Cable full life cycle prediction method and system based on convolutional neural network

Abstract

The invention discloses a cable full life cycle prediction method and system based on a convolutional neural network, and belongs to the technical field of cable full life cycle prediction, and the method comprises the steps: collecting target cable load state data, carrying out the judgment of a target cable load mode, analyzing a first state index of a target cable, and analyzing a second state index of the target cable, and analyzing the comprehensive life evaluation index of the target cable, collecting a comprehensive life evaluation index sequence of the target cable, inputting the comprehensive life evaluation index sequence into the convolutional neural network model, and outputting a full life cycle prediction result of the cable. According to the invention, the safety and stability of the power system can be improved, the health conditions of the cable in different operation stages are accurately grasped by accurately analyzing the load mode, each state index and the comprehensive life evaluation index, the full life cycle of the cable is accurately predicted by using the convolutional neural network, and the maintenance, overhaul and replacement plans of the cable are reasonably planned. Power interruption caused by cable faults is avoided, and the utilization efficiency of power system resources is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cable life cycle prediction, and in particular to a cable life cycle prediction method and system based on convolutional neural networks. Background Art

[0002] With the rapid advancement of smart grid construction, the operational status of power cables, the core carrier of urban underground power transmission networks, directly impacts power safety and the continuity of industrial production. A sudden cable failure not only disrupts power supply and causes massive economic losses, but can also lead to serious accidents such as fires and explosions due to short circuits and leakage. Therefore, accurate prediction of cable lifecycles has become a key requirement for ensuring the safe and stable operation of power grids.

[0003] For example, the invention patent announcement with announcement number: CN118709571B discloses a method and system for predicting the remaining life of a cable taking into account the aging trend, which includes: obtaining historical load data of the cable to be predicted, processing the historical load data through a trained temperature prediction model, and determining the predicted operating temperature of the cable to be predicted in the next time period; determining the effective service life of the cable to be predicted at the predicted operating temperature based on the predicted operating temperature and the aging trend of the cable to be predicted; obtaining the operating time of the cable to be predicted, and considering the service life loss of the cable to be predicted based on the historical load data and operating time of the cable to be predicted, determining the actual operating time of the cable to be predicted and combining it with the effective service life to obtain the predicted remaining life of the cable to be predicted.

[0004] For example, the invention patent publication number CN116361679B discloses a data-driven intelligent cable life prediction method and system. The method includes the following steps: first, obtaining multidimensional time series data from the cable's historical operation, and then obtaining cluster discreteness and optimization degree based on the distance characteristics of the data points in iterative clustering. A distance optimization factor for the data point is obtained based on the distance characteristics between the data point's neighborhood data points and the cluster center in the cluster space; and a time series optimization factor for the data point is obtained based on the distance characteristics between the data point's time series neighboring data points and the cluster center in the multidimensional time series data. A distance optimization value is obtained based on the data point's cluster discreteness, optimization degree, distance optimization factor, and time series optimization factor, and iterative clustering is improved based on the distance optimization value.

[0005] However, in the process of implementing the technical solutions of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0006] Current cable life prediction methods often use some of the cable's operating parameters to predict cable life. However, in actual power system operation scenarios, different degrees of load fluctuations, such as frequent load switching and large voltage jumps, will have different impacts on the cable aging process. If a universal method is used to predict the life of all cables, ignoring the differences in load fluctuations faced by different cables in actual operation, it may lead to significant deviations between the prediction results and the actual remaining life of the cable, thereby increasing the potential risks of power system operation and may even cause sudden cable failures. Summary of the Invention

[0007] The first aspect of the present invention provides a cable life cycle prediction method based on a convolutional neural network, comprising the following steps:

[0008] S1, within a preset monitoring window, collect target cable load status data and determine the target cable load mode.

[0009] S2: Analyze a first state indicator of the target cable based on a target cable load mode determination result.

[0010] S3, collecting life status data of the target cable, analyzing the second status index of the target cable, and analyzing the comprehensive life assessment index of the target cable in combination with the first status index of the target cable.

[0011] S4 collects the comprehensive life assessment indicator sequence of the target cable multiple times at fixed time intervals, and combines the time characteristics of the comprehensive life assessment indicators of the target cable to construct a time series input tensor, input it into the convolutional neural network model, and output the full life cycle prediction results of the cable.

[0012] The second aspect of the present invention provides a cable life cycle prediction system based on a convolutional neural network, comprising:

[0013] The cable load mode determination module is used to collect target cable load status data within a preset monitoring window and perform target cable load mode determination.

[0014] The cable first state index analysis module is used to analyze the target cable first state index based on the target cable load mode determination result.

[0015] The cable comprehensive life assessment index analysis module is used to collect the life status data of the target cable, analyze the second status index of the target cable, and analyze the comprehensive life assessment index of the target cable in combination with the first status index of the target cable.

[0016] The cable life cycle prediction result output module is used to collect the comprehensive life assessment indicator sequence of the target cable multiple times at fixed time intervals, and combine the time characteristics of the comprehensive life assessment indicators of the target cable to construct a time series input tensor, input it into the convolutional neural network model, and output the cable life cycle prediction results.

