Cable defect multi-parameter aging characteristic and service life evaluation method and system

Abstract

The invention belongs to the technical field of cable service life evaluation, and particularly discloses a cable defect multi-parameter aging characteristic and service life evaluation method and system, and the method comprises the steps: obtaining a plurality of first parameters which can affect the aging of a cable, and a second parameter which changes when each first parameter affects the aging characterization of the cable; constructing a calculation model containing a first incidence relation between each second parameter and the corresponding first parameter; constructing a prediction model for calculating the service life of the cable according to the calculation model, and configuring a correction factor for representing the degree of influence on the defect of the cable on each calculation model; determining the numerical value of each correction factor by adopting an optimization algorithm, and updating the correction factors according to the numerical values to optimize the prediction model; calculating the service life of the cable through the optimized prediction model; the method has the advantages that cable aging is comprehensively analyzed, modeling is clear, influence factor relations are defined, an algorithm is optimized to adjust correction factors, and prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cable life assessment, and in particular to a cable defect multi-parameter aging characteristic and life assessment method and system. Background Art

[0002] As urban distribution networks become larger and more complex, their reliability becomes a key issue. In recent years, large-scale power outages have occurred frequently, causing huge economic and social losses. Underground cables work in a harsh environment, and insulation aging is very likely to occur under multiple stresses such as electricity, heat, and mechanics, which eventually leads to line failures and endangers the reliable operation of distribution networks. This makes it a problem to accurately locate defects and determine the relevant service life.

[0003] Therefore, a cable defect multi-parameter aging characteristics and life assessment method and system are proposed to solve the above-mentioned problems. Summary of the invention

[0004] The present invention aims to provide a cable defect multi-parameter aging characteristics and life assessment method and system to solve or improve at least one of the above-mentioned technical problems.

[0005] In view of this, a first aspect of the present invention is to provide a cable defect multi-parameter aging characteristic and life assessment method.

[0006] A second aspect of the present invention is to provide a system.

[0007] A first aspect of the present invention provides a cable defect multi-parameter aging characteristic and life assessment method, comprising the following steps: Acquire a plurality of first parameters that can affect cable aging and a second parameter that changes in the cable aging characterization caused by each first parameter; Constructing a calculation model including a first association relationship between each second parameter and the corresponding first parameter; Constructing a prediction model for calculating the cable life according to the calculation model, and configuring a correction factor for characterizing the degree of influence of the cable defect on each of the calculation models; An optimization algorithm is used to determine the value of each correction factor, and the correction factor is updated according to the value to optimize the prediction model; the cable life is calculated using the optimized prediction model.

[0008] In any of the above technical solutions, the step of constructing a prediction model for calculating the cable life through all the calculation models includes: Each of the first parameters and the corresponding second parameter form a feature group, and the second parameter is used as the first parameter to calculate and obtain an intermediate variable of the cable life; Constructing the fuzzy algorithm model according to the second correlation relationship between the first parameter and the cable life; The prediction model is determined by the fuzzy algorithm model.

[0009] In any of the above technical solutions, the step of determining the prediction model through the fuzzy algorithm model specifically includes: Acquire a third association relationship for characterizing the second parameter and the cable life according to the first association relationship and the second association relationship; The prediction model of each feature group is determined by the third association relationship.

[0010] In any of the above technical solutions, the prediction model is the following formula: ; wherein said is the cable life, and is independent of each other among the feature groups; is the square root of the variance of the normal distribution; is a constant; is the second parameter obtained by the calculation model, and i is a positive integer; is the correction factor; for The mathematical expectation of is the exponential constant In any of the above technical solutions, the optimization algorithm is optimized by a first optimization formula and a second optimization formula, and the steps of the optimization algorithm include: setting the number of iterative features in the iteration according to the number of the correction factors, and taking the optimized value of the iterative feature as the numerical value of the correction factor; obtaining the optimized value of each of the iterative features respectively through a preset number of iterations or a judgment threshold; obtaining a judgment factor for judging the adaptability of the correction factor in characterizing the degree of influence of the cable defect in the iterative step; and judging whether to use the first optimization formula or the second optimization formula to update the optimization value according to the judgment factor.

[0011] In any of the above technical solutions, the step of optimizing the numerical value by using the first optimization formula or the second optimization formula according to the judgment factor specifically includes: judging whether the judgment factor is greater than a judgment threshold; if it is greater than, updating the optimization value by using the first optimization formula; if it is not greater than, updating the optimization value by using the second optimization formula.

[0012] In any of the above technical solutions, the first optimization formula is: ; wherein, wherein, the is the position of the whale at the current iteration, and t is the number of iterations; is the current optimal whale position; The is a random number uniformly distributed in (0, 1); Each dimension of is uniformly distributed in (- a , a ), and a is the boundary constant, and its value is .

[0013] In any of the above technical solutions, the second optimization formula is: ; wherein said is an exponential constant, and the is the shape parameter and takes the value of 0.5. is a random number uniformly distributed in (0, 1); is a constant; is a trigonometric function.

[0014] In any of the above technical solutions, the first parameter includes at least one parameter of the cable in each of the working environment, working loss, working load and material properties.

[0015] The second aspect of the present invention provides a system for implementing the multi-parameter aging characteristics and life assessment method of cable defects in any of the above-mentioned technical solutions, including: a data collection and processing module, used to collect data of the first parameter and the second parameter; a model construction module, which constructs the calculation model according to the first correlation relationship, and constructs the prediction model through the calculation model; an optimization parameter adjustment module, which uses the optimization algorithm to determine the value of each correction factor and optimizes the prediction model; a life prediction module, which obtains the service life of the cable according to the optimized prediction model.

[0016] Compared with the prior art, the present invention has the following beneficial effects: Four typical influencing parameters are selected for the cable life problem, including temperature, leakage current, load current, and cable impurities, to conduct a full-dimensional status assessment and analysis of the cable life problem. Compared with traditional single-parameter life analysis or purely theoretical calculations, the evaluation results will be closer to the on-site operation itself and the results will be more accurate.

[0017] The fuzzy algorithm model based on normal Gaussian distribution can quickly converge features from the big data of cable sampling, which reflects the robustness of the fuzzy algorithm.

[0018] The algorithm is used to optimize the weights of the characteristic parameters of the prediction model, giving it the advantages of strong global optimization ability, potential parallelism and easy implementation.

[0019] Additional aspects and advantages of embodiments according to the present invention will become apparent in the following description or may be learned through practice of embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 A roadmap of the method steps of the present invention; Figure 2 It is a specific implementation flow chart of the method of the present invention; Figure 3 It is a system logic block diagram of the present invention; Figure 4 The following is the working process of the system of the present invention. DETAILED DESCRIPTION

[0021] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0023] See also Figure 1-Figure 4 , the following describes a cable defect multi-parameter aging characteristics and life assessment method and system according to some embodiments of the present invention.

[0024] As described in the background technology, the insulation aging phenomenon is most likely directly related to the electrical parameters of the cable, so monitoring the electrical parameters can indirectly understand the operating health status of the cable.

