Method and device for analyzing the aging state of crosslinked polyethylene insulation material

Abstract

The application discloses a kind of crosslinking polyethylene insulation material aging state analysis method and device, it is related to electrical engineering discipline high voltage and insulation technical field, wherein the method includes: respectively measuring the parameter value of multiple preset characteristic parameters of the XLPE insulation material to be analyzed;Multiple preset characteristic parameter values are input into aging degree prediction model, determine the aging time interval corresponding to the parameter value of multiple preset characteristic parameters of the XLPE insulation material to be analyzed, and the aging degree prediction model is obtained by training BP neural network model according to the sample parameter value of multiple preset characteristic parameters of XLPE insulation material sample in different aging state;The parameter value of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and corresponding aging time interval are analyzed by fuzzy clustering algorithm, and the aging state of the XLPE insulation material to be analyzed is determined.The application can accurately and comprehensively analyze the aging state of XLPE insulation material.

Description

Technical Field

[0001] This invention relates to the field of high voltage and insulation technology in electrical engineering, and in particular to a method and apparatus for analyzing the aging state of cross-linked polyethylene insulation materials. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] Cross-linked polyethylene (XLPE) is the primary insulation material for high-voltage power cables, offering advantages such as light weight, good heat resistance, and excellent electrical properties. XLPE cables are typically designed for a lifespan of 20 to 30 years. However, these cables are susceptible to adverse factors such as poor heat dissipation, dense installation, and excessive current carrying capacity, leading to temperature increases and accelerated insulation aging. Consequently, cable accidents are becoming increasingly frequent. Furthermore, compared to faults in overhead transmission lines, fires caused by overheating, combustion, or explosions in power cables spread rapidly, easily triggering secondary electrical fires and explosions, resulting in large-scale power outages, prolonged power restoration times, and substantial economic losses. Therefore, studying the aging state of XLPE insulation materials is of great practical significance.

[0004] Currently, the analysis of the aging state of XLPE insulation materials is carried out through rapid measurement and analysis using a single characteristic parameter. The parameter only focuses on a certain aspect of the XLPE insulation material and cannot comprehensively and accurately reflect the aging state of the XLPE insulation material. Summary of the Invention

[0005] This invention provides a method for analyzing the aging state of cross-linked polyethylene insulation materials, which analyzes the aging state of cross-linked polyethylene insulation materials based on multiple characteristics, making the analysis results more accurate and comprehensive. The method includes:

[0006] The parameter values ​​of several preset characteristic parameters of the cross-linked polyethylene XLPE insulation material to be analyzed were measured respectively.

[0007] The parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed are input into the aging degree prediction model to determine the aging time interval corresponding to the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states.

[0008] By using a fuzzy clustering algorithm, the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals are clustered to determine the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0009] This invention also provides an aging state analysis device for cross-linked polyethylene insulation materials, used to analyze the aging state of cross-linked polyethylene insulation materials based on multiple characteristics, making the analysis results more accurate and comprehensive. The device includes:

[0010] The parameter measurement module is used to measure the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene XLPE insulation material to be analyzed.

[0011] The aging time interval prediction module is used to input the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed into the aging degree prediction model to determine the aging time interval corresponding to the parameter values ​​of the multiple preset feature parameters of the XLPE insulation material to be analyzed. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states.

[0012] The aging state analysis module is used to perform cluster analysis on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and the corresponding aging time intervals using a fuzzy clustering algorithm, thereby determining the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0013] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for analyzing the aging state of cross-linked polyethylene insulation materials.

[0014] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for analyzing the aging state of cross-linked polyethylene insulation materials.

[0015] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described method for analyzing the aging state of cross-linked polyethylene insulation materials.

