Optical cable fault rapid positioning method, system and equipment

Abstract

The invention provides an optical cable fault rapid positioning method, system and equipment, and the method comprises the steps: obtaining multi-dimensional attribute parameters of a target optical cable section, specifically including a static attribute, a dynamic attribute and an environment attribute for evaluating the operation state of an optical cable; inputting the multi-dimensional attribute parameters into the weighted analysis model and the trained machine learning prediction model in parallel; wherein the weighted analysis model calculates and outputs a first candidate fault position set based on an attribute weight data set, and the attribute weight data set comprises weight scores defined for different types of attribute parameters; analyzing the current multi-dimensional attribute parameters by using the trained machine learning prediction model, and outputting a second candidate fault position set; and fusing overlapped data of the first candidate fault position set and the second candidate fault position set to generate comprehensive fault positioning information. Based on the method, the invention also provides an optical cable fault rapid positioning system and equipment. According to the method, the accuracy of fault positioning and the decision robustness are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of optical cable fault positioning, and particularly relates to an optical cable fault rapid positioning method, system and device. BACKGROUND

[0002] Communication optical cable is a key infrastructure to ensure the safe and stable operation of the power system. Once a fault occurs, quickly and accurately locating the fault point is the prerequisite for shortening the repair time and restoring communication. At present, optical cable fault positioning mainly relies on two types of technologies: one is the test technology based on optical time domain reflectometer (OTDR), which can determine the approximate distance of the fault point, but cannot accurately associate to specific physical locations such as towers and pipe wells, and is greatly affected by the test blind area and environmental factors; the other is to rely on the experience of operation and maintenance personnel for judgment, which is highly subjective, low in efficiency, and the experience is difficult to effectively inherit and scale application.

[0003] There are schemes in the prior art that attempt to predict faults through a single algorithm, such as statistical analysis or simple weight scoring based only on historical data. However, the causes of optical cable faults are complex, with both regularity and randomness, and a single model often has limitations: a rule-based weight model is difficult to adapt to complex and changing environments, while a purely data-driven model lacks reliability when data is sparse or the scene changes suddenly. Therefore, there is an urgent need in the art for a fault positioning solution that can integrate expert experience and data intelligence to overcome the shortcomings of the prior art. SUMMARY

[0004] To solve the above technical problems, the present application proposes an optical cable fault rapid positioning method, system and device, which provides an intelligent fusion mechanism that can fuse rule weights and machine learning prediction results to improve the reliability of decision-making.

[0005] To achieve the above purpose, the present application adopts the following technical solutions: An optical cable fault rapid positioning method, comprising the following steps: Obtaining multi-dimensional attribute parameters of a target optical cable section, the multi-dimensional attribute parameters including static attributes, dynamic attributes and environmental attributes for evaluating the running state of the optical cable; Parallelly inputting the multi-dimensional attribute parameters into a weighted analysis model and a trained machine learning prediction model; wherein the weighted analysis model is based on an attribute weight data set to calculate and output a first candidate fault location set, the attribute weight data set including weight scores defined for different types of attribute parameters; the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters to output a second candidate fault location set; Fusing the overlapping data of the first candidate fault location set and the second candidate fault location set to generate comprehensive fault positioning information.

[0006] Further, the static attributes include cable type and operation time; the dynamic attributes include electrocorrosion state, external force damage state and maintenance record; and the environmental attributes include line voltage level and tower structure type.

[0007] Further, the weighted analysis model is based on the attribute weight dataset to calculate and output a first candidate fault location set, specifically: For any tower position on the cable path, according to the attribute weight dataset, the weight score of each attribute parameter of the tower position is multiplied by the state degree coefficient of the parameter, and the sum of the multiplication results of all attributes is obtained to obtain the fault risk value of the tower position; According to the fault risk value, all tower positions are sorted, and the top N positions with the highest risk value are selected to form the first candidate fault location set.

