High-temperature superconducting cable life prediction method and system combined with working condition behavior analysis

Abstract

The application discloses a high-temperature superconducting cable life prediction method and system combined with working condition behavior analysis, relates to the technical field of life prediction and reliability analysis of high-temperature superconducting cables, and comprises the following steps: collecting working condition data of a high-temperature superconducting cable in a running process; performing mode division on the running working condition based on the working condition data through a clustering method, and establishing a working condition mode transition probability matrix; performing equivalent processing on the running data in the working condition mode, and establishing a bidirectional coupling degradation model according to the equivalent processing result; and outputting a reliability function, a residual life of the cable system and a life contribution rate in each working condition mode according to the working condition mode transition probability matrix and the bidirectional coupling degradation model. Through working condition analysis and bidirectional coupling modeling, the application solves the problem of inaccurate cable life prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of life prediction and reliability analysis of high-temperature superconducting cables, and more particularly to a high-temperature superconducting cable life prediction method and system combined with working condition behavior analysis. BACKGROUND

[0002] High-temperature superconducting cables are becoming an important technology for improving power transmission efficiency and power quality in modern power systems due to their zero resistance, large power transmission capacity, and significant energy saving and emission reduction effects. However, in actual operation, high-temperature superconducting cables are affected by various working condition factors such as current fluctuation, temperature change, mechanical stress, and uneven electric field distribution, which causes complex degradation processes in the insulation materials and superconductors themselves. These degradations not only directly affect the reliability and safe operation of the cable, but also relate to the stability and economy of the power system. Therefore, accurately predicting the remaining life of high-temperature superconducting cables and identifying high-risk working condition modes are of great significance for extending the service life of the cable, optimizing operation and maintenance strategies, and reducing operation risks.

[0003] For example, the invention patent with publication number CN120493627A discloses an interactive simulation method for calculating the aging parameters of superconducting cables and a superconducting cable system. The system builds an electromagnetic transient model of the cable body, a transient thermal model of the cooling system, and a three-dimensional multi-field finite element model, inputs cable material properties, environmental parameters, and operating condition data, constructs a model interaction interface through a simulation platform, realizes data interaction and coupling between multiple models, dynamically adjusts simulation parameters, and ensures precise matching of model step size and system dynamic characteristics to calculate the aging parameters of the superconducting cable system under complex working conditions. The present application analyzes mixed data to evaluate the operating state of the superconducting cable and performs aging analysis of the superconducting cable system under multiple working conditions to provide decision support for system maintenance.

[0004] For example, the invention patent with publication number CN120405285A discloses a superconducting cable aging state evaluation system and method based on mixed data. The system includes a data analysis module, an aging evaluation module, a physicochemical correlation module, and a report generation module connected in communication. The data analysis module includes a simulation submodule, a data acquisition submodule, and a data processing submodule connected thereto. During the aging state evaluation process performed by the aging evaluation module, the superconducting cable system model parameters are output and calibrated through a physicochemical correlation model, and then the aging state evaluation fitting function is output through the data processing submodule, and the aging state evaluation report is automatically generated by the report generation module. The present application analyzes mixed data to evaluate the operating state of the superconducting cable and performs aging analysis of the superconducting cable system under multiple working conditions to provide decision support for system maintenance.

[0005] The above-mentioned technical solutions at least have the following technical problems:

[0006] Traditional technologies typically use a single constant operating condition or a simple load condition for life prediction, ignoring the complex and varied operating conditions of cables in actual operation. This results in low prediction accuracy and insufficient reliability. Furthermore, they often consider the effects of temperature, strain, or electric field factors on cable degradation in isolation, neglecting the interaction between these factors and failing to reflect the comprehensive degradation characteristics of high-temperature superconducting cables under actual operating conditions.

[0007] To address the above problems, this invention proposes a solution. Summary of the Invention

[0008] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method and system for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis. Through operating condition analysis and bidirectional coupling modeling, the problem of inaccurate cable lifespan prediction is solved.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] A method for predicting the lifespan of high-temperature superconducting cables based on operating condition behavior analysis includes: collecting operating condition data of the high-temperature superconducting cable during operation; classifying the operating conditions into patterns using clustering methods based on the operating condition data and establishing an operating condition pattern transition probability matrix; performing equivalent processing on the operating data under each operating condition pattern and establishing a two-way coupled degradation model based on the equivalent processing results; and outputting the reliability function, remaining lifespan, and lifespan contribution rate under each operating condition pattern of the cable system based on the operating condition pattern transition probability matrix and the two-way coupled degradation model.