[0016] In this embodiment of the invention, the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene (XLPE) insulation material to be analyzed are measured respectively. These parameter values ​​are then input into an aging degree prediction model to determine the aging time intervals corresponding to the parameter values. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset characteristic parameters of XLPE insulation material samples in different aging states. A fuzzy clustering algorithm is used to perform cluster analysis on the parameter values ​​of the multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals to determine the aging state of the cross-linked polyethylene insulation material. Compared with existing technical solutions that analyze the aging state of XLPE insulation materials using a single characteristic parameter, this embodiment of the invention analyzes and predicts the aging degree corresponding to the multi-dimensional characteristic parameters of XLPE insulation materials using a BP neural network model. This considers the aging state of XLPE insulation materials from multiple perspectives, and the fuzzy clustering method clusters the multi-dimensional characteristic parameters and predicted aging time intervals to determine the aging state of the XLPE insulation material, making the analysis results more accurate and comprehensive. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0018] Figure 1 This is a flowchart of an aging state analysis method for cross-linked polyethylene insulation material provided in an embodiment of the present invention;

[0019] Figure 2 This is a flowchart of a model training and testing method provided in an embodiment of the present invention;

[0020] Figure 3 This is a schematic diagram of the infrared spectral test results of XLPE insulation material samples under different aging states provided in the embodiments of the present invention;

[0021] Figure 4 This is a graph showing the relationship between the carbonyl index and aging time of the XLPE insulation material sample provided in this embodiment of the invention.

[0022] Figure 5 This is a graph showing the relationship between the elongation at break and aging time of the XLPE insulation material sample provided in the embodiments of the present invention.

[0023] Figure 6This is a graph showing the relationship between the dielectric loss tangent and frequency of XLPE insulation material samples under different aging states provided in this embodiment of the invention.

[0024] Figure 7 This is a graph showing the relationship between the Welbull distribution probability and the electric field strength of XLPE insulation material samples under different aging states provided in this embodiment of the invention.

[0025] Figure 8 A flowchart of another method for analyzing the aging state of cross-linked polyethylene insulation material provided in an embodiment of the present invention;

[0026] Figure 9 This is a schematic diagram of an aging state analysis device for cross-linked polyethylene insulation material provided in an embodiment of the present invention;

[0027] Figure 10 This is a schematic diagram of another cross-linked polyethylene insulation material aging state analysis device provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] In the description of this specification, the terms "comprising," "including," "having," and "containing" are open-ended terms, meaning that they include but are not limited to. The terms "an embodiment," "a specific embodiment," "some embodiments," and "for example," etc., refer to specific features, structures, or characteristics described in connection with that embodiment or example that are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. The order of steps involved in the various embodiments is used to illustrate the implementation of this application, and the order of steps is not limited and can be adjusted appropriately as needed.

[0030] Currently, methods for analyzing the aging state of cross-linked polyethylene (XLPE) using only a single characteristic parameter, whether it's the mechanical properties specified in national standards or the widely discussed thermal properties and dielectric response parameters. These parameters focus on only one aspect of the characteristic and cannot comprehensively and accurately reflect the aging state of the insulating material. Therefore, this invention provides a method for analyzing the aging state of XLPE insulating materials, analyzing its aging state based on multiple characteristics, resulting in more accurate and comprehensive analysis results.

[0031] like Figure 1 The diagram shows a flowchart of an aging state analysis method for cross-linked polyethylene insulation material provided by an embodiment of the present invention. The method includes the following steps:

[0032] Step 101: Measure the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene (XLPE) insulation material to be analyzed.

[0033] Step 102: Input the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed into the aging degree prediction model to determine the aging time interval corresponding to the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states.

[0034] Step 103: Using a fuzzy clustering algorithm, cluster analysis is performed on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals to determine the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0035] In this embodiment of the invention, the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene (XLPE) insulation material to be analyzed are measured respectively. These parameter values ​​are then input into an aging degree prediction model to determine the aging time intervals corresponding to the parameter values. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset characteristic parameters of XLPE insulation material samples in different aging states. A fuzzy clustering algorithm is used to perform cluster analysis on the parameter values ​​of the multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals to determine the aging state of the cross-linked polyethylene insulation material. Compared with existing technical solutions that analyze the aging state of XLPE insulation materials using a single characteristic parameter, this embodiment of the invention analyzes and predicts the aging degree corresponding to the multi-dimensional characteristic parameters of XLPE insulation materials using a BP neural network model. This considers the aging state of XLPE insulation materials from multiple perspectives, and the fuzzy clustering method clusters the multi-dimensional characteristic parameters and predicted aging time intervals to determine the aging state of the XLPE insulation material, making the analysis results more accurate and comprehensive.