[0008] Further, the state degree coefficient is used to quantify the specific state severity of the corresponding attribute parameter at a specific candidate tower position. The method for determining the state degree coefficient is that a value determined based on at least one objective data source of artificial inspection record, sensor monitoring data or historical operation and maintenance data is determined through a predefined quantification rule.

[0009] Further, the machine learning prediction model is a model trained from a historical fault dataset, and the historical fault dataset contains historical multi-dimensional attribute parameters and corresponding historical fault location labels.

[0010] Further, the trained machine learning prediction model analyzes the current multi-dimensional attribute parameters to output a second candidate fault location set, specifically: The feature vector is formed by performing feature preprocessing on the current multi-dimensional attribute parameters; The feature vector is input into the trained machine learning prediction model to obtain the fault prediction probability of each candidate tower position; the machine learning prediction model adopts a random forest model; According to the fault prediction probability, each candidate tower position is sorted, and the top M positions are selected to form the second candidate fault location set.

[0011] Further, the overlapping data of the first candidate fault location set and the second candidate fault location set is fused to generate comprehensive fault location information, specifically: All tower positions in the first candidate fault location set and all tower positions in the second candidate fault location set are uniformly sorted; For the overlapping tower positions in the two sets, the overlapping tower positions are given a higher priority to generate the comprehensive fault location information.

[0012] Further, the method further comprises: updating the attribute weight dataset using the historical fault dataset; retraining the machine learning prediction model using the historical fault dataset.

[0013] The application further provides an optical cable fault rapid positioning system, comprising a data acquisition module, a calculation and prediction module and a fault positioning module. The acquisition module is configured to acquire multi-dimensional attribute parameters of a target optical cable section, the multi-dimensional attribute parameters comprising static attributes, dynamic attributes and environmental attributes for evaluating the running state of the optical cable. The calculation and prediction module is configured to input the multi-dimensional attribute parameters into a weighted analysis model and a trained machine learning prediction model in parallel, wherein the weighted analysis model is based on an attribute weight dataset, and the weighted analysis model is configured to calculate and output a first candidate fault position set, the attribute weight dataset comprising weight scores defined for different kinds of attribute parameters; the trained machine learning prediction model is configured to analyze the current multi-dimensional attribute parameters, and output a second candidate fault position set. The fault positioning module is configured to fuse the coincident data of the first candidate fault position set and the second candidate fault position set, and generate comprehensive fault positioning information.

[0014] The application further provides an optical cable fault rapid positioning device, comprising at least one processor and a memory, the memory storing a computer program, and the computer program being executed by the at least one processor to implement the optical cable fault rapid positioning method.

[0015] The effects provided in the summary are only the effects of the embodiments, and are not all the effects of the application, and one of the above technical solutions has the following advantages or beneficial effects: The application provides a method, system and device for quickly locating optical cable faults, the method comprising the following steps: obtaining multi-dimensional attribute parameters of a target optical cable section, the multi-dimensional attribute parameters comprising static attributes, dynamic attributes and environmental attributes for evaluating the running state of the optical cable; inputting the multi-dimensional attribute parameters into a weighted analysis model and a trained machine learning prediction model in parallel; wherein the weighted analysis model is based on an attribute weight dataset comprising weight scores defined for different types of attribute parameters, and calculates and outputs a first candidate fault location set; the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters, and outputs a second candidate fault location set; and the overlapping data of the first candidate fault location set and the second candidate fault location set are fused to generate comprehensive fault location information. Based on the method for quickly locating optical cable faults, the application also provides a system and device for quickly locating optical cable faults. The application performs parallel calculation and result fusion on the weighted analysis model based on expert experience and the machine learning prediction model based on data driving, and forms a double-track comparison decision mechanism. The fusion effectively overcomes the limitations of a single model, and significantly improves the accuracy of fault location and the robustness of decision-making, especially through the consensus-priority fusion rule.

