A method for assessing the life of old cables based on big data

Abstract

The present invention relates to the technical field of aging analysis, and specifically discloses a method for evaluating the life of old cables based on big data, comprising the following steps: S1: evaluating the remaining life of the cable, calculating the remaining life difference, and determining the abnormal cable according to the remaining life difference; S2: determining the functional relationship between the first target image and the second target image; calculating the first change rate and the second change rate, and determining the abnormal period according to the first change rate and the second change rate, and calculating the abnormal proportion; S3: calculating the judgment value according to the functional relationship of the first target image, the functional relationship of the second target image and the abnormal proportion, determining the target cable according to the judgment value, and updating the remaining life of the target cable with a preset update cycle. The present invention can screen the cables whose remaining life needs to be updated according to the deviation between the actual decline of the remaining life of the cable and the initial and expected decline, thereby reducing the cost required for the overall update.

Description

Technical Field

[0001] The present invention relates to the technical field of aging analysis, and in particular to a method for evaluating the life of old cables based on big data. Background Art

[0002] Cable life refers to the length of time from the time the cable is put into use to the time it can no longer be used safely and effectively due to aging, damage or performance degradation under normal conditions. The cable life is affected by many factors, including material aging, environmental conditions (such as temperature, humidity, chemical corrosion), electrical load, mechanical stress, etc.

[0003] In the existing technology, most of the influencing factors within a period of time are collected and the cable life is predicted through a pre-established cable life prediction model. This method can accurately predict the remaining life of the cable. However, in actual situations, influencing factors such as load conditions and temperature changes may change significantly. These new changes may cause the aging of the cable to accelerate, resulting in a deviation between the actual remaining life of the cable and the predicted remaining life.

[0004] The deviation between the actual remaining life of the cable and the predicted remaining life may result in the staff being unable to replace the cable whose remaining life has expired in time, thereby increasing the probability of safety accidents. This problem can be effectively solved by updating the remaining life of the cable. However, in actual applications, the number of cables is large, and the computing resources required to update the remaining life of all cables are very large, which will lead to a significant increase in the cost of evaluating the cable life. Summary of the invention

[0005] The purpose of the present invention is to provide a method for assessing the life of old cables based on big data to solve the following technical problems:

[0006] The deviation between the actual remaining life of the cable and the predicted remaining life may result in the staff being unable to replace the cable whose remaining life has expired in time, thereby increasing the probability of safety accidents. This problem can be effectively solved by updating the remaining life of the cable. However, in actual applications, the number of cables is large, and the computing resources required to update the remaining life of all cables are very large, which will lead to a significant increase in the cost of evaluating the cable life.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for assessing the life of old cables based on big data includes the following steps:

[0009] S1: Within the preset monitoring period T, the remaining life S of the cable is periodically evaluated from the starting point of the monitoring period. The time interval between two consecutive evaluations of the remaining life of the cable is T. int, and calculate the remaining life difference Ssy i =S i -S i+1 , S i represents the remaining life of the cable at the i-th evaluation, S i+1 Indicates the remaining life of the cable at the i+1th evaluation. When the remaining life difference Ssy of the cable i ≥1.2T int When the cable is marked as abnormal cable;

[0010] S2: Generate data points (t i , S i ), t i Represents the time interval between the time of the i-th evaluation of the remaining life of the cable and the starting point of the monitoring period, connects two adjacent data points by a straight line to obtain a first target image, and determines the functional relationship f(t) of the first target image;

[0011] Generate coordinate points (t i ,S1-i*T int ), obtain a second target image by connecting two adjacent coordinate points through a straight line, and determine a functional relationship h(t) of the second target image;

[0012] Calculate the first rate of change and the second rate of change , t i+1 represents the time interval between the time of the i+1th evaluation of the remaining life of the cable and the starting point of the monitoring cycle, and when the first change rate C<Csec i When t i to i+1 Mark as abnormal period, and calculate the abnormal proportion P=Byc / (n-1), Byc represents the total number of abnormal periods;

[0013] S3: Calculate the judgment value K according to the functional relationship of the first target image, the functional relationship of the second target image and the abnormal proportion, determine the target cable according to the judgment value K, and update the remaining life of the target cable at a preset update period Tgx.