[0036] In step 101 above, the parameter values ​​of several preset characteristic parameters of the XLPE insulation material to be analyzed are measured respectively.

[0037] The aforementioned preset characteristic parameters may include carbonyl index, AC breakdown strength, elongation at break, and low-frequency dielectric loss tangent. Specifically, the carbonyl index is a characteristic parameter used to indicate the chemical properties of XLPE insulation material; the elongation at break is a characteristic parameter used to indicate the mechanical properties of XLPE insulation material; the AC breakdown strength and the low-frequency dielectric loss tangent are characteristic parameters used to indicate the dielectric properties of XLPE insulation material, and the low-frequency dielectric loss tangent can be the dielectric loss tangent at a frequency of 0.1 Hz.

[0038] In practice, the aforementioned carbonyl index can be obtained by performing infrared spectroscopy tests on the XLPE insulating material to be analyzed and calculating based on the infrared spectroscopy results. For example, the carbonyl index in this embodiment of the invention can be defined as a wavenumber of 1720 cm⁻¹. -1 Carbonyl absorption peak intensity and wavenumber at 2010 cm⁻¹ -1 The ratio of the absorption peak intensity; the above-mentioned elongation at break, AC breakdown strength and low-frequency dielectric loss tangent can be measured using methods commonly used in the field, and no specific limitations are imposed here.

[0039] It should be noted that the preset characteristic parameters in the embodiments of the present invention can also be set according to specific application scenarios, and are not limited to the carbonyl index, AC breakdown strength, elongation at break and low frequency dielectric loss tangent as described in the embodiments of the present invention.

[0040] In step 102 above, the parameter values ​​of multiple preset characteristic parameters are input into the aging degree prediction model, which can determine the aging time interval corresponding to the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed.

[0041] The aforementioned aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states.

[0042] like Figure 2 The diagram shown is a flowchart of a model training and testing method provided by an embodiment of the present invention, which may include the following steps:

[0043] Step 201: Obtain sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states as sample data, and construct training set and test set. Each sample parameter value in the sample data is marked with a sample aging time interval.

[0044] In practice, firstly, multiple XLPE insulation material samples can be subjected to thermal aging tests. The temperature can be set to 130℃, and the time can be set to 0 days, 14 days, 28 days, 42 days, 56 days, and 70 days, respectively, to obtain XLPE insulation material samples that are unaged, aged for 14 days, aged for 28 days, aged for 42 days, aged for 56 days, and aged for 70 days. The appearance, color, and other characteristics of the XLPE insulation material samples differ depending on the aging state.

[0045] Then, the carbonyl index, AC breakdown strength, elongation at break and low-frequency dielectric loss tangent of XLPE insulation material samples under different aging states were measured respectively, and each sample parameter value was marked with the sample aging time interval.

[0046] For example, XLPE insulation material samples in different aging states have different molecular structures and contents. Under infrared light irradiation with similar vibrational frequencies, different chemical bonds in the molecules undergo vibrational absorption at different frequencies and appear at different positions on the infrared spectrum. Figure 3 The image shows a schematic diagram of the infrared spectral test results for XLPE insulation material samples under different aging conditions. Figure 3 The carbonyl index of XLPE insulation material samples under different aging states can be calculated from the infrared spectroscopy test results. The carbonyl index can be represented by a wavenumber of 1720 cm⁻¹. -1 Carbonyl absorption peak intensity and wavenumber at 2010 cm⁻¹ -1 The ratio of the absorption peak intensities at each location. For example... Figure 4 The figure shown is a graph depicting the relationship between the carbonyl index and aging time of XLPE insulation material samples. Figure 4 It can be seen that the carbonyl index of the XLPE insulation material sample increases with the increase of aging time.

