Intelligent diagnosis and maintenance system for electric power communication optical cable

Abstract

The invention relates to the technical field of optical cable fault diagnosis, in particular to an intelligent diagnosis and maintenance system for an electric power communication optical cable. Comprising an optical cable monitoring module; a data processing module; the fault positioning module is used for positioning an optical cable fault position based on the data of the optical cable monitoring module and the physical structure and layout information of the optical cable, extracting fault features by comparing monitoring data changes before and after the fault, and identifying a fault type according to an optical cable performance evaluation model; and the maintenance strategy generation module is used for generating an optimal maintenance strategy according to the fault positioning result and the real-time state of the optical cable, and dynamically adjusting transmission power, wavelength allocation and a coding mode according to the performance change of the optical cable. According to the intelligent diagnosis and maintenance system for the electric power communication optical cable, the optical cable is monitored in real time, the transmission burden is reduced, the maintenance cost is reduced, the abnormal state and potential faults of the optical cable are intelligently diagnosed, and the stability of data transmission is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of optical cable fault diagnosis, and particularly to an intelligent diagnosis and maintenance system for power communication optical cables. Background Art

[0002] With the rapid development of the power communication network, as the cornerstone of information transmission, the stability and reliability of optical cables are directly related to the safe and efficient operation of the power grid. However, during long-term use, optical cables are susceptible to multiple factors such as external force damage, environmental erosion, and natural aging, resulting in performance degradation or even failures. Traditional optical cable maintenance methods rely on manual inspections, which are not only inefficient but also difficult to capture the abnormal status of optical cables in real time. Especially in complex or inaccessible environments, the discovery and handling of faults are often delayed. In addition, the communication quality of optical cables, such as optical fiber loss, directly affects the efficiency and stability of data transmission. Excessive loss not only increases the transmission burden but may also trigger system failures.

[0003] For example, in the patent with the publication number CN116819394A and the name "A Method and System for Aging Diagnosis and Monitoring of Power Cables", the grounding current of the outer sheath of the cable is monitored to determine whether the grounding current of the outer sheath is abnormal; when it is determined that the grounding current of the outer sheath is abnormal, the fault location of the cable is located; the temperature monitoring device corresponding to the fault location of the cable is started; according to the thermal imaging map feedback by the temperature monitoring device, the type of cable fault is determined; if the type of cable fault is a cable body fault, the test current of the cable when connected to a specified voltage is monitored; according to the test current, the aging degree of the cable is analyzed, and a corresponding maintenance strategy is generated. After determining the existence of a fault, the current between the cable core and the spiral steel tape is further measured. The disadvantage is that it cannot monitor optical cables in real time and has a large transmission burden. Summary of the Invention

[0004] Aiming at the problems in the existing power communication optical cable maintenance system that it cannot monitor optical cables in real time and has a large transmission burden, the present invention provides an intelligent diagnosis and maintenance system for power communication optical cables, which can monitor optical cables in real time, reduce the transmission burden, lower the maintenance cost, realize the intelligent diagnosis of the abnormal status and potential faults of optical cables, and ensure the stability of data transmission.

[0005] To achieve the above technical objectives, a technical solution provided by the present invention is an intelligent diagnosis and maintenance system for power communication optical cables, including: An optical cable monitoring module that monitors the temperature, stress, vibration, bending degree, and external environment parameters of the optical fiber in real time, captures real-time images of the surface of the optical cable through a camera, and identifies the abnormal status of the surface of the optical cable according to the image processing method; A data processing module that receives data transmitted by the optical cable monitoring module, preprocesses the data through machine learning algorithms, identifies abnormal states and potential fault types, establishes an optical cable performance evaluation model based on historical data and real-time monitoring data, and predicts potential fault points of the optical cable; A fault location module that locates the fault position of the optical cable based on the data of the optical cable monitoring module, the physical structure of the optical cable, and the layout information, extracts fault features by comparing the changes in monitoring data before and after the fault, and identifies the fault type according to the optical cable performance evaluation model; A maintenance strategy generation module that generates an optimal maintenance strategy according to the fault location result and the real-time state of the optical cable, and dynamically adjusts the transmission power, wavelength allocation, and coding method according to the performance change of the optical cable.

[0006] In this technical solution, by integrating distributed fiber optic sensing technology and image recognition capabilities, the system can continuously monitor the temperature, stress, vibration, bending degree of the optical fiber and external environmental changes. At the same time, it uses a camera to capture the surface details of the optical cable, instantly analyzes and identifies any abnormal signs to ensure a comprehensive grasp of the optical cable status. Using machine learning algorithms to deeply mine the monitoring data, it can not only identify the abnormal state and potential faults of the current optical cable, but also build an optical cable performance prediction model based on historical data and real-time information to predict potential fault points in the future, pre-position the maintenance work, and effectively reduce the risk of sudden faults. The fault location module locks the exact position of the optical cable fault based on the data analysis results, and identifies the fault type by comparing and analyzing the data changes before and after the fault, providing strong support for rapid repair. The maintenance strategy generation module intelligently generates an optimized maintenance plan based on the fault location and the real-time state of the optical cable. By dynamically adjusting the transmission power, wavelength allocation, and coding method, the system can effectively reduce the transmission burden of the optical cable, extend its service life, reduce maintenance costs, and ensure the efficient and stable data transmission.

