Communication line operation and maintenance method and system based on fuzzy logic, and storage medium

Abstract

The invention relates to the technical field of communication engineering, in particular to a communication line operation and maintenance method and system based on fuzzy logic and a storage medium. The method comprises the following steps: acquiring real-time communication line parameters; performing fault prediction on the real-time communication line parameters through the dynamic fuzzy learning model to obtain fault prediction data; inputting the fault prediction data into a fuzzy reasoning model for analysis to obtain a fault possibility; and performing communication line operation and maintenance on the fault possibility according to the alarm threshold. Faults can be accurately predicted, adaptability is improved, response time is shortened, and active fault prevention is achieved.

Description

Technical Field

[0001] The present application relates to the field of communication engineering technology, and in particular to a method, system and storage medium for communication line operation and maintenance based on fuzzy logic. Background Art

[0002] Existing communication line maintenance methods primarily rely on periodic inspections and response mechanisms, typically based on fixed parameter thresholds to monitor line status, such as temperature, humidity, and load. However, these methods are unable to adapt to rapidly changing network environments; they lack comprehensive analysis of historical and real-time data, making it difficult to accurately predict faults; and after detecting an anomaly, traditional methods require manual intervention to confirm the fault and take action, resulting in longer response times and increasing the impact of the fault on communication services. Furthermore, these methods lack the ability to proactively analyze and predict faults.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to propose a method, system and storage medium for communication line operation and maintenance based on fuzzy logic, which can accurately predict faults, improve adaptability and shorten response time, and achieve proactive fault prevention.

[0005] To achieve the above objectives, an embodiment of the present application provides a method for communication line operation and maintenance based on fuzzy logic, the method comprising the following steps:

[0006] Obtain real-time communication line parameters;

[0007] Performing fault prediction on the real-time communication line parameters by using a dynamic fuzzy learning model to obtain fault prediction data;

[0008] Inputting the fault prediction data into a fuzzy reasoning model for analysis to obtain the possibility of fault;

[0009] Perform communication line operation and maintenance based on the fault possibility according to the alarm threshold.

[0010] In some embodiments, the dynamic fuzzy learning model includes a dynamic fuzzy rule base module and a deep Q network module.

[0011] In some embodiments, the process of constructing the dynamic fuzzy rule base module includes the following steps:

[0012] Acquiring the real-time communication line parameters and historical line parameters;

[0013] dividing the historical line parameters into a plurality of fuzzy sets according to the urgency of fault prediction;

[0014] Determine implementation measures according to the plurality of fuzzy sets, and obtain execution parameters corresponding to the fuzzy sets;

[0015] Performing fuzzy description on the historical route parameters and the execution parameters by using Gaussian membership function;

[0016] Determine fuzzy rules by combining the plurality of fuzzy sets, the historical route parameters described in the fuzzy manner, and the execution parameters described in the fuzzy manner with experience to obtain an initial fuzzy rule base module;

[0017] The initial fuzzy rule base module is updated by using the real-time communication line parameters to obtain the dynamic fuzzy rule base module.

[0018] In some embodiments, updating the initial fuzzy rule base module using the real-time communication line parameters to obtain the dynamic fuzzy rule base module includes the following steps:

[0019] Preprocessing the acquired real-time communication line parameters;

[0020] The pre-processed real-time communication line parameters within a preset time period are statistically analyzed in combination with fuzzy sets to obtain the number of abnormality levels;

[0021] The weight of the initial fuzzy rule base module is adjusted according to the number of abnormality levels in combination with a weight adjustment strategy to obtain the dynamic fuzzy rule base module.

[0022] In some embodiments, performing fault prediction on the real-time communication line parameters using a dynamic fuzzy learning model to obtain fault prediction data includes the following steps:

[0023] Preprocessing the acquired real-time communication line parameters;

[0024] Performing fuzzy reasoning on the pre-processed real-time communication line parameters through the dynamic fuzzy rule base module to obtain a preliminary estimated output;

[0025] The preliminary estimated output is forward propagated through the deep Q network module to obtain fault prediction data.

[0026] In some embodiments, forward propagating the preliminary estimated output through the deep Q network module to obtain fault prediction data includes the following steps:

[0027] Passing the preliminary estimation output to the hidden layer through the input layer of the deep Q network module for feature extraction to obtain fault prediction features;

[0028] The fault prediction data is obtained by performing Q value calculation on the fault prediction feature through the output layer.

[0029] In some embodiments, inputting the fault prediction data into a fuzzy inference model for analysis to obtain the fault probability includes the following steps:

[0030] Converting the fault prediction data into fuzzy sets through a fuzzy logic-based reasoning model;

[0031] Performing reasoning calculation on the fault prediction data converted into the fuzzy set according to fuzzy rules of a reasoning model based on fuzzy logic to obtain an output fuzzy set;

[0032] Defuzzification is performed on the output fuzzy set to obtain the fault possibility.