[0025] By studying the insulation aging problem of XLPE (cross-linked polyethylene) cables during operation, the characteristic quantity change trend and data characteristics of cables with different insulation defect types are analyzed from a mechanism perspective, the defect position is located in time according to the test results, and the remaining service life of the cable is estimated through the life assessment method, which is of great help to the inspection and maintenance of the cable, and is also of great significance to ensure the safe and stable operation of the power grid and maximize the benefits of XLPE cables. In the early stage, domestic and foreign scholars mainly studied XLPE cables with voltage levels of 35kV and below. In terms of material property analysis, by doping different particulate materials, the physical and chemical properties and electrical properties of the modified XLPE materials were tested to find the most suitable doping material and concentration to achieve the purpose of optimizing XLPE performance; in terms of aging characteristics, the main research is on the changes in the characteristic quantities of insulation performance and mechanical properties of XLPE cables under the action of different aging factors such as electricity, heat, machinery and moisture, such as insulation resistance, relaxation current, breakdown voltage, partial discharge and elongation at break; in terms of life assessment method research, the main focus is on establishing a thermal aging life assessment model.

[0026] The structural setting of the semi-conductive layer of the 10kV cross-linked polyethylene cable makes the electric field on the surface of the metal electrode evenly distributed, and at the same time can avoid the problem of air gap caused by the inability of the insulation layer and the core conductor to be in close contact. The main insulation layer is the key structure to reliably isolate the high-voltage conductor from the ground electrode. It can withstand the heat generated by the heating of the conductor to a certain extent and maintain the proper voltage resistance; the existence of the metal shielding layer is to provide a low-voltage electrode for the high-voltage electric field, as a discharge path for capacitive current and fault current; the lining filler mainly plays the role of phase isolation insulation and moisture absorption; the main function of the external armor and sheath is mechanical protection, waterproofing, corrosion resistance and biological damage prevention. Most cable failure accidents occur at the middle joint or terminal joint of the cable. The reason for the failure of the cable joint is mostly due to the improper installation process during the cable laying process, and the external force damage causes defects in the joint.

[0027] The loss of insulating materials under the action of AC electric field is called dielectric loss, including conductivity loss and polarization loss. The dielectric loss P of the cable is related to the magnitude and frequency of the applied voltage and the material, so it is difficult to directly compare the dielectric loss of different materials. However, the dielectric loss tangent tan is the ratio of the active component to the reactive component of the current, which is independent of other external conditions and can reflect the dielectric loss of the material.

[0028] For electrical parameters, the insulation status of the cable can be estimated by measuring the magnitude of the common-mode leakage current. The common-mode leakage current is proportional to the change in dielectric loss. Choosing the common-mode leakage current as a parameter to characterize the aging status of cable insulation has the following advantages: the common-mode leakage current will not flow through the load, making the monitoring method based on the common-mode leakage current non-invasive; at the same time, the load change will not affect the common-mode leakage current, so it can achieve effective monitoring under different working conditions.

[0029] For non-electrical parameters, temperature change is an important parameter affecting the life of the cable. When the temperature rises, the polarization in the insulating medium intensifies, the conductivity increases, and the insulation resistance value decreases. The insulation resistance will decrease with the increase in temperature. In addition, temperature changes also affect the properties and structure of the insulating material. Different insulating materials have different forms of aging. The continuous effect of thermal stress causes the performance of the cable insulation to deteriorate, the aging phenomenon gradually deepens, and eventually leads to breakdown. The effect of temperature on the electron movement rate during electrical breakdown can be compared to the relationship between temperature and chemical reaction rate during a chemical reaction.

[0030] Regarding the cable structural parameters, the common impurities in cable insulation are generally pores, industrial impurities, biological impurities, inorganic impurities, etc. Different types of impurities have different effects on the life of the cable. However, no matter what kind of impurities exist in the cable insulation, they will have an adverse effect on the life of the cable insulation. The hazard coefficient to the cable insulation will also vary due to the shape, expansion coefficient, dielectric constant, etc. of the impurities. The edge shape of the conductive impurities will affect the field strength inside the cable; the dielectric constant is greater than the dielectric constant of the insulating material, which will cause the field strength to concentrate; during the extrusion process, the insulating material will be heated together with the impurities. Due to the difference in expansion coefficients, cracks or gaps will appear between the insulation and the impurities during the cooling process. The greater the difference between the expansion coefficient of the impurities and the expansion coefficient of the insulating material, the easier it is to produce cracks or gaps. It can be seen that the different sizes and shapes of the impurities will have different effects on the cable insulation.

[0031] Construction of cable life assessment model based on real-time data: When cable insulation is affected by two or more aging factors, the cable will age more rapidly, but its aging speed is not simply the algebraic sum of the individual influences of these factors. The interaction between different influencing factors is called synergistic effect. Since the degradation process of the cable is caused by multiple factors, and the changes in the dielectric and mechanical properties of the insulating material are nonlinear, if various factors are combined, the calculation process of the cable aging life is difficult to achieve, and the existing degradation data of the service cable is limited. It is difficult to establish an accurate mathematical model with traditional methods. Therefore, in order to better deal with the complex and nonlinear characteristics of cable degradation, a new cable aging model is established to predict the life of the cable by using the changes in the nonlinear characteristics of the insulating material with aging time.

[0032] In view of the above-mentioned technical problems, the embodiment of the first aspect of the present invention proposes a cable defect multi-parameter aging characteristic and life assessment method. In some embodiments of the present invention, such as Figure 1 As shown, the cable defect multi-parameter aging characteristics and life assessment method include: S101, obtaining a plurality of first parameters that can affect cable aging and a second parameter that changes in a cable aging characterization caused by each first parameter.

[0033] Here, the first parameter is the external factors that directly affect the aging of the cable, such as ambient temperature, voltage and current of the cable, mechanical pressure, chemical corrosion environment, etc. These parameters are usually obtained from real-time monitoring of the cable operating environment; the second parameter is the change of physical and chemical properties of the cable during aging, such as the decrease of insulation resistance, change of dielectric constant, attenuation of mechanical strength, etc. These parameters are used to characterize the state of cable aging; the collected data is cleaned, outliers and noise are removed, the data is formatted to meet the requirements of subsequent processing and analysis, and preliminary statistical analysis is performed on the data to determine the basic characteristics and trends of the data, providing a reference for model construction.

[0034] As can be seen from the above, a variety of sensors deployed in the cable and its environment continuously monitor the first parameter and the second parameter. For example, temperature sensors, current and voltage sensors, mechanical stress sensors, etc., these sensors send the collected data to the central data collection system; a centralized data receiving system is responsible for receiving data from various sensors, which may be transmitted through wired or wireless networks, and the data is stored in a database to support efficient data query and retrieval, ensuring the integrity and security of the data; data processing software (such as SQL database management, Python scripts, etc.) is used to perform necessary cleaning and conversion on the raw data, including scaling, normalization and other processing, so that the data is suitable for further analysis. Data preprocessing also includes identifying and processing missing values ​​or outliers to ensure the quality of the data.

[0035] In any of the above embodiments, the first parameter includes at least one parameter selected from each of the working environment, working loss, working load and material properties of the cable.

