Method, device, system and terminal equipment for predicting cable insulation breakdown field strength

Abstract

The application provides a cable insulation breakdown field strength prediction method, device, system and terminal equipment, relates to the high-voltage insulation technical field, and the method comprises the following steps: obtaining the characteristic parameters of a to-be-tested cable; taking the characteristic parameters as input, obtaining the insulation breakdown field strength prediction correction value of the to-be-tested cable through a breakdown field strength correction value prediction model; obtaining the insulation layer thickness of the to-be-tested cable, determining the insulation breakdown field strength theoretical value of the to-be-tested cable under the insulation layer thickness through an insulation breakdown field strength prediction model, and the insulation breakdown field strength prediction model is obtained through nonlinear fitting on the insulation breakdown field strength actual value of an insulation layer slice sample with different thicknesses; and determining the insulation breakdown field strength prediction value of the to-be-tested cable according to the insulation breakdown field strength prediction correction value and the insulation breakdown field strength theoretical value. The application can realize the prediction of the cable insulation breakdown field strength without destructive test, and meanwhile, the accuracy of the cable insulation breakdown field strength prediction is improved.

Description

Technical Field

[0001] This application relates to the field of high voltage insulation technology, specifically to a method for predicting the breakdown field strength of cable insulation, a device for predicting the breakdown field strength of cable insulation, a system for predicting the breakdown field strength of cable insulation, a machine-readable storage medium, and a terminal device. Background Technology

[0002] Insulation breakdown field strength is an important indicator reflecting the electrical withstand capability of cable insulation. As transmission voltage levels increase, the insulation thickness of power cables also increases accordingly. Taking ultra-high voltage cables as an example, their insulation thickness often exceeds 20mm, with 500kV XLPE cables reaching over 30mm. For such thick insulation layers, the output voltage of existing high-voltage generators is insufficient to cause breakdown, thus making it impossible to conduct insulation breakdown field strength tests.

[0003] Furthermore, for cables with relatively low voltage levels and relatively thin insulation layers, although the output voltage of existing high-voltage generating equipment is sufficient to cause breakdown, and the insulation breakdown field strength can be obtained through testing, this destructive testing is costly and often time-consuming, requiring significant manpower and resources. Therefore, there is an urgent need in this field for a non-destructive method that can accurately predict the insulation breakdown field strength of high-voltage cables based on small-sized sample testing. Summary of the Invention

[0004] The purpose of this application is to provide a method for predicting the breakdown field strength of cable insulation, a device for predicting the breakdown field strength of cable insulation, a system for predicting the breakdown field strength of cable insulation, a machine-readable storage medium, and a terminal device, so as to solve the problems of high testing cost and low prediction accuracy in the prior art.

[0005] To achieve the above objectives, the first aspect of this application provides a method for predicting the breakdown field strength of cable insulation, comprising:

[0006] Obtain the characteristic parameters of the cable under test;

[0007] Using the aforementioned feature parameters as input, the insulation breakdown field strength prediction correction value of the cable under test is obtained through the breakdown field strength correction value prediction model. The breakdown field strength correction value prediction model is obtained by training a preset deep learning algorithm with the feature parameters of different cables and the insulation breakdown field strength correction value of each cable. The insulation breakdown field strength correction value of each cable is obtained based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable.

[0008] The insulation layer thickness of the cable under test is obtained, and the theoretical value of the insulation breakdown field strength of the cable under test is determined by the insulation breakdown field strength prediction model. The insulation breakdown field strength prediction model is obtained by nonlinear fitting of the actual values ​​of insulation breakdown field strength of insulation layer slices with different thicknesses.

[0009] The predicted insulation breakdown field strength of the cable under test is determined based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength.

[0010] Optionally, the characteristic parameters include at least one of the structural characteristics, physicochemical properties, and processing characteristics of the cable under test;

[0011] The structural features include: insulation eccentricity;

[0012] The physicochemical properties include at least one of the following: crystallinity, degree of crosslinking, insulation breakdown field strength of the insulating layer slice sample, and volume resistivity.

[0013] The processing technology features include at least one of the following: by-product content and thermal elongation.

[0014] The insulation layer slice sample is made of the same material as the cable under test, and the insulation breakdown field strength of the insulation layer slice sample is obtained by performing an insulation breakdown field strength test on the pre-prepared insulation layer slice sample.

[0015] Optionally, the correction value for the insulation breakdown field strength of each cable is obtained based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable, including:

[0016] The actual insulation breakdown field strength of each cable was obtained by insulation breakdown field strength test.

[0017] The insulation layer thickness of each cable is obtained. Using the insulation layer thickness of each cable as input, the theoretical value of the insulation breakdown field strength of each cable is output through the insulation breakdown field strength prediction model.

