Grounding grid fault detection method, device, storage medium and computer program

Abstract

The application provides a grounding grid fault detection method, device, storage medium and computer program, which comprises the following steps: determining a regional grey correlation degree based on ground surface potential data above cross nodes of a grounding grid, and detecting whether the regional grey correlation degree is less than a corresponding correlation degree threshold; determining an abnormal region when the regional grey correlation degree is less than the correlation degree threshold; determining an abnormal grey correlation degree based on ground surface potential data above a middle part of a conductor of the abnormal region, and detecting whether the abnormal grey correlation degree is less than a corresponding correlation degree threshold; determining an abnormal part when the abnormal grey correlation degree is less than the correlation degree threshold; determining a state recognition result by using a state recognition model based on ground surface potential data and ground surface magnetic induction intensity of the abnormal part, and determining a health state of the grounding grid based on the state recognition result. The abnormal state of the grounding grid is effectively and rapidly recognized, and the fault region is accurately positioned, so as to provide an important reference basis for on-site operation and maintenance of a transformer substation.

Description

Technical Field

[0001] This disclosure relates to the field of power electrical equipment maintenance technology, and in particular to a grounding grid fault detection method, device, storage medium, and computer program. Background Technology

[0002] In my country, galvanized flat steel is generally used as the grounding conductor in the grounding grid of traction substations. The traction power supply system uses the rails and the earth as the main return path. During normal operation, a large traction ground return current is collected and flows back to the traction substation through the grounding grid.

[0003] In some technologies, existing methods for detecting grounding grid faults in power grid substations typically include the electrical network method and the electrochemical method. The electrical network method uses DC excitation injected into the grounding grid, but its diagnostic effect is not significant for the AC return current flowing into the grounding grid under normal operating conditions in traction substations. The electrochemical method studies the corrosion mechanism of electrochemical corrosion, but its scope of application is small, making it difficult to accurately identify the operating status of the grounding grid. Summary of the Invention

[0004] This disclosure provides a grounding grid fault detection method, device, storage medium, and computer program to effectively and quickly identify abnormal states of substation grounding grids and accurately locate fault areas, thereby providing important reference for substation on-site operation and maintenance.

[0005] In a first aspect, this disclosure provides a grounding grid fault detection method, including:

[0006] Acquire surface data of the test condition in the grounding grid, wherein the surface data includes surface potential data and surface magnetic induction intensity;

[0007] Based on the surface potential data above the intersection nodes of the grounding grid, the gray relational degree algorithm is used to determine the regional gray relational degree between the surface potential data and the reference value of the surface potential data, and to detect whether the regional gray relational degree is less than the corresponding relational degree threshold.

[0008] When the gray correlation degree of the region is less than the corresponding correlation degree threshold, the region in the grounding grid whose gray correlation degree is less than the corresponding correlation degree threshold is determined to be an abnormal region.

[0009] Based on the surface potential data above the middle part of the conductor in the abnormal region, the abnormal gray correlation degree between the surface potential data of the abnormal region and the reference value of the surface potential data is determined, and it is detected whether the abnormal gray correlation degree is less than the corresponding correlation degree threshold.

[0010] When the abnormal gray correlation degree is less than the corresponding correlation degree threshold, the part of the abnormal region where the abnormal gray correlation degree is less than the corresponding correlation degree threshold is determined as an abnormal part;

[0011] Based on the surface potential data and surface magnetic induction intensity of the abnormal location, a preset state identification model is used to determine the state identification result, and the health status of the grounding grid is determined based on the state identification result.

[0012] In some embodiments, the step of acquiring surface data of the test condition in the grounding grid includes:

[0013] Acquire initial surface data from multiple ground surfaces of the grounding grid;

[0014] The initial surface data of multiple sources are normalized to obtain initial surface data of each source with the same dimension;

[0015] Initial surface data of the same dimension are defined as surface data.

[0016] In some embodiments, the preset state recognition model is obtained through the following steps:

[0017] Construct a simulation model of the grounding grid;

[0018] The surface data of different operating states in the abnormal area are input into the grounding grid simulation model for training to obtain the state identification model. The operating states of the abnormal area include normal state, corrosion state and fracture state. The surface data also includes soil resistivity and traction return distribution data.

[0019] In some embodiments, after the step of constructing the grounding grid simulation model, the method further includes:

[0020] By changing the ratio of the change in the conductor radius of the grounding grid, surface data of different operating states in the abnormal area of ​​the grounding grid can be obtained;

[0021] Based on surface data of different operating states in the abnormal areas of the grounding grid, a sample database is constructed for training the grounding grid simulation model.

[0022] In some embodiments, the step of determining the health status of the grounding grid based on the state identification result includes:

[0023] Based on the state identification results, the conductor radius range of the abnormal region is obtained;

[0024] Based on a preset correspondence, the health status of the grounding grid is determined according to the conductor radius range of the abnormal area.

[0025] In some embodiments, it also includes:

[0026] The corrosion rate is averaged based on the conductor radius range of the abnormal region to obtain the conductor lifetime of the grounding grid.

[0027] Secondly, this disclosure provides a grounding grid fault detection device, comprising:

[0028] The acquisition module is used to acquire surface data of the test condition in the grounding grid, wherein the surface data includes surface potential data and surface magnetic induction intensity.

[0029] The regional correlation module is used to determine the regional gray correlation degree between the surface potential data and the reference value of the surface potential data based on the surface potential data above the cross nodes of the grounding grid, using a gray correlation algorithm, and to detect whether the regional gray correlation degree is less than the corresponding correlation degree threshold.

[0030] The first determining module is used to determine that when the gray correlation degree of the area is less than the corresponding correlation degree threshold, the area in the grounding grid with the gray correlation degree less than the corresponding correlation degree threshold is an abnormal area;

[0031] An anomaly correlation module is used to determine the anomaly gray correlation between the surface potential data of the anomaly region and the reference value of the surface potential data based on the surface potential data above the middle part of the conductor in the anomaly region, and to detect whether the anomaly gray correlation is less than the corresponding correlation threshold.