[0036] In this embodiment, the working environment includes temperature, humidity and chemical corrosion; Temperature: When the cable works in a high temperature or low temperature environment, the physical and chemical properties of its material may change, affecting its insulation and mechanical strength; Humidity: A high humidity environment may cause moisture to penetrate the cable insulation layer, accelerate the aging process, and increase the risk of leakage and short circuit; Chemical corrosion: If there are corrosive gases or liquids in the environment around the cable, it may corrode the insulating material or metal parts and shorten the life of the cable.

[0037] Working loss includes dielectric loss and conductor loss. Dielectric loss refers to the energy loss of the insulating material of the cable under the action of an alternating electric field, reflecting the degree of material aging. Conductor loss refers to the loss caused by current passing through the cable conductor, which is mainly heat loss and is related to the cable current carrying capacity and material resistance.

[0038] The working load includes current load and voltage level. Current load: the maximum current that the cable often bears, which affects its temperature and loss; voltage level: the voltage that the cable is designed to withstand, which determines its structure and the materials used.

[0039] Material properties include insulation material type and impurity content. Insulation material type: the type of insulation material used in the cable, such as polyethylene, rubber, etc., each with different aging characteristics and performance; impurity content: impurities may cause electric field unevenness, accelerate local aging, and affect the overall performance of the cable.

[0040] Specifically, the first parameters include temperature, leakage current, load current and cable impurities, and the second parameters include electrothermal aging, elongation at break retention rate, dielectric loss tangent value and insulation resistance.

[0041] As can be seen from the above, the parameters of the first parameter specifically include: temperature: directly affects the physical properties of cable insulation and conductor materials, and changes in temperature will accelerate the aging process; leakage current: when the cable insulation performance decreases, the leakage current increases, which is an early indicator of insulation damage; load current: the actual current that the cable bears during operation, which may cause overheating and accelerate the aging of insulation materials under overload conditions; cable impurities: impurities such as bubbles and metal particles will form an electric field concentration inside the cable insulation, causing partial discharge and accelerating aging. Data is collected through real-time monitoring equipment such as temperature sensors, current measuring equipment, and leakage current detectors. Real-time data is used to analyze the operating conditions of the cable and determine whether there are signs of overload, overheating, or insulation damage.

[0042] The parameters of the second parameter specifically include: electrothermal aging: measures the degree of aging of the cable under voltage and temperature; elongation retention rate at break: the attenuation of mechanical properties of cable materials due to aging, such as a decrease in elongation; dielectric loss tangent (tanδ): the dielectric loss of cable materials, reflecting material aging and changes in insulation performance; insulation resistance: directly reflects the health of the cable insulation layer, and a decrease in insulation resistance indicates a decrease in insulation performance; monitoring of the second parameter is usually obtained through regular diagnostic tests, such as using insulation resistance testers, dielectric loss test equipment, etc. These test results are used to establish and update the cable health status model and evaluate the remaining life of the cable.

[0043] All data collected from the first parameter and the second parameter are gathered in a central data processing center, which uses data analysis and machine learning methods to process these data, extract key trends and patterns, and establish and maintain a comprehensive cable health and aging model; based on the model output, the system predicts the remaining life of the cable and the possible failure time point, helping the operation and maintenance team to make data-based maintenance decisions. The prediction results provided by the system can be used to optimize maintenance plans and replacement strategies to minimize downtime and maintenance costs; with the continuous input of new data, the system can self-adjust and optimize the prediction model to improve prediction accuracy. Through continuous learning and adaptation, the system can better respond to environmental changes and dynamic changes in cable usage conditions.

[0044] S102, constructing a calculation model including a first association relationship between each second parameter and the corresponding first parameter.

[0045] Here, the purpose of associative model building is to develop one or more computational models that can describe and predict how to infer the aging state (second parameter) of the cable from its operating conditions (first parameter). For example, the model can predict how the insulation resistance of the cable changes under specific load current and temperature conditions; model parameterization is based on the actual collected data to parameterize the model to ensure that the model can accurately reflect the actual cable behavior; model verification and calibration verify and calibrate the model through experimental data or historical data to ensure the accuracy and reliability of its predictions.

[0046] From the above, we can see that data about the first parameter is collected from the cable system, such as ambient temperature, load current, leakage current and information about impurities inside the cable, and data about the second parameter is collected, such as electrothermal aging, elongation retention rate at break, dielectric loss tangent and insulation resistance; statistical analysis and machine learning methods (such as linear regression, decision tree, neural network) are used to analyze the relationship between the first parameter and the second parameter, and a mathematical model, such as a regression model or a more complex prediction model, is developed to describe the dependency between these parameters; the model is tested using a known data set to evaluate its prediction accuracy and response speed, and the model parameters are adjusted according to the test results to optimize the model structure to improve the prediction accuracy and generalization ability; the verified and calibrated model is deployed in the actual cable health monitoring system, and the model is used to predict the aging status and life of the cable in real time, providing decision support for cable maintenance and replacement.

[0047] Specifically, regarding the calculation model of electrothermal aging, the conventional electrothermal parameter model can accurately calculate the cable life at a specific temperature and a specific electric field strength, but the cable temperature fluctuates with the load cycle and ambient temperature, and has a certain randomness. Therefore, the present invention proposes a method for estimating the cumulative amount of electrothermal degradation of power cables based on a random forest process to estimate the accumulated electrothermal degradation of power cables in historical load cycles. Assuming that the cable life is reduced by a certain percentage ReduceΔti% at a specific cable temperature within a period of time: ; In the formula, It indicates a certain percentage reduction at a specific cable temperature over a period of time; dt Represents calculus with temperature as base; L Indicates the cable length; E Indicates energy; Indicates the changing temperature.

[0048] Cable degradation in one day d(n) for: ; In the formula, Indicates the amount of cable degradation in one day; i Indicates hours.

[0049] If the accumulation process of degradation is regarded as a linear process, then n+1 The aging amount of the cable is; ; In the formula, Indicates the amount of cable aging.

[0050] Specifically, regarding the calculation model of cable load heat, the relationship between the AC voltage applied to the cable and the time of insulation breakdown is called the vt characteristic. It follows a certain exponential law: ; In the formula, V is the AC voltage, n is the insulation aging coefficient, t is the time, and C is a constant. The higher the voltage, the shorter the insulation breakdown time, and the two are negatively correlated. The thermal effect when the current passes through the cable conductor will increase. At the same time, the skin effect of the charge, the eddy current loss of the steel, the insulation medium loss, etc. will also generate additional heat, making the cable temperature higher than normal use. Long-term overload operation will accelerate the aging of the insulation due to the excessively high internal temperature of the cable. The cable load is negatively correlated with the insulation breakdown time, and the relationship between the cable load and the cable life is the same as the relationship between temperature and cable life, which is negatively correlated.