[0016] The application constructs a complete system from data collection, intelligent analysis to result output. Meanwhile, by introducing a model updating mechanism based on feedback information, the system has the ability of self-learning and continuous optimization, and can continuously adapt to changes in the optical cable network, and maintain long-term effectiveness. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of a method for quickly locating optical cable faults is provided for the application embodiment 1; Figure 2 A schematic diagram of a system for quickly locating optical cable faults is provided for the application embodiment 2; Figure 3 A schematic diagram of a device for quickly locating optical cable faults is provided for the application embodiment 3. DETAILED DESCRIPTION

[0018] To clearly illustrate the technical features of the present application, the application will be described in detail below with reference to specific embodiments and drawings. The following disclosure provides many different embodiments or examples of implementing the application. In order to simplify the disclosure of the application, the components and settings of specific examples are described below. In addition, the application can repeatedly refer to numbers and / or letters in different examples. Such repetition is for the purpose of simplification and clarity, and does not in itself indicate the relationship between the various embodiments and / or settings being discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. The application omits the description of well-known components and processing techniques and processes to avoid unnecessarily limiting the application.

[0019] Embodiment 1 Embodiment 1 of the present application proposes a method for quickly locating optical cable faults, a weighted analysis model of expert experience and a data-driven machine learning prediction model are calculated in parallel and the results are fused to form a double-track comparison decision mechanism.

[0020] Figure 1 A flowchart of the method for quickly locating optical cable faults according to Embodiment 1 of the present application is shown in the figure; In step S110, the multi-dimensional attribute parameters of the target optical cable section are obtained, wherein the multi-dimensional attribute parameters include static attributes, dynamic attributes and environmental attributes for evaluating the running state of the optical cable.

[0021] In this application, the static attributes include the optical cable type and the service life; The dynamic attributes include the electrocorrosion state, the external force damage state and the maintenance record; The environmental attributes include the line voltage level and the tower structure type.

[0022] The scope of protection of the present application is not limited to the optical cable attributes listed in Embodiment 1.

[0023] The system server has pre-stored optical cable attribute weight tables, and Table 1 below is a weighted algorithm optical cable running attribute weight table, which divides the optical cable section hanging on the tower into 10 weighted items affecting the running state of the optical cable according to the optical cable running attributes.

[0024] Table 1: Weighted algorithm optical cable running attribute weight table

[0025] In step S120, the multi-dimensional attribute parameters are input in parallel to the weighted analysis model and the trained machine learning prediction model; wherein the weighted analysis model is based on an attribute weight data set to calculate and output a first candidate fault location set, and the attribute weight data set includes weight scores defined for different types of attribute parameters; the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters to output a second candidate fault location set. The weighted analysis model of the present application is based on an attribute weight data set to calculate and output a first candidate fault location set, and the attribute weight data set includes weight scores defined for different types of attribute parameters; For any tower position on the optical cable path, according to the attribute weight data set, the weight score of each attribute parameter of the tower is multiplied by the state degree coefficient of the parameter, and the sum of the multiplication results of all attributes is obtained to obtain the fault risk value of the tower; according to the fault risk value, all tower positions are sorted, and the top N positions with the highest risk value are selected to form the first candidate fault location set.

[0026] ; wherein, represents the total number of optical cable attributes; represents the weight score of each attribute parameter; represents the state degree coefficient of each attribute parameter; represents the fault risk score.

[0027] The state degree coefficient in the present application is used to quantify the specific state severity of the corresponding attribute parameter at a specific candidate tower location; the method for determining the state degree coefficient is: a value determined based on at least one objective data source, such as artificial inspection records, sensor monitoring data, or historical operation and maintenance data, through a predefined quantification rule.

[0028] First, 7 optical cable attributes, including galvanic corrosion, external force damage, service life, easy to be damaged, maintenance, key marking, and excess cable length, are selected from Table 1 for judgment, and the 7 optical cable attributes are processed according to Boolean values (1 or 0). 1 represents a poor state, i.e., high risk; 0 represents a good state, i.e., low risk.