[0017] One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

[0018] 1. The cable life cycle prediction method based on convolutional neural network provided by the present invention can improve the safety and stability of the power system. By accurately analyzing the load mode, various status indicators and comprehensive life assessment indicators, it can accurately grasp the health status of the cable at different operation stages. By utilizing the powerful feature extraction and learning capabilities of the convolutional neural network, it can accurately predict the entire life cycle of the cable, which is helpful to rationally plan cable maintenance, inspection and replacement plans, avoid power interruptions caused by cable failures, and improve the utilization efficiency of power system resources.

[0019] 2. The present invention analyzes the first state index of the target cable based on the target cable load mode determination result. By conducting targeted analysis of different load modes, the actual aging degree of the cable under various operating conditions can be accurately analyzed. For load-disturbed cables, the influence of thermal-mechanical alternating stress on their aging can be accurately assessed. For load-stable cables, the thermal-insulation degradation condition can be accurately measured, thereby more accurately reflecting the current health status of the cable and providing a more reliable data foundation for subsequent comprehensive evaluation. In subsequent analysis, corrections can be made that are more in line with the actual situation of the cable, making the comprehensive evaluation results closer to the actual state of the cable and improving the accuracy of cable life cycle prediction.

[0020] 3. By analyzing the comprehensive life assessment indicators of the target cable, the present invention can comprehensively and accurately reflect the actual operating status and aging degree of the cable, provide accurate data basis for subsequent analysis and decision-making, prevent sudden failures from causing power supply interruptions, and improve the reliability and stability of the power system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Flowchart of the cable life cycle prediction method based on convolutional neural network provided in the embodiment of the present application;

[0022] Figure 2 A schematic diagram of the structure of a cable life cycle prediction system based on a convolutional neural network provided in an embodiment of the present application;

[0023] Figure 3 A flow chart of a method for analyzing cable status based on cable load status according to an embodiment of the present application;

[0024] Figure 4 This is a flow chart of the convolutional neural network model output involved in the embodiments of the present application. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] Reference Figure 1 As shown, the first aspect of the present invention provides a cable life cycle prediction method based on a convolutional neural network, comprising the following steps:

[0027] See Figure 3 As shown, it is a flow chart of the method for performing cable status analysis based on cable load status according to an embodiment of the present application. The load mode is determined according to the analysis results of the target cable load status data, and targeted collection and analysis are performed according to the load mode.

[0028] S1, within a preset monitoring window, collect target cable load status data and determine the target cable load mode.

[0029] In this embodiment, the target cable load mode is determined, and the specific analysis method is as follows:

[0030] Based on the target cable load status data, the load status index of the target cable is analyzed.

[0031] In a specific embodiment, the load status indicator of the target cable is analyzed, and the specific analysis process is as follows:

[0032] Within the preset monitoring window, the target cable load status data is collected, including the target cable's average voltage, voltage fluctuation standard deviation, load switching frequency, and number of large jumps.

[0033] It should be noted that the average voltage, voltage fluctuation standard deviation, load switching frequency and number of large jumps of the target cable can be obtained directly from the SCADA system. Among them, the load switching frequency is the total number of times the load equipment connected to the cable is started, stopped and switched within the preset monitoring window, reflecting the frequency of load state changes during system operation. The number of large jumps is the number of times the voltage signal suddenly changes (the amplitude exceeds the set threshold) in a short period of time, reflecting the frequency of severe disturbance events caused by the load side.

[0034] Extract the ideal average voltage, reference voltage fluctuation standard deviation, reference load switching frequency, and reference large jump number stored in the database.

[0035] The average voltage distribution coefficient, voltage fluctuation standard deviation distribution coefficient, load switching frequency distribution coefficient and large jump number distribution coefficient preset in the database are extracted.

[0036] It should be noted that the value range of the average voltage distribution coefficient, the voltage fluctuation standard deviation distribution coefficient, the load switching frequency distribution coefficient and the large jump number distribution coefficient is between 0 and 1, and the sum of the average voltage distribution coefficient, the voltage fluctuation standard deviation distribution coefficient, the load switching frequency distribution coefficient and the large jump number distribution coefficient is 1. When used, the pre-set value can be directly extracted from the database. The specific extraction method is, for example, to construct a one-to-one mapping set with the average voltage, voltage fluctuation standard deviation, load switching frequency and large jump number distribution coefficient respectively. When used, the obtained average voltage, voltage fluctuation standard deviation, load switching frequency and large jump number are respectively input into the corresponding mapping set, so as to extract the average voltage distribution coefficient, voltage fluctuation standard deviation distribution coefficient, load switching frequency distribution coefficient and large jump number distribution coefficient.

[0037] The load status index of the target cable is analyzed based on the load status data of the target cable.

[0038] The load state index of the target cable is a quantitative indicator of the degree of influence of the target cable's average voltage, voltage fluctuation standard deviation, load switching frequency and number of large jumps on the target cable's load state. The specific analysis process is: the average voltage of the target cable is deviated from its ideal value, the voltage fluctuation standard deviation, load switching frequency and number of large jumps of the target cable are differentiated from the corresponding reference values respectively, and the results of the deviation processing and differentiation processing are coupled with the corresponding distribution coefficient to obtain the load state index of the target cable.