[0011] In a preferred embodiment, the operation mode is divided into patterns based on the operating condition data using a clustering method, specifically as follows: Operating condition data of the high-temperature superconducting cable during operation is collected; operating condition features are extracted from the data to generate operating condition feature vectors, and the critical current margin is output; the operating condition feature vectors are clustered using the K-means algorithm, and the weighted Euclidean distance between any two operating condition vectors is output during the clustering process, combined with the critical current margin, forming several clusters; through iterative calculation, for each operating condition point, the weighted Euclidean distance to the center of each cluster is calculated, and the point is assigned to the cluster with the smallest weighted Euclidean distance, until the cluster centers converge; for each operating condition point within a cluster, if the weighted Euclidean distance exceeds a preset threshold, the point is classified into the unclassified point set; otherwise, it is classified into the steady-state operating condition cluster; the density-based DBSCAN algorithm performs secondary clustering on the unclassified point set, outputting abnormal operating condition clusters, and each cluster is verified.

[0012] In a preferred embodiment, the verification of each cluster is specifically as follows: taking the steady-state operating condition cluster and the abnormal operating condition cluster as input, the mean and variance of the operating condition characteristics corresponding to each cluster are obtained, a cluster representative operating condition vector is generated, and the number and time distribution of operating condition points within the cluster are recorded; according to the occurrence time distribution of each cluster, the cluster representative operating conditions are arranged in the order of actual operating time to form an equivalent time series of operating conditions; the generated equivalent operating condition series is physically constrained and verified by the critical current-temperature characteristic curve of the high-temperature superconducting cable and the allowable temperature rise boundary.

[0013] In a preferred embodiment, the density-based DBSCAN algorithm performs secondary clustering on the unclassified point set, specifically as follows: Based on historical operating condition data, the distance threshold and minimum number of points required by the DBSCAN algorithm are adaptively calculated; according to the obtained distance threshold and minimum number of points, the unclassified point set is clustered, points with higher density are aggregated into the same cluster, and points with sparse neighborhoods that cannot form clusters are marked as noise, resulting in several abnormal operating condition clusters; for each abnormal operating condition cluster, the feature vector attribute of the cluster center is extracted, and the cluster type is determined to form an abnormal operating condition cluster.

[0014] In a preferred embodiment, the establishment of the operating condition mode transition probability matrix is ​​specifically as follows: the equivalent time series of operating conditions is divided into multiple time scale windows. Under each time scale, the module counts the number of transitions between adjacent time slices according to the sliding window to form an operating condition cluster transition count matrix under the time scale. Based on the transition count matrix, the transition probability matrix is ​​output to obtain the transition count matrix of each time scale. The operating condition mode transition probability matrix is ​​obtained by weighted fusion of the transition probability matrices of each time scale.

[0015] In a preferred embodiment, the process of equivalence processing of the operating data under each operating condition and establishing a bidirectional coupled degradation model based on the equivalence results is as follows: Strain and temperature data under each operating condition are acquired and rainflow counting is performed to decompose the strain and temperature waveforms into standard load cycles, obtaining the equivalent cycle number for strain and the equivalent cycle number for temperature, respectively; the equivalent cycle number for strain and the equivalent cycle number for temperature are normalized and weighted according to the critical current margin to obtain the comprehensive equivalent cycle number; the peak factor and harmonic content of the voltage signal under each operating condition are extracted, the harmonic correction factor is calculated, and it is introduced into the electric field calculation model to generate an equivalent electric field distribution, obtaining the equivalent electric field intensity of harmonic distortion; based on the equivalent cycle number and the equivalent electric field intensity, a critical current degradation function and an insulation damage degree evolution function are established; based on the bidirectional coupling relationship between the critical current degradation function and the insulation damage degree evolution function, a bidirectional coupled degradation model is established; the bidirectional coupled degradation model updates the critical current and insulation damage degree in real time through iterative calculation until convergence, obtaining the critical current degradation curve and the insulation damage degree evolution curve.

[0016] In a preferred embodiment, the step of outputting the reliability function, remaining lifetime, and lifetime contribution rate of the cable system under each operating condition mode based on the operating condition mode transition probability matrix and the bidirectional coupled degradation model is as follows: Based on the results output by the bidirectional coupled degradation model, a degradation state database for each operating condition mode is generated. This database stores degradation state data of the cable under multiple single constant operating conditions, including insulation damage and critical current values ​​evolving over time. Based on the current operating condition mode and the operating condition mode transition probability matrix, several future operating condition sequences are generated through Monte Carlo simulation, and each sequence is assigned a corresponding probability of occurrence. For each generated future operating condition sequence, the degradation state is determined chronologically. The corresponding degradation data is extracted from the database and input into the bidirectional coupled degradation model for iterative calculation to simulate the comprehensive degradation process of the cable under this sequence and obtain the degradation trajectory. The degradation trajectories of all operating condition sequences are weighted and averaged according to their occurrence probability to obtain the statistical average degradation curve of the cable system. Based on the preset failure threshold, the time distribution of the first crossing of the threshold for all paths is statistically analyzed to obtain the probability distribution of the cable's remaining life and reliability function. The cumulative dwell time and cumulative damage increment of each operating condition mode in all sequences are statistically analyzed. After probability weighting, the time contribution rate and damage contribution rate of each operating condition mode are calculated respectively. The calculated damage contribution rates are sorted to identify the high-risk operating condition modes that contribute the most to the total life loss.