[0033] In some embodiments, performing communication line operation and maintenance on the fault possibility according to the alarm threshold includes:

[0034] When the fault probability is less than or equal to the alarm threshold, record no alarm event and maintain the status quo;

[0035] When the fault probability is greater than the alarm threshold, early warning information is obtained and an alarm event is recorded;

[0036] Sending the warning information to the operation and maintenance personnel terminal, so that the operation and maintenance personnel terminal performs communication line operation and maintenance according to the warning information;

[0037] The warning information includes: warning level, fault possibility, possible cause and recommended measures.

[0038] To achieve the above objectives, another aspect of the present application provides a system for communication line operation and maintenance based on fuzzy logic, the system comprising:

[0039] The first module is used to obtain real-time communication line parameters;

[0040] The second module is used to perform fault prediction on the real-time communication line parameters through a dynamic fuzzy learning model to obtain fault prediction data;

[0041] The third module is used to input the fault prediction data into the fuzzy reasoning model for analysis to obtain the possibility of fault;

[0042] The fourth module is used to perform communication line operation and maintenance based on the fault possibility according to the alarm threshold.

[0043] To achieve the above-mentioned purpose, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above-mentioned method when executed by a processor.

[0044] The embodiments of this application include at least the following beneficial effects: This application provides a method, system, and storage medium for communication line operation and maintenance based on fuzzy logic. This solution uses a dynamic fuzzy learning model to predict faults based on real-time communication line parameters, generating fault prediction data. This fault prediction data is then input into a fuzzy inference model for analysis to determine the likelihood of a fault. This system can adjust and optimize operation and maintenance rules in real time to adapt to environmental changes and new data patterns, providing more accurate fault predictions while rapidly responding to changes in the network environment, reducing service interruptions caused by faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of a method for communication line operation and maintenance based on fuzzy logic provided in an embodiment of the present application;

[0046] Figure 2 It is a flowchart for constructing a dynamic fuzzy rule base module;

[0047] Figure 3 It is a diagram of some fuzzy rule tables;

[0048] Figure 4 yes Figure 2 Schematic diagram of step S206 in ;

[0049] Figure 5 yes Figure 1 Flowchart of step S200 in FIG.

[0050] Figure 6 This is a schematic diagram of some fault prediction data tables;

[0051] Figure 7 yes Figure 1 Flowchart of step S300 in FIG.

[0052] Figure 8 It is a tabular diagram of fuzzy reasoning analysis in some fuzzy reasoning models;

[0053] Figure 9 It is a flowchart of communication line operation and maintenance;

[0054] Figure 10 This is a flow chart of a system for communication line operation and maintenance based on fuzzy logic provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are merely examples of methods consistent with some aspects of the embodiments of the present application.

[0056] It will be understood that the terms "first", "second", etc. used in this application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0057] The terms "at least one", "plurality", "each", "any", etc. used in this application include "at least one", "two" or more, "plurality" or "each", "any" or "any one", "each" or "any one" in the context of the present invention, and "at least one" or "at least one" includes one, two or more, "plurality" or "any one" includes two or more, "each" or "each one" in the context of the present invention, and "any" or "any one

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0059] In related technologies, Figure 1 is a flow chart of a method for communication line operation and maintenance based on fuzzy logic provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S100 to S400:

[0060] Step S100: obtaining real-time communication line parameters;

[0061] Step S200: performing fault prediction on real-time communication line parameters using a dynamic fuzzy learning model to obtain fault prediction data;

[0062] Step S300: inputting the fault prediction data into the fuzzy reasoning model for analysis to obtain the fault probability;

[0063] Step S400: Perform communication line operation and maintenance based on the possibility of failure according to the alarm threshold.

[0064] In some embodiments, real-time communication line parameters are obtained. These parameters may be operating parameters of the communication line and include, but are not limited to, temperature, humidity, and load. A dynamic fuzzy learning model is used to perform fault prediction on the real-time communication line parameters to obtain fault prediction data. When the fault prediction data indicates a potential fault, a fuzzy inference model is used to perform detailed fuzzy inference analysis on the fault prediction data to determine the probability of a fault. Communication line maintenance is performed based on the fault probability according to an alarm threshold, and maintenance personnel are notified to address the issue.

[0065] In some embodiments, in step S100, real-time communication line parameters of the line are obtained in real time through various sensors deployed on the communication line. For example, the data collected by the temperature sensor has a range of -40°C to 85°C, the data collected by the humidity sensor has a range of 0% to 100%, and the data collected by the load sensor has a range of 0 to 100%.

[0066] In some embodiments, the dynamic fuzzy learning model includes a dynamic fuzzy rule base module and a deep Q network module. The dynamic fuzzy rule base module is a knowledge representation and reasoning system based on fuzzy logic, which contains a series of dynamically changing fuzzy rules. Unlike the traditional fuzzy rule base module, the dynamic fuzzy rule base module can adjust and update the rules in real time according to the system's operating status, environmental changes or new data feedback to adapt to different situations and improve the system's adaptability and accuracy. The deep Q network module is a reinforcement learning algorithm based on deep learning. It combines the powerful feature extraction capabilities of deep learning with the Q learning algorithm to solve complex decision-making problems. Fault prediction is performed through the joint collaboration of the dynamic fuzzy rule base module and the deep Q network module to obtain fault prediction data. By real-time monitoring and dynamic updating of the dynamic fuzzy rule base module, it is possible to quickly respond to changes in the network environment and reduce the service interruption time caused by faults. This proactive fault prevention mechanism enables operation and maintenance personnel to take measures before a fault occurs, avoiding the passive response mode commonly seen in traditional operation and maintenance systems.