[0039] In a specific embodiment, the load status indicator of the target cable is specifically expressed as follows:

[0040]

[0041] Among them, S load is the load status indicator of the target cable, is the average voltage of the target cable, σ U is the voltage fluctuation standard deviation of the target cable, f swith is the load switching frequency of the target cable, N jump is the number of large jumps of the target cable, U0 is the ideal average voltage, (σ U ) vef is the standard deviation of the reference voltage fluctuation, (f swith )vef is the reference load switching frequency, (N jump ) vef is the reference number of large jumps, α1 is the average voltage distribution coefficient, α2 is the voltage fluctuation standard deviation distribution coefficient, α3 is the load switching frequency distribution coefficient, and α4 is the large jump number distribution coefficient.

[0042] It should be noted that the average voltage of the target cable, the standard deviation of voltage fluctuations, the load switching frequency, and the number of large jumps directly influence each other. For example, frequent load switching causes voltage fluctuations, which in turn increases the standard deviation of voltage fluctuations. The more severe the fluctuations, the more events that exceed the jump threshold, that is, the greater the number of large jumps. Each load switching is accompanied by a transient mutation, which directly increases the number of jumps. If the voltage control is stable, the standard deviation of voltage fluctuations is low.

[0043] Extract the cable load status indicator threshold preset in the database.

[0044] If the load state index of the target cable is greater than the cable load state index threshold, the target cable load mode is recorded as a load disturbance cable.

[0045] If the target cable's load status indicator exceeds the cable load status indicator threshold, it indicates that the target cable's operating condition is relatively unstable. Its average voltage, voltage fluctuation standard deviation, load switching frequency, and number of large jumps are all outside the normal range. Frequent load switching, large voltage fluctuations, and numerous large jumps place additional stress on the cable. These stresses accelerate cable aging, increase the risk of failure, and affect the cable's service life and performance. Therefore, this load pattern is recorded as a load-disturbance cable.

[0046] If the load state index of the target cable is less than or equal to the cable load state index threshold, the target cable load mode is recorded as a load-stable cable.

[0047] If the target cable's load status indicator is less than or equal to the cable load status indicator threshold, the target cable is operating relatively smoothly. The connected load device is infrequently started, stopped, or switched, resulting in minimal voltage fluctuations and a low number of significant jumps. This indicates that the cable is operating in a relatively stable electrical environment. In this scenario, the cable is subject to minimal disturbances, and its aging process is primarily influenced by conventional thermal aging factors. Overall, its operating condition is good, so its load mode is recorded as a stable load cable.

[0048] S2: Analyze a first state indicator of the target cable based on a target cable load mode determination result.

[0049] In this embodiment, the first state indicator of the target cable is analyzed based on the target cable load mode determination result. The specific process is as follows:

[0050] The load state correction factor corresponding to the load state index interval of each target cable stored in the database is extracted, and the load state correction factor corresponding to the interval where the load state index of the target cable is located is mapped and recorded as the load state correction factor of the target cable.

[0051] It should be understood that the load state index of the target cable comprehensively reflects the electrical stability of the cable during operation. The larger the load state index of the target cable, the greater the disturbances such as load fluctuations and voltage changes that the cable is subjected to, and the more serious the impact on cable aging and life. In order to more accurately measure the aging degree and remaining life of the cable under different load states, more significant corrections to the relevant evaluation indicators are required in subsequent analysis, so the corresponding extracted load state correction factor is larger, so as to ensure that the subsequent cable state evaluation based on this correction factor is more accurate.

[0052] If the target cable load mode is a load disturbance cable, the disturbance stress data of the target cable is collected, the disturbance stress score of the target cable is analyzed, and the disturbance stress correction score of the target cable is analyzed in combination with the load state correction factor of the target cable. The disturbance stress correction score of the target cable is recorded as the first state indicator of the target cable.

[0053] It should be noted that if the target cable load mode is a load disturbance cable, it means that the target cable is frequently started and shut down or the voltage fluctuates greatly. At this time, thermal-mechanical alternating stress dominates aging, that is, the cable disturbance stress will affect the subsequent cable life.

[0054] In this embodiment, the disturbance stress correction score of the target cable is analyzed in detail as follows:

[0055] Within the preset time window, the disturbance stress data of the target cable is collected, including the temperature rise rate, thermal cycle frequency, strain peak value and start-stop times of the target cable.

[0056] It should be noted that the disturbance stress data of the target cable can be obtained directly from the SCADA system, where the temperature rise rate refers to the rate at which the cable temperature rises within a preset time window, reflecting the degree of mutation of the thermal stress to which the cable is subjected. The thermal cycle frequency refers to the number of thermal cycles within a preset time window. The temperature first rises and then falls, and a thermal cycle is recorded when the temperature difference exceeds the preset temperature in the database. The strain peak refers to the maximum mechanical stress experienced by the cable during operation within the preset time window. A large strain peak indicates that the cable is subjected to strong mechanical stretching, bending and other external forces, and long-term accumulation leads to metal fatigue or sheath cracking. The total number of starts and stops of the load equipment (such as motors, heaters, etc.) connected to the cable within the preset time window. The more starts and stops, the more frequent the cable is subjected to current shocks and hot-cold alternations, inducing thermal-mechanical stress cycles.