[0032] The second determining module is used to determine the part of the abnormal region where the abnormal gray correlation degree is less than the corresponding correlation degree threshold as an abnormal part when the abnormal gray correlation degree is less than the corresponding correlation degree threshold.

[0033] The status determination module is used to determine the status identification result based on the surface potential data and surface magnetic induction intensity of the abnormal location using a preset status identification model, and to determine the health status of the grounding grid based on the status identification result.

[0034] Thirdly, this disclosure provides a computer device including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the foregoing aspects.

[0035] Fourthly, this disclosure provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the methods described in the above aspects.

[0036] Fifthly, this disclosure provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the methods described in the foregoing aspects.

[0037] This disclosure provides a grounding grid fault detection method, device, storage medium, and computer program. Based on surface potential data above the intersection nodes of the grounding grid, a grey relational algorithm is used to determine the regional grey relational degree between the surface potential data and a reference value of the surface potential data. The algorithm then detects whether the regional grey relational degree is less than a corresponding relational degree threshold. When the regional grey relational degree is less than the corresponding relational degree threshold, the area in the grounding grid with a regional grey relational degree less than the corresponding relational degree threshold is identified as an abnormal area. Next, based on the surface potential data above the middle of the conductor in the abnormal area, an abnormal grey relational degree between the surface potential data in the abnormal area and the reference value of the surface potential data is determined. The algorithm detects whether the abnormal grey relational degree is less than a corresponding relational degree threshold. When the abnormal grey relational degree is less than the corresponding relational degree threshold, the part in the abnormal area with an abnormal grey relational degree less than the corresponding relational degree threshold is identified as an abnormal part. Subsequently, based on the surface potential data and surface magnetic induction intensity of the abnormal part, a preset state identification model is used to determine the state identification result. Based on the state identification result, the health status of the grounding grid is determined. In this way, without the need for an external excitation source, power outage, or excavation, the abnormal state of the substation grounding grid can be quickly identified and the fault area can be accurately located, thus providing an important reference for the on-site operation and maintenance of the substation. Attached Figure Description

[0038] The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings:

[0039] Figure 1 A schematic flowchart illustrating a grounding grid fault detection method provided in this embodiment of the disclosure;

[0040] Figure 2 This is a structural block diagram of a grounding grid fault detection device in one embodiment;

[0041] Figure 3 This is a diagram showing the internal structure of an electronic device in one embodiment.

[0042] Figure 4 This is a flowchart illustrating a grounding grid fault detection method in another embodiment;

[0043] Figure 5 This is a schematic diagram of the grounding system of a traction substation in one embodiment;

[0044] Figure 6 This is a schematic diagram of the structure of a data acquisition device in one embodiment;

[0045] Figure 7 This is a schematic diagram of the grounding grid correlation distribution in one embodiment;

[0046] Figure 8 This is a schematic diagram of the distribution of correlation degree in abnormal regions in one embodiment;

[0047] Figure 9 This is a schematic diagram illustrating the effect of accurate identification of the grounding grid status in one embodiment. Detailed Implementation

[0048] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0049] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0050] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0051] Example 1

[0052] Figure 1 This is a schematic flowchart illustrating a grounding grid fault detection method provided in an embodiment of this disclosure. Figure 1 As shown, a grounding grid fault detection method includes:

[0053] Step 110: Obtain the surface data of the test condition in the grounding grid, wherein the surface data includes surface potential data and surface magnetic induction intensity.

[0054] In this embodiment, the surface data of the grounding grid whose health status needs to be evaluated is obtained, that is, the surface data of the grounding grid under test is obtained. The under test refers to the current operating condition of the grounding grid whose health status needs to be evaluated.

[0055] Step 120: Based on the surface potential data above the intersection nodes of the grounding grid, the gray relational degree algorithm is used to determine the regional gray relational degree between the surface potential data and the reference value of the surface potential data, and to detect whether the regional gray relational degree is less than the corresponding relational degree threshold.

[0056] It should be understood that the reference value for ground potential data is determined through a grounding grid operating under normal conditions. Normal operating conditions refer to the operating conditions of a grounding grid under ideal conditions or historically well-functioning conditions. In other words, when designing a grounding grid, the reference value and correlation threshold for ground potential data are determined based on the ideal state or historically good operating conditions of the grounding grid. Normal operating conditions serve as a benchmark for evaluating the condition under test. The reference value for ground potential data is obtained by calculating the ground potential data based on the ground potential data under normal operating conditions. In this embodiment, a grey relational algorithm is used to calculate the grey relational degree between the ground potential data above the intersection nodes of the grounding grid and the reference value of the ground potential data, obtaining the regional grey relational degree of the condition under test, and detecting whether the regional grey relational degree is less than the corresponding correlation threshold. That is, by comparing the ground potential data above the intersection nodes of the grounding grid with the reference value of the ground potential data, it is detected whether the ground potential data above the intersection nodes of the grounding grid is close to or equal to the reference value of the ground potential data, in order to quickly determine whether there is an abnormal state in the grounding grid, thereby accurately locating the abnormal area of ​​the grounding grid.

[0057] Step 130: When the gray correlation degree of the region is less than the corresponding correlation degree threshold, the region in the grounding grid whose gray correlation degree is less than the corresponding correlation degree threshold is determined to be an abnormal region.

[0058] In this embodiment, when the gray correlation degree of a region is less than the corresponding correlation degree threshold, that is, when the surface potential data of the test condition is not close to the reference value of the surface potential data of the normal condition, it can be determined that the grounding grid is in an abnormal state, and the region in the grounding grid with a gray correlation degree less than the corresponding correlation degree threshold is identified as an abnormal region.