[0051] Specifically, regarding the calculation model of cable insulation aging current, common-mode leakage current is an important indicator to characterize the insulation state of electrical equipment. The common-mode voltage applied to each phase conductor is the same, so no current will be induced on the ungrounded load. The current generated by the common-mode voltage is the leakage current flowing through the cable insulation. Since the common-mode leakage current flows through the cable insulation, its magnitude is closely related to the electrical parameters of the cable insulation. A method of estimating the common-mode leakage current by using a magnetic ring made of high magnetic permeability material to wrap the cable to be tested, and placing a magnetic field sensor outside the magnetic ring to measure the magnetic induction intensity. The magnetic ring made of high magnetic permeability material makes the magnetic flux lines in the surrounding space distributed along the ring. At this time, the magnetic field generated by the load current with a sum of zero will cancel each other out inside the ring. The magnetic field measured outside the ring is mainly generated by the common-mode leakage current. The tangential magnetic induction intensity outside the ring is in a simple positive proportional relationship with the common-mode leakage current, as shown below: ; In the formula, B is the phasor form of the vector magnetic potential amplitude; I k is the current phasor in the kth conductor; For the k Polar coordinates of the root conductor; k total is the total number of conductors in the cable; μ 0 is the vacuum permeability; μ r is the magnetic permeability of the magnetic shielding ring; a is the inner radius of the magnetic shielding ring; c is the outer radius of the magnetic shielding ring; is the polar coordinate of the measurement point. The first term on the right side of the equal sign is the magnetic induction intensity generated by the common mode leakage current, and the second term represents the magnetic field fluctuation caused by the load current and the conductor structure. μ1 and approaches 0. An analytical expression between common-mode leakage current and cable insulation parameters is established based on the distributed parameter model to analyze the relationship between insulation aging and common-mode leakage current.

[0052] Specifically, regarding the calculation model of cable impurities, different types of impurities have different effects on the life of the cable, but no matter what kind of impurities are present in the cable insulation, they will have an adverse effect on the life of the cable insulation, and the hazard coefficient to the cable insulation will also vary due to the shape, expansion coefficient, dielectric constant, etc. of the impurities. The edge shape of the conductive impurities will affect the field strength inside the cable; the dielectric constant is greater than the dielectric constant of the insulating material, which will cause field strength concentration; during the extrusion process, the insulating material will be heated together with the impurities. Due to the difference in expansion coefficients, cracks or gaps will appear between the insulation and the impurities during the cooling process. The greater the difference between the expansion coefficient of the impurities and the expansion coefficient of the insulating material, the easier it is to produce cracks or gaps. It can be seen that the material size and shape of the impurities will have different effects on the cable insulation. The calculation formula for the discharge amount of impurities is: ; Where, q is the impurity discharge; Ca is the series impurity capacitance; Vb is the gap breakdown voltage specified by Paschen; is the dielectric constant of air; is the relative dielectric constant of cross-linked polyethylene; A is the columnar insulation area connected in series with the impurity; D is the insulation thickness; a is the gap thickness; b is the gap length; c is the gap width; and d is the impurity thickness.

[0053] Specifically, the impurity discharge amount q, the phasor form B of the vector magnetic potential amplitude, the AC voltage V and the aging amount of the cable on the n+1th day Considered as the second parameter .

[0054] S103, constructing a prediction model for calculating the cable life according to the calculation model, and configuring a correction factor for characterizing the degree of influence of the cable defect on each calculation model.

[0055] Here, a comprehensive cable life prediction model is formed by combining the calculation models between various first parameters and second parameters that affect cable aging. The prediction model needs to be able to accept real-time data input and predict the remaining life of the cable based on these inputs; correction factors are configured in each calculation model to adjust the model parameters to make it more suitable for specific operating conditions. The correction factors can reflect the actual impact of different operating environments (such as temperature, humidity, current intensity, etc.) on cable aging.

[0056] From the above, we can see that the single influencing factor model (such as the effect of temperature on insulation resistance, the effect of current load on electrothermal aging, etc.) is integrated into a multi-factor comprehensive prediction model, and the data from different sensors are uniformly processed using data fusion technology and provided to the prediction model; the system continuously collects the operation data of the cable through sensors and inputs it into the prediction model in real time. The prediction model calculates the expected life of the cable based on the current data and previous historical data using machine learning algorithms (such as regression analysis and neural networks); based on the data collected during actual operation, the correction factor is dynamically adjusted to optimize the prediction accuracy of the model. The correction factor can be adjusted based on the deviation between the model output and the actual observation results. For example, if the life predicted by the model under high temperature conditions is significantly different from the actual life, the temperature correction factor can be adjusted to reduce this deviation; the prediction results can be displayed to the maintenance team through a visual interface to help them make more accurate maintenance decisions. The feedback from the maintenance team and the actual performance data of the cable will be used to further calibrate and optimize the prediction model, forming a cycle of continuous learning and improvement.

[0057] Specifically, the steps of constructing a prediction model for calculating cable life through all calculation models include: Each first parameter and the corresponding second parameter are combined into a feature group, and the second parameter is used as the first parameter to calculate and obtain an intermediate variable of the cable life.

[0058] A fuzzy algorithm model is constructed according to the second correlation relationship between the first parameter and the cable life.

[0059] The prediction model is determined by the fuzzy algorithm model.

[0060] In view of the above specific description, the feature combination and intermediate variable calculation is to combine various first parameters that affect the performance of the cable with the second parameters that characterize the aging state of the cable into feature groups. These feature groups are the input of the prediction model, and the second parameters here act as intermediate variables calculated from the first parameters; various first parameters (such as temperature, leakage current, etc.) are monitored in real time through the data acquisition system, and the corresponding second parameters (such as insulation resistance, dielectric loss, etc.) are evaluated in combination with historical data or experimental data. These second parameters, as direct indicators of cable aging, reflect the specific impact of the first parameters on the cable status.

[0061] Fuzzy algorithm model construction is to use fuzzy logic algorithm to build a model based on the relationship between the first parameter and the cable life. Fuzzy algorithm can handle uncertainty and imprecision, and is suitable for complex system analysis, such as cable life prediction; fuzzy algorithm simulates the human decision-making process by defining fuzzy sets and fuzzy rules. In cable life prediction, fuzzy algorithm takes into account the uncertainty and fuzzy relationship between various parameters. For example, "higher" temperature and "increased" leakage current may lead to "accelerated" cable aging.

[0062] The determination and application of the prediction model is based on the fuzzy algorithm model to determine the prediction model of cable life, which is used for life prediction in practical applications; the model uses the fuzzy algorithm to calculate the expected life of the cable by inputting the first parameter data of real-time monitoring. The prediction results can help the maintenance team to perform preventive maintenance and timely replacement to avoid cable failures.

[0063] In summary, the system continuously collects data from the cable operating environment, including direct environmental measurements and the performance parameters of the cable itself; the model first converts the collected first parameter data (such as temperature, current, etc.) into fuzzy values. Then, according to the preset fuzzy rules (such as "if the temperature is high and the load is heavy, the aging is fast"), the model evaluates the aging speed and life of the cable; through fuzzy reasoning, the model integrates the influence of various inputs and outputs a clear prediction result, that is, the estimated remaining life of the cable; the obtained life prediction results will inform the maintenance team to make corresponding maintenance decisions, such as strengthening monitoring, planning downtime maintenance or replacing aging cables.