[0047] For example, Figure 5 The graph showing the relationship between elongation at break and aging time of XLPE insulation material samples is presented. Figure 5 It can be seen that the elongation at break of XLPE insulation material samples decreases with the increase of aging time. According to the national standard GB / Z18890.2, an elongation at break greater than or equal to 500% can be regarded as the good performance stage of XLPE insulation material samples; when the elongation at break decreases by 50%, it is considered that the XLPE insulation material has reached the end of its service life.

[0048] For example, Figure 6 The graph shows the relationship between the dielectric loss tangent and frequency for XLPE insulation material samples under different aging conditions. Figure 6 We can obtain the relationship between the dielectric loss tangent and the aging time at a frequency of 0.1Hz.

[0049] For example, the Weibull distribution can be used to analyze the reliability of insulating materials. Figure 7 This is a graph showing the relationship between the Welbull distribution probability and the electric field strength (i.e., AC breakdown strength) of XLPE insulation material samples under different aging conditions. Figure 7 The Welbull distribution probabilities for different electric field intensities shown can be used to obtain the relationship between AC breakdown strength and aging time for XLPE insulation material samples under different aging conditions.

[0050] The measured sample parameter values ​​are normalized and used as sample data. The sample size is divided according to a preset ratio to construct training and test sets.

[0051] For example, a set of sample data could be (carbonyl index, 1.4, 14-28), where 1.4 is the reference value for the carbonyl index, and 14-28 is the aging time range corresponding to a carbonyl index reference value of 1.4.

[0052] Step 202: Train the BP neural network model using the training set to obtain the aging degree prediction model.

[0053] In one embodiment of the present invention, before step 202 above, it is necessary to construct an initial BP neural network model, and then use the training set to train the initial BP neural network model to obtain an aging degree prediction model.

[0054] Specifically, constructing an initial BP neural network model may include the following steps:

[0055] The initial BP neural network model is set to have 4 layers.

[0056] The number of neurons in each layer was set to 4, 32, 16, and 4, respectively;

[0057] Set the initial weights, bias matrix, activation function, learning rate, and number of iterations.

[0058] In practical implementation, considering the balance between accuracy and computational load, the number of layers in the initial BP neural network model in this embodiment can be set to 4 layers, with the number of neurons in each layer set to 4, 32, 16 and 4 respectively. The initial weights and bias matrices can be set by random assignment. The learning rate determines the speed of weight updates, and can be initially set to 0.15. The number of iterations can be set to 500.

[0059] It should be noted that the initialization of the BP neural network model can be set according to the specific application scenario in the embodiments of the present invention.

[0060] In this embodiment of the invention, the BP neural network model is initialized using a training set. The weights and bias matrices can be gradually adjusted using gradient descent to reduce the error. Specifically, the output and error functions can be calculated based on the initial weights, bias matrices, and the set activation function. The gradient of the error function with respect to each weight and bias matrix is ​​calculated, and then the weights and bias matrices are updated according to the learning rate. This process is repeated iteratively until the required number of iterations is reached to obtain the aging degree prediction model.

[0061] Step 203: Test the aging prediction model using the test set.

[0062] Thus, in this embodiment of the invention, an initial BP neural network model is constructed, and the parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states are used as sample data to train the initial BP neural network model, so that the prediction results of the final aging degree prediction model are more accurate.

[0063] In step 103 above, a fuzzy clustering algorithm can be used to perform cluster analysis on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and the corresponding aging time intervals predicted by the aging degree prediction model, so as to determine the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0064] In one embodiment of the present invention, as Figure 8 As shown, step 103 above may specifically include the following steps:

[0065] Step 801: Determine the number of clusters for fuzzy clustering based on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and the corresponding aging time intervals.

[0066] Step 802: Set weights for the parameter values ​​of each preset characteristic parameter of the XLPE insulation material to be analyzed and the corresponding aging time interval, and generate an initial membership matrix;

[0067] Step 803: Based on the initial membership matrix and the number of clusters, iteratively calculate the cluster centers and membership matrix until the change in the membership matrix is ​​less than the preset error threshold, or the number of iterations reaches the preset number of iterations, and output the final membership matrix.

[0068] Step 804: Determine the aging state of the cross-linked polyethylene insulation material to be analyzed based on the final membership matrix.