[0059] Step 140: Based on the surface potential data above the middle part of the conductor in the abnormal region, determine the abnormal gray correlation degree between the surface potential data of the abnormal region and the reference value of the surface potential data, and detect whether the abnormal gray correlation degree is less than the corresponding correlation degree threshold.

[0060] In this embodiment, after identifying the abnormal region, a grey relational algorithm is used to calculate the grey relational degree between the surface potential data above the middle of the conductor in the abnormal region and the reference value of the surface potential data, obtaining the abnormal grey relational degree, and then detecting whether the abnormal grey relational degree is less than the corresponding relational degree threshold. That is, the surface potential data above the middle of the conductor in the abnormal region is compared with the reference value of the surface potential data to detect whether the surface potential data above the middle of the conductor in the abnormal region is close to or equal to the reference value of the surface potential data.

[0061] Step 150: When the abnormal gray correlation degree is less than the corresponding correlation degree threshold, the part of the abnormal region where the abnormal gray correlation degree is less than the corresponding correlation degree threshold is determined as an abnormal part.

[0062] In this embodiment, when the abnormal gray correlation degree is less than the corresponding correlation degree threshold, that is, when the ground potential data above the middle of the conductor in the abnormal region is not close to the reference value of the ground potential data, the abnormal parts in the abnormal region can be further determined. The parts in the abnormal region with an abnormal gray correlation degree less than the corresponding correlation degree threshold are identified as abnormal parts. In this embodiment, when determining the gray correlation degree of a region, the gray correlation degree of the region is obtained by performing gray correlation calculation on the ground potential data above the intersection node of the grounding grid, so as to quickly determine whether there is an abnormal state in the grounding grid, thereby accurately locating the abnormal region of the grounding grid. When determining the abnormal gray correlation degree, the abnormal gray correlation degree is obtained by performing gray correlation calculation on the ground potential data above the middle of the conductor in the abnormal region, so as to further narrow down the range based on the determination of the abnormal region, thereby accurately locating the abnormal parts in the abnormal region, effectively improving the accuracy and efficiency of fault location.

[0063] Step 160: Based on the surface potential data and surface magnetic induction intensity of the abnormal location, determine the state identification result using a preset state identification model, and determine the health status of the grounding grid based on the state identification result.

[0064] In this embodiment, the state identification model is used to obtain the health status of the grounding grid. After identifying the abnormal parts, the surface potential data and surface magnetic induction intensity of the abnormal parts are input into the state identification model to obtain the state identification results, and the health status of the grounding grid is determined based on the state identification results.

[0065] In the above embodiments, based on the surface potential data above the intersection nodes of the grounding grid, a gray relational algorithm is used to determine the regional gray relational degree between the surface potential data and the reference value of the surface potential data. It is then detected whether the regional gray relational degree is less than a corresponding relational degree threshold. When the regional gray relational degree is less than the corresponding relational degree threshold, the region in the grounding grid with a regional gray relational degree less than the corresponding relational degree threshold is determined as an abnormal region. Then, based on the surface potential data above the middle of the conductor in the abnormal region, an abnormal gray relational degree between the surface potential data in the abnormal region and the reference value of the surface potential data is determined. It is then detected whether the abnormal gray relational degree is less than a corresponding relational degree threshold. When the abnormal gray relational degree is less than the corresponding relational degree threshold, the part in the abnormal region with an abnormal gray relational degree less than the corresponding relational degree threshold is determined as an abnormal part. Subsequently, based on the surface potential data and surface magnetic induction intensity of the abnormal part, a preset state identification model is used to determine the state identification result. Based on the state identification result, the health status of the grounding grid is determined. In this way, without the need for an external excitation source, power outage, or excavation, the abnormal state of the substation grounding grid can be quickly identified and the fault area can be accurately located, thus providing an important reference for the on-site operation and maintenance of the substation.

[0066] In one embodiment, the step of acquiring the grounding grid surface data includes:

[0067] Acquire initial surface data from multiple ground surfaces of the grounding grid;

[0068] The initial surface data of multiple sources are normalized to obtain initial surface data of each source with the same dimension;

[0069] Initial surface data of the same dimension are defined as surface data.

[0070] In this embodiment, multiple initial surface data of the grounding grid are first acquired. These initial surface data include surface potential data, surface magnetic induction intensity, soil resistivity, and traction return current distribution data. Since the surface potential data, surface magnetic induction intensity, soil resistivity, and traction return current distribution data in the initial surface data have dimensional differences, it is necessary to standardize these data. This means converting the surface potential data, surface magnetic induction intensity, soil resistivity, and traction return current distribution data to the same dimension. Then, the initial surface data of the same dimension are identified as surface data. The surface data includes surface potential data, surface magnetic induction intensity, soil resistivity, and traction return current distribution data. The surface potential data, surface magnetic induction intensity, soil resistivity, and traction return current distribution data in the surface data are the data after standardization.

[0071] In some embodiments, initial surface data are normalized using the following relation:

[0072]

[0073] In the formula: X 1 This is the distribution matrix of measured data (surface potential data) above the grounding grid under normal operating conditions. X represents the measured surface potential data under normal operating conditions, marked in row m and column n above the grounding grid crossover node. 2 This is the distribution matrix of the measured data (surface potential data) of the grounding grid under the tested operating conditions. The measured value of the ground surface potential data for the test condition, marked as row m and column n directly above the grounding grid intersection node, is max(x′). i ) represents the maximum value in the i-th row, min(x' i ) is the minimum value of the i-th row (i = 0, 1, 2, ..., m; j = 0, 1, 2, ..., n).

[0074] In some embodiments, grey relational degree is calculated using the following formula:

[0075]

[0076] Δ 2 x′ ij =t′ i,j+1 -2t′ ij +t′ i,j-1 (9)

[0077]

[0078] in: The result of the normalization processing of the ground surface potential of the grounding grid. The result of the regularization processing of the ground surface potential under the test state of the grounding grid. For the total displacement difference, For first-order displacement difference, For second-order displacement difference, This represents the grey relational degree.