[0067] In some embodiments, as Figure 2 As shown, Figure 2 This is a flowchart for building a dynamic fuzzy rule base module. Figure 2 The method may include but is not limited to steps S201 to S206:

[0068] Step S201: Acquire real-time communication line parameters and historical line parameters;

[0069] Step S202: dividing historical line parameters into multiple fuzzy sets according to the urgency of fault prediction;

[0070] Step S203: determining implementation measures based on multiple fuzzy sets and obtaining execution parameters corresponding to the fuzzy sets;

[0071] Step S204: fuzzy description of historical route parameters and execution parameters using Gaussian membership function;

[0072] Step S205: Determine fuzzy rules by combining multiple fuzzy sets, fuzzy historical route parameters, and fuzzy execution parameters with experience to obtain an initial fuzzy rule base module;

[0073] Step S206: updating the initial fuzzy rule base module through the real-time communication line parameters to obtain a dynamic fuzzy rule base module.

[0074] In some embodiments, the rules in the fuzzy rule base module are dynamically adjusted based on real-time communication line parameters and historical line parameters. These rules define the degree to which different parameters affect the probability of failure. For example, if the temperature parameter deviates from the normal range, the fuzzy rule will increase the weight of this parameter to reflect its increased impact on the probability of failure. The update of the fuzzy rules in the dynamic fuzzy rule base module can be expressed by the following formula:

[0075] R′ i =R i w i ;

[0076] where R′ i represents the updated rule, R i represents the original rule, w i Represents the weight coefficient, which is dynamically adjusted according to the real-time communication line parameters.

[0077] In some embodiments, in step S201, real-time communication line parameters and historical line parameters are obtained, where the real-time communication line parameters and historical line parameters include but are not limited to temperature (T), humidity (H), and load (L); the normal range of the parameters is set at T: 0°C-40°C, H: 30%-70%, and L: 20%-80%; the parameter abnormality threshold is set at ±10% beyond the normal range.

[0078] In some embodiments, in step S202, historical line parameters are divided into multiple fuzzy sets according to the urgency of the fault prediction. Specifically, the fuzzy sets are divided into multiple but not limited to five fuzzy sets: very low (VL), low (L), medium (M), high (H), and very high (VH).

[0079] In some embodiments, in step S203, implementation measures are determined based on multiple fuzzy sets to obtain implementation parameters corresponding to the fuzzy sets. Specifically, after the fuzzy sets are classified into very low (VL), low (L), medium (M), high (H), and very high (VH), different implementation parameters are established for each different emergency category, where the implementation parameters include information such as the alarm level.

[0080] In some embodiments, in step S204, the historical line parameters and the execution parameters are fuzzy described by the Gaussian membership function. Specifically, the temperature deviation e is obtained by the difference between the actual parameters T, H, L in the historical line parameters and the normal range parameters. T , humidity deviation e H and load deviation e L The fuzzy processing is performed through the membership function, which includes the Gaussian membership function. The formula of the Gaussian membership function is expressed as:

[0081]

[0082] Where μ(x) represents the membership degree of the input variable x to a certain fuzzy set; x represents the input variable, namely the historical line parameters and execution parameters; a represents the center value parameter of the control curve, and b represents the width value parameter of the control curve.

[0083] According to the deviation e T 、e H and e L The range of variation is divided into seven deviation levels: -0.1, 0, 0.1, 0.2, 0.3, 0.4, and 0.5 (corresponding to ±10% of the normal range). The deviation level is calculated by taking the difference between the real-time communication line parameter or the historical line parameter above or below the center value of the normal range and the ratio of the center value. Positive and negative numbers indicate whether the real-time communication line parameter or the historical line parameter is above or below the center value of the normal range, respectively.

[0084] In some embodiments, in step S205, fuzzy rules are determined based on multiple fuzzy sets, fuzzy description of historical line parameters and execution parameters combined with experience to obtain an initial fuzzy rule base module. Specifically, for fuzzy rules, based on communication line maintenance experience and combined with Figure 3 , determine the following partial fuzzy rules:

[0085] When e T is negative (NB) and e H is negative (NB) and e L is negative (NB), then E is very high (VH);

[0086] When e T is negative small (NS) and eH is negative small (NS) and e L If it is negative small (NS), then E is low (L);

[0087] When e T 、e H and e L If any one of them is 0, then E is medium (M);

[0088] When e T is positively small (PS) and e H is positively small (PS) and e L is positively small (PS), then E is high (H);

[0089] When e T is positive (PB) and e H is positive (PB) and e L is positive (PB), then E is very high (VH).