[0029] Table 2: Optical cable operation attribute table;

[0030] The Boolean value (0 or 1) in Table 2 is mapped to a specific degree coefficient (δ).

[0031] For example, for the “galvanic corrosion” (P1=1) of the L1 tower, the degree coefficient δ takes the maximum value of 0.8; For the “service life” (P3=1), δ takes 0.8 (because it exceeds 5 years); For “maintenance” (P5=1), δ is fixed at 0.5; According to the formula the risk value of each tower is calculated.

[0032] For example, the L1 tower is calculated as follows: =(3*0.8)+(3*0)+(1*0.8)+(2*0)+(1*0.5)+(3*0)+(1*0.1)+(1*0.1)+(1*0.2)+(1*0.2)=2.4+0+0.8+0+0.5+0+0.1+0.1+0.2+0.2=4.3. (Note: Parameters such as voltage level, optical cable type, and tower structure are not directly given in this example, and are assumed to be fixed values for calculation.) After calculating and sorting all towers, assume that the three towers with the highest risk values are L6, L4, and L7. The first candidate fault location set is {L6, L4, L7}. This set represents the most likely fault points calculated based on expert experience rules.

[0033] The attribute weight table in the weighted analysis model in the present application is not fixed, but has the ability to optimize itself based on feedback, forming a dynamic knowledge base. The system designs a feedback optimization closed loop: after each fault is repaired and the real fault location is verified on site, the case is recorded as a feedback sample.

[0034] The trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters, and a second candidate fault location set is output; the machine learning prediction model is a model trained from a historical fault data set, and the historical fault data set contains historical multi-dimensional attribute parameters and corresponding historical fault location labels.

[0035] The trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters, and a second candidate fault location set is output, specifically: the current multi-dimensional attribute parameters are feature preprocessed to form a feature vector; the feature vector is input into the trained machine learning prediction model to obtain the fault prediction probability of each candidate tower location; the machine learning prediction model uses a random forest model; According to the fault prediction probability, each candidate tower location is sorted, and the top M locations are selected to form a second candidate fault location set.

[0036] The random forest model learns the mapping relationship between features and fault locations. In the present application, each column (representing a feature vector of a tower) in the optical cable operation attribute table in Table 2 is input into the trained random forest model. The fault probability of each tower is output.

[0037] The line towers are divided into L1-Ln levels, and the optical cable attributes are marked as P1-P7. In order to better demonstrate the algorithm, the towers are defined as 10 levels, and the optical cable attributes are randomly virtual. The relationship between the towers and the optical cable attributes is represented as the first machine learning example table shown in Table 3. Table 3: First machine learning example table

[0038] The optical cable attribute number is sorted as: Pi{8, 5, 5, 2, 6, 3, 3}. At this time, the optical cable attribute P4 with the smallest number is excluded to obtain the second machine learning example table shown in Table 4.

[0039] Table 4: Second machine learning example table

[0040] The two-dimensional correlation degree of the optical cable attributes is statistically sorted as: P1P2=3, P1P3=3, P1P5=5, P1P6=3, P2P3=3, P2P5=4, P3P5=3, obtaining the third machine learning example table shown in Table 5.

[0041] Table 5: Third machine learning example table

[0042] Since L2 is empty, L8 = 1, the possibility of the tower cable interruption of the 2nd and 8th levels can be directly excluded, and a simplified table 6 is obtained, which is a fourth machine learning example table.

[0043] Table 6: Fourth machine learning example table

[0044] Statistical sorting of the three-dimensional correlation degree of the optical cable attribute is: P1P2P3 = 1, P1P2P5 = 3, P1P3P5 = 2, P2P3P5 = 2. Since the associated optical cable attribute is less, the amount of data that can be obtained is not rich, therefore, the application degree of the application is three-dimensional, and the judgment of the optical cable interruption position finally obtains the pushing tower as L6, L7 and L10, and a second candidate fault position set is obtained.