[0079] In one embodiment, the preset state recognition model is obtained through the following steps:

[0080] Construct a simulation model of the grounding grid;

[0081] The surface data of different operating states in the abnormal area are input into the grounding grid simulation model for training to obtain the state identification model. The operating states of the abnormal area include normal state, corrosion state and fracture state. The surface data also includes soil resistivity and traction return distribution data.

[0082] In this embodiment, a simulation model of the grounding grid of the traction substation is established using electromagnetic software. Different operating states of the grounding grid are simulated by changing the ratio of conductor radius variation. The operating state refers to the health status of the grounding grid, including normal, corroded, and broken states. It should be noted that the conductor refers to the material of the grounding grid. As shown in Table 1, specifically, when the grounding grid is in a normal operating state, the conductor radius is r, meaning the conductor radius has not changed; when the grounding grid is in a corroded operating state, the conductor radius is within the range (0, r), meaning the conductor radius has changed; the corrosion state can be divided into multiple levels based on the specific corrosion condition of the conductor radius; when the grounding grid is in a broken operating state, the conductor radius is 0, meaning the conductor has completely broken.

[0083] Table 1. Classification of Grounding Grid Operation Status Levels and Sample Database

[0084] Grounding grid operating status conductor radius Sample Labels Sample data volume normal r 0 50 Corrosion I 0.8r～r 1 50 Corrosion II 0.6r～0.8r 2 50 Corrosion III 0.5r～0.6r 3 50 Corrosion IV 0.4r～0.5r 4 50 Corrosion V 0.3r～0.4r 5 25 Corrosion VI 0.2r～0.3r 6 25 Corrosion VII 0.1r～0.2r 7 25 Corrosion VIII 0～0.1r 8 25 fracture 0 9 50

[0085] In one embodiment, after the step of constructing the grounding grid simulation model, the method further includes:

[0086] By changing the ratio of the change in the conductor radius of the grounding grid, surface data of different operating states in the abnormal area of ​​the grounding grid can be obtained;

[0087] Based on surface data of different operating states in the abnormal areas of the grounding grid, a sample database is constructed for training the grounding grid simulation model.

[0088] In this embodiment, the operating status is divided according to the conductor radius of the grounding grid. Surface data of different operating statuses of the grounding grid in the abnormal area are collected. The surface data includes surface potential data, surface magnetic induction intensity, soil resistivity and traction return current distribution data. The surface data of different operating statuses are normalized and preprocessed to establish a sample database for training the grounding grid simulation model based on the normalized and preprocessed surface data. As shown in Table 1, the grounding grid status identification model is established based on the XGBoost algorithm. After the surface data is normalized and preprocessed, the sample database is divided into a test set and a training set in a ratio of 0.8:0.2. By inputting the mapping relationship between the feature parameters and the output labels in the training grounding grid simulation model, with the objective function of formula (14) as the minimum, the probability distribution [y1,y2,…,y] corresponding to each label is output. n-1 ,y n If the output category with the highest probability is y i Then its corresponding label d i This is determined as the identification result.

[0089]

[0090] P(di )=y i (12)

[0091] max(y1,y2,…,y n-1 ,y n )=max[P(d1),P(d2),…,P(d n-1 ), P(d n (13)

[0092] The objective function consists of the model's loss function L and a regularization term Ω that suppresses model complexity, and is defined as follows:

[0093]

[0094] A second-order Taylor expansion of relation (10) can be written as:

[0095]

[0096] In one embodiment, the step of determining the health status of the grounding grid based on the state identification result includes:

[0097] Based on the state identification results, the conductor radius range of the abnormal region is obtained;

[0098] Based on a preset correspondence, the health status of the grounding grid is determined according to the conductor radius range of the abnormal area.

[0099] In this embodiment, based on surface potential data and surface magnetic induction intensity, after determining the state identification results using a preset state identification model, the conductor radius range of the abnormal area is obtained based on the state identification results, as shown in Table 2. The conductor radius range includes the normal conductor radius r corresponding to the normal state, the corroded conductor radius range (0, r) corresponding to the corrosion state, and the fractured conductor radius corresponding to the fracture state. The corroded conductor radius range can be divided into multiple levels according to the specific corrosion condition, such as the light corrosion range (0.707r, r), the moderate corrosion range (0.316r, 0.577r), and the severe corrosion range (0, 0.316r). Then, the health status of the grounding grid is determined based on the conductor radius range of the abnormal area.

[0100] Table 2 Classification of Grounding Grid Health Status

[0101] Grounding grid health status conductor radius normal r Mild corrosion 0.707r～r Moderate corrosion 0.316r～0.577r Severe corrosion 0～0.316r fracture 0

[0102] In one embodiment, the grounding grid fault detection method further includes:

[0103] The corrosion rate is averaged based on the conductor radius range of the abnormal region to obtain the conductor lifetime of the grounding grid.

[0104] In this embodiment, the conductor lifetime of the grounding grid is calculated based on soil resistivity and traction return current distribution data. After determining the state identification results using a preset state identification model, the conductor radius range of the abnormal region is obtained based on the state identification results. Then, corrosion rate averaging is performed based on the conductor radius range of the abnormal region to obtain the conductor lifetime of the grounding grid. The conductor lifetime is obtained through the following formula:

[0105]

[0106] ω0=ρlπr0 2 (18)

[0107] ω i =ρlπr i 2 (19)

[0108]

[0109] Where: ARL is the remaining life of the grounding grid conductor, T is the time the grounding grid conductor has been in operation, ω0 is the mass of the conductor in normal condition, r0 is the conductor radius of the grounding grid in normal condition, ω i The mass of the conductor in the test state, l is the length of the grounding grid conductor in the abnormal area, and r is the mass of the conductor in the test state. i Here, S is the conductor radius under test in the grounding grid, S is the surface area of ​​the conductor in contact with the soil, t is the corrosion time, ρ is the conductor resistivity, and v is the grounding grid surface area under test. g The corrosion rate and v calculated by the conductor weight loss method L Let be the corrosion rate calculated from the conductor, a and b be the length and width of the grounding grid conductor, respectively, and d be the equivalent diameter of the grounding grid conductor.

