A method and system for detecting cabinet grounding current based on a current sensor

Abstract

This invention discloses a method and system for detecting cabinet grounding current based on a current sensor. The method includes real-time acquisition of current values ​​at each node, calculation of inter-node delay time, determination of synchronization strength and shunt attenuation rate, and division of main path and branch paths; obtaining the path peak frequency through Fast Fourier Transform and combining it with a frequency threshold to obtain the peak offset; monitoring the path current, extracting time-series data to obtain abnormal time periods; tracing back historical current data to generate a dynamic flow vector and obtain the complete flow trajectory of the abnormal current; locating the point of abrupt change in flow direction to determine the preliminary coordinate position of the abnormal source; extracting current data, removing noise and performing feature analysis to obtain accurate current characteristics, comparing with normal current characteristics, projecting the current vector onto a three-dimensional mesh model, simulating the diffusion path to determine minute leakage points. This method can achieve dynamic monitoring of grounding current changes, comprehensively capture dynamic current changes, and accurately trace the source.

Description

Technical Field

[0001] This invention relates to the field of current detection technology, and in particular to a method and system for detecting cabinet grounding current based on a current sensor. Background Technology

[0002] Currently, with the widespread application of intelligent and distributed systems, the detection of abnormal grounding currents has become a focus of industry attention. In the field of electrical engineering and equipment safety, the detection and management of cabinet grounding currents is a crucial link in ensuring stable system operation and personal safety. Research in this area is directly related to the reliability of power systems, equipment lifespan, and fault prevention capabilities.

[0003] In one existing technology, a grounding resistance tester is used to indirectly determine the grounding current by measuring the continuity resistance between the cabinet grounding point and the PE busbar. This method can effectively detect grounding continuity, but it is not sensitive enough to monitoring dynamic changes in grounding current. These limitations make it difficult to comprehensively capture dynamic changes in current in complex cabinet structures, especially when faced with multi-path distribution and abnormal leakage, often making accurate source tracing impossible. If these problems are not effectively addressed, it will lead to delayed fault location and even cause greater system risks.

[0004] In summary, existing technologies are not sensitive enough to monitor dynamic changes in grounding current, making it difficult to fully capture these changes and thus hindering accurate source tracing. Summary of the Invention

[0005] This invention provides a method and system for detecting cabinet grounding current based on a current sensor, in order to solve the problem of inaccurate tracing in cabinet grounding current detection.

[0006] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a cabinet grounding current detection method based on a current sensor, comprising:

[0007] The current values ​​of each node in the cabinet grounding system are acquired in real time by sensors, and the current dynamic sequence is obtained by recording the current values ​​collected per second.

[0008] The arrival delay time of current signals between different nodes is calculated based on the current dynamic sequence, and the synchronization strength and shunt attenuation rate of current signals between nodes are determined based on the delay time. The main path and branch path are divided based on the shunt attenuation rate.

[0009] The peak frequencies of the main path and the branch paths are extracted by fast Fourier transform, and the peak offset of each path is obtained by combining the preset frequency threshold.

[0010] The path current of the main path and the branch path is monitored by the peak offset. When the path current exceeds the abnormal trigger threshold, the time series data within the reconstruction time window of the path is extracted to obtain the abnormal time period.

[0011] Based on the abnormal time period, the historical current data in memory is traced back to generate a dynamic flow direction vector, and the complete flow trajectory of the abnormal current is obtained.

[0012] Based on the complete flow trajectory, the current change locations between nodes at the points of abrupt change in flow direction are used as the preliminary coordinate locations of the anomaly source.

[0013] Based on the preliminary coordinate positions, the sensor current data of the corresponding nodes are extracted, noise interference is removed, and feature analysis is performed to obtain the accurate current characteristics of the anomaly source.

[0014] The precise current characteristics are compared with the preset normal current characteristics to obtain the comparison results. Based on the comparison results, the current vector is projected onto the three-dimensional mesh model of the cabinet grounding system, and the diffusion path of the leakage current is simulated to determine the micro-leakage point.

[0015] In one optional implementation, determining the synchronization strength and shunt attenuation rate of the inter-node current signal based on the delay time includes:

[0016] The reciprocal of the delay time is calculated and normalized to obtain the synchronization strength of the inter-node current signal;

[0017] The convergence point of the path is determined by identifying the concentrated signal area. The attenuation change value is calculated based on the current signal strength of the convergence point and the current signal strength of other nodes connected to it, and the shunt attenuation rate is obtained.

[0018] In one optional implementation, the step of dividing the main path and branch paths according to the shunt attenuation rate includes:

[0019] The shunt attenuation rate is compared with a preset attenuation rate threshold, and the synchronization strength is compared with a preset synchronization strength threshold.