[0069] In this embodiment of the invention, the fuzzy clustering algorithm can be the fuzzy c-means clustering algorithm (FCM).

[0070] In specific implementation, in step 801 above, the aging state can be defined into four types based on the aging time interval: no aging, light aging, moderate aging, and severe aging. Therefore, the number of clusters mentioned above can be four. For example, the aging state with an aging time interval of 1 to 42 days can be defined as light aging; the aging state with an aging time interval of 42 to 56 days can be defined as moderate aging; and the aging state with an aging time interval of 56 to 70 days can be defined as severe aging.

[0071] In step 802 above, weights are assigned to the parameter values ​​and corresponding aging time intervals of each preset characteristic parameter of the XLPE insulation material to be analyzed. Random assignment can be used, and the weights represent the degree to which the parameter value and corresponding aging time interval of the preset characteristic parameter belong to a certain cluster. Based on the set weights, an initial membership matrix is ​​generated to satisfy the constraint condition, i.e., Formula 1 below, which indicates that the sum of the membership degrees of each preset characteristic parameter value and its corresponding aging time interval belonging to each cluster is 1.

[0072]

[0073] In step 803 above, after determining the initial membership matrix and the number of clusters, the objective function of the FCM clustering algorithm is defined as shown in Formula 2:

[0074]

[0075] Where m represents the number of clusters; x i This represents the parameter value of the i-th preset feature parameter and the corresponding aging time interval; c represents the membership degree of the i-th preset feature parameter value and its corresponding aging time interval to class j; j This represents the cluster center of class j.

[0076] The membership matrix and cluster centers are calculated iteratively using Formulas 3 and 4 below. Then, the calculated membership matrix and cluster centers are substituted into Formula 2 to make the objective function smaller and smaller, thus achieving the effect of clustering.

[0077]

[0078]

[0079] The iteration ends when the change in the membership matrix is ​​less than the preset error threshold or the number of iterations reaches the preset number of iterations, and the final membership matrix is ​​output. The final membership matrix can be shown in Formula 5 below.

[0080]

[0081] In step 804 above, the aging state of the cross-linked polyethylene insulation material to be analyzed can be determined based on the final membership matrix, which may specifically include:

[0082] The degree of aging corresponding to the maximum membership degree in the final membership matrix is ​​determined as the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0083] Thus, in this embodiment of the invention, the aging state of cross-linked polyethylene insulation material is classified and determined by fuzzy clustering algorithm, which can greatly improve the accuracy of analyzing the aging state of XLPE insulation.

[0084] This invention also provides an aging state analysis device for cross-linked polyethylene insulation materials, as described in the following embodiments. Since the principle by which this device solves the problem is similar to the aging state analysis method for cross-linked polyethylene insulation materials, the implementation of this device can refer to the implementation of the aging state analysis method for cross-linked polyethylene insulation materials; repeated details will not be elaborated further.

[0085] like Figure 9 The diagram shown is a schematic diagram of an aging state analysis device for cross-linked polyethylene insulation material provided in an embodiment of the present invention. The device includes:

[0086] The parameter measurement module 901 is used to measure the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene XLPE insulation material to be analyzed.

[0087] The aging time interval prediction module 902 is used to input the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed into the aging degree prediction model, and determine the aging time interval corresponding to the parameter values ​​of the multiple preset feature parameters of the XLPE insulation material to be analyzed. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states.

[0088] The aging state analysis module 903 is used to perform cluster analysis on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and the corresponding aging time intervals through fuzzy clustering algorithm, so as to determine the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0089] In one embodiment of the present invention, the plurality of preset characteristic parameters include carbonyl index, AC breakdown strength, elongation at break and low-frequency dielectric loss tangent.

[0090] In one embodiment of the present invention, as Figure 10As shown, it also includes a model training and testing module 1001, used to input the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed into the aging degree prediction model, and determine the aging time intervals corresponding to the parameter values ​​of the multiple preset characteristic parameters of the XLPE insulation material to be analyzed:

[0091] The sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples under different aging states are obtained as sample data to construct training sets and test sets. Each sample parameter value in the sample data is marked with a sample aging time interval.