[0007] The present invention is further set as: The image processing method includes: Perform Gaussian filtering denoising on the captured optical cable surface image; Enhance the contrast, adjust the brightness, detect the positions where the gray value changes violently in the image and the gray changes of the surface texture through an edge detection algorithm, and extract the contour and texture information of the optical cable surface; Detect the straight line segments and circular features on the optical cable surface through a shape analysis algorithm, extract the characteristic parameters related to the damaged state, and identify the abnormal shape on the optical cable surface.

[0008] In this technical solution, the system uses Gaussian filtering technology to denoise the captured image of the optical cable surface, ensuring that the image quality is clear and interference-free. Then, through contrast enhancement and brightness adjustment, the subtle damage marks become particularly obvious after adjustment, facilitating subsequent identification. Subsequently, the system uses an edge detection algorithm to extract the contour and texture information of the optical cable surface. Further, through a shape analysis algorithm, the system can automatically detect the straight line segments and circular features on the optical cable surface, which are closely related to the damage state of the optical cable. Finally, based on the extracted feature parameters, the system intelligently identifies abnormal shapes on the optical cable surface, such as cracks, abrasions, or deformations, providing strong support for timely discovery and handling of potential problems with the optical cable.

[0009] The present invention is further set as follows: The preprocessing of data, identification of abnormal states, and potential fault types through machine learning algorithms include: Clean the original data transmitted by the optical cable monitoring module, perform normalization or standardization processing on the data, and extract the statistical features, time-domain features, and frequency-domain features of the data to construct a feature vector. Use the training data set to train the machine learning algorithm to obtain an initial model, optimize the model through cross-validation and grid search methods, and select the optimal model according to the performance indicators of the model. Input the preprocessed data into the optimal model for anomaly detection. The model determines whether the data is abnormal based on the feature vector of the data. Analyze the features of the data marked as abnormal, and combine the actual situation and historical data of the optical cable to identify specific abnormal states and potential fault types.

[0010] In this technical solution, first, the system comprehensively cleans the original data transmitted by the optical cable monitoring module to ensure the purity and accuracy of the data. Subsequently, through normalization or standardization processing, the data is unified to the same dimension for subsequent analysis. On this basis, the system extracts the statistical features, time-domain features, and frequency-domain features of the data to construct a feature vector rich in information, providing rich input information for the machine learning model. Next, use the training data set to comprehensively train the selected machine learning algorithm to initially obtain a prediction model. To further improve the performance of the model, cross-validation and grid search optimization methods are used to adjust the model parameters, and the optimal model is selected according to the performance indicators of the model to ensure the accuracy and stability of the model in identifying abnormal states of the optical cable. Then, the system inputs the preprocessed data into this optimal model for efficient anomaly detection. The model intelligently determines whether the data is abnormal based on the feature vector of the data, providing immediate feedback on the health status of the optical cable. Finally, for the data marked as abnormal by the model, the system analyzes its features and combines the actual operating conditions and historical data of the optical cable to identify specific abnormal states and potential fault types.

[0011] The present invention is further configured such that the steps of establishing an optical cable performance evaluation model based on historical data and real-time monitoring data to predict potential fault points of the optical cable include: Integrate historical data and real-time monitoring data to form a data set containing time series, and perform cleaning and standardization processing on the data set; Extract the statistics of the features that affect the optical cable performance evaluation from the data set, and use a feature selection algorithm to screen the features related to the model prediction performance; Construct an optical cable performance evaluation model according to a machine learning algorithm, train the model with historical data, optimize the prediction performance of the model by adjusting the model parameters and structure, and use a cross-validation method to verify the model to evaluate the generalization ability of the model; Input the real-time monitoring data into the trained performance evaluation model to obtain the evaluation result of the current state of the optical cable, and analyze the potential fault points of the optical cable according to the evaluation result in combination with the physical characteristics and operating environment factors of the optical cable.

[0012] In this technical solution, first, the historical data and real-time monitoring data are seamlessly integrated to form a huge data set covering time series. To ensure data quality, strict cleaning and standardization processing are performed on the data set, laying a solid foundation for subsequent analysis. Then, the statistical quantities of the features crucial for the optical cable performance evaluation in the data set are mined. With the help of a feature selection algorithm, the key features closely related to the model prediction performance are screened out, providing an information cornerstone for model construction. On this basis, an optical cable performance evaluation model is constructed relying on a machine learning algorithm. By deeply training the model with rich historical data and continuously adjusting the model parameters and structure, the prediction performance of the model is optimized. At the same time, a cross-validation method is used to comprehensively verify the model to ensure its good generalization ability and its ability to handle various complex scenarios. Finally, the real-time monitoring data is continuously input into this trained performance evaluation model to immediately obtain a comprehensive evaluation result of the current state of the optical cable. Combining multi-dimensional information such as the physical characteristics and operating environment factors of the optical cable, the potential fault points of the optical cable are analyzed and predicted, providing strong support for taking timely maintenance measures and ensuring the stable operation of the power communication network.