[0110] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0111] Example 2

[0112] Based on the above embodiments, in this embodiment, as follows: Figure 4As shown, a grounding grid fault detection method is provided, including:

[0113] Step 1: Collect surface potential data, surface magnetic induction intensity, soil resistivity, and traction return distribution data.

[0114] Step 2: Using the surface potential data above the grounding grid intersection nodes as the driving force, calculate the regularized gray relational distribution between the test condition and the normal condition. When the difference between the correlation degrees of the two conditions at a certain point is less than a threshold, it is considered normal; otherwise, it is considered abnormal. In other words, the test condition refers to the current state of the grounding grid that needs to be evaluated, while the normal condition is the ideal state during the grounding grid design or a historically good operating state. Essentially, the normal condition serves as a benchmark. If the data for the test condition and the normal condition differ significantly, it indicates that the grounding grid may be abnormal. The gray relational degree is calculated by comparing the gray relational degree of the two conditions. If the gray relational degree of the test condition is close to or equal to the gray relational degree of the normal condition, it indicates that the grounding grid is in a normal state; otherwise, it indicates that the grounding grid is in an abnormal state.

[0115] Step 3: If the abnormality is determined in Step 2, the abnormal area is initially determined. Using the ground potential data above the middle of the grounding grid conductor in the area as the driving force, the regularized gray relational distribution of the test condition and the normal condition is calculated. When the difference between the relational degree of the two conditions at a certain point is less than the threshold, it is judged as normal; otherwise, it is judged as abnormal.

[0116] Step 4: Based on the distribution of gray relational degree in Step 2 and Step 3, accurately determine the abnormal location, collect surface electromagnetic parameter data (surface potential data, surface magnetic induction intensity, soil resistivity, and traction return distribution data) of the grounding grid under different operating states in the area (abnormal area), and perform normalization preprocessing to establish a sample database.

[0117] Step 5: Input the sample data into the XGBoost (machine learning) algorithm to establish a grounding grid status identification model. When the maximum value of the output probability distribution is greater than the threshold, the label with the maximum output probability is taken as the output result. Optimize relevant parameters to improve the diagnostic effect of the model.

[0118] Step 6: Based on the grounding grid status identification results, estimate the health status and remaining lifespan of the grounding grid.

[0119] Furthermore, the soil resistivity data in step 1 is obtained by testing the distribution of typical soil layers in different seasons of the substation using the four-electrode method.

[0120] Furthermore, the traction return current distribution data in step 1 can be collected by installing a sensor at the traction return current port of the centralized grounding box of the traction substation.

[0121] Furthermore, the surface electromagnetic data in step 1 can be collected by burying reference electrodes above the grounding grid cross nodes of the traction substation and above the conductors of important equipment to collect surface electromagnetic parameter data.

[0122] Furthermore, the improved grey relational algorithm for data regularization in steps 2 and 3 is as follows:

[0123] Data normalization

[0124]

[0125] In the formula: X 1 This is the distribution matrix of measured data (surface potential data) above the grounding grid under normal operating conditions. X represents the measured surface potential data under normal operating conditions, marked as row m and column n directly above the grounding grid intersection node. 2 This is the distribution matrix of the measured data (surface potential data) of the grounding grid under the tested operating conditions. The measured value of the ground surface potential data for the test condition, marked in row m and column n directly above the grounding grid intersection node, is max(x'). i ) represents the maximum value in the i-th row, min(x' i ) is the minimum value of the i-th row (i = 0, 1, 2, ..., m; j = 0, 1, 2, ..., n).

[0126] Grey relational degree calculation

[0127]

[0128] Δ 2 x′ ij =t′ i,j+1 -2t′ ij +t′ i,j-1 (9)

[0129]

[0130] in: The result of the normalization processing of the ground surface potential of the grounding grid. The result of the regularization processing of the ground surface potential under the test state of the grounding grid. For the total displacement difference, For first-order displacement difference, For second-order displacement difference, This represents the grey relational degree.

[0131] Furthermore, the electromagnetic parameter data from step 4 can be used to establish a simulation model of the traction substation grounding grid using relevant electromagnetic software, such as CDEGS. By changing the ratio of conductor radius variation, different operating states of the grounding grid are simulated, and simulation data of the grounding grid surface potential and surface magnetic induction intensity are obtained to establish a sample database. The specific division of the grounding grid operating states and the composition of the sample database are shown in Table 1.

[0132] Table 1. Classification of Grounding Grid Operation Status Levels and Sample Database

[0133] Grounding grid operating status conductor radius Sample Labels Sample data volume normal r 0 50 Corrosion I 0.8r～r 1 50 Corrosion II 0.6r～0.8r 2 50 Corrosion III 0.5r～0.6r 3 50 Corrosion IV 0.4r～0.5r 4 50 Corrosion V 0.3r～0.4r 5 25 Corrosion VI 0.2r～0.3r 6 25 Corrosion VII 0.1r～0.2r 7 25 Corrosion VIII 0～0.1r 8 25 fracture 0 9 50

[0134] Furthermore, in step 5, a grounding grid state identification model is established based on the XGBoost algorithm. After preprocessing the sample dataset, the test set and training set are divided in a ratio of 0.8:0.2. By training the model through the mapping relationship between the input feature parameters and the output labels, and with the objective function of formula (14) as the minimum, the probability distribution [y1, y2, ..., y] corresponding to each label is output. n-1 ,y n If the output category with the highest probability is y i Then its corresponding label d i This is determined as the identification result.

