are gradually adjusted according to the error gradient direction. The prediction error is minimized using the gradient descent algorithm, thus optimizing and updating the health assessment model.

[0126] Loss function: ;

[0127] Where n represents the training samples, which come from the repaired node data and historical node data; This is the predicted health index value based on the data of the i-th sample. This represents the true value of the health index based on the i-th sample data.

[0128] Through the above description of the embodiments, those skilled in the art can clearly understand that the various embodiments of this application can be implemented by means of software or software combined with necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware functions. Based on this understanding, the technical solution of this application, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions to cause a computer device, such as including but not limited to a personal computer, server, or network device, to execute all or part of the steps of the method described in any embodiment of this application.

[0129] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent operation and maintenance of optical fiber communication cables in port areas based on AI technology, characterized in that, Includes the following: S1. Real-time acquisition of acoustic vibration intensity parameters, environmental parameters, and positioning signal parameters of the optical cable; S2. Perform quantitative preprocessing on the acoustic vibration intensity parameter, the environmental parameter and the positioning signal parameter to generate the vibration energy entropy coefficient, the environmental stress coefficient and the positioning signal coefficient, respectively. S3. Establish an optical cable health assessment model that integrates the vibration energy entropy coefficient, the environmental stress coefficient, and the positioning confidence coefficient, and generate a health index; S4. Preset health index thresholds and response measures, determine the operation and maintenance nodes based on the health index values, and trigger the response measures; S5. If there are multiple maintenance nodes, a maintenance priority list is generated based on the fault impact range, service interruption level, environmental risk level and equipment availability status. Maintenance path planning is performed on the maintenance nodes. Node edge weights are generated by combining the maintenance priority, distance cost, time cost and risk cost of the access nodes. The maintenance path planning is implemented based on the Dijkstra algorithm. S6. Perform repairs on each of the maintenance nodes according to the maintenance path, and update the optical cable health assessment model after the processing is completed.

2. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 1, characterized in that: The acoustic vibration intensity parameter includes the signal amplitude. Statistics are performed on the M frames of vibration signals for each set of vibration data, and the vibration energy entropy coefficient is used as the parameter. Characterizing the energy distribution of the vibration signal: ； ； ； in, Let k be the energy probability of the m-th frame, and k be the energy sensitivity factor. Let be the energy of the m-th frame, L be the number of signal sampling points per frame, and s(n) be the signal amplitude of the n-th sampling point.

3. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 2, characterized in that: The environmental parameters include temperature and humidity. The impact of these environmental parameters on the optical cable is manifested as the cumulative stress caused by changes in temperature and humidity, expressed as the environmental stress coefficient. Perform comprehensive characterization: ； ； ； in, and These are the weighting coefficients for temperature and humidity, respectively. + =1; For temperature-accumulated stress, This is the temperature influence coefficient. for Instantaneous temperature value at a given time point. This refers to the standard operating temperature of the optical cable. For the current time, This is the initial time for the statistics; For humidity-accumulated stress, Humidity influence coefficient for Instantaneous humidity value at a given time.

4. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 3, characterized in that: Considering the inherent errors of the positioning terminal equipment, and calculating the positioning accuracy coefficient C based on signal strength and signal fluctuation: ； Signal strength influence factor ; Signal fluctuation influencing factors ; Equipment error impact factor ; in, , and These are the weight coefficients of each influencing factor, and + + =1; The signal strength of the current acquisition frame. This represents the maximum signal strength. The signal conditioning coefficient, The standard deviation of the signal fluctuation; This is the adjustment coefficient for the positioning terminal device. This refers to the inherent error of the positioning terminal device.

5. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 4, characterized in that: The optical cable health assessment model is as follows: ； in, The health index, It is a non-linear adjustment factor. For health threshold, , and These are the dynamic weighting coefficients for the vibration energy entropy coefficient, the environmental stress coefficient, and the positional confidence coefficient, respectively.

6. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 5, characterized in that: When the health index If the health index threshold is ≤, a level 3 compensation measure will be triggered, specifically: The first level is: activate the backup fiber optic link to switch data routes; The second level involves using thermal imaging technology to inspect along the line and search for fiber optic cable fault points. The third level is to perform local fiber optic splicing to repair the fault points in the optical cable.

7. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 5, characterized in that: It also includes fault analysis, collecting SDH alarm information, system error information, and optical power information data, and performing fault analysis by integrating the vibration energy entropy coefficient, the environmental stress coefficient, and the positioning signal coefficient. The analysis methods include: 1) If an R-Los alarm occurs, and the vibration energy entropy coefficient changes abruptly and the positioning information coefficient is abnormally high, and there is a breakage in the historical fault data of the fault reporting point, then the fault of the fault reporting point is determined to be a fiber optic cable breakage. 2) For system error alarms, calculate the bit error rate. If the bit error rate fluctuates periodically and is related to the change period of the environmental parameters, and combined with the change of the environmental stress coefficient during the time period, when the environmental stress coefficient exceeds the environmental stress threshold, the bit error fault is determined to be a fault caused by environmental influence. 3) For the system error alarm, if the bit error rate suddenly increases without obvious periodicity and no R-Los alarm occurs, but the optical power fluctuates abnormally, then the fault point is determined to be optical cable aging or optical cable fault.

8. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 6, characterized in that: The maintenance priority list is created based on the Mamdani fuzzy inference method. Fuzzy sets are assigned to the fault impact range, the service interruption level, the environmental risk level, and the equipment availability status. The rule base for the maintenance priority is established by combining the various fuzzy sets, and the maintenance priority is matched to each of the operation and maintenance nodes. Different maintenance priorities are matched with different cost influencing factors. A multi-objective cost function is established, with the cost value used as the node edge weight from the current location to the access node, and the operation and maintenance path is planned based on Dijkstra's algorithm. ； Where a, b, and c are the weighting coefficients for the distance cost, the time cost, and the risk cost, respectively. The physical distance from the current location to the access node. The time taken to reach the access node from the current location. This is the comprehensive risk assessment value during the process of reaching the access node from the current location.

9. The intelligent operation and maintenance method for port communication optical cables based on AI technology according to claim 5, characterized in that: After the maintenance node is repaired, the true health index of the optical cable is reassessed, a loss function is established, and the nonlinear adjustment factor, the health threshold, and the dynamic weight coefficient are gradually adjusted according to the gradient direction of the error. The prediction error is then minimized using the gradient descent algorithm. Optimize and update the optical cable health assessment model; wherein, Loss function: ; Where n represents the training samples, which come from the repaired node data and historical node data; This is the predicted health index value based on the data of the i-th sample. This represents the true value of the health index based on the i-th sample data.

10. An intelligent operation and maintenance system for optical fiber cables in port areas, characterized in that: It includes a data acquisition module, an AI analysis terminal, a database storage module, an intelligent operation and maintenance center, and an operation and maintenance module; among which, The data acquisition module includes: an acoustic wave sensing DNS device, an SDH transmission device, an environmental monitoring instrument, and a positioning terminal; The AI ​​analysis terminal is used to calculate the vibration energy entropy coefficient, environmental stress coefficient, and location information coefficient based on real-time monitoring data, and send them to the intelligent operation and maintenance center. The database storage module includes: a time-series database for receiving and storing the real-time monitoring data uploaded by the data acquisition module; a relational database for storing the operation and maintenance data uploaded by the operation and maintenance module, as well as the parameters and definitions of each algorithm model; and an optical cable status feature database, which includes optical cable abnormal event types, fault location mapping relationships, and a maintenance strategy knowledge base. The intelligent operation and maintenance center is used to receive and process the real-time monitoring data from the data acquisition module, calculate the health index based on the vibration energy entropy coefficient, the environmental stress coefficient, and the positioning information coefficient, and generate a maintenance priority list; the intelligent operation and maintenance center analyzes the causes of failures based on the coefficients and the information data from the acoustic wave sensor DNS device. The operation and maintenance module includes: a scheduling terminal, used to receive the maintenance priority list and the positioning information of the positioning terminal, and generate scheduling instructions according to the built-in cost analysis algorithm and operation and maintenance strategy; a drone swarm, used to perform thermal imaging inspection of optical cables according to the scheduling instructions; an intelligent robot, used to perform fiber optic splice repair work according to the scheduling instructions; and maintenance personnel, used to perform in-depth repair work according to the scheduling instructions.

Images