[0045] The application also includes updating the attribute weight data set by using the historical fault data set; and retraining the machine learning prediction model by using the historical fault data set.

[0046] In step S130, the overlapping data of the first candidate fault position set and the second candidate fault position set is fused to generate comprehensive fault positioning information; specifically: All tower positions in the first candidate fault position set and all tower positions in the second candidate fault position set are uniformly sorted; For the overlapping tower positions in the two sets, a higher priority is given to the overlapping tower positions to generate the comprehensive fault positioning information.

[0047] The fusion module receives two sets: {L7, L10, L6} and {L6, L4, L7}. The towers contained in both sets are L6 and L7. According to the "consensus priority" principle, they will be given a higher priority. All towers (L7, L10, L6, L4) in the two sets are fused and sorted. L6 and L7 are ranked first due to the "consensus reward points". Assuming that the final sorting is: L6, L7, L4, L10.

[0048] The final output comprehensive fault position recommendation list is [L6, L7, L4, L10]. The list clearly indicates that the tower L6 should be prioritized for investigation, followed by L7.

[0049] It should be noted that the scope of protection of the application is not limited to the specific values listed in Example 1. Those skilled in the art can apply the method disclosed in the application to fault positioning in different scenarios.

[0050] The fusion process in the application can also adopt a fusion algorithm based on weighted Borda ranking method, and the specific process includes: The towers in the first candidate fault location set and the second candidate fault location set are respectively arranged in descending order according to their original scores (fault risk value S or fault probability P).

[0051] The first ranked tower is assigned 3 points, the second is assigned 2 points, and the third is assigned 1 point (taking a set size of 3 as an example).

[0052] For the towers appearing in both sets, the final score is: (Borda score + Borda score 2) x consensus coefficient K.

[0053] Wherein, the consensus coefficient K is a real number greater than 1, which is used to significantly improve the ranking of the consensus tower.

[0054] For the towers appearing in only one set, the final score is the Borda score in the set.

[0055] All the towers involved are arranged in descending order according to the weighted total score to generate the final fault location recommendation list.

[0056] After each fault repair, the operation and maintenance personnel need to confirm the real fault tower on the APP. The feedback information is recorded and included in the historical fault data set. The system automatically performs an incremental training on the machine learning model using the new data, and adjusts the scores in the weight table according to the feedback to realize the continuous optimization of the system.

[0057] The optical cable fault rapid positioning method proposed in embodiment 1 of the application adopts a mode combining B / S and mobile application. The server side adopts a high-performance commercial server, the operating system is CentOS, and the application server software is Tomcat. The database adopts MySQL5.7, which is used to store all optical cable attribute data, historical fault records and algorithm intermediate results. The terminal is an industrial-grade explosion-proof handheld terminal loaded with Android9.0 and above versions, and the required tables are created in the MySQL database. The scope of protection of the application is not limited to the hardware in embodiment 1, and persons skilled in the art can make reasonable selection according to actual conditions.

[0058] The optical cable fault rapid positioning method proposed in embodiment 1 of the application constructs a complete system from data acquisition, intelligent analysis to result output. At the same time, by introducing a model updating mechanism based on feedback information, the system has the ability of self-learning and continuous optimization, can continuously adapt to the changes of the optical cable network, and maintains long-term effectiveness.