[0018] For each cable, the actual value and theoretical value of the insulation breakdown field strength of the current cable are used as inputs, and the corrected value of the insulation breakdown field strength of the current cable is obtained based on the insulation breakdown field strength correction value calculation model.

[0019] The calculation model for the insulation breakdown field strength correction value is constructed based on the equivalent relationship between the insulation breakdown field strength correction value of the insulating material and the ratio of the actual insulation breakdown field strength value and the theoretical insulation breakdown field strength value of the insulating material.

[0020] Optionally, the fitting process of the insulation breakdown field strength prediction model includes:

[0021] Multiple insulation layer slices of different thicknesses are prepared in advance, and the material of the insulation layer slices is the same as that of the cable to be tested.

[0022] Each insulation layer slice sample was placed in a test environment at the preset cable insulation operating temperature;

[0023] For each insulation layer slice sample, the insulation breakdown field strength test is performed on the current insulation layer slice sample by continuously increasing the voltage, so as to obtain the actual value of the insulation breakdown field strength of the current insulation layer slice sample under different breakdown voltages.

[0024] Based on the actual values ​​of insulation breakdown field strength of each insulation layer slice sample under different breakdown voltages, the pre-constructed functional relationship between insulation breakdown field strength and insulation layer thickness is nonlinearly fitted to obtain the insulation breakdown field strength prediction model.

[0025] The functional relationship between the insulation breakdown field strength and the insulation layer thickness is used to represent the mapping relationship in which the insulation breakdown field strength of the insulation layer decreases exponentially with the increase of its thickness.

[0026] Optionally, the preset deep learning algorithm includes the random forest algorithm, and the training steps of the breakdown field strength correction value prediction model include:

[0027] Multiple cables with different insulation layer thicknesses were prepared using the same material as the cable under test;

[0028] Each cable obtained is cut into multiple cable samples of equal length;

[0029] For each cable sample, obtain the characteristic parameters of the current cable sample and determine the insulation breakdown field strength correction value for each cable sample.

[0030] Using the characteristic parameters of each cable sample as input, and the insulation breakdown field strength correction value of the corresponding cable sample as the label vector, the random forest algorithm is trained to obtain the breakdown field strength correction value prediction model.

[0031] Optionally, determining the predicted insulation breakdown field strength of the cable under test based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength includes:

[0032] An insulation breakdown field strength correction value calculation model is obtained, which is constructed based on the equivalent relationship between the insulation breakdown field strength correction value of the insulating material and the ratio of the actual insulation breakdown field strength and the theoretical insulation breakdown field strength of the insulating material;

[0033] The insulation breakdown field strength prediction correction value is used as the insulation breakdown field strength correction value. Based on the insulation breakdown field strength prediction correction value and the insulation breakdown field strength theoretical value, the actual insulation breakdown field strength of the cable under test is obtained based on the insulation breakdown field strength correction value calculation model. The obtained actual insulation breakdown field strength is used as the insulation breakdown field strength prediction value of the cable under test.

[0034] A second aspect of this application provides a cable insulation breakdown field strength prediction device, comprising:

[0035] The data acquisition module is configured to acquire characteristic parameters of the cable under test;

[0036] The correction value prediction module is configured to take the feature parameters as input and obtain the predicted correction value of the insulation breakdown field strength of the cable under test through the breakdown field strength correction value prediction model. The breakdown field strength correction value prediction model is obtained by training a preset deep learning algorithm with the feature parameters of different cables and the insulation breakdown field strength correction value of each cable. The insulation breakdown field strength correction value of each cable is obtained based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable.

[0037] The insulation breakdown field strength theoretical value calculation module is configured to obtain the insulation layer thickness of the cable under test, and determine the theoretical value of the insulation breakdown field strength of the cable under test under the insulation layer thickness through the insulation breakdown field strength prediction model. The insulation breakdown field strength prediction model is obtained by nonlinear fitting of the actual insulation breakdown field strength values ​​of insulation layer slice samples with different thicknesses.

[0038] The insulation breakdown field strength prediction module is configured to determine the predicted value of the insulation breakdown field strength of the cable under test based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength.

[0039] A third aspect of this application provides a cable insulation breakdown field strength prediction system, comprising:

[0040] An insulation material testing device is used to test at least one of the structural characteristics, physicochemical characteristics, and processing characteristics of a cable under test to obtain characteristic parameters of the cable under test; and

[0041] Such as the cable insulation breakdown field strength prediction device mentioned above.

[0042] In a fourth aspect, this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the cable insulation breakdown field strength prediction method as described above.

[0043] In a fifth aspect, this application provides a terminal device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the cable insulation breakdown field strength prediction method described above.