[0064] In any of the above embodiments, the data affecting the cable aging and the data reflecting the cable aging characteristics are expressed in the form of an input and output data group ( x , z ).in, x represents the input vector, which consists of variables related to the output characteristic quantities of cable aging, such as temperature, leakage current, load current, impurities, etc.; z The output variable of the system is a characteristic parameter that describes the aging of the system, such as elongation at break retention rate, dielectric loss tangent value, insulation resistance, etc. The fuzzy algorithm model is as follows: ; In the formula, f() Through the second correlation relationship of the improved fuzzy algorithm model after identification, To predict the output, r is the relevant grid number.

[0065] In this embodiment, the fuzzy algorithm model is used to handle the uncertainty and complexity of the cable aging data. It takes the input variables x (such as temperature, leakage current, load current, and impurities inside the cable) into output variables z (such as service life, etc.), these output variables characterize the aging status of the cable.

[0066] enter xReal-time monitoring data or historical data, including the possible impact of environmental and operating conditions on cables, are collected through sensors and monitoring systems and input into the prediction model in digital form; the precise input data x is converted into fuzzy sets, which define the fuzziness of the input data, such as "high temperature", "medium temperature" and "low temperature"; fuzzy rules are established based on expert experience and historical data, such as "if the temperature is high and the leakage current increases, the insulation resistance is expected to decrease significantly", and fuzzy rules are applied to infer the output variables in combination with the fuzzified input data z fuzzy value; convert the fuzzy reasoning results into accurate prediction outputs, such as specific fracture elongation retention rate or dielectric loss angle values, and use various defuzzification techniques (such as centroid method, maximum membership method, etc.) to determine a clear output value; predict the results z It can be used to assess the current health of the cable and predict its remaining life. This information helps the operation and maintenance team develop more effective maintenance strategies and preventive measures to extend the service life of the cable and prevent failures.

[0067] The fuzzy algorithm model can handle uncertain and ambiguous data and is suitable for complex systems such as cable aging analysis. It improves the flexibility and adaptability of predictions and can adjust the output according to changing environmental conditions.

[0068] In any of the above embodiments, according to t The effective information obtained before the moment, predict t Special parameters of time . Let the predicted feature parameters Equal to the preset failure threshold r, the time at this time is obtained by calculation t w , then the cable is t t-m Remaining life at a moment RUL t-m It can be obtained by the following formula: ; In the formula, when m = 1 is a one-step prediction. The essence of cable aging modeling is to identify the system model based on the existing observation data, predict the characteristic parameters that characterize the cable aging, and thus predict the cable aging amount and remaining life.

[0069] In this embodiment, the main function of the cable life prediction model is to use historical and real-time data to predict the remaining life (RUL) of the cable. The model uses time series data, combined with the operating parameters and aging indicators of the cable, to predict the time point when it reaches the preset failure threshold.

[0070] The model first needs to collect various data related to cable aging (first parameters), such as temperature, leakage current, load current and impurities, and collect second parameters that characterize the aging state, such as elongation retention rate at break, dielectric loss tangent value and insulation resistance; use the collected data to calculate characteristic parameters (such as the rate of change of insulation resistance or the growth trend of dielectric loss), which are used to describe the current aging state of the cable and set one or more failure thresholds ( r ), for example, the insulation resistance drops to a certain critical value, indicating that the cable is about to fail; use historical data and statistical analysis methods (which may include regression analysis, time series analysis, or machine learning algorithms) to build a prediction model that predicts when the cable will reach the failure threshold based on current characteristic parameters and historical trends. tw ; Based on predicted failure time tw , calculated from the current time t The time difference to failure time is the remaining life RUL t-m ,when m =1, it indicates a one-step prediction, that is, predicting the state of the cable in the next time unit starting from the current moment; the model is continuously verified and optimized to ensure the accuracy and reliability of its predictions, and new observation data are used to continuously adjust and calibrate the model parameters to improve the prediction results.

[0071] Specifically, the steps of determining the prediction model through the fuzzy algorithm model include: A third correlation relationship for characterizing the second parameter and the cable life is obtained according to the first correlation relationship and the second correlation relationship.

[0072] The prediction model of each feature group is determined through the third association relationship.

[0073] In response to the above specific description, the third correlation relationship is established by integrating the first correlation relationship and the second correlation relationship to form a third correlation relationship between the second parameter (cable aging characteristics) and the cable life. By accurately describing the relationship between the cable aging characteristics and the life, the accuracy and reliability of the life prediction are improved; the fuzzy algorithm model is constructed using fuzzy logic to deal with the uncertainty and ambiguity in the data, forming a flexible prediction model that can adapt to different operating conditions, and converting complex cable data (including multiple variables and uncertainties) into clear prediction results that can be used for actual decision-making.

[0074] In summary, data integration and analysis collects the first parameters of the cable (such as temperature, load current, etc.) through real-time monitoring and historical data, and calculates the second parameters (such as insulation resistance, dielectric loss tangent, etc.). These data directly characterize the aging state of the cable, analyze the relationship between the first parameter and the second parameter, and how they affect the life of the cable individually and together; fuzzy logic application converts numerical data into fuzzy sets and fuzzy variables, such as dividing the temperature range into "low", "medium" and "high", and establishing fuzzy rules, such as: "If the temperature is high and the insulation resistance drops rapidly, the cable life is greatly reduced", and a fuzzy reasoning system is used to evaluate the expected life of the cable, which involves fuzzy matching and rule reasoning; the prediction model is determined based on the third association relationship, that is, the impact of the first parameter on the second parameter and the impact of the second parameter on the life are comprehensively considered, and the fuzzy algorithm is used to determine the final prediction model. For each feature group, the model can output a predicted value for the remaining life of the cable; prediction output and model optimization prediction results can be fed back to the maintenance team in real time for planning maintenance and preventive replacement. The model will be continuously optimized based on newly acquired data and feedback to improve the accuracy and applicability of the prediction.

[0075] In any of the above embodiments, the prediction model is the following formula: ; in, is the cable life and is independent between feature groups; is the square root of the variance of the normal distribution; is a constant; is the second parameter obtained by the calculation model, and i is a positive integer; is the correction factor; for The mathematical expectation of is an exponential constant.

[0076] In this embodiment, Adjustment The influence of some observations makes some observations more or less important in the overall evaluation. can more accurately represent the actual importance of observations under specific conditions or situations; calculate and The square of the difference is used to quantify the deviation between the observed value and the expected value. The square term amplifies the deviation, so that larger deviations are punished more severely. This is a common practice in many statistical models to emphasize the impact of larger deviations. Describes the effect of the severity of the deviation on the probability density. This form ensures that the deviation from the mean Observations farther away have lower probability density, which conforms to the property of normal distribution, that is, most data should be concentrated near the mean; controlling the width of the probability distribution, larger Indicates that the data is more widely distributed, whereas inversely, it is more concentrated. Directly affects the sensitivity of the exponential function, thereby adjusting the model's tolerance for deviations. This formula provides a powerful tool for evaluation and analysis by accurately calculating the probability density corresponding to the deviation between the weighted observed value and the expected value. In practical applications, this method can be used for error analysis of various forecasting models, helping researchers and engineers understand the variability and uncertainty of the data and their potential impact on the forecast results. This not only helps to optimize model parameters, but also improves the quality of the decision-making process.

[0077] S104, using an optimization algorithm to determine the value of each correction factor, updating the correction factor according to the value to optimize the prediction model; and calculating the cable life through the optimized prediction model.