[0014] As a further solution of the present invention: in step S3, the process of determining the target cable specifically includes:

[0015] The judgment value K is calculated by a formula, and the formula specifically includes:

[0016] ;

[0017] Wherein, ε represents a preset correction coefficient, P represents the abnormal proportion, C represents the first change rate, T represents the length of the monitoring period, Byc represents the total number of abnormal time periods, Cyc b represents the second rate of change corresponding to the bth abnormal period, g(t)=h(t)-f(t);

[0018] When the judgment value K≥Kpd, the corresponding abnormal cable is marked as a target cable, and Kpd represents a preset judgment value threshold.

[0019] As a further solution of the present invention: in the step S3, the process of updating the remaining life of the target cable at a preset update cycle specifically includes:

[0020] Let f(t) = T safe , solve to get the predicted time Tys, T safe Indicates the preset target cable replacement safety time;

[0021] The update period Tgx = Tys / K is calculated, and the remaining life of the target cable is evaluated when the time interval from the end of the monitoring period reaches Tgx.

[0022] As a further solution of the present invention: when the cable is at the starting point T of the monitoring cycle sta The remaining life on A≥γ*T safe When γ is the preset value, the subsequent steps are not executed, and the starting point of the monitoring cycle is adjusted to T sta +T int , and repeat the above steps until the starting point of the monitoring cycle is adjusted once, and the remaining life at the starting point of the monitoring cycle is A<γ*T safe .

[0023] As a further solution of the present invention: Determine the remaining life A<γ*T safe The starting point of the monitoring period Ts sta , and let the monitoring period be Ts sta -T to Ts sta , and follow the next steps.

[0024] As a further solution of the present invention: in the step S1, when the remaining life of the cable S < T safe When the error occurs, the subsequent steps are not executed and an early warning message is sent to remind the staff.

[0025] As a further solution of the present invention: in step S1, the process of evaluating the remaining life of the cable specifically includes:

[0026] Establishing a database, wherein the database stores influencing factors of the marked remaining life variation, wherein the influencing factors are factors that affect the remaining life of the cable;

[0027] Establish a remaining life change prediction model through deep learning model, and train and verify it through database;

[0028] Determine the time tpre of the last evaluation of the remaining life of the cable and the remaining life Spre of the cable at the last evaluation, and convert the time tpre to the time tpre+T int The influencing factors collected in the system are input into the remaining life change prediction model, the remaining life change prediction value Syc is output, and tpre+T is determined. int The remaining life of the cable is SSpre = Spre-Syc-T int .

[0029] As a further solution of the present invention: in the step S1, when the total time tsy of the cable being put into use is less than 0.2Tsy, its remaining life is not evaluated until the total time tsy of the cable being put into use is greater than or equal to 0.2Tsy, where Tsy represents the designed service life of the cable.