[0020] When the shunt attenuation rate is less than or equal to the attenuation rate threshold and the synchronization strength is greater than or equal to the synchronization strength threshold, the inter-node current signal transmission path is determined to be the mainstream path.

[0021] When the shunt attenuation rate is greater than the attenuation rate threshold or the synchronization strength is less than the synchronization strength threshold, the inter-node current signal transmission path is determined to be the branch path.

[0022] In one optional implementation, the step of extracting the peak frequencies of the main path and the branch paths using Fast Fourier Transform, and combining this with a preset frequency threshold to obtain the peak offset of each path, includes:

[0023] Fast Fourier transform is performed on the current dynamic sequences of the main path and the branch path respectively to convert the time domain signal into the frequency domain signal and obtain the spectrum.

[0024] Calculate the amplitude of each frequency component in the spectrum;

[0025] The frequency point with the largest amplitude is identified as the peak frequency in the spectrum.

[0026] When the peak frequency exceeds a preset frequency threshold, the difference between the peak frequency and the frequency threshold is calculated to obtain the peak offset.

[0027] In one optional implementation, when the path current exceeds the abnormal trigger threshold, the timing data within the reconstruction time window of the path is extracted to obtain the abnormal time period, including:

[0028] When the path current exceeds the abnormal trigger threshold, the timing data of all nodes of the path within the reconstruction time window are retrieved from memory, wherein the reconstruction time window includes a preset fixed window and an adaptive window.

[0029] The standard deviation of the time series data is calculated. When the standard deviation of a continuous time interval exceeds the fluctuation threshold, the continuous time intervals are merged to obtain the abnormal time period.

[0030] In one optional implementation, the step of generating a dynamic flow direction vector by backtracking historical current data in memory based on the abnormal time period to obtain the complete flow trajectory of the abnormal current includes:

[0031] Based on the abnormal time period, extract the historical current value corresponding to the time period from memory;

[0032] Calculate the current flow direction and intensity between each node to generate a dynamic flow direction vector;

[0033] By analyzing the changing trend of the dynamic flow direction vector, the complete flow trajectory of the abnormal current is obtained.

[0034] In one optional implementation, the step of determining the current change location between nodes at the point of abrupt change in flow direction, based on the complete flow trajectory, as the preliminary coordinate location of the anomaly source, includes:

[0035] Based on the complete flow trajectory, the rate of change of current between a node and its connected adjacent nodes is calculated. When the rate of change of current exceeds a preset current change threshold, it is marked as a flow direction change point. When the flow direction change point is detected, the current change location is determined by combining the dynamic flow direction vector corresponding to the flow direction change point, which serves as the preliminary coordinate location of the anomaly source.

[0036] In one optional implementation, the step of extracting sensor current data of the corresponding node based on the preliminary coordinate position, removing noise interference, and performing feature analysis to obtain the precise current characteristics of the anomaly source includes:

[0037] Extract the sensor current data of the corresponding node based on the preliminary coordinate position;

[0038] By combining the reconstruction error range, noise interference is removed from the sensor current data to obtain effective current data;

[0039] The current time-domain characteristics of the anomaly source are obtained by calculating the mean and variance of the effective current data.

[0040] The effective current data is analyzed in the frequency domain using the Fast Fourier Transform algorithm to obtain the current frequency domain characteristics of the anomaly source.

[0041] The amplitude, frequency, and phase information in the current time-domain features and the current frequency-domain features are used as precise current features of the anomaly source.

[0042] In one optional implementation, the step of comparing the precise current characteristics with preset normal current characteristics to obtain a comparison result, projecting the current vector onto a three-dimensional mesh model of the cabinet grounding system based on the comparison result, and simulating the diffusion path of leakage current to determine minute leakage points includes:

[0043] The precise current characteristics are compared with the preset normal current characteristics. When the amplitude exceeds the amplitude threshold and the frequency offset exceeds the offset threshold, the corresponding dynamic flow vector is extracted to determine the three-dimensional coordinates of each node in the cabinet grounding system and obtain the current vector.

[0044] The current vector is projected onto the three-dimensional mesh model;

[0045] By combining the Euler-Lagrange particle tracking method, the diffusion path of leakage current is simulated. When the current density exceeds a critical value at a certain coordinate point, that coordinate point is determined to be a micro-leak point.

[0046] Secondly, the present invention provides a cabinet grounding current detection device based on a current sensor, comprising:

[0047] The current value acquisition module is used to acquire the current value of each node in the cabinet grounding system in real time through sensors, and record the current dynamic sequence by collecting the current value per second.

[0048] The path division module is used to calculate the arrival delay time of current signals between different nodes based on the current dynamic sequence, and to determine the synchronization strength and shunt attenuation rate of current signals between nodes based on the delay time, and to divide the main path and branch paths based on the shunt attenuation rate.