[0044] This application, based on the physical and chemical properties, processing performance and other characteristic parameters of cable insulation, uses machine learning algorithms to obtain a predictive model for the quantitative relationship between the multi-dimensional performance parameters of cable insulation and its insulation breakdown field strength. This model can predict the cable insulation breakdown field strength, thereby bypassing the actual cable breakdown test and avoiding destructive testing of the cable. It achieves a quantitative description of the dielectric strength of the cable insulation layer, while avoiding the cost of performing a breakdown test on the cable insulation.

[0045] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0046] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings:

[0047] Figure 1 A flowchart illustrating the method for predicting the breakdown field strength of cable insulation provided in a preferred embodiment of this application;

[0048] Figure 2 This is a schematic diagram of the cable insulation breakdown field strength prediction logic provided in a preferred embodiment of this application;

[0049] Figure 3 A schematic diagram of a cable insulation breakdown field strength prediction device provided in a preferred embodiment of this application;

[0050] Figure 4 A schematic diagram of a terminal device provided in a preferred embodiment of this application.

[0051] Explanation of reference numerals in the attached figures

[0052] 10 - Terminal device, 100 - Processor, 101 - Memory, 102 - Computer program. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0054] It should be noted that the technical solutions of the various embodiments of this application can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by this application.

[0055] To solve the above problems, such as Figure 1 As shown, the first aspect of this application provides a method for predicting the breakdown field strength of cable insulation, comprising:

[0056] S100: Obtain the characteristic parameters of the cable under test;

[0057] S200. Using the characteristic parameters as input, the insulation breakdown field strength prediction correction value of the cable under test is obtained through the breakdown field strength correction value prediction model. The breakdown field strength correction value prediction model is obtained by training a preset deep learning algorithm with the characteristic parameters of different cables and the insulation breakdown field strength correction value of each cable. The insulation breakdown field strength correction value of each cable is obtained based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable.

[0058] S300. Obtain the insulation layer thickness of the cable under test, and determine the theoretical value of the insulation breakdown field strength of the cable under test under the insulation layer thickness through the insulation breakdown field strength prediction model. The insulation breakdown field strength prediction model is obtained by nonlinear fitting of the actual values ​​of insulation breakdown field strength of insulation layer slices with different thicknesses.

[0059] S400. Determine the predicted value of the insulation breakdown field strength of the cable under test based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength.

[0060] Thus, this application, based on the physical and chemical properties, processing properties and other characteristic parameters of cable insulation, obtains a predictive model through machine learning algorithms to describe the quantitative relationship between the multi-dimensional performance parameters of cable insulation and its insulation breakdown field strength, thereby predicting the cable insulation breakdown field strength. This bypasses the actual cable breakdown test, avoids destructive testing of the cable, and achieves a quantitative description of the dielectric strength of the cable insulation layer, while avoiding the cost of performing a breakdown test on the cable insulation.

[0061] In step S100, the characteristic parameters of this application include at least one of the structural characteristics, physicochemical properties, and processing characteristics of the cable under test; wherein, the structural characteristics include: insulation eccentricity (%); the physicochemical properties include at least one of: crystallinity (DSC method, %), degree of crosslinking (gel content, %), insulation breakdown field strength of the insulation layer slice sample (kV / mm), and volume resistivity (Ω·m); the processing characteristics include at least one of: by-product content (ppm) and thermal elongation (%); wherein, the insulation layer slice sample is made of the same material as the cable under test, and the insulation breakdown field strength of the insulation layer slice sample is obtained by testing the insulation breakdown field strength of a pre-prepared insulation layer slice sample. The insulation layer slice sample can be prepared using cross-linked polyethylene (XLPE) insulation material, and the thickness of the insulation layer slice sample can be 0.2 mm.

[0062] In step S200, the correction value for the insulation breakdown field strength of each cable is obtained based on the actual value and theoretical value of the insulation breakdown field strength of each cable, including:

[0063] S210. Obtain the actual value of the insulation breakdown field strength of each cable through insulation breakdown field strength test.

[0064] Specifically, in this application, the same XLPE insulation material as the cable under test was used to prepare three AC model cables with different insulation thicknesses by extrusion. The insulation thicknesses of the three model cables were 3.0 mm, 4.5 mm, and 6.0 mm, respectively. Each thickness of model cable was continuously extruded for 400 m, with 100 m reserved for constructing the test simulation circuit, etc. The remaining portion was evenly cut into 10 segments, each 30 m long, thus obtaining 10 cable samples for each thickness. Insulation breakdown tests were performed on all 30 model cable samples, and the actual insulation breakdown field strength (Eac) of each sample was measured.