[0090] like Figure 3 As shown, Figure 3 This is a partial fuzzy rule table diagram, which lists three conditions of temperature, humidity and load. More specifically, the normal range of temperature (T) is 0℃-40℃, and the deviation calculation formula is e T =T-20℃, where e T Indicates the temperature deviation. T represents the temperature value in the real-time communication line parameter. 20°C represents the median of the normal range. When the deviation level is -0.2, the deviation level indicates that the temperature value in the real-time communication line parameter is -20% of the normal range, which is expressed as negative (NB). The normal range of humidity (H) is 30%-70%. The deviation calculation formula is e H =H-50%, where e H Indicates the humidity deviation. H represents the humidity value in the real-time communication line parameter. 50% represents the median of the normal range. When the deviation level is 0.1, the deviation level indicates that the deviation value of the humidity value in the real-time communication line parameter is 10% of the normal range. This is expressed as positive minimum (PS). The normal range of load (L) is 20%-80%. The deviation calculation formula is e L = L - 50%, where e L Indicates the load deviation, L represents the load value in the real-time communication line parameter, and 50% represents the median of the normal range. When the deviation level is -0.1, the deviation level indicates that the deviation value of the load value in the real-time communication line parameter is -10% of the normal range.

[0091] Fuzzy sets and deviation levels describe parameters holistically from the perspective of fault urgency, while deviation levels focus on the numerical deviation of the parameters themselves. Comparing the parameters to the normal range, they are divided into five levels: negative (large), negative (small), positive (large), positive (small), and zero. This intuitively reflects the degree and direction of the parameter's deviation from the normal state and provides a more specific description of the parameter's value. When constructing a fuzzy rule base, fuzzy sets and deviation levels are often used together as the conditional components of the rules. Fuzzy sets provide more macroscopic context for deviation levels, explaining the position and importance of parameter deviations within the overall fault urgency framework. Deviation levels, on the other hand, provide more specific parameter numerical basis for fuzzy sets, helping to further refine and clarify the definition of fuzzy sets. During the fuzzy reasoning process, the information carried by fuzzy sets and deviation levels is integrated to reach the final conclusion. Fuzzy sets and deviation levels describe parameters from different perspectives. They work together and synergistically in the construction of the dynamic fuzzy rule base module and fuzzy reasoning, providing the system with a more comprehensive and accurate decision-making basis to better address complex practical problems.

[0092] In some embodiments, in step S206, the initial fuzzy rule base module is updated by the real-time communication line parameters to obtain a dynamic fuzzy rule base module. Figure 4 As shown, Figure 4 yes Figure 2 In the schematic diagram of step S206 in FIG, real-time communication line parameters are obtained, where the real-time communication line parameters have been preprocessed. The dynamic fuzzy rule base module determines whether the real-time communication line parameters are abnormal. If it is determined to be abnormal, the corresponding fuzzy rule weight is adjusted; if it is determined to be normal, the original rule is maintained. The initial fuzzy rule base module is updated by adjusting the corresponding fuzzy rule weight. Specifically, for example, if high load (L>80%) has caused multiple failures in the past week, the weight of the load parameter in failure prediction is increased. The weight adjustment formula is as follows:

[0093] w′ L =w L ·f(L);

[0094] Among them, w′ L represents the updated weight; w L represents the original weight, and f(L) represents the adjustment function of the load parameter.

[0095] In some embodiments, another method is provided for updating the initial fuzzy rule base module through the real-time communication line parameters to obtain a dynamic fuzzy rule base module. Specifically, the acquired real-time communication line parameters are preprocessed, and the preprocessed real-time communication line parameters within a preset time period are combined with fuzzy sets to obtain the number of abnormality levels; the weight of the initial fuzzy rule base module is adjusted according to the number of abnormality levels in combination with the weight adjustment strategy to obtain a dynamic fuzzy rule base module. According to actual needs, the time period for counting the number of abnormality levels is determined so as to accurately calculate the number of abnormality levels caused by specific parameters within the time period. According to the set time period, the parameter data and fault records within the time period are filtered out from the data storage system, each parameter is analyzed, and the number of abnormality levels caused by the parameter is counted. According to the distribution of the number of abnormality levels, the number of abnormality levels is divided into different levels, and the threshold value of each level is set according to the actual situation. For different levels of abnormality, corresponding rule weight adjustment strategies are formulated. For example, when the abnormality level is "low", the rule weight remains unchanged; when the abnormality level is "medium", the weight of the rules related to the parameter is appropriately reduced; when the abnormality level is "high", the weight of the rules related to the parameter is significantly reduced. By modifying the weight coefficient of each rule in the rule base, the weight of the rules related to the parameter in the rule base is adjusted according to the abnormality level and the corresponding adjustment strategy. The adjusted rule weights are updated to the dynamic fuzzy rule base module, and the dynamic fuzzy rule base module is verified to ensure that the adjusted rule base can work properly and improve the performance of the system. Repeat the above steps regularly, continuously collect new parameter data and fault records, re-count the abnormality level, and adjust the rule weights to ensure that the dynamic fuzzy rule base module can continuously adapt to the dynamic changes of the system, thereby obtaining a dynamic fuzzy rule base module.