[0059] Embodiment 2 Based on the optical cable fault rapid positioning method proposed in embodiment 1 of the present application, the optical cable fault rapid positioning system proposed in embodiment 2 of the present application, Figure 2 A schematic diagram of the optical cable fault rapid positioning system proposed in embodiment 2 of the present application; the system comprises a data acquisition module, a calculation and prediction module, and a fault positioning module. The data acquisition module is used to acquire the multi-dimensional attribute parameters of the target optical cable section, and the multi-dimensional attribute parameters include static attributes, dynamic attributes, and environmental attributes used to evaluate the running state of the optical cable. The calculation and prediction module is used to input the multi-dimensional attribute parameters into a weighted analysis model and a trained machine learning prediction model in parallel; wherein the weighted analysis model is based on an attribute weight data set to calculate and output a first candidate fault location set, and the attribute weight data set includes weight scores defined for different types of attribute parameters; the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters to output a second candidate fault location set. The fault positioning module is used to fuse the overlapping data of the first candidate fault location set and the second candidate fault location set to generate comprehensive fault positioning information.

[0060] In the data acquisition module of the present application, The static attributes include the optical cable type and the operation time limit; The dynamic attributes include the electrocorrosion state, the external damage state, and the maintenance record; The environmental attributes include the line voltage level and the tower structure type.

[0061] In the calculation and prediction module, The weighted analysis model is based on an attribute weight data set to calculate and output a first candidate fault location set, specifically: for any tower position on the optical cable path, according to the attribute weight data set, the weight score of each attribute parameter of the tower is multiplied by the state degree coefficient of the parameter, and the sum of the multiplication results of all attributes is obtained to get the fault risk value of the tower; according to the fault risk value, all tower positions are sorted, and the top N positions with the highest risk value are selected to form the first candidate fault location set.

[0062] The state degree coefficient is used to quantify the specific state severity of the corresponding attribute parameter at a specific candidate tower position; the method for determining the state degree coefficient is: through a predefined quantification rule, a numerical value determined based on at least one objective data source in the artificial inspection record, the sensor monitoring data, or the historical operation and maintenance data.

[0063] The machine learning prediction model is a model trained from a historical fault data set, and the historical fault data set contains historical multi-dimensional attribute parameters and their corresponding historical fault location labels.

[0064] The trained machine learning prediction model analyzes current multi-dimensional attribute parameters and outputs a second candidate fault location set, specifically: performing feature preprocessing on the current multi-dimensional attribute parameters to form a feature vector; inputting the feature vector into the trained machine learning prediction model to obtain fault prediction probabilities of each candidate tower location; the machine learning prediction model adopts a random forest model; sorting each candidate tower location according to the fault prediction probability, and selecting M positions with higher probabilities to form the second candidate fault location set.

[0065] and updating the attribute weight data set by using the historical fault data set; and retraining the machine learning prediction model by using the historical fault data set.

[0066] In the fault location module, all tower locations in the first candidate fault location set and all tower locations in the second candidate fault location set are uniformly sorted; for the overlapping tower locations in the two sets, a higher priority is assigned to the overlapping tower locations to generate comprehensive fault location information.

[0067] Embodiment 2 of the present application proposes a cable fault rapid positioning system, which constructs a complete system from data acquisition, intelligent analysis to result output. At the same time, by introducing a model updating mechanism based on feedback information, the system has the ability of self-learning and continuous optimization, can continuously adapt to the changes of the cable network, and maintains long-term effectiveness.

[0068] Embodiment 3 The present application also proposes a device, Figure 3 A schematic diagram of a cable fault rapid positioning device according to Embodiment 3 of the present application is shown in the figure, which includes: A memory for storing a computer program; A processor for executing the computer program to realize the method steps as follows: In step S110, the multi-dimensional attribute parameters of the target cable section are obtained, wherein the multi-dimensional attribute parameters include static attributes, dynamic attributes and environmental attributes for evaluating the running state of the cable.

[0069] In the present application, the static attributes include cable type and operation time; The dynamic attributes include electrocorrosion state, external damage state and maintenance record; The environmental attributes include line voltage level and tower structure type.

[0070] In step S120, the multi-dimensional attribute parameters are input into a weighted analysis model and a trained machine learning prediction model in parallel; wherein the weighted analysis model is based on an attribute weight dataset including weight scores defined for different kinds of attribute parameters, calculates and outputs a first candidate fault location set; and the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters and outputs a second candidate fault location set; In step S130, the overlapping data of the first candidate fault location set and the second candidate fault location set are fused to generate comprehensive fault location information.