[0065] S220. Obtain the insulation layer thickness of each cable. Using the insulation layer thickness of each cable as input, output the theoretical value of the insulation breakdown field strength of each cable through the insulation breakdown field strength prediction model.

[0066] like Figure 2 As shown, the fitting process of the insulation breakdown field strength prediction model in this application includes:

[0067] Step 1: Prepare multiple insulation layer slices of different thicknesses in advance. The material of the insulation layer slices is the same as that of the cable to be tested. For example, using cross-linked polyethylene (XLPE) insulation material, and employing a flat vulcanizing machine under the same cross-linking process conditions, prepare three flat samples of different thicknesses: 0.2 mm, 0.6 mm, and 1.0 mm, each with a length and width of 100 mm × 100 mm. To ensure statistical validity, prepare 10 replicate samples for each thickness.

[0068] Step 2: Place each insulation layer slice sample in a test environment at the preset cable insulation operating temperature. Specifically, use a thickness gauge to measure the actual thickness of the flat sample, and then place the flat sample in silicone oil at 90°C to simulate the actual operating temperature conditions of the cable insulation and prevent surface flashover of the sample.

[0069] Step 3: For each insulation layer slice sample, the insulation breakdown field strength is tested sequentially by continuously increasing the voltage to obtain the actual insulation breakdown field strength of the current insulation layer slice sample under different breakdown voltages. Record the insulation breakdown voltage U of each plate sample. break And calculate the actual value of its insulation breakdown field strength E. break The actual values ​​of the insulation breakdown field strength of the three thicknesses of the flat plate samples are shown in Table 1.

[0070] Table 1

[0071]

[0072] Step 4: Based on the actual values ​​of the insulation breakdown field strength of each insulation layer slice sample under different breakdown voltages, perform nonlinear fitting on the pre-constructed functional relationship between the insulation breakdown field strength and the insulation layer thickness to obtain the insulation breakdown field strength prediction model; wherein, the functional relationship between the insulation breakdown field strength and the insulation layer thickness is used to represent the mapping relationship in which the insulation breakdown field strength of the insulation layer decreases exponentially with the increase of its thickness. Furthermore, in the decay process of the insulation breakdown field strength decreasing exponentially with the increase of its thickness, the decay reference amplitude and decay rate of the insulation breakdown field strength are determined based on at least one coefficient related to the properties of the insulation material.

[0073] Understandably, the breakdown field strength of insulating materials exhibits a size effect, meaning there is a functional relationship between the thickness of the insulating sample and its breakdown field strength. This functional relationship typically follows an inverse power law. In this application, the functional relationship between the thickness of the insulating sample and its breakdown field strength can be specifically expressed as follows:

[0074] E(d) = kd -n

[0075] Where E(d) is the breakdown field strength of the insulation sample, d is the thickness of the insulation sample, and k and n are coefficients related to the properties of the insulation material. The attenuation reference amplitude of the insulation breakdown field strength is determined by the value of k, and the attenuation rate is determined by the value of n.

[0076] Based on Table 1, the thickness d is compared with the corresponding actual value of the insulation breakdown field strength E. break Substituting the data into the inverse power law formula E(d)=kd -nNonlinear least squares fitting was performed. After fitting, the intrinsic parameters of the XLPE insulation material were obtained as k=81.2 and n=0.266. Based on this, a size effect model M_p, i.e., an insulation breakdown field strength prediction model, was established: E(d)=81.2d -0.266 (Unit: kV / mm), where the goodness of fit of the model is R2=0.985, indicating that the fitting effect is good.

[0077] S230. For each cable, using the actual and theoretical values ​​of the insulation breakdown field strength as inputs, the corrected insulation breakdown field strength value is obtained based on the insulation breakdown field strength correction value calculation model. The calculation model is constructed based on the equivalence relationship between the corrected insulation breakdown field strength value of the insulation material and the ratio of the actual and theoretical insulation breakdown field strength values ​​of that material. In this application, the insulation breakdown field strength correction value f... E Defined as the ratio of the actual value of the insulation breakdown field strength Eac to the theoretical value Eth output through M_p, i.e., f E =Eac / Eth.

[0078] As mentioned above, the inverse power law of insulation breakdown field strength mainly considers the thickness of the insulation sample as the main parameter affecting the breakdown field strength, but it cannot take into account the anisotropy of the insulation sample's structure and material properties. For the cross-linked polyethylene (XLPE) insulation layer of a real cable, it has a coaxial structure rather than a simple flat structure. Furthermore, in reality, the cable insulation layer may have eccentricity, local defects, and other manufacturing issues. These problems make it difficult for the mathematical model defined in the above formula to accurately predict the insulation breakdown field strength of actual cable insulation layers. Therefore, this application, based on considering the influence of the anisotropy of the insulation sample's structure and material properties on the insulation breakdown field strength, further trains a deep learning algorithm on the characteristic parameters of different cables and their corresponding insulation breakdown field strength correction values ​​to construct a prediction model for the breakdown field strength correction value, representing the mapping relationship between the characteristic parameters of the insulation material and the insulation breakdown field strength correction value.