[0013] The present invention is further configured such that the steps of positioning the optical cable fault location based on the data of the optical cable monitoring module, the physical structure of the optical cable, and the layout information include: Extract the features related to the optical cable fault from the data preprocessed by the data processing module, identify the fault features from the extracted features according to a machine learning algorithm, and classify the fault features in combination with the physical structure, layout information, and historical fault data of the optical cable; The extracted fault features are input into the fault location algorithm to determine the range of the fault point and generate an optical cable fault distribution map.

[0014] In this technical solution, first, the system extracts key features closely related to optical cable faults from the massive data preprocessed by the data processing module. With the help of machine learning algorithms, the fault signals in these features can be identified, providing a basis for fault location. Subsequently, these fault features are deeply integrated and compared with the physical structure, layout information, and historical fault data of the optical cable. The fault features are carefully classified, not only clarifying the type of the fault but also revealing the background and reasons for the possible occurrence of the fault, providing strong support for subsequent fault handling. Finally, the system inputs the extracted fault features into the fault location algorithm, which can lock the approximate range of the fault point and generate an intuitive optical cable fault distribution map. It not only clearly shows the geographical location of the fault point but also presents the severity and influence range of the fault through colors and markings, facilitating rapid response and precise repair by maintenance personnel.

[0015] The present invention is further set as follows: extracting fault features by comparing the changes in monitoring data before and after the fault, and identifying the fault type according to the optical cable performance evaluation model includes: Obtain the monitoring data before and after the fault from the data processing module and align the obtained data in terms of time; Combined with the physical characteristics and operating environment of the optical cable, analyze the potential correlation between the data changes before and after the fault and the optical cable fault; extract the features related to the fault type from the data changes, and construct a fault type recognition model according to the machine learning algorithm; Use the monitoring data of known fault types to train the fault type recognition model, input the change features of the monitoring data before and after the fault into the trained fault type recognition model, and obtain the recognition result of the fault type; According to the recognition result, classify the fault into breakage, abrasion, corrosion, and external interference.

[0016] In this technical solution, first, the system obtains the monitoring data before and after the fault from the data processing module, and performs time alignment to ensure the time consistency between the data, laying a solid foundation for subsequent analysis. Next, closely combining the physical characteristics of the optical cable and the actual operating environment, the potential laws of data changes before and after the fault are deeply explored. By analyzing the intrinsic connection between these data and the optical cable fault, a scientific basis is provided for the identification of the fault type. On this basis, the system extracts the key features closely related to the fault type from the data changes, and uses the machine learning algorithm to build an accurate fault type identification model. In order to further improve the recognition ability of the model, the model is fully trained using the monitoring data of known fault types to ensure its stability and accuracy in practical applications. Subsequently, the characteristics of the monitoring data changes before and after the fault are input into this trained model, and the fault type identification results can be quickly obtained. Finally, according to the recognition results of the model, the fault is accurately classified into types such as fracture, wear, corrosion, and external interference. It not only provides clear fault information for maintenance personnel, but also helps to formulate targeted repair plans, thereby improving maintenance efficiency and ensuring the stability and smoothness of the power communication network.

[0017] The present invention is further configured as follows: generating an optimal maintenance strategy according to the fault location result and the real-time status of the optical cable comprises: Receive and analyze the fault location information provided by the fault location module, and visualize the fault location and surrounding optical cable layout in combination with the optical cable network topology diagram; Obtain the current status data of the optical cable from the real-time monitoring module and use the data analysis algorithm to evaluate the real-time health status of the optical cable; Determine the resources required for maintenance based on fault location results and real-time status assessment; Generate resource allocation strategies based on the availability and distribution of maintenance resources; Based on the fault type, fault location, real-time status of the optical cable and maintenance resource allocation strategy, the optimal maintenance strategy is generated using a decision support algorithm.

[0018] In this technical solution, first, the system receives and analyzes the detailed fault location information provided by the fault location module. Combined with the optical cable network topology diagram, the system displays the fault location and the layout of the surrounding optical cables in an intuitive and visual way. Then, the system obtains the current status data of the optical cable from the real-time monitoring module in real time, which fully reflects the real-time operation status of the optical cable. Using the data analysis algorithm, the system evaluates the health status of the optical cable, providing a scientific basis for formulating maintenance strategies. On the basis of clarifying the fault location and the real-time status of the optical cable, the system intelligently determines the various resources required for maintenance, including personnel, materials and tools, to ensure the smooth progress of maintenance work. At the same time, the system fully considers the availability and distribution of maintenance resources, generates resource allocation strategies, and realizes the optimal allocation and efficient utilization of resources. Finally, based on the multi-dimensional information of fault type, fault location, real-time status of optical cable and maintenance resource allocation strategy, the system uses decision support algorithms to comprehensively weigh various factors and intelligently generate the optimal maintenance strategy, providing maintenance personnel with a clear and feasible maintenance plan to ensure that the fault is handled in a timely and effective manner.