[0096] like Figure 5 As shown, Figure 5 yes Figure 1 Flowchart of step S200 in FIG. Figure 1 Step S200 includes but is not limited to steps S210 to S230:

[0097] Step S210: pre-processing the acquired real-time communication line parameters;

[0098] Step S220: performing fuzzy reasoning on the pre-processed real-time communication line parameters through a dynamic fuzzy rule base module to obtain a preliminary estimated output;

[0099] Step S230: forward propagating the preliminary estimated output through the deep Q network module to obtain fault prediction data.

[0100] In some embodiments, in step S210, the acquired real-time communication line parameters are preprocessed. First, data cleaning is performed on the real-time communication line parameters to remove noise, outliers, and missing values from the collected data to improve data quality. Normalization is then performed to normalize parameter data of varying ranges and magnitudes to specific intervals to facilitate subsequent processing and module calculations. Acquiring real-time communication line parameters is the starting point of the entire process, providing raw data for subsequent processing. Preprocessing ensures data quality and consistency, making it more suitable for processing by the dynamic fuzzy rule base and laying the foundation for the accurate application of fuzzy rules.

[0101] In some embodiments, in step S220, fuzzy reasoning is performed on the pre-processed real-time communication line parameters through the dynamic fuzzy rule base module to obtain a preliminary estimated output. Specifically, the pre-processed real-time communication line parameters are mapped to different fuzzy sets according to the corresponding membership functions, and the membership of each parameter in each fuzzy set is determined. Fuzzy rules that match the fuzzified parameters are searched in the dynamic fuzzy rule base. By matching the fuzzy set of the input parameters with the fuzzy set in the fuzzy rules, fuzzy rules that meet the conditions are found. For the matched fuzzy rules, reasoning calculations are performed based on their conclusions to obtain a preliminary estimated output. This may involve comprehensive processing of the conclusions of multiple matching rules, such as using weighted average, maximum membership, etc., to determine the final preliminary estimated output result.

[0102] Fuzzification converts the preprocessed numerical real-time communication line parameters into fuzzy concepts for matching with the rules in the dynamic fuzzy rule base module. Rule matching searches for relevant rules in the rule base based on the fuzzified results, while inference calculation derives a preliminary estimated output based on the matched rules. This series of steps, based on the principles of fuzzy logic, can handle information with uncertainty and ambiguity. It provides a more realistic analysis and processing of the preprocessed real-time communication line parameters, resulting in a preliminary estimate that provides the foundation for further processing using the Deep Q Network.

[0103] In some embodiments, in step S230, the preliminary estimated output is forward propagated through the deep Q network module to obtain fault prediction data. Specifically, the obtained preliminary estimated output data is normalized to the interval [0,1] to meet the input requirements of the deep Q network module. For example, for temperature T, if the real-time temperature of the preliminary estimated output is 45°C, its normalized value is If it is outside the normal range, special treatment is required.

[0104] The Deep Q Network module includes the DQN algorithm. Specifically, the DQN algorithm is configured with a learning rate of α = 0.001, a discount factor of γ = 0.9, and an exploration rate of ε = 0.1. An experience replay pool is initialized to store the four-tuple of state, action, reward, and next state. The state space is defined as the normalized values of the real-time communication line parameters T, H, and L in the preliminary estimated output. The state vector is represented as: S = [T norm , H norm , L norm ]. Define the action space as possible operation and maintenance operations, such as no operation (0), check the line (1), adjust the load (2). For the design of the reward function, the reward function R = -|e T |-|e H |-|e L |, where e T 、e H and e L They are the deviation values of temperature, humidity, and load respectively. The negative sign indicates that the larger the deviation, the smaller the reward, thereby guiding the model to reduce parameter deviation.

[0105] For the training of the DQN algorithm, the DQN algorithm is used to train and update the Q value table or Q network. The Q-learning update formula is:

[0106]

[0107] Where Q(s, a) represents the value function of the state-action pair, α represents the learning rate, γ represents the discount factor, r represents the reward, and S′ represents the next state.

[0108] For each state S t =[T norm , H norm , L norm ], the DQN algorithm selects an action a t And execute, after executing the action, observe the next state S t+1 and reward r t , through the experience replay pool of four tuples (S t , a t , r t , S t+1 )Update the Q value.

[0109] Regularly evaluate the performance of the DQN algorithm and validate it using a test set. Adjust parameters such as the learning rate and discount factor based on model performance to optimize model performance. Through these steps, the DQN algorithm learns to extract features from the real-time communication line parameters in the preliminary estimate output and predict potential failures. For example, if the temperature T suddenly rises to 45°C, exceeding the normal range, the DQN algorithm will predict the likelihood of failure based on historical line parameters and the current real-time communication line parameters, and select appropriate operational and maintenance actions to reduce the risk of failure. This approach can improve the adaptability and accuracy of the intelligent communication line operation and maintenance system and reduce service interruptions caused by failures.