[0030] Beneficial effects of the present invention: In the present invention, firstly, abnormal cables are screened according to the difference in the remaining life of the cables; it can be understood that the abnormal cables are screened from all the cables to reduce the amount of data to be processed later and improve the overall processing efficiency. int When the remaining life of the cable decreases, it means that the remaining life of the cable is within an acceptable range, so there is no need to process it, so that the remaining life difference is greater than 1.2T intThe reasons include but are not limited to load conditions and temperature changes. For example, when the cable is under high load or overload operation for a long time, the heat generated by the current passing through the cable will increase. If the cable carries a current exceeding its design capacity for a long time, the cable's insulation material will age, harden, or even crack due to the high temperature, which will increase the rate of decrease in the remaining life of the cable. The length T of the monitoring period and the time interval between two adjacent assessments of the remaining life of the cable can be set based on experience; then determine the functional relationship between the first target image and the second target image, and calculate the abnormal proportion based on the functional relationship between the first target image and the second target image; it is worth noting that the functional relationship of the first target image reflects the actual decrease in the remaining life of the abnormal cable, and the functional relationship of the second target image reflects the theoretical decrease in the remaining life of the abnormal cable. In the subsequent process, g(t) is used to quantify the deviation between the actual decrease and the initial and expected decrease; it should be noted that the determination of the abnormal period is to determine which periods The actual decline in the remaining life of the cable deviates from expectations, and the abnormal ratio reflects the proportion of the period in which the remaining life decline deviates from expectations. The larger the abnormal ratio, the more attention should be paid to the cable to prevent the cable life from declining more due to lack of timely attention, which may cause safety hazards; finally, the judgment value is calculated according to the functional relationship of the first target image, the functional relationship of the second target image and the abnormal ratio, and the target cable is determined according to the judgment value, and the remaining life of the target cable is updated with a preset update cycle; it can be understood that the abnormal cable may only have a residual life difference exceeding expectations in a certain period of time, and the residual life decline in the remaining periods meets the expected situation. The judgment value is calculated by the functional relationship of the first target image, the functional relationship of the second target image and the abnormal ratio to comprehensively judge whether the abnormal cable is the target cable, which reduces the probability of misjudgment, and secondary screening from the abnormal cables can reduce the number of cables that need to update the remaining life, thereby reducing the cost of overall update. The present invention can screen the cables that need to update the remaining life according to the deviation between the actual decline in the remaining life of the cable and the initial and expected decline, thereby reducing the cost required for overall update. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The present invention will be further described below in conjunction with the accompanying drawings.

[0032] Figure 1 It is a flow chart of a method for assessing the life of old cables based on big data according to the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1 As shown, the present invention is a method for assessing the life of old cables based on big data, comprising the following steps:

[0035] S1: Within the preset monitoring period T, the remaining life S of the cable is periodically evaluated from the starting point of the monitoring period. The time interval between two consecutive evaluations of the remaining life of the cable is T. int , and calculate the remaining life difference Ssy i =S i -S i+1 , S i represents the remaining life of the cable at the i-th evaluation, S i+1 Indicates the remaining life of the cable at the i+1th evaluation. When the remaining life difference Ssy of the cable i ≥1.2T int When the cable is marked as abnormal cable;

[0036] S2: Generate data points (t i , S i ), t i Represents the time interval between the time of the i-th evaluation of the remaining life of the cable and the starting point of the monitoring period, connects two adjacent data points by a straight line to obtain a first target image, and determines the functional relationship f(t) of the first target image;

[0037] Generate coordinate points (t i ,S1-i*T int ), obtain a second target image by connecting two adjacent coordinate points through a straight line, and determine a functional relationship h(t) of the second target image;

[0038] Calculate the first rate of change and the second rate of change , t i+1 represents the time interval between the time of the i+1th evaluation of the remaining life of the cable and the starting point of the monitoring cycle, and when the first change rate C<Csec i When t i to i+1 Mark as abnormal period, and calculate the abnormal proportion P=Byc / (n-1), Byc represents the total number of abnormal periods;

[0039] S3: Calculate the judgment value K according to the functional relationship of the first target image, the functional relationship of the second target image and the abnormal proportion, determine the target cable according to the judgment value K, and update the remaining life of the target cable at a preset update period Tgx.