[0079] Specifically, the deep learning algorithm preset in this application includes the random forest algorithm, and the training steps of the breakdown field strength correction value prediction model include:

[0080] Step 1: Prepare multiple cables with different insulation thicknesses using the same material as the cable under test. For example, using the same XLPE insulation material as the cable under test, prepare three AC model cables with different insulation thicknesses by extrusion. The insulation thicknesses of the three model cables are 3.0 mm, 4.5 mm, and 6.0 mm, respectively.

[0081] Step 2: Cut each prepared cable into multiple cable samples of equal length. Specifically, continuously extrude 400m of the model cable of each thickness prepared in the above steps, retain 100m for building test simulation circuits, and cut the remaining part evenly into 10 segments, each 30m long, thus obtaining 10 cable samples of each thickness.

[0082] Step 3: For each cable segment, obtain the characteristic parameters of the current cable sample and determine the insulation breakdown field strength correction value for each cable sample. Insulation breakdown tests are performed on all 30 model cable samples, and the actual insulation breakdown field strength value Eac for each sample is measured. Simultaneously, for each model cable sample, the measured insulation thickness d is substituted into the model M_p to calculate its theoretical breakdown field strength value Eth = 81.2d. -0.266 Subsequently, the unique insulation breakdown field strength correction value f for each sample was calculated. E =Eac / Eth. The calculation results of the breakdown field strength and insulation breakdown field strength correction values ​​for all model cable samples are shown in Table 2.

[0083] Table 2

[0084]

[0085] The broken-down model cable sample was sliced ​​to prepare standard test specimens, for example, 0.2 mm thick slices. Multiple performance tests were performed on each slice to obtain the insulation eccentricity (%), crystallinity (DSC method, %), crosslinking degree (gel content, %), insulation breakdown field strength (kV / mm), volume resistivity (Ω·m), by-product content (ppm), and thermal elongation (%) of each slice.

[0086] The results of the slice test of 30 model cable samples are shown in Table 3.

[0087] Table 3

[0088]

[0089] For each model cable sample, all characteristic parameters (sample characteristics) are compared with their corresponding insulation breakdown field strength correction value f. E (Sample labels) are associated to form a data record. Ultimately, a training database containing 30 data records is obtained.

[0090] Step 4: Using the characteristic parameters of each cable sample as input, and the insulation breakdown field strength correction value of the corresponding cable sample as the label vector, train the random forest algorithm to obtain the breakdown field strength correction value prediction model.

[0091] Specifically, the feature parameters (eccentricity, crystallinity, degree of crosslinking, slice breakdown field strength, resistivity, by-product content, and thermal elongation) in the aforementioned database are used as input features (x), and the insulation breakdown field strength correction value f is used. E The sample label (y) is used. The RandomForestRegressor algorithm from the Scikit-learn library is used for training. The specific steps are as follows:

[0092] 1) Parameter configuration: Set n_estimators=100 (number of decision trees), keep the other parameters at their default values;

[0093] 2) Data splitting: With a total of 30 samples, divide the data into 24 training sets and 6 internal test sets using the train_test_split function at a ratio of 80% / 20%;

[0094] 3) Model training: The fit() method is called based on the training set to complete the model fitting.

[0095] The process of the forest regression algorithm is existing technology, and this application does not limit it.

[0096] In this application, the prediction results of the model on the test set are shown in Table 4.

[0097] Table 4

[0098]

[0099] Based on Table 4, the key performance indicators of the prediction model were calculated, and the results are as follows:

[0100] The mean absolute error (MAE) of the prediction model is 0.0093, meaning that the model's prediction of the numerical solution fE of the correction function differs from the true value by an average of only 0.0093; the coefficient of determination (R2) is 0.952, meaning that the model can explain 95.2% of the fluctuations in the correction value, indicating an extremely high good fit.

[0101] The above metrics demonstrate that, based on the provided 30 sets of data, the random forest model M_f trained in this application possesses the required accuracy and generalization ability for target prediction, fully meeting the requirements for predicting the cable insulation breakdown field strength correction function. After training and internal validation, the breakdown field strength correction value prediction model M_f is obtained. This model can output a predicted breakdown field strength correction value f based on the input feature parameter array. E .