[0057] It should also be noted that the target cable's temperature rise rate, thermal cycle frequency, strain peak, and number of starts and stops are interrelated. A higher temperature rise rate often leads to more frequent thermal cycles. This is because after a rapid temperature rise, the cable is more likely to experience a temperature drop during system operation, resulting in thermal cycles and an increase in thermal cycle frequency. Furthermore, frequent starts and stops lead to frequent current fluctuations, which in turn cause cable temperature fluctuations, altering the temperature rise rate. This also creates thermal expansion and contraction, increasing the thermal cycle frequency. Furthermore, larger strain peaks are often accompanied by changes in mechanical stress during the start-stop process. The greater the number of starts and stops, the more frequent the mechanical stress cycles experienced by the cable, and the greater the likelihood of generating high strain peaks. These parameters work together to influence the cable's disturbed stress state and aging process.

[0058] The reference temperature rise rate, reference thermal cycle frequency, reference strain peak value and reference start-stop times stored in the database are extracted.

[0059] The temperature rise rate distribution factor, thermal cycle frequency distribution factor, strain peak distribution factor and start-stop number distribution factor preset in the database are extracted.

[0060] It should be noted that the temperature rise rate distribution factor, thermal cycle frequency distribution factor, strain peak distribution factor and start-stop number distribution factor all have value ranges between 0 and 1, and the sum of the temperature rise rate distribution factor, thermal cycle frequency distribution factor, strain peak distribution factor and start-stop number distribution factor is 1. When used, the pre-set values can be directly extracted from the database. The specific extraction method is, for example: the temperature rise rate, thermal cycle frequency, strain peak value and start-stop number are respectively mapped to the temperature rise rate distribution factor, thermal cycle frequency distribution factor, strain peak value distribution factor and start-stop number distribution factor to construct a one-to-one mapping set. The method of use is to input the obtained temperature rise rate, thermal cycle frequency, strain peak value and start-stop number into the corresponding mapping set, thereby extracting the temperature rise rate distribution factor, thermal cycle frequency distribution factor, strain peak value distribution factor and start-stop number distribution factor.

[0061] The disturbance stress score of the target cable is analyzed based on the disturbance stress data of the target cable.

[0062] The disturbance stress score of the target cable is a quantitative indicator of the degree to which the temperature rise rate, thermal cycle frequency, strain peak value and start-stop number of the target cable jointly affect the disturbance stress state of the target cable. The specific analysis process is as follows: the temperature rise rate, thermal cycle frequency, strain peak value and start-stop number of the target cable are differentiated from their corresponding reference values, and the differentiation processing results are coupled with the corresponding distribution factors to obtain the disturbance stress score of the target cable.

[0063] In a specific embodiment, the disturbance stress score of the target cable is specifically expressed as follows:

[0064]

[0065] Among them, S dis is the disturbance stress score of the target cable, R T is the temperature rise rate of the target cable, f th is the thermal cycle frequency of the target cable, ε max is the peak strain of the target cable, N ss is the start and stop times of the target cable, (R T ) vef is the reference temperature rise rate, (f th ) vef is the reference thermal cycle frequency, (ε max ) vef is the reference strain peak, (N ss ) vef is the reference start-stop number, β1 is the temperature rise rate distribution factor, β2 is the thermal cycle frequency distribution factor, β3 is the strain peak distribution factor, and β4 is the start-stop number distribution factor.

[0066] The disturbance stress correction score of the target cable is analyzed according to the disturbance stress score of the target cable and the load state correction factor of the target cable.

[0067] The specific analysis process of the disturbance stress correction score of the target cable is: the disturbance stress score of the target cable is corrected based on the load state correction factor of the target cable, so as to obtain the disturbance stress correction score of the target cable, which is specifically expressed as: the disturbance stress score of the target cable is multiplied by the numerical result of the load state correction factor of the target cable as the disturbance stress correction score of the target cable.

[0068] If the target cable load mode is a load-stable cable, the thermal-insulation status data of the target cable is obtained, the thermal-insulation degradation comprehensive index of the target cable is analyzed, and the thermal-insulation degradation correction index of the target cable is analyzed in combination with the load status correction factor of the target cable. The thermal-insulation degradation correction index of the target cable is recorded as the first status indicator of the target cable.

[0069] It should be noted that if the target cable load mode is a stable load cable, it means that the cable is under a constant or slowly changing load for a long time, the cable is mainly thermally aged, and the insulation material gradually deteriorates, that is, the thermal-insulation state of the cable affects the subsequent cable life.

[0070] In a specific embodiment, the thermal-insulation degradation correction index of the target cable is analyzed in the following manner:

[0071] Obtain the thermal-insulation status data of the target cable, including the operating time of the target cable, the conductor resistivity change rate, and the operating voltage deviation rate.

[0072] It should be noted that the thermal-insulation status data of the target cable can be obtained directly from the SCADA system, where the operating time of the target cable refers to the cumulative operating time of the target cable, the conductor resistivity change rate refers to the rate of change of the resistivity per unit length of the cable conductor relative to the initial value, and the operating voltage offset rate is the relative deviation between the actual operating voltage and the rated voltage of the cable.

[0073] Extract the reference operating duration, reference conductor resistivity change rate, and reference operating voltage deviation rate stored in the database.

[0074] Extract the operating time allocation factor, conductor resistivity change rate allocation factor and operating voltage deviation rate allocation factor preset in the database.