[0040] First, the abnormal cables are screened according to the difference in the remaining life of the cables; it can be understood that the abnormal cables are screened from all the cables to reduce the amount of data to be processed later and improve the overall processing efficiency. int When the remaining life of the cable decreases, it means that the remaining life of the cable is within an acceptable range, so there is no need to process it, so that the remaining life difference is greater than 1.2T int The reasons include but are not limited to load conditions and temperature changes. For example, when the cable is under high load or overload operation for a long time, the heat generated by the current passing through the cable will increase. If the cable carries a current exceeding its design capacity for a long time, the cable's insulation material will age, harden, or even crack due to the high temperature, which will increase the rate of decrease in the remaining life of the cable. The length T of the monitoring period and the time interval between two adjacent assessments of the remaining life of the cable can be set based on experience; then determine the functional relationship between the first target image and the second target image, and calculate the abnormal proportion based on the functional relationship between the first target image and the second target image; it is worth noting that the functional relationship of the first target image reflects the actual decrease in the remaining life of the abnormal cable, and the functional relationship of the second target image reflects the theoretical decrease in the remaining life of the abnormal cable. In the subsequent process, g(t) is used to quantify the deviation between the actual decrease and the initial and expected decrease; it should be noted that the determination of the abnormal period is to determine which periods The actual decline in the remaining life of the cable deviates from the expectation, and the abnormal ratio reflects the proportion of the time period in which the decline in the remaining life deviates from the expectation. The larger the abnormal ratio, the more attention should be paid to the cable to prevent the cable life from declining too much due to lack of timely attention, which may cause safety hazards. Finally, the judgment value is calculated according to the functional relationship of the first target image, the functional relationship of the second target image and the abnormal ratio, and the target cable is determined according to the judgment value, and the remaining life of the target cable is updated with a preset update cycle. It can be understood that the abnormal cable may have a remaining life difference that exceeds expectations in a certain period of time, and the decline in the remaining life in the remaining periods meets the expectations. The judgment value is calculated by the functional relationship of the first target image, the functional relationship of the second target image and the abnormal ratio to comprehensively judge whether the abnormal cable is the target cable, thereby reducing the probability of misjudgment, and secondary screening from the abnormal cables can reduce the number of cables that need to update the remaining life, thereby reducing the overall update cost.

[0041] In another preferred embodiment of the present invention, in step S3, the process of determining the target cable specifically includes:

[0042] The judgment value K is calculated by a formula, and the formula specifically includes:

[0043] ;

[0044] Wherein, ε represents a preset correction coefficient, P represents the abnormal proportion, C represents the first change rate, T represents the length of the monitoring period, Byc represents the total number of abnormal time periods, Cyc b represents the second rate of change corresponding to the bth abnormal period, g(t)=h(t)-f(t);

[0045] When the judgment value K≥Kpd, the corresponding abnormal cable is marked as a target cable, and Kpd represents a preset judgment value threshold.

[0046] In another preferred embodiment of the present invention, in step S3, the process of updating the remaining life of the target cable at a preset update period specifically includes:

[0047] Let f(t) = T safe , solve to get the predicted time Tys, T safe Indicates the preset target cable replacement safety time;

[0048] The update period Tgx = Tys / K is calculated, and the remaining life of the target cable is evaluated when the time interval from the end of the monitoring period reaches Tgx.

[0049] Understandably, T safe It is used to ensure that the target cable will not exceed or reach the limit of its safe service life. Before that, the remaining life of the target cable is evaluated again, and the remaining life of the cable is updated with the evaluated remaining life, so as to prevent safety accidents caused by the failure to replace the cable that has reached the upper limit of its life in time; for cables that are not target cables, the remaining life at the end of the monitoring period is used to remind the staff to replace them in time, and the remaining life at the end of the monitoring period is not directly used as the remaining life of all cables. The reasons include: if the difference between the decrease of the remaining life of the target cable during the monitoring period and the expectation is large, it may not actually reach the remaining life evaluated at the end of the monitoring period, and intervention measures need to be taken in advance, so as to avoid the safety hazards caused by the failure to replace the target cable whose actual remaining life has reached the replacement condition in time;

[0050] It is worth noting that in this scheme, the first image is obtained by connecting two adjacent data points by a straight line, which means that the first image is a line graph. At the same time, when the remaining life evaluated each time is sorted in the order of the time axis, the monotonicity of the sorting is monotonically decreasing, which further indicates that the first target image is monotonically decreasing. In summary, f(t) is a monotonically decreasing line graph. On this basis, the predicted time Tys (i.e., determining the remaining life to reach T) is obtained. safe time), solve f(t)=T safe Then, f(t) decreases monotonically, which means that the above formula (i.e. f(t) = T safe ) There is only one unique solution, which is the prediction time Tys.