[0071] The optical cable fault rapid positioning equipment provided in Embodiment 3 of the present application performs parallel calculation and result fusion based on a weighted analysis model based on expert experience and a machine learning prediction model based on data driving, forming a double-track comparison decision mechanism. This fusion effectively overcomes the limitations of a single model, and especially through a consensus priority fusion rule, significantly improves the accuracy of fault location and the robustness of decision-making.

[0072] It should be noted that the technical scheme of the present application also provides an electronic device, which comprises: a communication interface capable of interacting with other devices such as network devices; a processor connected with the communication interface to realize information interaction with other devices, used to run a computer program to execute the optical cable fault rapid positioning method provided by one or more technical schemes, and the computer program is stored on a memory. Of course, in actual application, various components in the electronic device are coupled together through a bus system. It can be understood that the bus system is used to realize the connection and communication between the components. In addition to the data bus, the bus system also includes a power bus, a control bus and a state signal bus. The memory in the embodiment of the present application is used to store various types of data to support the operation of the electronic device. Examples of these data include any computer programs used to operate on the electronic device. It can be understood that the memory can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM, Read-Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), an erasable programmable read-only memory (EPROM, Erasable Programmable Read-Only Memory), an electrically erasable programmable read-only memory (EEPROM, Electrically Erasable Programmable Read-Only Memory), a ferromagnetic random access memory (FRAM, ferromagnetic random access memory), a flash memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM, Compact Disc Read-Only Memory); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM, Random Access Memory) used as an external cache.By way of example, and not limitation, many forms of RAM can be used, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDR SDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), Direct Rambus Random Access Memory (DRRAM). The memory described in the embodiments of the present application is intended to include, but not be limited to, these and any other suitable types of memory. The methods disclosed in the embodiments of the present application can be applied in or implemented by a processor. The processor can be an integrated circuit chip chip with a processing capability. In the implementation process, each step of the above method can be completed by integrated logic circuits or instructions in the form of software in the processor. The processor described above can be a general-purpose processor, a DSP (Digital Signal Processing, i.e., a chip capable of implementing digital signal processing technology), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The processor can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or any conventional processor. In combination with the steps of the method disclosed in the embodiments of the present application, the hardware decoding processor can be directly embodied to execute the completion, or the hardware and software modules in the decoding processor are combined to execute the completion. The software module can be located in a storage medium, which is located in the memory, and the processor reads the program in the memory to complete the steps of the above method in combination with the hardware. The processor executes the program to implement the corresponding flow in each method of the embodiments of the present application, and for the sake of brevity, it will not be repeated here.

[0073] The description of the related parts of the optical cable fault rapid positioning system and the device provided by the embodiments 2 and 3 of the present application can refer to the detailed description of the corresponding parts of the optical cable fault rapid positioning method provided by the embodiment 1 of the present application, and will not be repeated here.

[0074] It should be noted that, in this paper, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device inherent in the series of elements. Without more limitations, the element defined by the sentence "includes a" does not exclude the existence of other same elements in the process, method, article or device including the element. In addition, the above technical solutions provided by the embodiments of the present application have not been described in detail, which are consistent with the implementation principles of the corresponding technical solutions in the prior art, so as not to be too repetitive.

[0075] The above describes the specific embodiments of the present application in combination with the drawings, but is not a limitation on the protection scope of the present application. Based on the above description, those skilled in the art can make other different forms of modifications or changes. Here, it is not necessary and impossible to enumerate all the embodiments. Various modifications or changes made by those skilled in the art on the basis of the technical solutions of the present application without creative labor are still within the protection scope of the present application.