[0075] It should be noted that the operating time allocation factor, the conductor resistivity change rate allocation factor and the operating voltage offset rate allocation factor all have value ranges between 0 and 1, and the sum of the operating time allocation factor, the conductor resistivity change rate allocation factor and the operating voltage offset rate allocation factor is 1. When used, the pre-set value can be directly extracted from the database. The specific extraction method is, for example: constructing a one-to-one mapping set with the operating time, the conductor resistivity change rate and the operating voltage offset rate respectively, and inputting the obtained operating time, conductor resistivity change rate and operating voltage offset rate into the corresponding mapping set respectively when used, so as to extract the operating time allocation factor, the conductor resistivity change rate allocation factor and the operating voltage offset rate allocation factor.

[0076] Based on the thermal-insulation status data of the target cable, the thermal-insulation degradation comprehensive index of the target cable is analyzed.

[0077] The comprehensive thermal-insulation degradation index of the target cable is a quantitative indicator of the degree to which the operating time, conductor resistivity change rate and operating voltage offset rate of the target cable jointly affect the thermal-insulation degradation state of the target cable. The specific analysis process is: the absolute values of the operating time, conductor resistivity change rate and operating voltage offset rate of the target cable are differentiated with their corresponding reference values, and the differentiation results are coupled with the corresponding distribution factors to obtain the comprehensive thermal-insulation degradation index of the target cable.

[0078] In a specific embodiment, the thermal-insulation degradation comprehensive index of the target cable is specifically expressed as follows:

[0079]

[0080] Among them, I deg is the thermal-insulation degradation comprehensive index of the target cable, t is the operating time of the target cable, |Rρ | is the conductor resistivity change rate of the target cable, R U is the operating voltage deviation rate of the target cable, t vef is the reference running time, (R ρ ) vef is the resistivity change rate of the reference conductor, (R U ) vef is the reference operating voltage deviation rate, γ1 is the operating time allocation factor, γ2 is the conductor resistivity change rate allocation factor, and γ3 is the operating voltage deviation rate allocation factor.

[0081] It should be noted that the operating time, conductor resistivity change rate, and operating voltage excursion rate of the target cable are interrelated. As operating time increases, the cable conductor is subjected to the long-term effects of current thermal effects and chemical corrosion, which often increases the conductor resistivity change rate. This is because the conductor material gradually deteriorates and its resistance increases, and the operating voltage excursion rate also affects this. If the operating voltage is too high for a long time, it will increase conductor heat generation, accelerate material aging, and increase the rate of resistivity change. On the other hand, if the operating voltage is too low for a long time, although heat generation is reduced, it may affect the cable's performance stability and also affect the conductor resistivity change to a certain extent. In addition, as operating time increases, the cable's insulation performance deteriorates, becoming more susceptible to voltage fluctuations, and the operating voltage excursion rate may also change more frequently. These parameters work together to affect the cable's thermal and insulation degradation state.

[0082] The thermal-insulation degradation correction index of the target cable is analyzed according to the thermal-insulation degradation comprehensive index of the target cable and the load state correction factor of the target cable.

[0083] The specific analysis process of the thermal-insulation degradation correction index of the target cable is: the thermal-insulation degradation comprehensive index of the target cable is corrected based on the load state correction factor of the target cable, so as to obtain the thermal-insulation degradation correction index of the target cable, which is specifically expressed as: the thermal-insulation degradation comprehensive index of the target cable is multiplied by the load state correction factor of the target cable as the numerical result of the thermal-insulation degradation correction index of the target cable.

[0084] S3, collecting life status data of the target cable, analyzing the second status index of the target cable, and analyzing the comprehensive life assessment index of the target cable in combination with the first status index of the target cable.

[0085] In this embodiment, the comprehensive life assessment index of the target cable is analyzed, and the specific analysis process is as follows:

[0086] The second status indicator of the target cable is analyzed based on the target cable life status data.

[0087] In a specific embodiment, the second status indicator of the target cable is analyzed, and the specific analysis steps are as follows:

[0088] Collect target cable life status data within the preset time window.

[0089] The target cable life status data includes the target cable's maximum discharge capacity, discharge frequency, and dielectric loss angle change rate.

[0090] It should be noted that the maximum discharge amount, discharge frequency and dielectric loss angle change rate of the target cable can be obtained directly from the SCADA system. Among them, the maximum discharge amount refers to the maximum charge of a single partial discharge within a preset time window, the discharge frequency refers to the number of discharges within a preset time window, and the dielectric loss angle change rate refers to the rate of change of the loss tangent value of the cable insulation medium over time.

[0091] It should be noted that the maximum discharge capacity, discharge frequency, and dielectric loss angle change rate of the target cable are interrelated. When defects appear in the cable's internal insulation, an increase in the discharge frequency means that local discharge activity at the insulation defect becomes more frequent. Each discharge consumes a certain amount of energy, which in turn leads to increased dielectric loss and a larger dielectric loss angle change rate. At the same time, frequent discharges may gradually expand the scope and extent of the insulation defect, allowing subsequent discharges to release more energy, resulting in an increase in the maximum discharge capacity. Conversely, a larger maximum discharge capacity also indicates a more severe insulation defect, which often makes discharge more likely to occur, thereby increasing the discharge frequency, further exacerbating dielectric loss, and causing the dielectric loss angle change rate to continue to rise.

[0092] The reference maximum discharge capacity, reference discharge frequency, and reference dielectric loss angle change rate stored in the database are extracted.