[0051] In a preferred embodiment of the present invention, when the cable is at the starting point T of the monitoring cycle, sta The remaining life on A≥γ*T safe When γ is the preset value, the subsequent steps are not executed, and the starting point of the monitoring cycle is adjusted to T sta +T int , and repeat the above steps until the starting point of the monitoring cycle is adjusted once, and the remaining life at the starting point of the monitoring cycle is A<γ*T safe .

[0052] It should be noted that this step can avoid unnecessary evaluation and adjustment of the cable too frequently. When the remaining life of the cable is still significantly higher than the safety threshold, extending the monitoring period can reduce the frequency of unnecessary operations, optimize the use of resources, and avoid unnecessary processing of cables that are still in a safe state, thereby improving the operating efficiency of the entire system and saving time and costs.

[0053] In a preferred case of this embodiment, the remaining life is determined as A<γ*T safe The starting point of the monitoring period Ts sta , and let the monitoring period be Ts sta -T to Ts sta , and follow the next steps.

[0054] In a preferred embodiment of the present invention, in step S1, when the remaining life of the cable S<T safe When the error occurs, the subsequent steps are not executed and an early warning message is sent to remind the staff.

[0055] It is understandable that when the remaining life of the cable S<T safeWhen the cable is close to or has reached its safe use limit, if the usual monitoring and evaluation process is continued, the cable may continue to work without timely treatment, increasing the risk of failure. By sending early warning information, the staff can be immediately reminded to take necessary measures, such as further checking the cable status, emergency replacement or deactivating the cable to avoid potential safety hazards. When the remaining life of the cable is already below the safety threshold, it may not make sense to continue with subsequent steps (such as further calculation, evaluation or adjustment of the monitoring cycle) because the cable has been judged to be in an unsafe state. The top priority is to ensure safety, not to optimize the monitoring cycle.

[0056] In another preferred embodiment of the present invention, in step S1, the process of evaluating the remaining life of the cable specifically includes:

[0057] Establishing a database, wherein the database stores influencing factors of the marked remaining life variation, wherein the influencing factors are factors that affect the remaining life of the cable;

[0058] Establish a remaining life change prediction model through deep learning model, and train and verify it through database;

[0059] Determine the time tpre of the last evaluation of the remaining life of the cable and the remaining life Spre of the cable at the last evaluation, and convert the time tpre to the time tpre+T int The influencing factors collected in the system are input into the remaining life change prediction model, the remaining life change prediction value Syc is output, and tpre+T is determined. int The remaining life of the cable is SSpre = Spre-Syc-T int .

[0060] It is worth noting that through the deep learning model, a large amount of historical data can be used to accurately predict the change in the remaining life of the cable, thereby improving the accuracy of the assessment; influencing factors include but are not limited to ambient temperature, humidity, load conditions, etc. Since there may be complex nonlinear relationships between these factors, and the deep learning model has powerful data processing capabilities, it can automatically extract features from a large amount of data and consider the impact of multiple factors on life, thereby giving a more reliable prediction and avoiding premature or late replacement of the cable.

[0061] In another preferred embodiment of the present invention, in step S1, when the total time the cable is put into use tsy<0.2Tsy, its remaining life is not evaluated until the total time the cable is put into use tsy≥0.2Tsy, where Tsy represents the designed service life of the cable.

[0062] It is worth noting that, usually, cables have a relatively stable operation stage in the early stage of use, during which the performance of the cables changes little, and the aging and wear rate is also slow. Therefore, during this period, the necessity of remaining life assessment is low, because the state of the cable usually does not change significantly; frequent life assessment in the early stage of cable operation may waste monitoring resources and manpower and material resources. Since the cable is still in the early stage of its service life, early assessment may not provide valuable information, but will increase unnecessary workload and cost.