[0017] In a preferred embodiment, obtaining the remaining lifetime probability distribution and reliability function of the cable is specifically as follows: For each operating path, the time point at which its degradation trajectory first reaches or exceeds a preset failure threshold is determined and denoted as the remaining lifetime corresponding to that path; the remaining lifetime of all paths and their corresponding path occurrence probabilities are used as a sample set to construct an empirical probability distribution function for the remaining lifetime; the probability distribution function is obtained by dividing the time axis into intervals and calculating the sum of the path probabilities where the remaining lifetime value falls within each interval; based on the probability distribution function, the cumulative distribution function, i.e., the reliability function, is calculated.

[0018] In a preferred embodiment, the degradation trajectory acquisition step is as follows: for each generated working condition sequence, the corresponding degradation data is extracted from the degradation state database in chronological order; at each working condition switching point, a bidirectional coupled degradation model is used for iterative calculation, taking the degradation state at the end of the previous working condition as the initial condition of the new working condition, and calculating the degradation increment in this period under the stress environment of the new working condition, thereby obtaining the complete degradation trajectory under this path.

[0019] The system for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis includes a data acquisition module, a mode segmentation module, a bidirectional coupling degradation module, and an output module, with connections between the modules. The data acquisition module collects operating condition data of the high-temperature superconducting cable during operation. The mode segmentation module uses clustering methods to segment operating conditions based on the operating condition data and establishes an operating condition mode transition probability matrix. The bidirectional coupling degradation module performs equivalent processing on the operating data under each operating condition mode and establishes a bidirectional coupling degradation model based on the equivalent processing results. The output module outputs the reliability function, remaining lifespan, and lifespan contribution rate of each operating condition mode of the cable system based on the operating condition mode transition probability matrix and the bidirectional coupling degradation model.

[0020] The technical effects and advantages of the high-temperature superconducting cable life prediction method and system combining operating condition behavior analysis of the present invention are as follows:

[0021] 1. This invention employs a clustering-based operating condition pattern classification method, combining K-means and density clustering (DBSCAN) algorithms, to perform pattern recognition of the operating conditions of high-temperature superconducting cables. By using weighted Euclidean distance and critical current margin to accurately cluster the feature vectors of operating conditions, it effectively distinguishes between steady-state and abnormal operating conditions and generates equivalent time series of operating conditions. This method not only reflects the diversity and complexity of the actual operating states of the cable but also identifies and classifies abnormal operating conditions, providing a crucial reference for cable reliability analysis.

[0022] 2. This invention performs multi-timescale window analysis on equivalent time series of operating conditions, counts the number of transitions between operating condition modes, and obtains a global transition probability matrix through weighted fusion, thereby achieving a quantitative description of the dynamic characteristics of operating condition behavior. This matrix can be used to predict the changing trends of future operating conditions, thus improving the timeliness and accuracy of lifespan prediction. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the process for predicting the lifespan of high-temperature superconducting cables based on operating condition behavior analysis, as described in this invention.

[0024] Figure 2 This is a schematic diagram of the system structure of the high-temperature superconducting cable life prediction method that combines operating condition behavior analysis according to the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0026] Example 1, Figure 1 This invention presents a method for predicting the lifespan of high-temperature superconducting cables by combining operational behavior analysis, including:

[0027] S1. Collect operating condition data of high-temperature superconducting cable during operation. Based on the operating condition data, use clustering method to classify the operating conditions and establish an operating condition mode transition probability matrix.

[0028] In this embodiment, operating condition data of the high-temperature superconducting cable during operation are collected, as follows:

[0029] Various types of sensors are arranged in different structural layers of the high-temperature superconducting cable to collect operating condition data of the high-temperature superconducting cable during operation. The operating condition data includes current, magnetic field, temperature, mechanical strain, partial discharge signal and voltage harmonic data.

[0030] A high-frequency current sensor is installed in the conductor layer to obtain the waveforms of rated current, overload current and short-circuit fault current;

[0031] A fiber optic grating sensor is embedded in the insulating layer to collect distributed temperature and strain fields;

[0032] An ultrasonic partial discharge sensor and an electromagnetic partial discharge sensor are installed in the sheath layer to jointly collect partial discharge signals of the insulation layer.

[0033] A flow sensor and a temperature sensor are installed in the cooling circuit to monitor the coolant status and cable temperature rise;

[0034] The operating condition features are extracted from the operating condition data to generate an operating condition feature vector and output the critical current margin. The operating condition features include: extracting the effective value, peak factor, harmonic content and AC loss index from the current signal; extracting the mean, variance, temperature gradient and coolant undercooling from the temperature signal; extracting the main strain direction and strain amplitude from the strain signal; and extracting the pulse count, amplitude distribution and discharge energy from the insulation partial discharge signal.