[0078] Here, the optimization correction factor uses an optimization algorithm (such as genetic algorithm, particle swarm optimization, or gradient descent) to automatically adjust the correction factor in the model so that the output of the prediction model is as close as possible to the actual observation value. By adjusting these factors, it is ensured that the model can accurately predict the aging and life of the cable under different environmental and load conditions; the update prediction model updates the prediction model according to the optimized correction factor to make it better reflect the actual aging behavior of the cable, improve the generalization ability of the model, reduce prediction errors, and provide more reliable decision support.

[0079] As can be seen from the above, initially, the correction factor is set based on theoretical analysis, prior knowledge or early experimental data. These factors may include environmental factors that affect the cable aging rate (such as temperature and humidity) and the physical properties of the cable itself (such as material anti-aging performance); the system collects key parameters that affect the cable life in real time through sensors and monitoring equipment, such as temperature, current, voltage, etc., and at the same time, regularly measures the actual aging status of the cable, such as insulation resistance, dielectric loss, etc.; the collected data is processed using an optimization algorithm, and the algorithm attempts to find the best correction factor so that the error between the cable life predicted by the model and the actual measured value is minimized. This process usually involves a large number of iterative calculations. Each iteration adjusts the correction factor, evaluates the accuracy of the model prediction results, and seeks further improvements; based on the optimization results, the value of the correction factor is updated, and the updated correction factor is reintegrated into the life prediction model to ensure that the model output can reflect the latest equipment status and environmental conditions; using the optimized prediction model, based on real-time data and updated correction factors, the remaining life of the cable is predicted. The prediction results can help the cable management team make decisions on maintenance, replacement or repair, optimize operating costs and reduce the risk of failure.

[0080] Specifically, the optimization algorithm is optimized by a first optimization formula and a second optimization formula, and the steps of the optimization algorithm include: The number of iterative features in the iteration is set according to the number of correction factors, and the optimized value of the iterative feature is used as the value of the correction factor.

[0081] The optimized value of each iterative feature is obtained through the preset number of iterations or judgment threshold.

[0082] In the iteration step, a judgment factor is obtained for judging the adaptability of the correction factor in representing the degree of influence of the cable defect.

[0083] The first optimization formula or the second optimization formula is used to update the optimization value according to the judgment factor.

[0084] In response to the above specific description, the iterative feature setting sets the corresponding number of iterative features according to the number of correction factors. These features represent the adjustable parameters in the model, ensuring that each correction factor can be optimized through the iterative process to achieve the best performance of the prediction model; the iterative optimization executes the preset number of iterations, or until the judgment threshold (such as error minimization, fitness maximization, etc.) is reached, and the optimal value of each iterative feature is found through a continuous optimization process, so as to accurately adjust the correction factor; fitness judgment and formula selection calculate a fitness judgment factor in each iteration, and choose to use the first optimization formula or the second optimization formula for iterative update based on this factor to adapt to different optimization scenarios and data characteristics, and flexibly respond to the performance of the model under different conditions.

[0085] In summary, before the optimization begins, the iteration features are initialized, usually based on previous experience or randomly generated initial values, and the maximum number of iterations or termination conditions are set, such as the fitness reaching a certain threshold; in each iteration, the model output, such as the predicted cable life, is calculated based on the current iteration feature (the current value of the correction factor), and the fitness of the current model output is calculated using an evaluation function (which may be an error function or other performance indicator); according to the value of the fitness judgment factor, an appropriate optimization formula is selected for the next iteration update. For example, the first optimization formula may be suitable for fast convergence scenarios, while the second optimization formula is suitable for high-precision adjustments; the values ​​of the iteration features are updated using the selected optimization formula. These formulas may involve different mathematical or heuristic methods such as gradient descent, Newton's method, and genetic algorithms. The updated iteration features are used for the next round of iterations until the termination conditions are met; once the upper limit of the number of iterations is reached or the fitness meets the preset threshold, the iteration terminates, and the value of the final iteration feature is used as the optimal value of the correction factor in the prediction model to calculate the final life prediction of the cable.

[0086] Specifically, the step of determining whether to use the first optimization formula or the second optimization formula to optimize the value according to the judgment factor specifically includes: Determine whether the judgment factor is greater than the judgment threshold; the initial value of the judgment factor is a random number between 0 and 1.

[0087] If it is greater than, the optimization value is updated using the first optimization formula.

[0088] If it is not greater than, the optimization value is updated by the second optimization formula.

[0089] In view of the above specific description, the judgment factor plays a decisive role and is used to select the specific formula to be applied when updating the optimization value. Its value is randomly generated between 0 and 1, representing a probability or decision threshold, which is used to dynamically adjust the model response. The value of the judgment factor directly affects which optimization formula is adopted, thereby indirectly affecting the adjustment range and direction of the prediction model; the judgment threshold is a fixed standard or parameter, which is compared with the value of the judgment factor to determine the selection of the optimization formula and interacts with the judgment factor. The judgment threshold determines the judgment limit of the judgment factor value and is the benchmark for selecting the optimization strategy.

[0090] It is used when the judgment factor is greater than the judgment threshold. It is usually designed as a more aggressive update strategy, which may be intended to quickly adjust the model to adapt to significantly changing environments or data. Combined with the judgment factor, it can respond quickly to dynamic changes in the environment or data, affecting the immediate performance of the model. It is used when the judgment factor is not greater than the judgment threshold. It is usually designed as a more conservative or detailed update strategy, which may be intended to fine-tune, maintain the stability of the model and reduce overfitting. Combined with the judgment factor, it is used to refine the model when the environment or data does not change much, affecting the long-term stability and accuracy of the model.

[0091] In summary, the judgment factor and the judgment threshold jointly determine which optimization formula to use, which affects the way the optimization value is updated. The first optimization formula and the second optimization formula each correspond to different scenarios and requirements, allowing the model to flexibly adapt to various changes and ensure the accuracy of the prediction and the adaptability of the model. This dynamic selection mechanism enables the prediction model to not only quickly adapt to environmental changes in the short term, but also maintain long-term stability and reliability.

[0092] In any of the above embodiments, the first optimization formula is: ; Among them, the is the position of the whale at the current iteration, and t is the number of iterations; is the current optimal whale position; The is a random number uniformly distributed in (0, 1); Each dimension of is uniformly distributed in (- a ,a ), and a is the boundary constant.

[0093] In this embodiment, and The differences drive the update of the entire system. Set a goal, and is the state actually achieved, and the deviation between the two is the driving force for adjustment. and Acting together on this difference, the sensitivity and magnitude of the adjustment are determined, thus ensuring that the system is moving towards In the process, there is enough power without losing stability due to excessive response. In this way, the model can gradually approach the ideal state or target while maintaining stability, and can be effectively applied to cable aging and life assessment, helping the prediction model to more accurately reflect the actual situation and make reasonable predictions. This dynamic update mechanism makes the model highly adaptable and robust in complex and changing practical application environments.

[0094] In any of the above embodiments, the second optimization formula is: ; in, is an exponential constant, and is the shape parameter and takes the value of 0.5, is a random number uniformly distributed in (0, 1); is a constant; is a trigonometric function.