[0092] A backpropagation neural network model is trained using the training set to obtain an aging prediction model;

[0093] The aging prediction model was tested using a test set.

[0094] In one embodiment of the present invention, the model training and testing module 1001 is further configured to:

[0095] Constructing the initial BP neural network model includes:

[0096] The initial BP neural network model is set to have 4 layers.

[0097] The number of neurons in each layer was set to 4, 32, 16, and 4, respectively;

[0098] Set the initial weights, bias matrix, activation function, learning rate, and number of iterations.

[0099] In one embodiment of the present invention, the aging state analysis module 903 is specifically used for:

[0100] Based on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and the corresponding aging time intervals, the number of clusters for fuzzy clustering is determined.

[0101] Set weights for the parameter values ​​of each preset characteristic parameter of the XLPE insulation material to be analyzed and the corresponding aging time intervals, and generate an initial membership matrix;

[0102] Based on the initial membership matrix and the number of clusters, the cluster centers and membership matrix are iteratively calculated until the change in the membership matrix is ​​less than the preset error threshold, or the number of iterations reaches the preset number of iterations, and the final membership matrix is ​​output.

[0103] Based on the final membership matrix, the aging state of the cross-linked polyethylene insulation material to be analyzed is determined.

[0104] In one embodiment of the present invention, the aging state analysis module 903 is specifically used for:

[0105] The degree of aging corresponding to the maximum membership degree in the final membership matrix is ​​determined as the aging state of the cross-linked polyethylene insulation material to be analyzed.

[0106] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for analyzing the aging state of cross-linked polyethylene insulation materials.

[0107] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for analyzing the aging state of cross-linked polyethylene insulation materials.

[0108] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described method for analyzing the aging state of cross-linked polyethylene insulation materials.

[0109] In this embodiment of the invention, the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene (XLPE) insulation material to be analyzed are measured respectively. These parameter values ​​are then input into an aging degree prediction model to determine the aging time intervals corresponding to the parameter values. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset characteristic parameters of XLPE insulation material samples in different aging states. A fuzzy clustering algorithm is used to perform cluster analysis on the parameter values ​​of the multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals to determine the aging state of the cross-linked polyethylene insulation material. Compared with existing technical solutions that analyze the aging state of XLPE insulation materials using a single characteristic parameter, this embodiment of the invention analyzes and predicts the aging degree corresponding to the multi-dimensional characteristic parameters of XLPE insulation materials using a BP neural network model. This considers the aging state of XLPE insulation materials from multiple perspectives, and the fuzzy clustering method clusters the multi-dimensional characteristic parameters and predicted aging time intervals to determine the aging state of the XLPE insulation material, making the analysis results more accurate and comprehensive.

[0110] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention 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.

[0111] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0112] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0113] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0114] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for analyzing the aging state of cross-linked polyethylene insulation materials, characterized in that, include: The parameter values ​​of several preset characteristic parameters of the cross-linked polyethylene XLPE insulation material to be analyzed were measured respectively. The parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed are input into the aging degree prediction model to determine the aging time interval corresponding to the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states. By using a fuzzy clustering algorithm, the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals are clustered to determine the aging state of the cross-linked polyethylene insulation material to be analyzed. The preset characteristic parameters include carbonyl index, AC breakdown strength, elongation at break, and low-frequency dielectric loss tangent. Using a fuzzy clustering algorithm, cluster analysis is performed on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals to determine the aging state of the cross-linked polyethylene insulation material to be analyzed. This includes: determining the number of clusters in the fuzzy clustering based on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals. Weights are assigned to the parameter values ​​of each preset characteristic parameter of the XLPE insulation material to be analyzed and the corresponding aging time intervals, generating an initial membership matrix. Based on the initial membership matrix and the number of clusters, the cluster centers and membership matrix are iteratively calculated until the change in the membership matrix is ​​less than the preset error threshold, or the number of iterations reaches the preset number of iterations, and the final membership matrix is ​​output. Based on the final membership matrix, the aging state of the cross-linked polyethylene insulation material to be analyzed is determined.