[0110] like Figure 6 As shown, Figure 6 This is a diagram of some fault prediction data tables. The table corresponds to each state S t =[T norm , H norm , L norm ] for different situations, the action, reward, next state and Q value are updated for each situation. For example, when the state is [0.8, 0.7, 0.6], it is necessary to check the communication line, and the reward is -5 at this time. The larger the deviation is, the smaller the corresponding reward is, thereby reducing the parameter deviation. According to the reward, the next state is [0.75, 0.65, 0.55], and the Q value is updated at the same time. The DQN algorithm enhances the overall prediction ability. By learning patterns in historical data, the DQN algorithm can predict possible failures under specific conditions. This solution provides more accurate fault prediction through the combination of fuzzy logic and DQN algorithms. This predictive ability can not only reduce the probability of failure, but also reduce the economic losses caused by failures and the impact of service interruptions on users. In addition, accurate fault prediction can also optimize resource allocation, reduce unnecessary maintenance operations, and thus reduce operation and maintenance costs.

[0111] In some embodiments, the preliminary estimated output is forward propagated through the deep Q network module to obtain fault prediction data. Specifically, the preliminary estimated output is passed through the input layer of the deep Q network module to the hidden layer for feature extraction to obtain fault prediction features; the fault prediction features are passed to the output layer through the hidden layer, and the Q value of the fault prediction features is calculated through the output layer to obtain fault prediction data.

[0112] In some embodiments, the deep Q network module takes the preliminary estimated output of the dynamic fuzzy rule base as input, and uses its powerful learning ability to further extract features and process the data to obtain more accurate fault prediction data. Network training and reasoning are the core processes of module learning and application. By continuously optimizing the module algorithm, it can better capture the potential patterns in the data and improve the accuracy and reliability of the prediction. The processing of the deep Q network module is carried out on the basis of the processing of the dynamic fuzzy rule base, further improving the analysis and prediction capabilities of real-time parameters to adapt to more complex system and task requirements. The entire process starts with the acquisition of real-time communication line parameters, and then improves the data quality through preprocessing. The dynamic fuzzy rule base module is used to perform fuzzy reasoning to obtain a preliminary estimated output. Finally, the deep Q network module is used to perform deep processing on the preliminary estimated output to obtain predicted data. Each step is closely connected. The previous step provides data and basis for the next step, gradually realizing the effective processing and accurate prediction of real-time parameters.

[0113] like Figure 7 As shown, Figure 7 yes Figure 1 Flowchart of step S300 in FIG. Figure 1 Step S300 includes but is not limited to steps S310 to S330:

[0114] Step S310: converting the fault prediction data into a fuzzy set through a fuzzy logic-based reasoning model;

[0115] Step S320: performing reasoning calculation on the fault prediction data converted into a fuzzy set according to the fuzzy rules of the reasoning model based on fuzzy logic to obtain an output fuzzy set;

[0116] Step S330: Defuzzify the output fuzzy set to obtain the fault possibility.

[0117] In some embodiments, when a dynamic fuzzy learning model detects a potential fault, a fuzzy inference model is activated. The fuzzy inference model uses fuzzy logic to perform a detailed analysis of the potential fault and determine the likelihood and urgency of the fault. For example, when a load parameter suddenly increases, fuzzy inference will assess the impact of this change on the stability of the communication line and provide a score for the likelihood of a fault. Fuzzy inference calculations can be implemented using fuzzy logic-based inference models, including the Mamdani model. Its basic steps include: fuzzification: converting input data into a fuzzy set; inference calculation: applying fuzzy rules to infer the fuzzy set; and defuzzification: converting the resulting output fuzzy set into a clear value. Specific implementation conditions and parameters include: communication line parameters, normal parameter ranges, parameter abnormality thresholds, and fuzzy inference parameters. Among them, communication line parameters include temperature (T), humidity (H) and load (L); the normal range of parameters includes T: 0℃-40℃, H: 30%-70%, L: 20%-80%; the abnormal threshold of parameters includes ±10% exceeding the normal range; the fuzzy inference parameters include the membership function parameters of the fuzzy set, such as the center (c) and standard deviation (σ) of the Gaussian membership function.

[0118] In some embodiments, in step S310, the fault prediction data is converted into a fuzzy set using a fuzzy logic-based inference model. Specifically, the input data to the Mamdani model is normalized to the interval [0, 1]. Data outside the normal range requires special processing. A Gaussian membership function is applied to each parameter for fuzzification. For example, for temperature T, assuming the center of the Gaussian function c = 40 and the standard deviation σ = 5, the calculation formula is:

[0119]

[0120] Where T is the actual temperature, c is the center of the Gaussian function, and σ is the standard deviation.