[0095] In this embodiment, Indicates the current state, and Represents the target state, the difference between the two Directly affects the direction and magnitude of state updates. This difference is the driving force that pushes the system toward the target state and is the basis for state updates. By adjusting the values ​​of b and c, the rate of growth or decay can be controlled, which is the key to adjusting the update step size. Introducing periodic adjustments increases the model's adaptability to time changes. The combination of exponential and cosine functions makes the update not only dependent on the state difference, but also affected by periodic and exponential adjustments, so that the model can show different update behaviors at different stages, such as smoother or more intense adjustments. b adjusts the sensitivity and rate of the update amplitude by affecting the base of the exponential function, and c affects the periodicity of the cosine function and adjusts the periodic characteristics of the update. b and c jointly determine The update strategy allows the model to be flexibly adjusted according to different operating environments and target requirements. The design of this update formula takes into account the dynamics and adaptability of state updates. By reasonably adjusting the values ​​of b and c, fine control of model behavior can be achieved. This method is particularly suitable for systems that need to be highly sensitive to environmental changes, such as cable aging models, to ensure that the prediction model can effectively adapt to environmental changes and improve the accuracy of predictions and the stability of the system.

[0096] In any of the above embodiments, the multi-parameter prediction model design is based on the above four typical algorithm models associated with cable aging defects, and an improved fuzzy algorithm is introduced. The outputs of the four algorithms are used as the input of the prediction function. , as the input of the reaction characteristic of the cable aging characteristics. The Gaussian equation of the normal distribution is selected as the metric function. The normal distribution is a bell-shaped curve, and the feature focusing convergence is strong, which can better identify the characteristic parameters. x represents the input vector, z Represents the feature output, and the prediction function is as follows: ; Since the characteristics of each input vector xi have different influences on cable defects, if the algorithm is designed according to the average influence, the prediction result will deviate from the actual situation, especially when the environment is quite different. Therefore, the present invention proposes a whale algorithm to optimize weights. The whale algorithm is an observation process of the hunting behavior of humpback whale groups. It has the advantages of strong global optimization ability, potential parallelism and easy implementation. The algorithm includes two processes: shrinking and surrounding and spiral updating position. Different processes are selected according to different parameter value ranges. Specific description of the algorithm: 1) Initialize the basic parameters of the algorithm: cable feature population size n Take 4, the maximum number of iterations T max Take 100, shape parameter b take 0.5, and number of iterations t =1; 2) The iterative optimization calculation formula of the whale algorithm is as follows: Shrinkage encirclement iteration formula: ; Spiral update position iteration formula: ; renew A, b, c The value of controls the search range parameter, which decreases with iteration, simulating the behavior of whales gradually approaching their prey and generating a judgment factor between [0, 1] P。

[0097] 3) If PIf greater than 0.5, the prey is randomly surrounded according to the shrinking and surrounding iterative formula; P If it is less than 0.5, the whale position is updated according to the spiral position iteration formula.

[0098] 4) Determine whether to end the optimization process. t achieve T max , or when the convergence accuracy requirement is reached, the optimization process ends, otherwise go to (2) for the next search.

[0099] 5) Output the optimal weight vector .

[0100] This optimizes the following formula: ; Let the predicted feature parameters Equal to the preset failure threshold r , and the time at this time is obtained by calculating t w , then the cable is t t-m Remaining life at a moment RUL t-m It can be obtained by the following formula.

[0101] ; In the formula, when m = 1 is a one-step prediction. The essence of cable aging modeling is to identify the system model based on the existing observation data, predict the characteristic parameters that characterize the cable aging, and thus predict the cable aging amount and remaining life.

[0102] The present invention provides a cable defect multi-parameter aging characteristic and life assessment method. By obtaining multiple first parameters affecting cable aging and their corresponding second parameters (cable aging characterization), the present invention can more comprehensively analyze the dynamic process of cable aging, and solve the problem that traditional methods fail to accurately identify and quantify all factors affecting cable aging; construct a calculation model including a first correlation relationship between each second parameter and the corresponding first parameter, so that the relationship between the influencing factors and results of cable aging is clearer and traceable; and use an optimization algorithm to dynamically determine and adjust the value of the correction factor, so that the prediction model can instantly reflect the latest operating status and environmental conditions.

[0103] Another embodiment of the first aspect of the present invention provides a specific implementation process of a cable defect multi-parameter aging characteristic and life assessment method. In some embodiments of the present invention, Figure 2 As shown, the specific implementation process includes: Step 201, cable test and evaluation request: initiate a request for cable test and evaluation. The system receives a request for test and evaluation and starts the processing flow.

[0104] Step 202, read line_eval table: read relevant information from line_eval database table. The system retrieves necessary data through eval_id and line_id.

[0105] Step 203, determine whether the information exists: check whether the line_eval table contains relevant information. If the information exists, continue; if not, the process terminates.

[0106] Step 204, more detailed data check: verify the more detailed information in the line_eval table and verify whether other necessary information is complete.

[0107] Step 205, determine whether to continue: determine whether to continue based on more detailed data check results. If the data check passes, proceed to the next step; otherwise, the process ends.

[0108] Step 206, update the database using the information: use the acquired information to update the records in the database. The system updates the database based on the new evaluation information to keep the data up to date.

[0109] Step 207, perform cable assessment: perform a detailed assessment of the cable. The system uses the collected data and updated database information to perform a detailed assessment of the cable status.

[0110] Step 208, execution decision: make corresponding decisions based on the cable evaluation results, and perform necessary maintenance, replacement or repair operations according to the evaluation results.

[0111] Step 209, outputting the evaluation results: outputting the detailed results of the cable evaluation and recording the evaluation results in a database for subsequent query and analysis.

[0112] The specific implementation process of the system provided by the present invention starts from receiving an evaluation request, automatically triggers the test and evaluation process, quickly reads and verifies the data in the line_eval table to ensure the integrity and accuracy of the information, and continuously updates the database information to ensure that the data basis for the evaluation is always the latest, so that each evaluation is based on the most accurate cable status. In addition, potential risks and problems are identified through a meticulous evaluation process, providing a scientific basis to support preventive maintenance or timely repair of cables. Based on the evaluation results, the maintenance team can make wise decisions, such as repairing, replacing or further monitoring the cables, optimizing resource allocation and maintenance plans. Ultimately, the evaluation results are recorded in detail in the database, which not only supports immediate operational decisions, but also provides valuable data support for long-term equipment management and maintenance strategies. Overall, it not only improves the efficiency and effectiveness of cable management, but also helps reduce operating costs, extend the service life of cables, and enhances the reliability and safety of the entire power system.

[0113] The second aspect of the present invention provides a system for implementing the cable defect multi-parameter aging characteristics and life assessment method in any of the above embodiments. Figure 3 As shown, the prediction system 2 includes: The data collection and processing module 201 is used to collect data of the first parameter and the second parameter.

[0114] The model building module 202 builds a calculation model according to the first association relationship, and builds a prediction model through the calculation model.

[0115] The optimization parameter adjustment module 203 uses an optimization algorithm to determine the value of each correction factor and optimize the prediction model.