[0093] Extract the maximum discharge capacity distribution coefficient, discharge frequency distribution coefficient and dielectric loss angle change rate distribution coefficient preset in the database.

[0094] It should be noted that the value ranges of the maximum discharge capacity distribution coefficient, the discharge frequency distribution coefficient, and the dielectric loss angle change rate distribution coefficient are all between 0 and 1, and the sum of the maximum discharge capacity distribution coefficient, the discharge frequency distribution coefficient, and the dielectric loss angle change rate distribution coefficient is 1. When used, the pre-set values can be directly extracted from the database. The specific extraction method is as follows: a one-to-one mapping set is constructed for the maximum discharge capacity, the discharge frequency, and the dielectric loss angle change rate, respectively, with the maximum discharge capacity distribution coefficient, the discharge frequency distribution coefficient, and the dielectric loss angle change rate distribution coefficient. When used, the obtained maximum discharge capacity, discharge frequency, and dielectric loss angle change rate are respectively input into the corresponding mapping set, thereby extracting the maximum discharge capacity distribution coefficient, the discharge frequency distribution coefficient, and the dielectric loss angle change rate distribution coefficient.

[0095] The second status indicator of the target cable is analyzed based on the target cable life status data.

[0096] The second state index of the target cable is the quantitative data of the degree of influence of the maximum discharge capacity, discharge frequency and dielectric loss angle change rate of the target cable on the target cable state. The specific analysis process is: comparing the reference value of the maximum discharge capacity of the target cable with the maximum discharge capacity, comparing the discharge frequency and dielectric loss angle change rate of the target cable, and coupling the comparison results with the corresponding distribution coefficient to obtain the second state index of the target cable.

[0097] In a specific embodiment, the second state indicator of the target cable is specifically expressed as follows:

[0098]

[0099] Among them, S it is the second state indicator of the target cable, E ins is the maximum discharge of the target cable, f PD is the discharge frequency of the target cable, δ is the change rate of the dielectric loss angle of the target cable, (E ins ) vef is the reference maximum discharge capacity, (f PD ) vef is the reference discharge frequency, δ vef is the reference medium loss angle change rate, is the maximum discharge capacity distribution coefficient, is the discharge frequency distribution coefficient, is the dielectric loss angle change rate distribution coefficient.

[0100] The first state correction index corresponding to the first state index interval of each target cable stored in the database is extracted, and the first state correction index corresponding to the interval where the first state index of the target cable is located is mapped and extracted, and recorded as the first state correction index of the target cable.

[0101] It should be understood that the larger the first state index of the target cable, the more significant the factors affecting the cable's aging and lifespan. For example, if it is a load-disturbance cable, a larger first state index means that the disturbance stress has caused a high degree of damage to the cable; if it is a load-stable cable, it indicates that the thermal-insulation degradation is relatively severe. In order to more accurately assess the comprehensive life state of the cable, when analyzing the comprehensive life assessment index, for cables that are more seriously affected, the subsequent correction of the second state index based on the first state correction index requires a more significant adjustment of the assessment results. Therefore, the larger the first state correction index of the corresponding extracted target cable, the more accurately the comprehensive life assessment index can reflect the actual condition of the cable, thereby improving the accuracy of the cable's full life cycle prediction.

[0102] The comprehensive life assessment index of the target cable is analyzed based on the second state index of the target cable and the first state correction index of the target cable.

[0103] The specific analysis method of the comprehensive life assessment index of the target cable is: based on the first state correction index of the target cable, the second state index of the target cable is corrected to obtain the comprehensive life assessment index of the target cable, which is specifically expressed as: multiplying the second state index of the target cable by the numerical result of the first state correction index of the target cable as the comprehensive life assessment index of the target cable.

[0104] See Figure 4 The figure shows the output flow chart of the convolutional neural network model involved in the embodiment of this application. Specifically, the model sets the interval and number of collection times for the comprehensive life assessment indicators, collects the indicators multiple times according to the set interval, forms a sequence, obtains the time features corresponding to these indicators, integrates the indicator sequence and time features into a time series input tensor, and inputs this tensor into the convolutional neural network model. The model extracts local features through the convolution layer, reduces the data dimension through the pooling layer, and finally outputs the prediction results through the fully connected layer, thereby obtaining the full life cycle prediction of the cable.

[0105] S4 collects the comprehensive life assessment indicator sequence of the target cable multiple times at fixed time intervals, and combines the time characteristics of the comprehensive life assessment indicators of the target cable to construct a time series input tensor, input it into the convolutional neural network model, and output the full life cycle prediction results of the cable.

[0106] In this embodiment, the full life cycle prediction results of the output cable are analyzed in the following specific steps:

[0107] A1, collecting and analyzing the comprehensive life assessment indicators of the target cable multiple times at fixed time intervals to obtain a comprehensive life assessment indicator sequence of the target cable, wherein the time interval and the number of collection times are pre-set in the database.

[0108] A2, obtain the collection period corresponding to the comprehensive life assessment indicator sequence of the target cable, record it as the time feature of the comprehensive life assessment indicator of the target cable, integrate the comprehensive life assessment indicator sequence of the target cable and the time feature of the comprehensive life assessment indicator of the target cable, use the time feature as an additional channel to splice with the original indicator sequence, reshape the integrated data into a tensor form suitable for the convolutional neural network model, and thus obtain the time series input tensor.