[0049] The peak module is used to extract the peak frequencies of the main path and the branch path through fast Fourier transform, and to obtain the peak offset of each path by combining it with a preset frequency threshold.

[0050] The abnormal time period acquisition module is used to monitor the path current of the main path and the branch path through the peak offset. When the path current exceeds the abnormal trigger threshold, the time series data within the reconstruction time window of the path is extracted to obtain the abnormal time period.

[0051] The backtracking module is used to backtrack historical current data in memory based on the abnormal time period, generate a dynamic flow direction vector, and obtain the complete flow trajectory of the abnormal current.

[0052] The positioning module is used to locate the current change position between nodes at the point of abrupt change in flow direction based on the complete flow trajectory, as the preliminary coordinate position of the anomaly source.

[0053] The feature analysis module is used to extract the sensor current data of the corresponding node based on the preliminary coordinate position, remove noise interference and perform feature analysis to obtain the accurate current characteristics of the abnormal source.

[0054] The leakage point output module is used to compare the precise current characteristics with the preset normal current characteristics to obtain the comparison result. Based on the comparison result, the current vector is projected onto the three-dimensional mesh model of the cabinet grounding system, and the diffusion path of the leakage current is simulated to determine the tiny leakage point.

[0055] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the cabinet grounding current detection method based on a current sensor as described above.

[0056] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the cabinet grounding current detection method based on a current sensor as described above.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] (1) The present invention collects current values ​​in real time through sensors and, based on the current dynamic sequence recorded every second, can capture the real-time changes of current and ensure high sensitivity monitoring of dynamic changes.

[0059] (2) By calculating the delay time and shunt attenuation rate of current signals between different nodes, the present invention divides the main path and branch path, thereby enabling a comprehensive analysis of the multi-path distribution of current and overcoming the limitations of the prior art.

[0060] (3) This invention extracts the peak frequencies of the main path and branch paths through Fast Fourier Transform and obtains the peak offset by combining it with a preset frequency threshold. This frequency domain analysis method can capture subtle changes in the current signal and provide a basis for anomaly warning.

[0061] (4) By retrospectively analyzing abnormal time periods and generating dynamic flow vectors, this invention can completely reconstruct the flow path of abnormal currents, providing a clear trajectory for fault location.

[0062] (5) By locating the flow change point and analyzing the characteristics of sensor data, combined with the simulation of a three-dimensional mesh model, the present invention can accurately locate the source of the anomaly and provide detailed current characteristics.

[0063] (6) By projecting precise current characteristics onto a three-dimensional mesh model and simulating the diffusion path of leakage current, this invention can intuitively present the propagation path of abnormal current and provide guidance for fault repair.

[0064] In summary, this invention comprehensively addresses the shortcomings of existing technologies in grounding current monitoring and anomaly tracing through real-time dynamic monitoring, multi-path analysis, frequency domain feature extraction, anomaly tracing, precise location, and three-dimensional simulation, significantly improving the monitoring sensitivity and fault location accuracy of the cabinet grounding system. Attached Figure Description

[0065] Figure 1 This is a schematic diagram of a cabinet grounding current detection method based on a current sensor provided in the first embodiment of the present invention;

[0066] Figure 2 This is a schematic diagram of a cabinet grounding current detection system based on a current sensor, provided in the second embodiment of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Reference Figure 1 The first embodiment of the present invention provides a method for detecting cabinet grounding current based on a current sensor, including the following steps:

[0069] S11 uses sensors to acquire the current values ​​of each node in the cabinet grounding system in real time, and records the current dynamic sequence by collecting the current values ​​per second.

[0070] S12, calculate the arrival delay time of current signals between different nodes based on the current dynamic sequence, and determine the synchronization strength and shunt attenuation rate of current signals between nodes based on the delay time, and divide the main path and branch path based on the shunt attenuation rate.

[0071] S13, extract the peak frequencies of the main path and the branch path through fast Fourier transform, and obtain the peak offset of each path by combining the preset frequency threshold.

[0072] S14, monitor the path current of the main path and the branch path through the peak offset. When the path current exceeds the abnormal trigger threshold, extract the time series data within the reconstruction time window of the path to obtain the abnormal time period.

[0073] S15, based on the abnormal time period, backtrack the historical current data in memory, generate a dynamic flow direction vector, and obtain the complete flow trajectory of the abnormal current.

[0074] S16, Based on the complete flow trajectory, locate the current change position between nodes at the point of sudden change in flow direction, and use it as the preliminary coordinate position of the anomaly source.

[0075] S17. Based on the preliminary coordinate position, extract the sensor current data of the corresponding node, remove noise interference and perform feature analysis to obtain the accurate current characteristics of the abnormal source.

[0076] S18, compare the precise current characteristics with the preset normal current characteristics to obtain the comparison result, project the current vector onto the three-dimensional mesh model of the cabinet grounding system according to the comparison result, and simulate the diffusion path of the leakage current to determine the micro-leakage point.