[0135]

[0136] P(d i )=y i (12)

[0137] max(y1,y2,…,y n-1 ,y n )=max[P(d1),P(d2),…,P(d n-1 ), P(d n )](13)

[0138] The objective function consists of the model's loss function L and a regularization term Ω that suppresses model complexity, and is defined as follows:

[0139]

[0140] A second-order Taylor expansion of relation (10) can be written as:

[0141]

[0142] Furthermore, in step 6, the health status of the grounding grid is predicted by obtaining the radius range of the conductor in the abnormal area based on the identification results in step (5), and evaluating it with reference to the classification of the health status of the grounding grid. The specific classification is shown in Table 2.

[0143] Table 2 Classification of Grounding Grid Health Status

[0144] Grounding grid health status conductor radius normal r Mild corrosion 0.707r～r Moderate corrosion 0.316r～0.577r Severe corrosion 0～0.316r fracture 0

[0145] Furthermore, in step 6, the remaining lifespan of the grounding grid is estimated by obtaining the radius range of the conductor in the abnormal area based on the identification results in step (5), averaging the corrosion rate, and calculating the remaining lifespan of the grounding grid conductor.

[0146] Remaining life estimate:

[0147]

[0148]

[0149] ω0=ρlπr0 2 (18)

[0150] ω i =ρlπr i 2 (19)

[0151]

[0152] Where: ARL is the remaining life of the grounding grid conductor, T is the time the grounding grid conductor has been in operation, ω0 is the mass of the conductor in normal condition, r0 is the conductor radius of the grounding grid in normal condition, ω i The mass of the conductor in the test state, l is the length of the grounding grid conductor in the abnormal area, and r is the mass of the conductor in the test state. i Here, S is the conductor radius under test in the grounding grid, S is the surface area of ​​the conductor in contact with the soil, t is the corrosion time, ρ is the conductor resistivity, and v is the grounding grid surface area under test. g The corrosion rate and v calculated by the conductor weight loss method L Let be the corrosion rate calculated from the conductor, a and b be the length and width of the grounding grid conductor, respectively, and d be the equivalent diameter of the grounding grid conductor.

[0153] In this embodiment, the goal is to monitor and diagnose the health status of the grounding grid. Using traction return current as an excitation, surface electromagnetic parameter data is collected. Based on a data-regularized grey relational algorithm, grounding grid anomalies are identified and located. A classification model is constructed using machine learning algorithms to identify the operating status of the grounding grid, achieving grounding grid health status monitoring and fault location without power outages, excavation, or external excitation. This method effectively utilizes the characteristics of traction return current and combines relational algorithms with artificial intelligence algorithms, resulting in stronger feature extraction capabilities and the ability to identify weak features of grounding grid corrosion, thus achieving grounding grid anomaly identification and location. Therefore, using traction return current as an excitation source, without power outages or excavation, and driven by surface electromagnetic parameter data, to identify, assess, and locate the health status of the traction substation grounding grid is of great significance for ensuring the safe and stable operation of electrified railways and efficient on-site maintenance.

[0154] Specifically, the grounding system structure of the traction substation under the direct power supply method addressed in this invention is as follows: Figure 5 As shown; based on the four-electrode method, the soil resistivity data was collected near the grounding grid of the traction substation, and sensors were installed to collect traction return current data at each terminal; reference electrodes were placed directly above the grounding grid intersection nodes and the middle of the conductor to collect surface potential distribution data under normal operating conditions. The required data are as follows: Figure 6 As shown.

[0155] Driven by the ground surface potential data above the grounding grid intersection node, the correlation between the ground surface potential under test and the normal operating condition is calculated based on the gray relational algorithm of data regularization.

[0156] The steps for calculating the grey relational degree of surface potential data regularization are as follows:

[0157] (1) Let X be the distribution matrix of the surface potential data above the grounding grid under normal operating conditions. 1 for:

[0158]

[0159] In the formula: The measured value of the ground surface potential under normal operating conditions, marked in row m and column n above the grounding grid intersection node;

[0160] Distribution matrix X of surface potential data above the grounding grid under unknown operating conditions 2 for:

[0161]

[0162] In the formula: The measured value of the unknown working condition ground surface potential is marked in row m and column n above the grounding grid intersection node;

[0163] (2) Data elements To perform normalization, firstly, mean normalization is performed:

[0164]

[0165] Then, range normalization is performed:

[0166]

[0167] max(x' i — The maximum value in the i-th row;

[0168] min(x' i — The minimum value in the i-th row;

[0169] In the formula: i = 0, 1, 2, ..., m; j = 0, 1, 2, ..., n.

[0170] (3) Calculate the total displacement difference of the data:

[0171]

[0172] (4) Calculate the first-order slope difference of the data:

[0173]

[0174] (5) Calculate the second slope difference of the data:

[0175]

[0176] Δ 2 x′ ij =t′ i,j+1 -2tv ij +t′ i,j-1

[0177] (6) Calculate the grey relational degree of the data:

[0178]

[0179] in The result of the surface potential regularization process for the normal operation of the grounding grid. The result of the regularization processing of the ground surface potential under the test state of the grounding grid. For the total displacement difference, For first-order displacement difference, For second-order displacement difference, This represents the grey relational degree.

[0180] By combining formulas (1) to (10), the grey relational degree is calculated, and the results are as follows: Figure 7 and Figure 8 As shown.

[0181] 3. If the judgment in step (2) is abnormal, the abnormal area is initially determined according to the distribution of gray relational degree, and the ground surface potential above the middle part of the grounding grid conductor in the area is collected; similarly, the correlation between the ground surface potential of the test condition above the middle part of the grounding grid conductor and the distribution of the ground surface potential of the normal condition is calculated according to formulas (1) to (10), and the abnormal part of the abnormal area is further determined.