[0109] A3 inputs the time series input tensor into the convolutional neural network model. The convolutional neural network model extracts and learns features from the input data through convolutional layers, pooling layers, and fully connected layers, and outputs the full life cycle prediction results of the cable.

[0110] The convolutional layer slides convolution kernels of varying sizes across the input tensor to extract local features. The pooling layer reduces data dimensionality and computational complexity while retaining important features. The fully connected layer integrates the previously extracted features and outputs a full-lifecycle prediction result for the cable.

[0111] It should be noted that the full life cycle prediction results of the cable include the remaining life of the cable and the fault risk level.

[0112] See Figure 2 As shown, the second aspect of the present invention provides a cable life cycle prediction system based on a convolutional neural network, comprising:

[0113] The cable load mode determination module is used to collect target cable load status data within a preset monitoring window and perform target cable load mode determination.

[0114] The cable first state index analysis module is used to analyze the target cable first state index based on the target cable load mode determination result.

[0115] The cable comprehensive life assessment index analysis module is used to collect the life status data of the target cable, analyze the second status index of the target cable, and analyze the comprehensive life assessment index of the target cable in combination with the first status index of the target cable.

[0116] The cable life cycle prediction result output module is used to collect the comprehensive life assessment indicator sequence of the target cable multiple times at fixed time intervals, and combine the time characteristics of the comprehensive life assessment indicators of the target cable to construct a time series input tensor, input it into the convolutional neural network model, and output the cable life cycle prediction results.

[0117] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0118] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0119] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0121] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0122] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A cable life cycle prediction method based on convolutional neural network, characterized by: The following steps are involved: S1, within a preset monitoring window, collect target cable load status data and determine the target cable load mode; S2, analyzing a first state indicator of the target cable based on a target cable load mode determination result; S3, collecting life status data of the target cable, analyzing the second status index of the target cable, and combining the first status index of the target cable to analyze the comprehensive life assessment index of the target cable; S4 collects the comprehensive life assessment indicator sequence of the target cable multiple times at fixed time intervals, and combines the time characteristics of the comprehensive life assessment indicators of the target cable to construct a time series input tensor, input it into the convolutional neural network model, and output the full life cycle prediction results of the cable.

2. The cable life cycle prediction method based on convolutional neural network according to claim 1, characterized in that: The specific analysis method for determining the target cable load mode is as follows: Analyzing the load status index of the target cable based on the load status data of the target cable; Extracting the cable load status indicator threshold value preset in the database; If the load state index of the target cable is greater than the cable load state index threshold, the target cable load mode is recorded as a load disturbance cable; If the load state index of the target cable is less than or equal to the cable load state index threshold, the target cable load mode is recorded as a load-stable cable.

3. The cable life cycle prediction method based on convolutional neural network according to claim 2, characterized in that: The load status indicator of the target cable is analyzed, and the specific analysis process is as follows: Within the preset monitoring window, the target cable load status data is collected, including the target cable's average voltage, voltage fluctuation standard deviation, load switching frequency, and number of large jumps; Analyzing the load status index of the target cable based on the load status data of the target cable; The load state index of the target cable is a quantitative index of the degree of influence of the average voltage, voltage fluctuation standard deviation, load switching frequency and number of large jumps of the target cable on the load state of the target cable. The specific analysis process is: the average voltage of the target cable is deviated from its ideal value, the voltage fluctuation standard deviation, load switching frequency and number of large jumps of the target cable are differentiated from the corresponding reference values respectively, and the results of the deviation processing and differentiation processing are coupled with the corresponding distribution coefficient to obtain the load state index of the target cable.

4. The cable life cycle prediction method based on convolutional neural network according to claim 1, characterized in that: The specific process of analyzing the first state indicator of the target cable based on the target cable load mode determination result is as follows: Extracting the load state correction factor corresponding to the load state index interval of each target cable stored in the database, and mapping the extracted load state correction factor corresponding to the interval where the load state index of the target cable is located, and recording it as the load state correction factor of the target cable; If the target cable load mode is a load disturbance cable, the target cable disturbance stress data is collected, the target cable disturbance stress score is analyzed, and the target cable disturbance stress correction score is analyzed in combination with the target cable load state correction factor. The target cable disturbance stress correction score is recorded as the first state indicator of the target cable. If the target cable load mode is a load-stable cable, the thermal-insulation status data of the target cable is obtained, the thermal-insulation degradation comprehensive index of the target cable is analyzed, and the thermal-insulation degradation correction index of the target cable is analyzed in combination with the load status correction factor of the target cable. The thermal-insulation degradation correction index of the target cable is recorded as the first status indicator of the target cable.