[0102] In step S400, the predicted insulation breakdown field strength of the cable under test is determined based on the predicted correction value and the theoretical value of the insulation breakdown field strength. This includes: obtaining a calculation model for the insulation breakdown field strength correction value, which is constructed based on the equivalent relationship between the corrected insulation breakdown field strength of the insulation material and the ratio of the actual insulation breakdown field strength of the insulation material to the theoretical insulation breakdown field strength; using the predicted correction value of the insulation breakdown field strength as the actual insulation breakdown field strength correction value; obtaining the actual insulation breakdown field strength of the cable under test based on the predicted correction value and the theoretical value of the insulation breakdown field strength, and using the obtained actual insulation breakdown field strength as the predicted insulation breakdown field strength of the cable under test. Specifically, in this application, the breakdown field strength correction prediction model is the ratio between the actual value and the theoretical value of the breakdown field strength. Therefore, by multiplying M_p and M_f, we can obtain the cable insulation breakdown field strength prediction model M_c, that is: M_c=M_p×M_f.

[0103] The following specific experiment verifies the cable insulation breakdown field strength prediction model of this application:

[0104] Preparation of validation samples: Using the same process, model cables with insulation thicknesses different from those in the training set were extruded, for example, a model cable with a thickness of 9.0 mm. Ten duplicate samples were prepared and tested, and their actual insulation breakdown field strength (Eac-9) was measured. The same performance tests were performed on slices of these samples to obtain the feature parameter dataset X.

[0105] Prediction and Comparison: Using M_p and the measured thickness of each sample, the theoretical value of the insulation breakdown field strength Eth-9 for a 9mm insulated model cable is calculated. The dataset X is input into the trained model M_f to obtain the predicted and corrected value of the insulation breakdown field strength, which is then used as the corrected value f for the insulation breakdown field strength. E The predicted breakdown field strength was then calculated as Epr⁻⁹ = Eth⁻⁹ × f. E Then, the Epr-9 and Eac-9 models of 10 validation samples were compared. The calculated average relative error was 3.8%, and the maximum relative error was less than 6.5%. This error is much smaller than the ±10% deviation range, indicating that the accuracy of model M_c in predicting the breakdown field strength of real-type cable insulation has been validated.

[0106] The method of this application is illustrated below with a specific example:

[0107] Taking the application scenario of predicting the insulation breakdown field strength of a 500kV AC cable as an example, a 1m long sample is first cut from the 500kV AC cable. The insulation layer is sliced, and its characteristic parameter dataset is obtained: X500kV = [eccentricity: 5.1%, crystallinity: 48.2%, crosslinking degree: 75.5%, actual insulation breakdown field strength of the slice: 105.6kV / mm, resistivity: 5.6×10¹⁶Ω·m, byproduct content: 120ppm, thermal elongation: 55%]. Simultaneously, the average insulation thickness is accurately measured to be d500kV = 31.5mm.

[0108] Based on the calculations of model M_p:

[0109] Eth=M_p(d500kV)=81.2×(31.5) -0.266 ≈32.43kV / mm

[0110] Based on the predictions of model M_f:

[0111] f(X500kV)=0.92

[0112] As mentioned above, the model predicts that due to issues such as eccentricity and byproducts, the actual performance of this cable is approximately 92% of the ideal flat plate model. In summary, the predicted actual value of the insulation breakdown field strength of the 500kV AC cable is as follows:

[0113] Epr=M_c=Eth×0.92=32.43×0.92≈29.83kV / mm

[0114] As can be seen from the above, the insulation breakdown field strength of the 500kV AC cable is predicted to be approximately 29.83kV / mm using model M_c.

[0115] In summary, the prediction method proposed in this application considers the structural characteristics, physicochemical properties, and processing performance characteristics of a full-scale cable. Based on a combination of machine learning and experimental testing, it achieves non-destructive prediction of the cable insulation breakdown field strength. Therefore, it eliminates the need for insulation breakdown testing of the full-scale cable; only a series of performance tests on cable insulation samples are required. Based on the test results, the accurate prediction of the insulation breakdown field strength of the full-scale cable can be achieved. Compared with existing methods, this method has the advantages of saving time and avoiding the significant manpower and material costs associated with conducting full-scale cable insulation breakdown field strength tests.

[0116] like Figure 3 As shown, in a second aspect of this application, a cable insulation breakdown field strength prediction device is provided, comprising:

[0117] The data acquisition module is configured to acquire characteristic parameters of the cable under test;

[0118] The correction value prediction module is configured to take the feature parameters as input and obtain the predicted correction value of the insulation breakdown field strength of the cable under test through the breakdown field strength correction value prediction model. The breakdown field strength correction value prediction model is obtained by training a preset deep learning algorithm with the feature parameters of different cables and the insulation breakdown field strength correction value of each cable. The insulation breakdown field strength correction value of each cable is based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable.