[0116] The service life prediction module 204 obtains the service life of the cable according to the optimized prediction model.

[0117] The present invention provides a system, wherein the data collection and processing module 201 provides a systematic way to collect key operation and environmental data, including first parameters such as cable temperature, current, and voltage, and second parameters such as insulation resistance and dielectric loss. Through automated data collection, human errors are reduced, the accuracy and timeliness of data are guaranteed, and high-quality input is provided for subsequent analysis. The model building module 202 uses the collected data to build accurate calculation models, which can reflect the performance and aging status of cables under different conditions. Through these models, complex data can be converted into easy-to-understand prediction results to help technicians understand the dynamics of cable aging. The optimization parameter adjustment module 203 uses advanced optimization algorithms (such as genetic algorithms, particle swarm optimization, etc.) to dynamically adjust the correction factors in the prediction model to ensure the optimal performance of the model under different environments. This dynamic adjustment mechanism allows the model to adapt to environmental changes by itself, thereby improving the accuracy and reliability of the prediction. The life prediction module 204 accurately predicts the remaining service life of the cable based on the optimized prediction model, helping the operation and maintenance team to formulate more effective maintenance strategies. The prediction results can be used for preventive maintenance and planned downtime, reducing the risk of unexpected downtime and extending the service life of cables and related equipment.

[0118] Another embodiment of the second aspect of the present invention provides a system workflow. In some embodiments of the present invention, such as Figure 4 As shown, the workflow includes: 2. Read the evaluation model parameters and perform integrity check: Read the model parameters used for cable life evaluation from the system configuration or database, verify the integrity and validity of the read model parameters, and ensure the accuracy and consistency of the evaluation model during operation.

[0119] 3. Read cable static parameters and perform integrity check: Collect static parameters related to the cable, such as cable type, installation date, manufacturer, etc., and ensure that these static parameters are complete and correct, as they are key factors in evaluating the cable life.

[0120] 4. Read the cable monitoring data required for evaluation through the data interface: obtain the cable operation data such as current, temperature and load in real time from the monitoring system or sensor network to ensure that the evaluation model has sufficient real-time data for analysis and improve the real-time and accuracy of the evaluation results.

[0121] 5. Preprocess the acquired monitoring data: Format, clean and filter the collected data to remove noise and outliers, prepare clean and accurate data to input into the evaluation model, and improve the efficiency of data processing and the reliability of analysis results.

[0122] 6. Input into the evaluation model for analysis: Input the pre-processed data into the life evaluation model for comprehensive analysis and processing. The data is processed and parsed by advanced algorithms to derive the health status and predicted life of the cable.

[0123] 7. Output analysis results to database: Store the cable life prediction results and other relevant analysis data into the database, save the evaluation results for historical comparison and future reference, and support other system functions such as report generation and trend analysis.

[0124] The workflow of the system provided by the present invention not only ensures the systematic and scientific nature of cable life assessment, but also provides continuous monitoring of data and tracking of assessment results, making cable maintenance and replacement plans more accurate and timely.

[0125] In the description of the present invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0126] The above embodiments are only descriptions of the preferred modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A cable defect multi-parameter aging characteristics and life assessment method, characterized in that: The steps include: Acquire a plurality of first parameters that can affect cable aging and a second parameter that changes in the cable aging characterization caused by each first parameter; Constructing a calculation model including a first association relationship between each second parameter and the corresponding first parameter; Constructing a prediction model for calculating the life of the cable according to the calculation model, and configuring a correction factor for characterizing the degree of influence of the cable defect on each of the calculation models; An optimization algorithm is used to determine the value of each correction factor, and the correction factor is updated according to the value to optimize the prediction model; the cable life is calculated using the optimized prediction model.

2. The cable defect multi-parameter aging characteristics and life assessment method according to claim 1 is characterized in that: The step of constructing a prediction model for calculating the cable life through all the calculation models comprises: Each of the first parameters and the corresponding second parameter form a feature group, and the second parameter is used as the first parameter to calculate and obtain an intermediate variable of the cable life; Constructing the fuzzy algorithm model according to the second correlation relationship between the first parameter and the cable life; The prediction model is determined by the fuzzy algorithm model.

3. The cable defect multi-parameter aging characteristics and life assessment method according to claim 2 is characterized in that: The step of determining the prediction model through the fuzzy algorithm model specifically includes: Acquire a third association relationship for characterizing the second parameter and the cable life according to the first association relationship and the second association relationship; The prediction model of each feature group is determined by the third association relationship.

4. The cable defect multi-parameter aging characteristics and life assessment method according to claim 3 is characterized in that: The prediction model is the following formula: ； Among them, the is the cable life, and is independent of each other among the feature groups; is the square root of the variance of the normal distribution; is a constant; is the second parameter obtained by the calculation model, and i is a positive integer; is the correction factor; for The mathematical expectation of is an exponential constant.

5. The cable defect multi-parameter aging characteristics and life assessment method according to claim 2, characterized in that: The optimization algorithm is optimized by a first optimization formula and a second optimization formula, and the steps of the optimization algorithm include: The number of iterative features in the iteration is set according to the number of the correction factors, and the optimized value of the iterative feature is used as the value of the correction factor; Obtaining the optimization value of each of the iterative features through a preset number of iterations or a judgment threshold; In the iteration step, a judgment factor is obtained for judging the adaptability of the correction factor in representing the degree of influence of the cable defect; The first optimization formula or the second optimization formula is used to update the optimization value according to the judgment factor.

6. The cable defect multi-parameter aging characteristics and life assessment method according to claim 5, characterized in that: The step of determining, according to the judgment factor, to use the first optimization formula or the second optimization formula to optimize the value specifically includes: Determine whether the judgment factor is greater than a judgment threshold; If it is greater than, updating the optimization value by the first optimization formula; If it is not greater than, the optimization value is updated by the second optimization formula.

7. The cable defect multi-parameter aging characteristics and life assessment method according to claim 5, characterized in that: The first optimization formula is: ； Among them, the is the optimized value under the current number of iterations, and t is the number of iterations; is the current optimal whale position; The is a random number uniformly distributed in (0, 1); Each dimension of is uniformly distributed in (- a , a ), and a is the boundary constant, and its value is .

8. The cable defect multi-parameter aging characteristics and life assessment method according to claim 7, characterized in that: The second optimization formula is: ； Among them, the is an exponential constant, and the is the shape parameter and takes the value of 0.

5. is a random number uniformly distributed in (0, 1); is a constant; is a trigonometric function.

9. The cable defect multi-parameter aging characteristics and life assessment method according to claims 1-8, characterized in that: The first parameter includes at least one parameter of the cable selected from each of working environment, working loss, working load and material property.

10. A system for implementing the cable defect multi-parameter aging characteristics and life assessment method according to any one of claims 1 to 9, characterized in that: include: A data collection and processing module, used for collecting data of the first parameter and the second parameter; A model building module, which builds the calculation model according to the first association relationship, and builds the prediction model through the calculation model; An optimization parameter adjustment module, which uses the optimization algorithm to determine the value of each correction factor and optimize the prediction model; The service life prediction module obtains the service life of the cable according to the optimized prediction model.

Images