[0121] In some embodiments, in step S320, the fault prediction data converted into a fuzzy set is inferred and calculated according to the fuzzy rules of the reasoning model based on fuzzy logic to obtain an output fuzzy set, that is, the fuzzy rules are applied for reasoning. For example, when the membership of temperature T and humidity H are both very high, but the membership of load L is low, a specific fault warning rule may be triggered. Reasoning is performed through the Mamdani model of fuzzy logic, and for each input parameter, its membership in each fuzzy set is calculated, and then aggregated according to the fuzzy rules to obtain an output fuzzy set. For example, when the membership of temperature T is 0.8 (high), the membership of humidity H is 0.7 (high), and the membership of load L is 0.3 (low), then according to the fuzzy rules, the membership of the fault possibility E is calculated as:

[0122] μE =min(μ T , μ H ,max(1-μ L ));

[0123] Among them, μ E Indicates the membership degree of fault possibility, μ T Indicates the membership degree of temperature fault, μ H It represents the membership degree of humidity fault and the opposite membership degree of load fault.

[0124] In some embodiments, in step S330, the output fuzzy set is defuzzified to obtain the fault probability. Specifically, the output fuzzy set is converted into a clear value to facilitate decision-making. The defuzzification method is the Center of Gravity method. For example, when the fuzzy set of the fault probability E is {0.2, 0.5, 0.8}, and the corresponding fault level is {low, medium, high}, the defuzzification calculation formula is:

[0125]

[0126] Among them, x i It represents the value of the fault level, μ i Indicates the corresponding membership degree.

[0127] In some embodiments, as Figure 8 As shown, Figure 8 This table illustrates the fuzzy reasoning analysis of a partial fuzzy reasoning model. The table lists the membership of temperature, humidity, and load at specific actual values. For example, when the actual temperature (T) is 42°C, its normalized value is 1.05 after preprocessing. Fuzzification using the Gaussian membership function yields a fuzzy set of high (0.8). The defuzzified calculation formula calculates the fault probability E as low (L), resulting in a corresponding membership of 0.2. The same applies to the humidity and load examples.

[0128] In some embodiments, in step S400, communication line operation and maintenance are performed based on the fault probability according to the alarm threshold. Specifically, based on the results of the fuzzy inference model analysis, if the fault probability exceeds a preset threshold, the operation and maintenance personnel are notified via the alarm notification module. This notification can be provided via SMS, email, or a system pop-up window. For example, if the calculated fault probability is 0.75, which exceeds the preset threshold of 0.6, the system will automatically notify the operation and maintenance personnel through the preset alarm mechanism to conduct an inspection.

[0129] In some embodiments, based on the results of fuzzy reasoning analysis, a membership value of the fault probability E is obtained. Assume that after defuzzification processing, the current fault probability E value is 0.75. Compare the fault probability E value with the preset alarm threshold. If the E value is greater than or equal to the alarm threshold (for example, 0.6), the alarm notification process is triggered. When the alarm is triggered, the system will generate detailed alarm information, which includes but is not limited to: alarm level, fault probability, possible causes and recommended measures. Specifically: Alarm level: high; Failure probability: 75%; Possible cause: Too high temperature; Current temperature: 42°C; Recommended measures: Check the heat dissipation of the communication line, and consider reducing the load or adding heat dissipation equipment.

[0130] Alerts will be sent to maintenance personnel via SMS, email, or system pop-up notifications based on the pre-set alarm notification method. For emergency alerts: "Communication line failure is highly likely. Current temperature is 42°C, exceeding the normal range." Please immediately check the line's heat dissipation and consider reducing the load or adding cooling equipment. The system records detailed information for each alarm event, including the alarm time, alarm severity, failure probability, and alarm notification method, to facilitate subsequent maintenance and analysis. After implementing maintenance measures, maintenance personnel will provide feedback to the system, which will update the status and reassess the failure probability. If the failure probability decreases to a safe range, the system will clear the alert. If the failure probability remains high or continues to increase, the system will maintain the alert and escalate the severity. The real-time communication line parameters after maintenance measures are fed into a dynamic fuzzy learning model, combined with a fuzzy inference model, and processed repeatedly until the failure probability is less than or equal to the alarm threshold, at which point the alert will be cleared.

[0131] In some embodiments, as Figure 9 As shown, Figure 9 This is a flowchart of the communication line operation and maintenance process. Specifically, when the probability of failure is less than or equal to the alarm threshold, a record of no alarm event is made and the status quo is maintained; when the probability of failure is greater than the alarm threshold, an early warning message is obtained and an alarm event is recorded; the early warning message is sent to the operation and maintenance personnel terminal, so that the operation and maintenance personnel terminal processes the early warning message and obtains the real-time communication line parameters of the early warning feedback; the real-time communication line parameters of the early warning feedback are input into the communication line operation and maintenance method based on fuzzy logic for cyclic processing until the probability of failure is less than or equal to the alarm threshold. This method can improve the system's adaptability. In a changing network environment, the system can automatically adjust its operation and maintenance strategy to adapt to new challenges. This adaptive capability makes the system not only effective in the current environment, but also adaptable to new situations that may arise in the future, ensuring the long-term effectiveness and reliability of the system.