[0119] The insulation breakdown field strength theoretical value calculation module is configured to obtain the insulation layer thickness of the cable under test, and determine the theoretical value of the insulation breakdown field strength of the cable under test under the insulation layer thickness through the insulation breakdown field strength prediction model. The insulation breakdown field strength prediction model is obtained by nonlinear fitting of the actual insulation breakdown field strength values ​​of insulation layer slice samples with different thicknesses.

[0120] The insulation breakdown field strength prediction module is configured to determine the predicted value of the insulation breakdown field strength of the cable under test based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength.

[0121] It is understood that those skilled in the art will clearly recognize that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0122] A third aspect of this application provides a cable insulation breakdown field strength prediction system, comprising:

[0123] An insulation material testing device is used to test at least one of the structural characteristics, physicochemical characteristics, and processing characteristics of a cable under test to obtain characteristic parameters of the cable under test; and

[0124] Such as the cable insulation breakdown field strength prediction device mentioned above.

[0125] In a fourth aspect, this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the cable insulation breakdown field strength prediction method as described above.

[0126] In a fifth aspect, this application provides a terminal device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the cable insulation breakdown field strength prediction method described above.

[0127] like Figure 4 The diagram shown is a schematic representation of a terminal device provided in an embodiment of this application. Figure 4 As shown, the terminal device 10 of this embodiment includes a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, it implements the steps in the above method embodiments. Alternatively, when the processor 100 executes the computer program 102, it implements the functions of each module / unit in the above device embodiments.

[0128] For example, computer program 102 may be divided into one or more modules / units, one or more of which are stored in memory 101 and executed by processor 100 to complete this application. One or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 102 in terminal device 10.

[0129] Terminal device 10 may be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. Terminal device 10 may include, but is not limited to, a processor 100 and a memory 101. Those skilled in the art will understand that... Figure 4 This is merely an example of terminal device 10 and does not constitute a limitation on terminal device 10. It may include more or fewer components than shown, or combine certain components, or different components. For example, terminal device may also include input / output devices, network access devices, buses, etc.

[0130] Processor 100 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0131] The memory 101 can be an internal storage unit of the terminal device 10, such as a hard disk or RAM of the terminal device 10. The memory 101 can also be an external storage device of the terminal device 10, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard equipped on the terminal device 10. Furthermore, the memory 101 can include both internal and external storage units of the terminal device 10. The memory 101 is used to store computer programs and other programs and data required by the terminal device 10. The memory 101 can also be used to temporarily store data that has been output or will be output.

[0132] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0134] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for predicting the breakdown field strength of cable insulation, characterized in that, include: Obtain the characteristic parameters of the cable under test; Using the aforementioned feature parameters as input, the insulation breakdown field strength prediction correction value of the cable under test is obtained through the breakdown field strength correction value prediction model. The breakdown field strength correction value prediction model is obtained by training a preset deep learning algorithm with the feature parameters of different cables and the insulation breakdown field strength correction value of each cable. The insulation breakdown field strength correction value of each cable is obtained based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable. The insulation layer thickness of the cable under test is obtained, and the theoretical value of the insulation breakdown field strength of the cable under test is determined by the insulation breakdown field strength prediction model. The insulation breakdown field strength prediction model is obtained by nonlinear fitting of the actual values ​​of insulation breakdown field strength of insulation layer slices with different thicknesses. The predicted insulation breakdown field strength of the cable under test is determined based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength.

2. The method for predicting the breakdown field strength of cable insulation according to claim 1, characterized in that, The characteristic parameters include at least one of the structural characteristics, physicochemical properties, and processing characteristics of the cable under test; The structural features include: insulation eccentricity; The physicochemical properties include at least one of the following: crystallinity, degree of crosslinking, insulation breakdown field strength of the insulating layer slice sample, and volume resistivity. The processing technology features include at least one of the following: by-product content and thermal elongation. The insulation layer slice sample is made of the same material as the cable under test, and the insulation breakdown field strength of the insulation layer slice sample is obtained by performing an insulation breakdown field strength test on the pre-prepared insulation layer slice sample.

3. The method for predicting the breakdown field strength of cable insulation according to claim 1, characterized in that, The correction values ​​for the insulation breakdown field strength of each cable are obtained based on the actual values ​​and theoretical values ​​of the insulation breakdown field strength of each cable, including: The actual insulation breakdown field strength of each cable was obtained by insulation breakdown field strength test. The insulation layer thickness of each cable is obtained. Using the insulation layer thickness of each cable as input, the theoretical value of the insulation breakdown field strength of each cable is output through the insulation breakdown field strength prediction model. For each cable, the actual value and theoretical value of the insulation breakdown field strength of the current cable are used as inputs, and the corrected value of the insulation breakdown field strength of the current cable is obtained based on the insulation breakdown field strength correction value calculation model. The calculation model for the insulation breakdown field strength correction value is constructed based on the equivalent relationship between the insulation breakdown field strength correction value of the insulating material and the ratio of the actual insulation breakdown field strength value and the theoretical insulation breakdown field strength value of the insulating material.