[0182] 4. Based on the XGBoost algorithm, a grounding grid state identification model is established. After preprocessing the sample database data, the test set and training set are divided into a ratio of 0.8:0.2. With the objective function of formula (14) as the goal, the mapping relationship between the input feature parameters and the output labels in the training model is established, and the output is the probability distribution [y1,y2,…,y] corresponding to each label. n-1 ,y n If the output category with the highest probability is y. i Then its corresponding label d i This is determined as the identification result.

[0183]

[0184] P(d i )=y i

[0185] max(y1,y2,…,y n-1 ,y n )=max[P(d1),P(d2),…,P(d n-1 ), P(d n )]

[0186] The objective function consists of the model's loss function L and a regularization term Ω that suppresses model complexity, and is defined as follows:

[0187]

[0188] A second-order Taylor expansion of relation (10) can be written as:

[0189]

[0190] The grounding grid condition identification model was trained, and its diagnostic performance was verified using a test set. The accuracy rate reached 97.92%, demonstrating precise identification of the grounding grid condition. Specific diagnostic results are as follows: Figure 9 As shown.

[0191] 5. Based on the identification results of the grounding grid status and relevant standards and specifications, predict the health status and remaining life of the conductors in the abnormal area of ​​the grounding grid. According to the identification results in step (4), obtain the radius range of the conductors in the abnormal area r1≤r≤r2, average the corrosion rate, and calculate the remaining life of the grounding grid conductors.

[0192] Remaining life estimate:

[0193]

[0194] ω0=ρlπr0 2

[0195] ω i =ρlπr i 2

[0196]

[0197] Where: ARL is the remaining life of the grounding grid conductor, T is the time the grounding grid conductor has been in operation, ω0 is the mass of the conductor in normal condition, r0 is the conductor radius of the grounding grid in normal condition, ω i The mass of the conductor under test, l is the length of the grounding grid conductor in the abnormal area, and r i Here, S is the conductor radius under test in the grounding grid, S is the surface area of ​​the conductor in contact with the soil, t is the corrosion time, ρ is the conductor resistivity, and v is the grounding grid surface area under test. g The corrosion rate and v calculated by the conductor weight loss method L Let be the corrosion rate calculated for the conductor, a and b be the length and width of the grounding grid conductor, respectively, and d be the equivalent diameter of the grounding electrode.

[0198] The remaining lifespan of the grounding grid conductor in the abnormal area can be estimated based on the above relationship.

[0199] Example 3

[0200] Based on the above embodiments, in this embodiment, as follows: Figure 2 As shown, a grounding grid fault detection device is provided, comprising:

[0201] The acquisition module 210 is used to acquire surface data of the test condition in the grounding grid, wherein the surface data includes surface potential data and surface magnetic induction intensity.

[0202] The correlation module 220 is used to determine the regional gray correlation degree between the ground potential data and the reference value of the ground potential data based on the ground potential data above the cross node of the grounding grid, and to detect whether the regional gray correlation degree is less than the corresponding correlation degree threshold.

[0203] The first determining module 230 is used to determine that when the gray correlation degree of the region is less than the corresponding correlation degree threshold, the region in the grounding grid with the gray correlation degree of the region being less than the corresponding correlation degree threshold is an abnormal region.

[0204] Anomaly correlation module 240 is used to determine the anomalous gray correlation degree between the surface potential data of the anomalous region and the reference value of the surface potential data based on the surface potential data above the middle part of the conductor in the anomalous region, and to detect whether the anomalous gray correlation degree is less than a corresponding threshold.

[0205] The second determining module 250 is used to determine the part of the abnormal region where the abnormal gray correlation degree is less than the corresponding correlation degree threshold as an abnormal part when the abnormal gray correlation degree is less than the correlation degree threshold corresponding to the correlation degree.

[0206] The status determination module 260 is used to determine the status identification result based on the surface potential data and the surface magnetic induction intensity of the abnormal location using a preset status identification model, and to determine the health status of the grounding grid based on the status identification result.

[0207] In one embodiment, the acquisition module 210 includes:

[0208] The acquisition unit is used to acquire multiple initial surface data of the grounding grid;

[0209] A normalization processing unit is used to normalize multiple initial surface data to obtain initial surface data of the same dimension.

[0210] The data determination unit is used to determine the initial surface data of various regions with the same dimension as surface data.

[0211] In one embodiment, the grounding grid fault detection device further includes:

[0212] The model building module is used to build simulation models of grounding grids.

[0213] The training module is used to input surface data of different operating states in the abnormal area into the grounding grid simulation model for training to obtain a state identification model. The operating states of the abnormal area include normal state, corrosion state and fracture state. The surface data also includes soil resistivity and traction return distribution data.

[0214] In one embodiment, the grounding grid fault detection device further includes:

[0215] The proportional adjustment module is used to obtain surface data of different operating states in the abnormal area of ​​the grounding grid by changing the change ratio of the conductor radius of the grounding grid;

[0216] The database construction module is used to construct a sample database for training the grounding grid simulation model based on surface data of different operating states in the abnormal areas of the grounding grid.

[0217] In one embodiment, the state determination module 260 includes:

[0218] A conductor radius determination unit is used to obtain the conductor radius range of the abnormal region based on the state identification result;

[0219] The status determination unit is used to determine the health status of the grounding grid based on a preset correspondence and the conductor radius range of the abnormal area.

[0220] In one embodiment, the grounding grid fault detection device further includes:

[0221] The lifetime determination module is used to perform corrosion rate averaging based on the conductor radius range of the abnormal region to obtain the conductor lifetime of the grounding grid.

[0222] Example 4

[0223] Based on the above embodiments, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the above embodiments.

[0224] In some embodiments of this example, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0225] In some embodiments of this example, a computer program product is provided, including a computer program / instructions, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0226] The processor may include, but is not limited to, one or more processors or microprocessors. Each processor may be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic component, for executing the methods in the above embodiments.