[0035] The z-score method is used to normalize the feature vectors of the working conditions to ensure that different physical quantities are in the same dimension. The normalized data is then input into the dimensionality reduction module, where principal component analysis is used to extract the main components and remove noise and redundant information. When the data dimension exceeds a preset threshold, the t-SNE method is called to perform nonlinear dimensionality reduction while preserving the relative distribution relationship between the working conditions.

[0036] The formula for calculating the critical current margin is as follows:

[0037]

[0038] In the formula: It is the critical current of a superconducting cable at temperature T. It is the actual temperature of the conductor during operation. It is the critical temperature of superconducting materials, that is, the turning point at which the material loses its superconductivity. This is a reference temperature. These are empirical fitting coefficients for the effect of temperature, used to adjust the steepness of the temperature curve. This is the nominal critical current of the superconducting cable at a reference temperature (usually liquid nitrogen temperature, 77K). It is the external magnetic field around the conductor. It is the characteristic magnetic field constant, used to characterize the rate at which the critical current decreases with respect to the magnetic field. It is a magnetic field sensitivity index, used to control the power-law relationship of magnetic field attenuation. It is the critical current at a specific temperature T and an applied magnetic field B. It is mechanical strain under operating conditions. It is the optimal strain point, used as the center of symmetry to determine the location of the vertex of the strain curve. Strain sensitivity coefficient, used to determine Deviation with strain The decay rate at that time It is the instantaneous critical current under the coupling of temperature, magnetic field, and strain. It is the instantaneous current waveform during operation. It is the integration period calculated from the root mean square of the operating current, typically taken as one grid fundamental cycle or the length of the operating window. It is the critical current margin. This is the actual operating current.

[0039] In this embodiment, operating conditions are classified into patterns based on operating condition data using a clustering method, as follows:

[0040] The K-means algorithm is used to cluster the operating condition feature vectors. During the clustering process, the critical current margin is combined to output the weighted Euclidean distance between any two operating condition vectors, forming several clusters.

[0041] Through iterative calculation, for each working point, the weighted Euclidean distance to each cluster center is calculated, and it is assigned to the cluster with the smallest weighted Euclidean distance, until the cluster center converges or the preset number of iterations is reached;

[0042] For each operating point within a cluster, if its weighted Euclidean distance exceeds a preset threshold, the point is assigned to the unclassified point set; otherwise, it is assigned to the steady-state operating condition cluster (including the rated operating cluster and the light-load operating cluster).

[0043] The density-based DBSCAN algorithm performs secondary clustering on the unclassified point set and outputs abnormal operating condition clusters, including short-circuit impact operating condition clusters, frequent start-stop operating condition clusters, harmonic distortion operating condition clusters, and other patterns.

[0044] Using steady-state and abnormal operating condition clusters as input, the mean and variance of the operating condition characteristics corresponding to each cluster are obtained, a cluster representative operating condition vector is generated, and the number and time distribution of operating condition points within the cluster are recorded.

[0045] Based on the time distribution of each cluster, the operating conditions represented by the clusters are arranged in the order of actual operating time to form an equivalent time series of operating conditions;

[0046] Using the critical current-temperature characteristic curve and allowable temperature rise boundary of the high-temperature superconducting cable, the generated equivalent operating condition sequence is physically constrained. If a certain equivalent operating condition in the sequence exceeds the allowable critical current margin or temperature rise boundary, the weight of that operating condition is adjusted or the cluster representative vector is reselected. This adjustment is repeated until the equivalent time sequence of the operating condition meets the physical constraint conditions.

[0047] The formula for calculating the weighted Euclidean distance is as follows:

[0048]

[0049] In the formula: For weighted Euclidean distance, The characteristic weighting coefficients are obtained through a linear mapping transformation of the critical current margin. Let k be the feature component of the i-th working condition vector. It is the k-th feature component of the j-th working condition vector.

[0050] The density-based DBSCAN algorithm performs secondary clustering on the unclassified point set, as follows:

[0051] Based on historical operating data, the distance threshold and minimum number of points required by the DBSCAN algorithm are adaptively calculated.

[0052] In the DBSCAN algorithm, the distance threshold is adaptively adjusted based on the variance of historical operating data, and the minimum number of points is determined according to the event frequency within the monitoring time window.

[0053] The average Euclidean distance and variance of the feature space are calculated based on historical working condition data samples, and a scaling factor is set according to the clustering sensitivity requirements to achieve adaptive adjustment of the distance threshold.

[0054] Obtain the total number of operating events and the frequency of event occurrence within the monitoring time window, and set the scaling factor to output the minimum number of points;

[0055] Based on the obtained distance threshold and minimum number of points, a second clustering is performed on the unclassified point set. Points with higher density are clustered into the same cluster, and points with sparse neighborhoods that cannot form clusters are marked as noise, resulting in several abnormal working condition clusters.