[0019] The present invention is further configured as follows: the method of dynamically adjusting the transmission power, wavelength allocation and encoding method according to the performance change of the optical cable includes: Real-time monitoring of the transmission performance parameters of the optical cable, using data analysis algorithms to analyze the changing trends and abnormal fluctuations of the performance parameters; dynamically adjusting the transmission power of the optical signal according to the attenuation changes and transmission distance of the optical cable; Analyze the transmission performance of different wavelengths in optical cables based on mutual interference between wavelengths; According to the performance analysis results, dynamically adjust the wavelength allocation and select the wavelength with the best performance or the least interference for signal transmission; according to the noise characteristics of the optical cable and the transmission requirements, select or adjust the appropriate encoding method; Combined with the adjustment of transmission power, wavelength allocation and coding method, a multi-objective optimization algorithm is used to minimize the transmission burden and improve the overall transmission performance.

[0020] In this technical solution, first, the system monitors the transmission performance parameters of the optical cable in real time. By using advanced data analysis algorithms, it analyzes the change trends and abnormal fluctuations of the performance parameters, providing data support for subsequent dynamic adjustments. Next, the system adjusts the transmission power of the optical signal according to the attenuation change and transmission distance of the optical cable to ensure that the signal remains strong and stable during transmission. At the same time, the system analyzes the transmission performance of different wavelengths in the optical cable and the mutual interference between wavelengths, enabling the system to accurately identify the wavelength with the best performance or the least interference and perform dynamic allocation, improving the clarity and efficiency of transmission. In addition, the system also intelligently selects or adjusts a suitable coding method according to the noise characteristics and transmission requirements of the optical cable, effectively resisting the interference of noise and ensuring the integrity and accuracy of data. Finally, the system combines the adjustments of transmission power, wavelength allocation, and coding method, and uses a multi-objective optimization algorithm to seek the best balance in multiple dimensions to minimize the transmission burden and improve the overall transmission performance.

[0021] The present invention is further configured such that: the monitoring data before and after the occurrence of the fault includes the voltage, current, optical power, temperature, stress, and vibration of the optical cable.

[0022] The present invention is further configured such that: the selection or adjustment of a suitable coding method includes: when the transmission conditions deteriorate or the noise increases, dynamically switching to a more anti-interference coding method to improve the reliability and stability of transmission.

[0023] The beneficial effects of the present invention are as follows: (1) Monitor the optical cable in real time, reduce the transmission burden, lower the maintenance cost, realize the intelligent diagnosis of the abnormal state and potential faults of the optical cable, and ensure the stability of data transmission; (2) By integrating distributed optical fiber sensing technology and image recognition capabilities, the system can continuously monitor the temperature, stress, vibration, bending degree of the optical fiber and external environment changes. At the same time, it uses a camera to capture the surface details of the optical cable, instantly analyzes and identifies any abnormal signs to ensure a comprehensive grasp of the optical cable state. By using machine learning algorithms to deeply mine the monitoring data, it can not only identify the current abnormal state and potential faults of the optical cable, but also build an optical cable performance prediction model based on historical data and real-time information to predict possible future fault points, bringing forward the maintenance work and effectively reducing the risk of sudden faults; (3) The fault location module locks the exact location of the optical cable fault based on the data analysis results, and identifies the fault type by comparing and analyzing the data changes before and after the fault, providing strong support for rapid repair. The maintenance strategy generation module intelligently generates an optimized maintenance plan based on the fault location and the real-time state of the optical cable. By dynamically adjusting the transmission power, wavelength allocation, and coding method, the system can effectively reduce the transmission burden of the optical cable, extend its service life, and at the same time reduce the maintenance cost to ensure the efficient and stable data transmission. Description of the Drawings

[0024] Figure 1It is a schematic structural diagram of an intelligent diagnosis and maintenance system for power communication optical cables. Detailed implementation manners

[0025] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific implementation manners described herein are only a best embodiment of the present invention, which is only used to explain the present invention and does not limit the protection scope of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] As Figure 1 shown, as the first embodiment of the present invention, an intelligent diagnosis and maintenance system for power communication optical cables includes: an optical cable monitoring module, which monitors the temperature, stress, vibration, bending degree and external environment parameters of the optical fiber in real time, captures real-time images of the surface of the optical cable through a camera, and identifies abnormal states on the surface of the optical cable according to image processing methods; a data processing module, which receives the data transmitted by the optical cable monitoring module, preprocesses the data through machine learning algorithms, identifies abnormal states and potential fault types, establishes an optical cable performance evaluation model based on historical data and real-time monitoring data, and predicts potential fault points of the optical cable; a fault location module, which locates the fault position of the optical cable based on the data of the optical cable monitoring module, the physical structure of the optical cable and layout information, extracts fault features by comparing the changes in monitoring data before and after the fault, and identifies the fault type according to the optical cable performance evaluation model; a maintenance strategy generation module, which generates an optimal maintenance strategy according to the fault location result and the real-time state of the optical cable, and dynamically adjusts the transmission power, wavelength allocation and coding mode according to the performance change of the optical cable.