5. The cable life cycle prediction method based on convolutional neural network according to claim 4, characterized in that: The specific analysis process of the disturbance stress correction score of the target cable is as follows: Within a preset time window, the disturbance stress data of the target cable is collected, including the temperature rise rate, thermal cycle frequency, strain peak value, and start-stop times of the target cable; Analyzing the disturbance stress score of the target cable based on the disturbance stress data of the target cable; The disturbance stress score of the target cable is a quantitative indicator of the degree to which the temperature rise rate, thermal cycle frequency, strain peak value, and start-stop number of the target cable jointly affect the disturbance stress state of the target cable. The specific analysis process is: the temperature rise rate, thermal cycle frequency, strain peak value, and start-stop number of the target cable are differentiated from their corresponding reference values, and the differentiation results are coupled with the corresponding distribution factors to obtain the disturbance stress score of the target cable; Analyze the disturbance stress correction score of the target cable according to the disturbance stress score of the target cable and the load state correction factor of the target cable; The specific analysis process of the disturbance stress correction score of the target cable is: correcting the disturbance stress score of the target cable based on the load state correction factor of the target cable, thereby obtaining the disturbance stress correction score of the target cable.

6. The cable life cycle prediction method based on convolutional neural network according to claim 4, characterized in that: The specific analysis process of the thermal-insulation degradation correction index of the target cable is as follows: Obtain thermal-insulation status data of the target cable, including the operating time of the target cable, the conductor resistivity change rate, and the operating voltage deviation rate; Based on the thermal-insulation status data of the target cable, analyze the thermal-insulation degradation comprehensive index of the target cable; The thermal-insulation degradation comprehensive index of the target cable is a quantitative indicator of the degree to which the operating time, conductor resistivity change rate, and operating voltage offset rate of the target cable jointly affect the thermal-insulation degradation state of the target cable. The specific analysis process is as follows: the operating time, conductor resistivity change rate, and operating voltage offset rate absolute values of the target cable are differentiated from their corresponding reference values, and the differentiation results are coupled with the corresponding distribution factors to obtain the thermal-insulation degradation comprehensive index of the target cable; Analyze the thermal-insulation degradation correction index of the target cable according to the thermal-insulation degradation comprehensive index of the target cable and the load state correction factor of the target cable; The specific analysis process of the thermal-insulation degradation correction index of the target cable is: based on the load state correction factor of the target cable, the thermal-insulation degradation comprehensive index of the target cable is corrected to obtain the thermal-insulation degradation correction index of the target cable.

7. The cable life cycle prediction method based on convolutional neural network according to claim 1, characterized in that: The comprehensive life assessment index of the target cable is analyzed, and the specific analysis process is as follows: Analyzing a second status indicator of the target cable based on the target cable life status data; Extracting the first state correction index corresponding to the first state index interval of each target cable stored in the database, and mapping the extracted first state correction index corresponding to the interval where the first state index of the target cable is located, and recording it as the first state correction index of the target cable; Analyzing a comprehensive life assessment index of the target cable based on a second state index of the target cable and a first state correction index of the target cable; The specific analysis method of the comprehensive life assessment index of the target cable is: based on the first state correction index of the target cable, the second state index of the target cable is corrected to obtain the comprehensive life assessment index of the target cable.

8. The cable life cycle prediction method based on convolutional neural network according to claim 7, characterized in that: The specific analysis steps for analyzing the second status indicator of the target cable are as follows: Collect target cable life status data within a preset time window; The target cable life status data includes the maximum discharge amount, discharge frequency and dielectric loss angle change rate of the target cable; Analyzing a second status indicator of the target cable based on the target cable life status data; The second state index of the target cable is quantitative data on the degree of influence of the maximum discharge amount, discharge frequency and dielectric loss angle change rate of the target cable on the target cable state. The specific analysis process is: comparing the reference value of the maximum discharge amount of the target cable with the maximum discharge amount, comparing the discharge frequency and dielectric loss angle change rate of the target cable, and coupling the comparison results with the corresponding distribution coefficient to obtain the second state index of the target cable.

9. The cable life cycle prediction method based on convolutional neural network according to claim 1, characterized in that: The full life cycle prediction results of the output cable are analyzed in the following specific steps: A1, collecting and analyzing the comprehensive life assessment indicators of the target cable multiple times at fixed time intervals to obtain a comprehensive life assessment indicator sequence of the target cable; A2, obtaining the acquisition period corresponding to the comprehensive life assessment indicator sequence of the target cable, recording it as the time feature of the comprehensive life assessment indicator of the target cable, integrating the comprehensive life assessment indicator sequence of the target cable with the time feature of the comprehensive life assessment indicator of the target cable, splicing the time feature with the original indicator sequence as an additional channel, and reshaping the integrated data into a tensor form suitable for the convolutional neural network model, thereby obtaining a time series input tensor; A3 inputs the time series input tensor into the convolutional neural network model. The convolutional neural network model extracts and learns features from the input data through convolutional layers, pooling layers, and fully connected layers, and outputs the full life cycle prediction results of the cable.

10. A system using the cable life cycle prediction method based on a convolutional neural network as described in any one of claims 1 to 9, characterized in that: include: The cable load mode determination module is used to collect target cable load status data within a preset monitoring window and perform target cable load mode determination; a cable first state index analysis module, configured to analyze a target cable first state index based on a target cable load mode determination result; The cable comprehensive life assessment index analysis module is used to collect the life status data of the target cable, analyze the second status index of the target cable, and analyze the comprehensive life assessment index of the target cable in combination with the first status index of the target cable; The cable life cycle prediction result output module is used to collect the comprehensive life assessment indicator sequence of the target cable multiple times at fixed time intervals, and combine the time characteristics of the comprehensive life assessment indicators of the target cable to construct a time series input tensor, input it into the convolutional neural network model, and output the cable life cycle prediction results.

Images