4. The method for predicting the breakdown field strength of cable insulation according to claim 1, characterized in that, The fitting process of the insulation breakdown field strength prediction model includes: Multiple insulation layer slices of different thicknesses are prepared in advance, and the material of the insulation layer slices is the same as that of the cable to be tested. Each insulation layer slice sample was placed in a test environment at the preset cable insulation operating temperature; For each insulation layer slice sample, the insulation breakdown field strength test is performed on the current insulation layer slice sample by continuously increasing the voltage, so as to obtain the actual value of the insulation breakdown field strength of the current insulation layer slice sample under different breakdown voltages. Based on the actual values ​​of insulation breakdown field strength of each insulation layer slice sample under different breakdown voltages, the pre-constructed functional relationship between insulation breakdown field strength and insulation layer thickness is nonlinearly fitted to obtain the insulation breakdown field strength prediction model. The functional relationship between the insulation breakdown field strength and the insulation layer thickness is used to represent the mapping relationship in which the insulation breakdown field strength of the insulation layer decreases exponentially with the increase of its thickness.

5. The method for predicting the breakdown field strength of cable insulation according to claim 1, characterized in that, The preset deep learning algorithm includes the random forest algorithm, and the training steps of the breakdown field strength correction value prediction model include: Multiple cables with different insulation layer thicknesses were prepared using the same material as the cable under test; Each cable obtained is cut into multiple cable samples of equal length; For each cable sample, obtain the characteristic parameters of the current cable sample and determine the insulation breakdown field strength correction value for each cable sample. Using the characteristic parameters of each cable sample as input, and the insulation breakdown field strength correction value of the corresponding cable sample as the label vector, the random forest algorithm is trained to obtain the breakdown field strength correction value prediction model.

6. The method for predicting the breakdown field strength of cable insulation according to claim 1, characterized in that, The predicted insulation breakdown field strength of the cable under test is determined based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength, including: An insulation breakdown field strength correction value calculation model is obtained, which is constructed based on the equivalent relationship between the insulation breakdown field strength correction value of the insulating material and the ratio of the actual insulation breakdown field strength and the theoretical insulation breakdown field strength of the insulating material; The insulation breakdown field strength prediction correction value is used as the insulation breakdown field strength correction value. Based on the insulation breakdown field strength prediction correction value and the insulation breakdown field strength theoretical value, the actual insulation breakdown field strength of the cable under test is obtained based on the insulation breakdown field strength correction value calculation model. The obtained actual insulation breakdown field strength is used as the insulation breakdown field strength prediction value of the cable under test.

7. A device for predicting the breakdown field strength of cable insulation, characterized in that, include: The data acquisition module is configured to acquire characteristic parameters of the cable under test; The correction value prediction module is configured to take the feature parameters as input and obtain the predicted correction value of the insulation breakdown field strength of the cable under test through the breakdown field strength correction value prediction model. The breakdown field strength correction value prediction model is obtained by training a preset deep learning algorithm with the feature parameters of different cables and the insulation breakdown field strength correction value of each cable. The insulation breakdown field strength correction value of each cable is obtained based on the actual value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength of each cable. The insulation breakdown field strength theoretical value calculation module is configured to obtain the insulation layer thickness of the cable under test, and determine the theoretical value of the insulation breakdown field strength of the cable under test under the insulation layer thickness through the insulation breakdown field strength prediction model. The insulation breakdown field strength prediction model is obtained by nonlinear fitting of the actual insulation breakdown field strength values ​​of insulation layer slice samples with different thicknesses. The insulation breakdown field strength prediction module is configured to determine the predicted value of the insulation breakdown field strength of the cable under test based on the predicted correction value of the insulation breakdown field strength and the theoretical value of the insulation breakdown field strength.

8. A cable insulation breakdown field strength prediction system, characterized in that, include: An insulation material testing device is used to test at least one of the structural characteristics, physicochemical characteristics, and processing characteristics of a cable under test to obtain characteristic parameters of the cable under test; and The cable insulation breakdown field strength prediction device as described in claim 7.

9. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the cable insulation breakdown field strength prediction method as described in any one of claims 1-6.

10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the cable insulation breakdown field strength prediction method as described in any one of claims 1-6.

Images