[0056] For each abnormal operating condition cluster, the feature vector attributes of the cluster center are extracted, and the cluster is classified into abnormal operating condition clusters. The classification includes: a sudden increase in current and a sharp decrease in critical current margin, which is a short-circuit impact operating condition cluster; frequent current fluctuations accompanied by multiple start-stop signals, which is a frequent start-stop operating condition cluster; and an increase in voltage distortion rate and high-order harmonic content, which is a harmonic distortion operating condition cluster.

[0057] The adaptive adjustment formula for the distance threshold is as follows:

[0058]

[0059] The formula for calculating the minimum number of points is as follows:

[0060]

[0061] In the formula: It is a distance threshold. It is the average Euclidean distance of historical operating condition data samples in the feature space, used to reflect the overall similarity level between different operating condition points. This is the adjustment coefficient for clustering sensitivity. It is the standard deviation of the Euclidean distance of historical operating condition data samples, used to reflect the degree of dispersion in the distribution of operating condition points. It is the frequency of event occurrence. This is a scaling factor used to convert event frequency into a minimum number of points; a recommended value range is 2 to 5. To round down, a minimum of 3 is used when events are sparse, and the frequency of events increases as the events occur more frequently. Increase and increase, It is the minimum number of points.

[0062] In this embodiment, a working condition mode transition probability matrix is ​​established, as follows:

[0063] The equivalent time series of operating conditions is divided into multiple time scale windows, and under each time scale... The module counts the number of transitions between adjacent time slices using a sliding window to form a time scale. The following is a transition counting matrix for the working condition cluster. ;

[0064] Based on the transition count matrix, the transition probability matrix is ​​output to obtain the transition count matrix at each time scale. Then, by weighted fusion of the transition probability matrices at each time scale, the transition probability matrix of the operating condition mode is obtained.

[0065] The transition probability matrix is ​​as follows:

[0066]

[0067] In the formula: In order to be on a time scale Below, operating condition cluster Transfer to operating condition cluster The probability, On a time scale Below, operating condition cluster Transfer to operating condition cluster The number of transitions is obtained by traversing the equivalent time series of the working conditions through a sliding window. Each time scale corresponds to a set of transition counting matrices. This indicates the length of the sliding time window used in the operating condition analysis.

[0068] S3 performs equivalent processing on the operating data under the working condition mode, and establishes a two-way coupled degradation model based on the equivalent results;

[0069] In this embodiment, the operating data under the working condition mode is equivalentized, and a two-way coupled degradation model is established based on the equivalentization result, as follows:

[0070] Obtain a sequence of operating condition patterns based on operating condition clustering, wherein the operating condition patterns include a steady-state operating condition cluster and an abnormal operating condition cluster;

[0071] Strain and temperature data for each operating condition are acquired and rainflow counting is performed to decompose complex strain and temperature waveforms into standard load cycles, thereby obtaining the equivalent number of strain cycles and the equivalent number of temperature cycles.

[0072] The strain equivalent cycle number and temperature equivalent cycle number are normalized and weighted according to the critical current margin to obtain the comprehensive equivalent cycle number.

[0073] The peak factor and harmonic content of the voltage signal under each operating condition are extracted, the harmonic correction factor is calculated, and it is introduced into the electric field calculation model to generate the equivalent electric field distribution and obtain the equivalent electric field intensity of harmonic distortion.

[0074] Based on the equivalent cycle number and equivalent electric field strength, the critical current degradation function and the insulation damage degree evolution function are established;

[0075] Based on the bidirectional coupling relationship between the critical current degradation function and the insulation damage degree evolution function, a bidirectional coupling degradation model is established. The bidirectional coupling relationship is as follows: on the one hand, the decrease in critical current will lead to an increase in operating current density, which in turn corrects the electric field strength; on the other hand, the increase in insulation damage degree will cause local electric field distortion, which will further amplify the electric field stress.

[0076] The bidirectional coupled degradation model updates the critical current and insulation damage degree in real time through iterative calculation until convergence, thus obtaining the critical current degradation curve and the insulation damage degree evolution curve.

[0077] The formula for calculating the comprehensive equivalent cycle number is as follows:

[0078]

[0079] In the formula: It is the critical current margin. It is the comprehensive equivalent cycle number. It is the equivalent strain cycle number, representing the total number of equivalent load cycles obtained after cyclic decomposition of the original strain data using the rainflow counting method. It is used to characterize mechanical fatigue effects. This is the equivalent temperature cycle number, representing the total number of equivalent thermal cycles obtained after cyclic decomposition of the temperature waveform using the rainflow counting method. It is used to characterize the effect of thermal stress. and These are weighting coefficients, corresponding to the importance of strain and temperature cycling, respectively. The weight values ​​are dynamically adjusted based on the critical current margin. When the critical current margin is small (insufficient margin, cable nearing failure boundary), the weight of temperature or strain cycling is increased to amplify the degradation effect. ,and , and The proportionality coefficient was determined based on the results of lifetime sensitivity analysis. It is an adjustment factor used to control the steepness of the function curve. It is the margin weighting factor.