[0027] In this embodiment, by integrating distributed fiber optic sensing technology with image recognition capabilities, the system can continuously monitor the temperature, stress, vibration, bend, and external environment changes of the optical fiber. At the same time, it uses a camera to capture the details of the optical cable surface, instantly analyze and identify any abnormal signs to ensure comprehensive control of the optical cable status. By applying machine learning algorithms to deeply mine the monitoring data, it can not only identify the abnormal status and potential faults of the current optical cable, but also build a prediction model for the optical cable performance based on historical data and real-time information to predict possible future fault points, advance the maintenance work, and effectively reduce the risk of sudden failures. The fault location module locates the exact position of the optical cable fault based on the data analysis results, and identifies the fault type by comparing and analyzing the data changes before and after the fault, providing strong support for quick repair. The maintenance strategy generation module intelligently generates the most optimized maintenance plan based on the fault location and the real-time status of the optical cable. By dynamically adjusting the transmission power, wavelength allocation, and coding method, the system can effectively reduce the transmission burden of the optical cable, extend its service life, while reducing the maintenance cost and ensuring efficient and stable data transmission.

[0028] It can be understood that the external environment parameters include humidity, wind speed, and air pressure.

[0029] It can be understood that the abnormal states of the optical cable surface include creases, abrasions, and cracks.

[0030] In an embodiment of the present invention, the image processing method includes: Perform Gaussian filtering denoising on the captured optical cable surface image; Enhance the contrast, adjust the brightness, detect the positions where the gray value changes violently and the gray changes of the surface texture in the image through the edge detection algorithm, and extract the contour and texture information of the optical cable surface; Detect the straight line segments and circular features on the optical cable surface through the shape analysis algorithm, extract the characteristic parameters related to the damaged state, and identify the abnormal shape on the optical cable surface.

[0031] In this technical solution, the system uses Gaussian filtering technology to denoise the captured optical cable surface image to ensure clear and interference-free image quality. Then, through contrast enhancement and brightness adjustment, the subtle damage marks become particularly obvious after adjustment, facilitating subsequent identification. Subsequently, the system uses the edge detection algorithm to extract the contour and texture information of the optical cable surface. Further, through the shape analysis algorithm, the system can automatically detect the straight line segments and circular features on the optical cable surface, which are closely related to the damaged state of the optical cable. Finally, the system intelligently identifies the abnormal shape on the optical cable surface, such as cracks, abrasions, or deformations, providing strong support for timely discovery and handling of potential problems of the optical cable.

[0032] It can be understood that the edge detection algorithm is the Sobel operator algorithm or the Canny operator algorithm.

[0033] It can be understood that the shape analysis algorithm is the Hough transform algorithm or the contour detection algorithm.

[0034] In an embodiment of the present invention, the preprocessing of data, identifying abnormal states and potential fault types through machine learning algorithms includes: Cleaning the original data transmitted by the optical cable monitoring module, performing normalization or standardization processing on the data, and extracting the statistical features, time-domain features and frequency-domain features of the data to construct a feature vector; Training the machine learning algorithm using the training data set to obtain an initial model, optimizing the model through cross-validation and grid search methods, and selecting the optimal model according to the performance indicators of the model; Inputting the preprocessed data into the optimal model for anomaly detection, and the model determines whether the data is abnormal according to the feature vector of the data; Analyzing the features of the data marked as abnormal, and combining the actual situation and historical data of the optical cable to identify specific abnormal states and potential fault types.

[0035] In this technical solution, the system comprehensively cleans the original data transmitted by the optical cable monitoring module to ensure the purity and accuracy of the data. Subsequently, through normalization or standardization processing, the data is unified to the same dimension for subsequent analysis. On this basis, the system extracts the statistical features, time-domain features and frequency-domain features of the data, constructs an information-rich feature vector, and provides rich input information for the machine learning model. Next, using the training data set, the selected machine learning algorithm is comprehensively trained to initially obtain a prediction model. In order to further improve the performance of the model, cross-validation and grid search optimization methods are adopted to adjust the model parameters, and according to the performance indicators of the model, the optimal model is selected to ensure the accuracy and stability of the model in identifying the abnormal state of the optical cable. Then, the system inputs the preprocessed data into this optimal model for efficient anomaly detection. The model intelligently determines whether the data is abnormal based on the feature vector of the data, providing immediate feedback on the health status of the optical cable. Finally, for the data marked as abnormal by the model, the system analyzes its features and combines the actual operating conditions and historical data of the optical cable to identify specific abnormal states and potential fault types.

[0036] It can be understood that the statistical features, time-domain features and frequency-domain features of the data include mean, standard deviation, peak-to-peak value, frequency and period.

[0037] In an embodiment of the present invention, the establishment of an optical cable performance evaluation model based on historical data and real-time monitoring data to predict potential fault points of the optical cable includes: Integrate historical data and real-time monitoring data to form a dataset containing time series, and perform cleaning and standardization processing on the dataset; Extract the statistics of the features that affect the optical cable performance evaluation from the dataset, and use the feature selection algorithm to screen the features related to the model prediction performance; Construct an optical cable performance evaluation model according to the machine learning algorithm, use historical data to train the model, adjust the model parameters and structure to optimize the prediction performance of the model, and use the cross-validation method to verify the model to evaluate the generalization ability of the model; Input the real-time monitoring data into the trained performance evaluation model to obtain the evaluation result of the current state of the optical cable. According to the evaluation result, combined with the physical characteristics and operating environment factors of the optical cable, analyze the potential fault points of the optical cable.