[0132] like Figure 10As shown in FIG, the system for communication line operation and maintenance based on fuzzy logic includes:

[0133] The first module is used to obtain real-time communication line parameters;

[0134] The second module is used to predict faults of real-time communication line parameters through a dynamic fuzzy learning model to obtain fault prediction data;

[0135] The third module is used to input the fault prediction data into the fuzzy reasoning model for analysis to obtain the fault probability;

[0136] The fourth module is used to perform communication line operation and maintenance based on the possibility of failure according to the alarm threshold.

[0137] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned method of communication line operation and maintenance based on fuzzy logic.

[0138] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiment, the functions specifically implemented by the present storage medium embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0139] The present application also provides an electronic device comprising a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for communication line operation and maintenance based on fuzzy logic. The electronic device can be any intelligent terminal, such as a tablet computer or an in-vehicle computer.

[0140] It can be understood that the contents of the above method embodiments are applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0141] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0142] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0143] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0144] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0145] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0146] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0147] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A method for communication line operation and maintenance based on fuzzy logic, characterized in that: The method comprises the following steps: Obtain real-time communication line parameters; Performing fault prediction on the real-time communication line parameters by using a dynamic fuzzy learning model to obtain fault prediction data; Inputting the fault prediction data into a fuzzy reasoning model for analysis to obtain the possibility of fault; Perform communication line operation and maintenance based on the fault possibility according to the alarm threshold.

2. The method according to claim 1, characterized in that The dynamic fuzzy learning model includes a dynamic fuzzy rule base module and a deep Q network module.

3. The method according to claim 2, characterized in that The construction process of the dynamic fuzzy rule base module includes the following steps: Acquiring the real-time communication line parameters and historical line parameters; dividing the historical line parameters into a plurality of fuzzy sets according to the urgency of fault prediction; Determine implementation measures according to the plurality of fuzzy sets, and obtain execution parameters corresponding to the fuzzy sets; Performing fuzzy description on the historical route parameters and the execution parameters by using Gaussian membership function; Determine fuzzy rules by combining the plurality of fuzzy sets, the historical route parameters described in the fuzzy manner, and the execution parameters described in the fuzzy manner with experience to obtain an initial fuzzy rule base module; The initial fuzzy rule base module is updated by using the real-time communication line parameters to obtain the dynamic fuzzy rule base module.

4. The method according to claim 3, characterized in that The updating of the initial fuzzy rule base module by using the real-time communication line parameters to obtain the dynamic fuzzy rule base module comprises the following steps: Preprocessing the acquired real-time communication line parameters; The pre-processed real-time communication line parameters within a preset time period are statistically analyzed in combination with fuzzy sets to obtain the number of abnormality levels; The weight of the initial fuzzy rule base module is adjusted according to the number of abnormality levels in combination with a weight adjustment strategy to obtain the dynamic fuzzy rule base module.

5. The method according to claim 2, characterized in that The method of performing fault prediction on the real-time communication line parameters by using a dynamic fuzzy learning model to obtain fault prediction data includes the following steps: Preprocessing the acquired real-time communication line parameters; Performing fuzzy reasoning on the pre-processed real-time communication line parameters through the dynamic fuzzy rule base module to obtain a preliminary estimated output; The preliminary estimated output is forward propagated through the deep Q network module to obtain fault prediction data.

6. The method according to claim 5, characterized in that The method of forward-propagating the preliminary estimated output through the deep Q network module to obtain fault prediction data includes the following steps: Passing the preliminary estimation output to the hidden layer through the input layer of the deep Q network module for feature extraction to obtain fault prediction features; The fault prediction data is obtained by performing Q value calculation on the fault prediction feature through the output layer.

7. The method according to claim 1, characterized in that The inputting of the fault prediction data into the fuzzy inference model for analysis to obtain the fault probability comprises the following steps: Converting the fault prediction data into fuzzy sets through a fuzzy logic-based reasoning model; Performing reasoning calculation on the fault prediction data converted into the fuzzy set according to fuzzy rules of a reasoning model based on fuzzy logic to obtain an output fuzzy set; Defuzzification is performed on the output fuzzy set to obtain the fault possibility.

8. The method according to claim 1, characterized in that The performing communication line operation and maintenance based on the fault possibility according to the alarm threshold includes: When the fault probability is less than or equal to the alarm threshold, record no alarm event and maintain the status quo; When the fault probability is greater than the alarm threshold, early warning information is obtained and an alarm event is recorded; Sending the warning information to the operation and maintenance personnel terminal, so that the operation and maintenance personnel terminal performs communication line operation and maintenance according to the warning information; The warning information includes: warning level, fault possibility, possible cause and recommended measures.

9. A system for communication line operation and maintenance based on fuzzy logic, characterized in that: The system comprises: The first module is used to obtain real-time communication line parameters; The second module is used to perform fault prediction on the real-time communication line parameters through a dynamic fuzzy learning model to obtain fault prediction data; The third module is used to input the fault prediction data into the fuzzy reasoning model for analysis to obtain the possibility of fault; The fourth module is used to perform communication line operation and maintenance based on the fault possibility according to the alarm threshold.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images