[0080] The formula for calculating the equivalent electric field strength is as follows:

[0081]

[0082] In the formula: It is a harmonic correction factor, used when the voltage waveform has severe harmonic distortion or impulses. > 1, which amplifies the electric field strength. It is the voltage peak factor, expressed as the ratio of the peak value of the voltage waveform to its root mean square value. Total harmonic distortion (THD) represents the relative abundance of harmonic components to the fundamental component in a voltage signal. and Yes, it is used to adjust the contribution of peak factor and harmonic distortion to the correction factor. It is the equivalent electric field strength. It is the reference electric field strength, usually taken as the electric field value under rated operating conditions.

[0083] The bidirectional coupled degradation model is as follows:

[0084]

[0085] In the formula: It is the critical current, representing the maximum current that a superconducting cable or conductor can carry without losing its quench. It is time. and Here, represents the material degradation coefficient, corresponding to the critical current degradation rate constants caused by mechanical cyclic fatigue and electric field stress, respectively. and Characterizing the sensitivity of cyclic fatigue and electric field stress to the decay of critical current, The insulation damage level is represented by a value in the range [0,1], where 0 indicates intact insulation and 1 indicates complete failure. and , where represents the insulation aging coefficient, corresponding to the accelerating effects of mechanical / thermal stress and electrical stress on insulation aging, respectively. and This is an empirical index used to characterize the nonlinear response of damage accumulation to load and electric field. It is the current density feedback coupling coefficient, characterizing the degree to which a decrease in the critical current leads to an increase in current density, thereby exacerbating the electric field stress. It is the insulation damage coupling coefficient, representing the amplification effect of insulation damage on electric field distortion and partial discharge. It is the final equivalent electric field after bidirectional coupling correction. It is the equivalent electric field strength after the critical current degradation correction. It is the comprehensive equivalent cycle number. It is the equivalent electric field strength.

[0086] S4, based on the operating condition mode transition probability matrix and the two-way coupled degradation model, outputs the reliability function, remaining life, and life contribution rate of each operating condition mode of the cable system.

[0087] In this embodiment, based on the operating condition mode transition probability matrix and the bidirectional coupled degradation model, the reliability function, remaining lifetime, and lifetime contribution rate of the cable system under each operating condition mode are output as follows:

[0088] Based on the output of the bidirectional coupled degradation model, a degradation state library is generated for each operating condition. The state library stores the degradation state data of the cable under multiple single constant operating conditions. The degradation state data includes the insulation damage degree and critical current value that evolve over time.

[0089] Based on the current operating condition mode and the operating condition mode transition probability matrix, several future operating condition sequences are generated through Monte Carlo simulation, and each sequence is assigned a corresponding probability of occurrence.

[0090] For each generated future operating condition sequence, the corresponding degradation data is extracted from the degradation state database in chronological order and input into the bidirectional coupled degradation model for iterative calculation to simulate the comprehensive degradation process of the cable under the sequence and obtain the degradation trajectory.

[0091] The degradation trajectories of all operating condition sequences are weighted and averaged according to their occurrence probability to obtain the statistical average degradation curve of the cable system. Based on the preset failure threshold, the time distribution of the first crossing of the threshold for all paths is statistically analyzed to obtain the probability distribution of the cable's remaining life and the reliability function.

[0092] The cumulative dwell time and cumulative damage increment of each working condition mode in all sequences were statistically analyzed. After probability weighting, the time contribution rate and damage contribution rate of each working condition mode were calculated respectively.

[0093] Based on the calculated damage contribution rate, the high-risk operating conditions that contribute the most to the total lifespan loss are identified and corresponding operational optimization or limitation suggestions are proposed.

[0094] The remaining lifetime probability distribution and reliability function of the cable are obtained as follows:

[0095] For each working condition path, determine the time point when its degradation trajectory first reaches or exceeds the preset failure threshold, and record it as the remaining lifetime corresponding to that path.

[0096] Using the remaining lifetime of all paths and their corresponding probabilities of occurrence as a sample set, an empirical probability distribution function for the remaining lifetime is constructed.

[0097] The probability distribution function is obtained by dividing the time axis into intervals and calculating the sum of the path probabilities in which the remaining lifetime value falls within each interval;

[0098] Based on the probability distribution function, the cumulative distribution function, i.e. the reliability function, is calculated. The reliability function is derived from the cumulative distribution function, specifically: the reliability function equals 1 minus the cumulative distribution function.