[0038] In this technical solution, first, the historical data and real-time monitoring data are seamlessly integrated to form a huge dataset covering time series. To ensure data quality, strict cleaning and standardization processing are performed on the dataset, laying a solid foundation for subsequent analysis. Then, the statistical quantities of the features that are crucial for the optical cable performance evaluation are mined from the dataset. With the help of the feature selection algorithm, the key features closely related to the model prediction performance are screened out, providing an information cornerstone for model construction. On this basis, relying on the machine learning algorithm, an optical cable performance evaluation model is constructed. By using rich historical data to deeply train the model and continuously adjusting the model parameters and structure, the prediction performance of the model is optimized. At the same time, the cross-validation method is used to comprehensively verify the model to ensure its good generalization ability and its ability to handle various complex scenarios. Finally, the real-time monitoring data is continuously input into this trained performance evaluation model to immediately obtain a comprehensive evaluation result of the current state of the optical cable. Combining multi-dimensional information such as the physical characteristics and operating environment factors of the optical cable, analyze and predict the potential fault points of the optical cable, providing strong support for timely taking maintenance measures and ensuring the stable operation of the power communication network.

[0039] It can be understood that the statistics of the features that affect the optical cable performance evaluation include mean, standard deviation, maximum value, minimum value, trend, and periodicity.

[0040] Further, the positioning of the optical cable fault location based on the data of the optical cable monitoring module, the physical structure of the optical cable, and the layout information includes: Extract the features related to the optical cable fault from the data preprocessed by the data processing module, identify the fault features from the extracted features according to the machine learning algorithm, and classify the fault features in combination with the physical structure, layout information, and historical fault data of the optical cable; The extracted fault features are input into the fault location algorithm to determine the scope of the fault point and generate an optical cable fault distribution map.

[0041] In this technical solution, first, the system extracts key features closely related to optical cable faults from the massive data preprocessed by the data processing module. With the help of machine learning algorithms, the fault signals in these features can be identified, providing a basis for fault location. Subsequently, these fault features are deeply integrated and compared with the physical structure, layout information, and historical fault data of the optical cable, and the fault features are carefully classified, not only clarifying the type of fault but also revealing the background and causes of the possible occurrence of the fault, providing strong support for subsequent fault handling. Finally, the system inputs the extracted fault features into the fault location algorithm, which can lock the approximate range of the fault point and generate an intuitive optical cable fault distribution map. It not only clearly shows the geographical location of the fault point but also presents the severity and impact range of the fault through colors and markings, facilitating the rapid response and precise repair of maintenance personnel.

[0042] It can be understood that the fault location algorithm is time domain reflectometry (TDR) or optical time domain reflectometry (OTDR) or correlation analysis method.

[0043] In an embodiment of the present invention, the extraction of fault features by comparing the changes in monitoring data before and after the fault and the identification of fault types according to the optical cable performance evaluation model include: Obtain the monitoring data before and after the fault from the data processing module and align the obtained data in time; Combined with the physical characteristics and operating environment of the optical cable, analyze the potential correlation between the data changes before and after the fault and the optical cable fault; extract the features related to the fault type from the data changes and construct a fault type identification model according to the machine learning algorithm; Use the monitoring data with known fault types to train the fault type identification model, input the monitoring data change features before and after the fault into the trained fault type identification model, and obtain the identification result of the fault type; According to the identification result, classify the fault into breakage, abrasion, corrosion, and external interference.

[0044] In this technical solution, first, the system obtains the monitoring data before and after the fault from the data processing module, and performs time alignment to ensure the time consistency between the data, laying a solid foundation for subsequent analysis. Next, closely combining the physical characteristics of the optical cable and the actual operating environment, the potential laws of data changes before and after the fault are deeply explored. By analyzing the intrinsic connection between these data and the optical cable fault, a scientific basis is provided for the identification of the fault type. On this basis, the system extracts the key features closely related to the fault type from the data changes, and uses the machine learning algorithm to build an accurate fault type identification model. In order to further improve the recognition ability of the model, the model is fully trained using the monitoring data of known fault types to ensure its stability and accuracy in practical applications. Subsequently, the characteristics of the monitoring data changes before and after the fault are input into this trained model, and the fault type identification results can be quickly obtained. Finally, according to the recognition results of the model, the fault is accurately classified into types such as fracture, wear, corrosion, and external interference. It not only provides clear fault information for maintenance personnel, but also helps to formulate targeted repair plans, thereby improving maintenance efficiency and ensuring the stability and smoothness of the power communication network.

[0045] Generating the optimal maintenance strategy according to the fault location result and the real-time status of the optical cable includes: Receive and analyze the fault location information provided by the fault location module, and visualize the fault location and surrounding optical cable layout in combination with the optical cable network topology diagram; Obtain the current status data of the optical cable from the real-time monitoring module and use the data analysis algorithm to evaluate the real-time health status of the optical cable; Determine the resources required for maintenance based on fault location results and real-time status assessment; Generate resource allocation strategies based on the availability and distribution of maintenance resources; Based on the fault type, fault location, real-time status of the optical cable and maintenance resource allocation strategy, the optimal maintenance strategy is generated using a decision support algorithm.