[0227] Computer-readable storage media can be implemented by any type of volatile or non-volatile storage device or a combination thereof. Computer-readable storage media may include, but are not limited to, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, and computer storage media (e.g., hard disks, floppy disks, solid-state drives, removable disks, CD-ROMs, DVD-ROMs, Blu-ray discs, etc.).

[0228] Computer-readable storage media may also store at least one computer-executable program / instruction, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0229] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0230] The processor can communicate with external devices via the I / O bus through wired or wireless networks.

[0231] In one embodiment, the at least one computer-executable instruction may also be compiled into or comprise a software product / computer program product, wherein one or more computer-executable instructions are executed by a processor to perform the steps of the various functions and / or methods in the embodiments described herein.

[0232] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0233] It should be noted that, in this disclosure, 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 a process, method, article, or apparatus. Without further limitation, an element limited 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.

[0234] While the embodiments disclosed herein are as described above, the foregoing content is merely for the purpose of facilitating understanding of this disclosure and is not intended to limit this disclosure. Any person skilled in the art to which this disclosure pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope of this disclosure; however, the scope of patent protection of this disclosure shall still be determined by the scope defined in the appended claims.

Claims

1. A method of ground grid fault detection, characterized by, The method comprises the following steps: acquiring ground surface data of a to-be-tested working condition in a grounding network, wherein the ground surface data comprises ground surface potential data and ground surface magnetic induction intensity; based on the ground surface potential data above the intersection nodes of the grounding network, determining the regional grey correlation degree between the ground surface potential data and a reference value of the ground surface potential data by using a grey correlation degree algorithm, and detecting whether the regional grey correlation degree is less than a corresponding correlation degree threshold value; when the regional grey correlation degree is less than the corresponding correlation degree threshold value, determining that a region in the grounding network, in which the regional grey correlation degree is less than the corresponding correlation degree threshold value, is an abnormal region; based on the ground surface potential data above the middle part of the conductor of the abnormal region, determining the abnormal grey correlation degree between the ground surface potential data of the abnormal region and the reference value of the ground surface potential data, and detecting whether the abnormal grey correlation degree is less than a corresponding correlation degree threshold value; when the abnormal grey correlation degree is less than the corresponding correlation degree threshold value, determining that a part in the abnormal region, in which the abnormal grey correlation degree is less than the corresponding correlation degree threshold value, is an abnormal part; based on the ground surface potential data and the ground surface magnetic induction intensity of the abnormal part, determining a state recognition result by using a preset state recognition model, and determining the health state of the grounding network based on the state recognition result; the preset state recognition model is obtained through the following steps: constructing a grounding network simulation model; inputting the ground surface data of different operating states in the abnormal region into the grounding network simulation model for training to obtain a state recognition model, wherein the operating states of the abnormal region comprise a normal state, a corrosion state and a fracture state, and the ground surface data further comprises soil resistivity and traction return current distribution data.

2. The ground net fault detection method according to claim 1, characterized by, The step of acquiring the ground surface data of the to-be-tested working condition in the grounding network comprises: acquiring a plurality of ground surface initial data of the grounding network; normalizing the plurality of ground surface initial data to obtain ground surface initial data of the same dimension; determining the ground surface initial data of the same dimension as the ground surface data.

3. The ground net fault detection method according to claim 1, characterized by, After the step of constructing the grounding network simulation model, the following steps are further included: obtaining the ground surface data of different operating states in the abnormal region of the grounding network by changing the change ratio of the conductor radius of the grounding network; based on the ground surface data of different operating states in the abnormal region of the grounding network, constructing a sample database for training the grounding network simulation model.

4. The method of ground grid fault detection according to any one of claims 1 to 3, characterized in that, The step of determining the health state of the grounding network based on the state recognition result comprises: based on the state recognition result, obtaining the conductor radius range of the abnormal region; based on a preset corresponding relationship, determining the health state of the grounding network according to the conductor radius range of the abnormal region.

5. The ground net fault detection method according to claim 4, characterized by, Further comprising: based on the conductor radius range of the abnormal region, performing corrosion rate averaging processing to obtain the conductor life of the grounding network.

6. A ground grid fault detection apparatus characterized by comprising: The method comprises the following steps: an acquisition module is configured to acquire ground surface data of a to-be-tested working condition in a grounding network, wherein the ground surface data comprises ground surface potential data and ground surface magnetic induction intensity; The area correlation degree module is configured to determine an area grey correlation degree between the ground potential data and a reference value of the ground potential data based on the ground potential data above the cross nodes of the grounding net by using a grey correlation degree algorithm, and detect whether the area grey correlation degree is less than a corresponding correlation degree threshold value; The first determination module is configured to determine that a region in which the area grey correlation degree is less than the corresponding correlation degree threshold value is an abnormal region when the area grey correlation degree is less than the corresponding correlation degree threshold value; The abnormal correlation degree module is configured to determine an abnormal grey correlation degree between the ground potential data of the abnormal region and a reference value of the ground potential data based on the ground potential data above the conductor middle part of the abnormal region, and detect whether the abnormal grey correlation degree is less than a corresponding threshold value; The second determination module is configured to determine that a part in which the abnormal grey correlation degree is less than the corresponding correlation degree threshold value is an abnormal part when the abnormal grey correlation degree is less than the corresponding correlation degree threshold value; The state determination module is configured to determine a state recognition result by using a preset state recognition model based on the ground potential data and the ground magnetic induction intensity of the abnormal part, and determine a health state of the grounding net based on the state recognition result. The device further comprises: The model construction module is configured to construct a grounding net simulation model; The training module is configured to input ground data of different operating states of the abnormal region into the grounding net simulation model for training to obtain a state recognition model, wherein the operating states of the abnormal region include a normal state, a corrosion state and a fracture state, and the ground data further includes soil resistivity and traction return current distribution data.

7. A computer device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program comprises instructions that, when executed by the processor, cause the processor to perform the method of any one of claims 1-6. The processor executes the computer program to implement the steps of the method of any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 5. The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 5.

Images