[0099] The steps for obtaining the degradation trajectory are as follows:

[0100] For each generated operating condition sequence, the corresponding degradation data is extracted from the degradation state database in chronological order.

[0101] At each working condition switching point, a bidirectional coupled degradation model is used for iterative calculation. The degradation state at the end of the previous working condition is used as the initial condition of the new working condition. The degradation increment in this period is calculated under the stress environment of the new working condition, so as to obtain the complete degradation trajectory under this path.

[0102] Example 2, Figure 2 The present invention provides a system for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis, comprising a data acquisition module, a mode division module, a bidirectional coupling degradation module, and an output module, with connections between the modules;

[0103] The data acquisition module is used to collect operating condition data of the high-temperature superconducting cable during operation.

[0104] The pattern segmentation module is used to segment operating conditions into patterns based on operating condition data using clustering methods, and to establish an operating condition pattern transition probability matrix.

[0105] The bidirectional coupling degradation module is used to perform equivalent processing on the operating data under the working condition mode, and to establish a bidirectional coupling degradation model based on the equivalent results.

[0106] The output module is used to output the reliability function, remaining life, and life contribution rate of the cable system under each operating condition mode, based on the operating condition mode transition probability matrix and the two-way coupled degradation model.

[0107] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0108] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0109] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0110] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0111] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0112] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting the lifespan of high-temperature superconducting cables based on operating condition behavior analysis, characterized in that: include: Collect operating condition data of high-temperature superconducting cables during operation; Based on operating condition data, clustering methods are used to classify operating conditions into patterns, and a pattern transition probability matrix is ​​established. The operating data under the working condition mode is equivalently processed, and a two-way coupled degradation model is established based on the equivalent results. Specifically, the critical current degradation function and the insulation damage degree evolution function are established based on the comprehensive equivalent cycle number and the equivalent electric field intensity. Based on the two-way coupling relationship between the critical current degradation function and the insulation damage degree evolution function, a two-way coupled degradation model is established. Based on the operating condition mode transition probability matrix and the two-way coupled degradation model, the reliability function, remaining lifetime, and lifetime contribution rate of the cable system under each operating condition mode are output.

2. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 1, characterized in that, The operating conditions are classified into patterns using clustering methods based on operating condition data, as detailed below: The system collects operating condition data of high-temperature superconducting cables during operation, extracts operating condition features from the data, generates an operating condition feature vector, and outputs the critical current margin. The K-means algorithm is used to cluster the operating condition feature vectors. During the clustering process, the critical current margin is combined to output the weighted Euclidean distance between any two operating condition vectors, forming several clusters. Through iterative calculation, for each operating point, the weighted Euclidean distance to each cluster center is calculated, and the point is assigned to the cluster with the smallest weighted Euclidean distance, until the cluster centers converge. For each operating point within a cluster, if the weighted Euclidean distance exceeds a preset threshold, the point is assigned to the unclassified point set; otherwise, it is assigned to the steady-state operating condition cluster. The density-based DBSCAN algorithm performs secondary clustering on the unclassified point set, outputs abnormal condition clusters, and examines each cluster.

3. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 2, characterized in that, The specific steps for verifying each cluster are as follows: Using steady-state and abnormal operating condition clusters as input, the mean and variance of the operating condition characteristics corresponding to each cluster are obtained, a cluster representative operating condition vector is generated, and the number and time distribution of operating condition points within the cluster are recorded. Based on the time distribution of each cluster, the cluster-representing operating condition vectors are arranged in order of actual running time to form an equivalent time series of operating conditions; The physical constraints of the generated equivalent time series of operating conditions are verified by using the critical current-temperature characteristic curve of the high-temperature superconducting cable and the allowable temperature rise boundary.

4. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 3, characterized in that, The density-based DBSCAN algorithm performs secondary clustering on the unclassified point set, as follows: Based on historical operating data, the distance threshold and minimum number of points required by the DBSCAN algorithm are adaptively calculated. Based on the obtained distance threshold and minimum number of points, a second clustering is performed on the unclassified point set. Points with higher density are clustered into the same cluster, and points with sparse neighborhoods that cannot form clusters are marked as noise, resulting in several abnormal working condition clusters. For each abnormal operating condition cluster, the feature vector attributes of the cluster center are extracted, the cluster type is determined, and an abnormal operating condition cluster is formed.

5. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 4, characterized in that, The establishment of the working condition mode transition probability matrix is ​​as follows: The equivalent time series of operating conditions is divided into multiple time scale windows. Under each time scale, the module counts the number of transitions between adjacent time slices according to the sliding window, forming a working condition cluster transition counting matrix under the time scale. Based on the transition counting matrix, the transition probability matrix is ​​output to obtain the transition probability matrix at each time scale. By weighted fusion of the transition probability matrices at each time scale, the operating condition mode transition probability matrix is ​​obtained.

6. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 5, characterized in that, The process involves performing equivalent processing on the operating data under the working condition mode, and establishing a two-way coupled degradation model based on the equivalent processing results, as detailed below: Strain and temperature data for each operating condition are acquired and rainflow counting is performed to decompose the strain and temperature waveforms into standard load cycles, thereby obtaining the equivalent number of strain cycles and the equivalent number of temperature cycles, respectively. The strain equivalent cycle number and temperature equivalent cycle number are normalized and weighted according to the critical current margin to obtain the comprehensive equivalent cycle number. The peak factor and harmonic content of the voltage signal under each operating condition are extracted, the harmonic correction factor is calculated, and it is introduced into the electric field calculation model to generate the equivalent electric field distribution and obtain the equivalent electric field intensity of harmonic distortion. Based on the comprehensive equivalent cycle number and equivalent electric field strength, the critical current degradation function and the insulation damage degree evolution function are established; A two-way coupled degradation model is established based on the two-way coupling relationship between the critical current degradation function and the insulation damage degree evolution function; The bidirectional coupled degradation model updates the critical current and insulation damage degree in real time through iterative calculation until convergence, thus obtaining the critical current degradation curve and the insulation damage degree evolution curve.

7. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 6, characterized in that, Based on the operating condition mode transition probability matrix and the two-way coupled degradation model, the reliability function, remaining lifetime, and lifetime contribution rate of the cable system under each operating condition mode are output as follows: Based on the output of the bidirectional coupled degradation model, a degradation state library is generated for each operating condition mode. The state library stores the degradation state data of the cable under multiple single constant operating conditions. Based on the current operating condition mode and the operating condition mode transition probability matrix, several future operating condition sequences are generated through Monte Carlo simulation, and each future operating condition sequence is assigned a corresponding probability of occurrence. For each generated future operating condition sequence, the corresponding degradation data is extracted from the operating condition degradation state database in chronological order and input into the bidirectional coupled degradation model for iterative calculation to simulate the comprehensive degradation process of the cable under the future operating condition sequence and obtain the degradation trajectory. The degradation trajectories of all future operating condition sequences are weighted and averaged according to their occurrence probability to obtain the statistical average degradation curve of the cable system. Based on the preset failure threshold, the time distribution of the first crossing of the threshold for all paths is statistically analyzed to obtain the remaining lifetime probability distribution and reliability function of the cable. The cumulative dwell time and cumulative damage increment of each working mode in all future working mode sequences are statistically analyzed. After probability weighting, the time contribution rate and damage contribution rate of each working mode are calculated separately. Based on the calculated damage contribution rate, the high-risk operating conditions that contribute the most to the total lifespan loss are identified and sorted.

8. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 7, characterized in that, The remaining lifetime probability distribution and reliability function of the cable are obtained as follows: For each working condition path, determine the time point when its degradation trajectory first reaches or exceeds the preset failure threshold, and record it as the remaining life corresponding to that path. Using the remaining lifetime of all paths and their corresponding probabilities of occurrence as a sample set, an empirical probability distribution function for the remaining lifetime is constructed. The probability distribution function is obtained by dividing the time axis into intervals and calculating the sum of the path probabilities in which the remaining lifetime value falls within each interval; Based on the probability distribution function, the cumulative distribution function, i.e. the reliability function, is calculated.

9. The method for predicting the lifespan of high-temperature superconducting cables by combining operating condition behavior analysis according to claim 8, characterized in that, The steps for obtaining the degradation trajectory are as follows: For each generated future operating condition sequence, the corresponding degradation data is extracted from the degradation state database in chronological order. At each working condition switching point, a bidirectional coupled degradation model is used for iterative calculation. The degradation state at the end of the previous working condition is used as the initial condition of the new working condition. The degradation increment in this period is calculated under the stress environment of the new working condition, so as to obtain the complete degradation trajectory under this path.

10. A system using the high-temperature superconducting cable lifetime prediction method combined with operating condition behavior analysis as described in any one of claims 1-9, characterized in that, It includes a data acquisition module, a pattern partitioning module, a bidirectional coupling degradation module, and an output module, with connections between the modules; The data acquisition module is used to collect operating condition data of the high-temperature superconducting cable during operation. The pattern segmentation module is used to segment operating conditions into patterns based on operating condition data using clustering methods, and to establish an operating condition pattern transition probability matrix. The bidirectional coupling degradation module is used to perform equivalent processing on the operating data under the operating condition mode, and to establish a bidirectional coupling degradation model based on the equivalent results. Specifically, based on the comprehensive equivalent cycle number and equivalent electric field intensity, the critical current degradation function and the insulation damage degree evolution function are established, and the bidirectional coupling relationship between the critical current degradation function and the insulation damage degree evolution function is used to establish the bidirectional coupling degradation model. The output module is used to output the reliability function, remaining life, and life contribution rate of the cable system under each operating condition mode, based on the operating condition mode transition probability matrix and the two-way coupled degradation model.

Images