[0046] In this technical solution, first, the system receives and analyzes the detailed fault location information provided by the fault location module. Combined with the optical cable network topology diagram, the system displays the fault location and the layout of the surrounding optical cables in an intuitive and visual way. Then, the system obtains the current status data of the optical cable from the real-time monitoring module in real time, which fully reflects the real-time operation status of the optical cable. Using the data analysis algorithm, the system evaluates the health status of the optical cable, providing a scientific basis for formulating maintenance strategies. On the basis of clarifying the fault location and the real-time status of the optical cable, the system intelligently determines the various resources required for maintenance, including personnel, materials and tools, to ensure the smooth progress of maintenance work. At the same time, the system fully considers the availability and distribution of maintenance resources, generates resource allocation strategies, and realizes the optimal allocation and efficient utilization of resources. Finally, based on the multi-dimensional information of fault type, fault location, real-time status of optical cable and maintenance resource allocation strategy, the system uses decision support algorithms to comprehensively weigh various factors and intelligently generate the optimal maintenance strategy, providing maintenance personnel with a clear and feasible maintenance plan to ensure that the fault is handled in a timely and effective manner.

[0047] It can be understood that the dynamically adjusting transmission power, wavelength allocation and encoding method according to the performance change of the optical cable includes: real-time monitoring of the transmission performance parameters of the optical cable, using data analysis algorithms to analyze the change trend and abnormal fluctuation of the performance parameters; dynamically adjusting the transmission power of the optical signal according to the attenuation change and transmission distance of the optical cable; Analyze the transmission performance of different wavelengths in optical cables based on mutual interference between wavelengths; According to the performance analysis results, dynamically adjust the wavelength allocation and select the wavelength with the best performance or the least interference for signal transmission; according to the noise characteristics of the optical cable and the transmission requirements, select or adjust the appropriate encoding method; Combined with the adjustment of transmission power, wavelength allocation and coding method, a multi-objective optimization algorithm is used to minimize the transmission burden and improve the overall transmission performance.

[0048] First, the system monitors the transmission performance parameters of the optical cable in real time. Using advanced data analysis algorithms, it analyzes the changing trends and abnormal fluctuations of the performance parameters, providing data support for subsequent dynamic adjustments. Next, the system adjusts the transmission power of the optical signal according to the attenuation change and transmission distance of the optical cable to ensure that the signal remains strong and stable during transmission. At the same time, the system analyzes the transmission performance of different wavelengths in the optical cable and the mutual interference between wavelengths, enabling the system to accurately identify the wavelength with the best performance or the least interference and perform dynamic allocation, improving the clarity and efficiency of transmission. In addition, the system also intelligently selects or adjusts the appropriate coding method according to the noise characteristics and transmission requirements of the optical cable, effectively resisting the interference of noise and ensuring the integrity and accuracy of data. Finally, the system combines the adjustments of transmission power, wavelength allocation, and coding method, and uses a multi-objective optimization algorithm to seek the best balance in multiple dimensions to minimize the transmission burden and improve the overall transmission performance.

[0049] It can be understood that the formula for the optimal maintenance strategy is: Minimize[Cost(x), Downtime(x), Burden(x)]; Subject to x∈X; Where x is a vector representing the maintenance strategy, X is the set of all possible strategies, and Cost(x), Downtime(x), and Burden(x) are the cost of the maintenance strategy, the resulting downtime of the optical cable, and the transmission burden respectively. It is transformed into a multi-objective optimization problem for solution.

[0050] The monitoring data before and after the fault includes the voltage, current, optical power, temperature, stress, and vibration of the optical cable.

[0051] The selection or adjustment of the appropriate coding method includes: when the transmission conditions deteriorate or the noise increases, dynamically switching to a more interference-resistant coding method to improve the reliability and stability of transmission.

[0052] It can be understood that the machine learning algorithm is a support vector machine (SVM) or a decision tree or a random forest or a neural network or an isolation forest or a local outlier factor (LOF).

[0053] The above embodiments' specific description of the present invention is only for further illustration of the present invention and cannot be understood as a limitation on the protection scope of the present invention. Any non-essential improvements and adjustments made by those skilled in the art based on the content of the above invention fall within the protection scope of the present invention.

Claims

1. Intelligent diagnosis and maintenance system for power communication optical cables, characterized in that: include: The optical cable monitoring module monitors the temperature, stress, vibration, curvature and external environmental parameters of the optical fiber in real time, captures the real-time image of the optical cable surface through the camera, and identifies the abnormal state of the optical cable surface based on the image processing method; The data processing module receives the data transmitted by the optical cable monitoring module, pre-processes the data through machine learning algorithms, identifies abnormal conditions and potential fault types, establishes an optical cable performance evaluation model based on historical data and real-time monitoring data, and predicts potential fault points of the optical cable; The fault location module locates the fault location of the optical cable based on the data of the optical cable monitoring module, the physical structure and layout information of the optical cable, extracts the fault characteristics by comparing the changes in monitoring data before and after the fault, and identifies the fault type according to the optical cable performance evaluation model; The maintenance strategy generation module generates the optimal maintenance strategy based on the fault location results and the real-time status of the optical cable, and dynamically adjusts the transmission power, wavelength allocation and encoding method according to the performance changes of the optical cable.

2. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 1 is characterized in that: The image processing method comprises: Perform Gaussian filtering to remove noise on the captured optical cable surface image; Enhance contrast, adjust brightness, detect locations with drastic grayscale value changes and grayscale changes of surface textures in the image through edge detection algorithms, and extract the contour and texture information of the cable surface; The shape analysis algorithm is used to detect the straight line segments and circular features on the surface of the optical cable, extract the characteristic parameters related to the damage state, and identify the abnormal shape of the optical cable surface.

3. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 2 is characterized in that: The method of preprocessing data and identifying abnormal conditions and potential fault types by machine learning algorithms includes: Clean the raw data transmitted by the optical cable monitoring module, normalize or standardize the data, extract the statistical features, time domain features and frequency domain features of the data to construct a feature vector; Use the training data set to train the machine learning algorithm to obtain the initial model, optimize the model through cross-validation and grid search methods, and select the optimal model based on the model's performance indicators; The preprocessed data is input into the optimal model for anomaly detection. The model determines whether the data is abnormal based on the feature vector of the data. The data marked as abnormal are analyzed for features, and combined with the actual situation and historical data of the optical cable, specific abnormal conditions and potential fault types are identified.

4. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 3 is characterized in that: The method of establishing an optical cable performance evaluation model based on historical data and real-time monitoring data to predict potential failure points of the optical cable includes: integrating historical data and real-time monitoring data to form a data set containing a time series, and cleaning and standardizing the data set; Extract statistics of features that have an impact on the performance evaluation of optical cables from the data set, and use feature selection algorithms to screen features related to model prediction performance; Construct an optical cable performance evaluation model based on machine learning algorithms, use historical data to train the model, adjust model parameters and structure to optimize the model's prediction performance, and use cross-validation methods to validate the model to evaluate the model's generalization ability; The real-time monitoring data is input into the trained performance evaluation model to obtain the evaluation results of the current status of the optical cable. Based on the evaluation results, combined with the physical characteristics of the optical cable and operating environment factors, the potential fault points of the optical cable are analyzed.

5. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 4 is characterized in that: The method of locating the optical cable fault position based on the data of the optical cable monitoring module, the physical structure and layout information of the optical cable comprises: Extract features related to optical cable faults from the data preprocessed by the data processing module, identify fault features from the extracted features based on the machine learning algorithm, and classify the fault features based on the physical structure and layout information of the optical cable and historical fault data; The extracted fault features are input into the fault location algorithm to determine the range of the fault point and generate the optical cable fault distribution map.

6. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 3, 4 or 5, characterized in that: The method of extracting fault features by comparing changes in monitoring data before and after the fault and identifying the fault type according to the optical cable performance evaluation model includes: acquiring monitoring data before and after the fault occurs from a data processing module and performing time alignment on the acquired data; Combined with the physical characteristics and operating environment of the optical cable, the potential correlation between the data changes before and after the failure and the optical cable failure is analyzed; Extract features related to fault types from data changes and build a fault type recognition model based on machine learning algorithms; Use monitoring data of known fault types to train the fault type recognition model, input the change characteristics of the monitoring data before and after the fault into the trained fault type recognition model, and obtain the fault type recognition result; According to the identification results, the faults are classified into fracture, wear, corrosion and external interference.

7. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 1 is characterized in that: Generating the optimal maintenance strategy according to the fault location result and the real-time status of the optical cable includes: Receive and analyze the fault location information provided by the fault location module, and visualize the fault location and surrounding optical cable layout in combination with the optical cable network topology diagram; Obtain the current status data of the optical cable from the real-time monitoring module and use the data analysis algorithm to evaluate the real-time health status of the optical cable; Determine the resources required for maintenance based on fault location results and real-time status assessment; Generate resource allocation strategies based on the availability and distribution of maintenance resources; Based on the fault type, fault location, real-time status of the optical cable and maintenance resource allocation strategy, the optimal maintenance strategy is generated using a decision support algorithm.

8. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 7 is characterized in that: The method of dynamically adjusting transmission power, wavelength allocation and encoding according to the performance change of the optical cable includes: Real-time monitoring of the transmission performance parameters of optical cables, and using data analysis algorithms to analyze the changing trends and abnormal fluctuations of performance parameters; Dynamically adjust the transmission power of the optical signal according to the attenuation change of the optical cable and the transmission distance; Analyze the transmission performance of different wavelengths in optical cables based on mutual interference between wavelengths; According to the performance analysis results, the wavelength allocation is dynamically adjusted to select the wavelength with the best performance or the least interference for signal transmission; Select or adjust the appropriate encoding method based on the noise characteristics and transmission requirements of the optical cable; Combined with the adjustment of transmission power, wavelength allocation and coding method, a multi-objective optimization algorithm is used to minimize the transmission burden and improve the overall transmission performance.

9. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 6, characterized in that: The monitoring data before and after the fault occurs include the voltage, current, optical power, temperature, stress and vibration of the optical cable.

10. The intelligent diagnosis and maintenance system for power communication optical cables according to claim 8, characterized in that: The selecting or adjusting the appropriate coding method includes: when the transmission condition deteriorates or the noise increases, dynamically switching to a coding method with better anti-interference ability to improve the reliability and stability of the transmission.

Images