2. The method as described in claim 1, characterized in that, Before inputting the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed into the aging degree prediction model, and determining the aging time intervals corresponding to the parameter values ​​of the multiple preset characteristic parameters of the XLPE insulation material to be analyzed, the following steps are also included: The sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples under different aging states are obtained as sample data to construct training sets and test sets. Each sample parameter value in the sample data is marked with a sample aging time interval. A backpropagation neural network model is trained using the training set to obtain an aging prediction model; The aging prediction model was tested using a test set.

3. The method as described in claim 2, characterized in that, Before training a BP neural network model using the training set to obtain an aging prediction model, the following steps are also included: Constructing the initial BP neural network model includes: The initial BP neural network model is set to have 4 layers. The number of neurons in each layer was set to 4, 32, 16, and 4, respectively; Set the initial weights, bias matrix, activation function, learning rate, and number of iterations.

4. The method as described in claim 1, characterized in that, Based on the final membership matrix, the aging state of the cross-linked polyethylene insulation material to be analyzed is determined, including: The degree of aging corresponding to the maximum membership degree in the final membership matrix is ​​determined as the aging state of the cross-linked polyethylene insulation material to be analyzed.

5. An aging state analysis device for cross-linked polyethylene insulation materials, characterized in that, include: The parameter measurement module is used to measure the parameter values ​​of multiple preset characteristic parameters of the cross-linked polyethylene XLPE insulation material to be analyzed. The aging time interval prediction module is used to input the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed into the aging degree prediction model to determine the aging time interval corresponding to the parameter values ​​of the multiple preset feature parameters of the XLPE insulation material to be analyzed. The aging degree prediction model is obtained by training a BP neural network model based on the sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples in different aging states. The aging state analysis module is used to perform cluster analysis on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and the corresponding aging time intervals through fuzzy clustering algorithm, so as to determine the aging state of the cross-linked polyethylene insulation material to be analyzed. The preset characteristic parameters include carbonyl index, AC breakdown strength, elongation at break, and low-frequency dielectric loss tangent. The aging state analysis module is specifically used for: determining the number of clusters in fuzzy clustering based on the parameter values ​​of multiple preset characteristic parameters of the XLPE insulation material to be analyzed and their corresponding aging time intervals; setting weights for the parameter values ​​of each preset characteristic parameter of the XLPE insulation material to be analyzed and their corresponding aging time intervals, and generating an initial membership matrix; iteratively calculating the cluster centers and membership matrix based on the initial membership matrix and the number of clusters until the change in the membership matrix is ​​less than a preset error threshold, or the number of iterations reaches a preset number of iterations, and outputting the final membership matrix; and determining the aging state of the cross-linked polyethylene insulation material to be analyzed based on the final membership matrix.

6. The apparatus as claimed in claim 5, characterized in that, It also includes a model training and testing module, used to input the parameter values ​​of multiple preset feature parameters of the XLPE insulation material to be analyzed into the aging degree prediction model before determining the aging time interval corresponding to the parameter values ​​of the multiple preset feature parameters of the XLPE insulation material to be analyzed: The sample parameter values ​​of multiple preset feature parameters of XLPE insulation material samples under different aging states are obtained as sample data to construct training sets and test sets. Each sample parameter value in the sample data is marked with a sample aging time interval. A backpropagation neural network model is trained using the training set to obtain an aging prediction model; The aging prediction model was tested using a test set.

7. The apparatus as claimed in claim 6, characterized in that, The model training and testing module is also used for: Constructing the initial BP neural network model includes: The initial BP neural network model is set to have 4 layers. The number of neurons in each layer was set to 4, 32, 16, and 4, respectively; Set the initial weights, bias matrix, activation function, learning rate, and number of iterations.

8. The apparatus as claimed in claim 5, characterized in that, The aging status analysis module is specifically used for: The degree of aging corresponding to the maximum membership degree in the final membership matrix is ​​determined as the aging state of the cross-linked polyethylene insulation material to be analyzed.

9. A computer 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 method of any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 4.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 4.

Images