Claims

1. A method for rapid fault location of an optical cable, characterized in that, The method comprises the following steps: acquiring multi-dimensional attribute parameters of a target optical cable section, the multi-dimensional attribute parameters comprising static attributes, dynamic attributes and environmental attributes used for evaluating an optical cable running state; inputting the multi-dimensional attribute parameters into a weighted analysis model and a trained machine learning prediction model in parallel; the weighted analysis model is based on an attribute weight dataset comprising weight scores defined for different types of attribute parameters, and calculates and outputs a first candidate fault location set; the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters and output a second candidate fault location set; fusing coincident data of the first candidate fault location set and the second candidate fault location set to generate comprehensive fault location information.

2. The method of claim 1, wherein, The static attributes comprise optical cable types and service life; the dynamic attributes comprise electrocorrosion states, external damage states and maintenance records; and the environmental attributes comprise line voltage levels and tower structure types.

3. The method of claim 1, wherein, The weighted analysis model is based on the attribute weight dataset, and calculates and outputs the first candidate fault location set, specifically as follows: for any tower position on the optical cable path, according to the attribute weight dataset, the weight score of each attribute parameter of the tower position is multiplied by a state degree coefficient of the parameter, and the sum of the multiplication results of all attributes is obtained to obtain a fault risk value of the tower position; all tower positions are sorted according to the fault risk value, and the top N positions with the highest risk values are selected to form the first candidate fault location set.

4. The method of claim 3, wherein, The state degree coefficient is used to quantify the specific state severity of the corresponding attribute parameter at a specific candidate tower position. The state degree coefficient is determined by a method as follows: a value determined based on at least one objective data source, such as artificial inspection records, sensor monitoring data or historical operation and maintenance data, through a predefined quantification rule.

5. The method of claim 1, wherein, The machine learning prediction model is a model trained from a historical fault dataset, and the historical fault dataset contains historical multi-dimensional attribute parameters and corresponding historical fault location labels.

6. The method of claim 1, wherein, The trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters and output the second candidate fault location set, specifically as follows: feature preprocessing is performed on the current multi-dimensional attribute parameters to form a feature vector; the feature vector is input into the trained machine learning prediction model to obtain fault prediction probabilities of each candidate tower position; the machine learning prediction model adopts a random forest model; each candidate tower position is sorted according to the fault prediction probability, and the top M positions with high probabilities are selected to form the second candidate fault location set.

7. The method of claim 1, wherein, The coincident data of the first candidate fault location set and the second candidate fault location set are fused to generate comprehensive fault location information, specifically as follows: all tower positions in the first candidate fault location set and all tower positions in the second candidate fault location set are uniformly sorted; for the coincident tower positions in the two sets, the coincident tower positions are given higher priority to generate the comprehensive fault location information.

8. The method of claim 1, wherein, The method further comprises: updating the attribute weight dataset using the historical fault dataset; retraining the machine learning prediction model using the historical fault dataset.

9. A rapid optical cable fault location system characterized by, The method comprises: a data acquisition module, a calculation and prediction module and a fault location module. The data acquisition module is configured to acquire multi-dimensional attribute parameters of the target optical cable section, the multi-dimensional attribute parameters including static attributes, dynamic attributes and environmental attributes used for evaluating the running state of the optical cable; The calculation and prediction module is configured to input the multi-dimensional attribute parameters into a weighted analysis model and a trained machine learning prediction model in parallel; the weighted analysis model is based on an attribute weight dataset including weight scores defined for different types of attribute parameters, and calculates and outputs a first candidate fault location set; the trained machine learning prediction model is used to analyze the current multi-dimensional attribute parameters, and outputs a second candidate fault location set; The fault location module is configured to fuse coincident data of the first candidate fault location set and the second candidate fault location set, and generate comprehensive fault location information.

10. An optical cable fault rapid location device comprising at least one processor and a memory, the memory storing a computer program, characterized in that, The computer program is executed by the at least one processor to implement the optical cable fault rapid positioning method according to any one of claims 1 to 9.

Images