[0063] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for assessing the life of old cables based on big data, characterized in that: The following steps are involved: S1: Within the preset monitoring period T, the remaining life S of the cable is periodically evaluated from the starting point of the monitoring period. The time interval between two consecutive evaluations of the remaining life of the cable is Tint, and the remaining life difference Ssyi = Si-Si is calculated. +1 , Si represents the remaining life of the cable at the i-th evaluation, Si +1 Indicates the remaining life of the cable at the i+1th evaluation. When the remaining life difference of the cable Ssyi ≥ 1.2Tint, the cable is marked as an abnormal cable; S2: Generate data points (ti, Si), where ti represents the time interval between the time of the i-th assessment of the remaining life of the cable and the starting point of the monitoring period, connect two adjacent data points by a straight line to obtain a first target image, and determine the functional relationship f(t) of the first target image; Generate a coordinate point (ti, S1-i*Tint), connect two adjacent coordinate points with a straight line to obtain a second target image, and determine a functional relationship h(t) of the second target image; Calculate the first rate of change and the second rate of change ti +1 represents the time interval between the time of the i+1th evaluation of the remaining life of the cable and the starting point of the monitoring cycle, and when the first change rate C<Cseci, ti to ti +1 Mark as abnormal time period, and calculate the abnormal proportion P=Byc / (n-1), Byc represents the total number of abnormal time periods; S3: Calculate a judgment value K according to the functional relationship of the first target image, the functional relationship of the second target image and the abnormal ratio, determine the target cable according to the judgment value K, and update the remaining life of the target cable at a preset update period Tgx; The process of determining the target cable specifically includes: The judgment value K is calculated by a formula, and the formula specifically includes: Wherein, ε represents a preset correction coefficient, P represents the abnormal proportion, C represents the first change rate, T represents the length of the monitoring period, Byc represents the total number of abnormal time periods, Cycb represents the second change rate corresponding to the b-th abnormal time period, and g(t)=h(t)-f(t); When the judgment value K≥Kpd, the corresponding abnormal cable is marked as a target cable, and Kpd represents a preset judgment value threshold.

2. The old cable life assessment method based on big data according to claim 1 is characterized in that: In step S3, the process of updating the remaining life of the target cable at a preset update cycle specifically includes: Let f(t) = Tsafe, and solve to get the predicted time Tys, where Tsafe represents the preset target cable replacement safety time; The update period Tgx=Tys / K is calculated, and the remaining life of the target cable is evaluated when the time interval from the end of the monitoring period reaches Tgx.

3. The old cable life assessment method based on big data according to claim 2 is characterized in that: When the remaining life A of the cable at the starting point Tsta of the monitoring cycle is ≥γ*Tsafe, γ is a preset value, the subsequent steps are not executed, and the starting point of the monitoring cycle is adjusted to Tsta+Tint, and the above steps are repeated until the starting point of the monitoring cycle is adjusted again, and the remaining life A at the starting point of the monitoring cycle is <γ*Tsafe.

4. The old cable life assessment method based on big data according to claim 3 is characterized in that: Determine the starting point Tssta of the monitoring period when the remaining life A<γ*Tsafe, set the monitoring period to Tssta-T to Tssta, and execute subsequent steps.

5. The old cable life assessment method based on big data according to claim 4 is characterized in that: In the step S1, when the remaining life of the cable S<Tsafe in the monitoring period, the subsequent steps are not executed and an early warning message is sent to remind the staff.

6. The old cable life assessment method based on big data according to claim 1 is characterized in that: In step S1, the process of evaluating the remaining life of the cable specifically includes: Establishing a database, wherein the database stores influencing factors of the marked remaining life variation, wherein the influencing factors are factors that affect the remaining life of the cable; Establish a remaining life change prediction model through deep learning model, and train and verify it through database; Determine the time tpre when the remaining life of the cable was evaluated last and the remaining life Spre of the cable at the last adjacent evaluation, and input the influencing factors collected from time tpre to time tpre+Tint into the remaining life change prediction model, output the remaining life change prediction value Syc, and determine that the remaining life of the cable SSpre=Spre-Syc-Tint at tpre+Tint.

7. The old cable life assessment method based on big data according to claim 1 is characterized in that: In the step S1, when the total time tsy of the cable being put into use is less than 0.2Tsy, the remaining life of the cable is not evaluated until the total time tsy of the cable being put into use is greater than or equal to 0.2Tsy, where Tsy represents the designed service life of the cable.

Images