[0077] In step S11, the real-time acquisition of current values ​​at each node in the cabinet grounding system via sensors, and the recording of the current dynamic sequence at a rate of current values ​​collected per second, includes:

[0078] Sensor nodes are deployed through a distributed network to collect the current values ​​of each node in the cabinet grounding system;

[0079] The current value is sampled at high frequency and recorded at a rate of one current value per second. A dynamic current sequence reflecting the dynamic changes is generated using a sliding window.

[0080] It should be noted that sensor nodes are deployed through a distributed network to collect current values ​​at each node in the rack grounding system. Specifically, non-invasive, high-precision current sensors are used to avoid interference with the existing grounding system. Sensors are deployed at key physical topology nodes of the rack grounding system (such as grounding bus connection points and branch path intersections). The minimum deployment density is determined according to the rack size and structural complexity, following a grid-like principle. Redundant sensors are added to critical high-risk areas (such as easily corroded or high-load nodes) to improve fault tolerance. By deploying sensor nodes through a distributed network, all critical paths in the rack grounding system can be covered, avoiding the blind spots of traditional single-point detection and enabling real-time capture of dynamic changes in current across multiple paths.

[0081] It should be noted that the current value is sampled at high frequency, recorded at a rate of one current value per second. The continuous time-series data is divided into fixed-length windows using a sliding window, and the data is processed in segments. A set of features, including mean, frequency, and phase components, is generated according to a preset step size. High-frequency sampling can completely record the details of the current waveform, avoiding the omission of anomalies. Segmented data processing reduces computational complexity, and processing only one set of data at a time enables rapid capture of transient anomalies.

[0082] For example, in a cabinet grounding system, there is a node N1. A sensor acquires the current value of node N1 in real time, performing high-frequency sampling at a frequency of 1000 times per second. Within 1 second, sensor N1 collects 1000 current values ​​(unit: mA) [5.21, 5.18, 5.23, ..., 5.15]. The preset sliding window step size is 0.1 seconds, dividing the current value of 1 second into 0.1 seconds / window, including 100 sampling points, for a total of 10 windows. The resulting current dynamic sequence is: timestamp: 58.0 seconds, N1: [[Window 1: [5.20, 50, 0.5]][Window 2: [5.19, 50, 0.48]]...[Window 10: [5.18, 50, 0.49]]].

[0083] In step S12, the arrival delay time of current signals between different nodes is first calculated based on the current dynamic sequence. Specifically, a normalized cross-correlation function is used to analyze the current dynamic sequence of different nodes. When the cross-correlation function reaches its maximum value, the corresponding time offset is the delay time. The normalized cross-correlation function is defined as:

[0084]

[0085] in, and Let be the current dynamic sequences of nodes A and B, respectively, where t = 1, 2, ..., N, and N is the signal length. This is the time offset. and for and The mean, after normalization ∈[-1,1]. Delay time is a core parameter for dynamic path analysis. By revealing the time propagation characteristics of current signals, it provides a key basis for path division, anomaly detection, and fault tracing, thereby overcoming the problems of insensitive dynamic monitoring and difficult tracing in existing technologies.

[0086] For example, in a cabinet grounding system, the current signal propagation paths of node A and node B are different, resulting in a time difference in the signals. The dynamic current sequence of node A is as follows: =[10.0,12.0,15.0,14.0,13.0] (unit: mA) The current dynamic sequence of node B is =[10.0,10.0,12.0,11.0,10.0] (unit: mA). Calculations show that when... =0.001 seconds, =0.814, The maximum value indicates that the signal at node B is delayed by 0.001 seconds compared to that at node A.

[0087] In one implementation, determining the synchronization strength and shunt attenuation rate of the inter-node current signal based on the delay time includes:

[0088] The reciprocal of the delay time is calculated and normalized to obtain the synchronization strength of the inter-node current signal;

[0089] The convergence point of the path is determined by identifying the concentrated signal area. The attenuation change value is calculated based on the current signal strength of the convergence point and the current signal strength of other nodes connected to it, and the shunt attenuation rate is obtained.

[0090] It should be noted that in this embodiment, the synchronization strength of the inter-node current signal is obtained by calculating the reciprocal of the delay time and normalizing it. in, Let be the delay time for node A and node B. The maximum allowable latency for the system is determined based on the rack size and signal propagation speed. Less than Synchronization strength is a quantitative indicator of the consistency of current signal propagation time between nodes. In essence, it infers the temporal consistency of the current propagation path by delaying the time. It can optimize the anomaly detection range, enhance the ability to capture dynamic anomalies, and provide both temporal and path evidence for fault tracing, overcoming the limitations of traditional static methods.

[0091] It should be noted that in the cabinet grounding system, there is a node in the signal concentration area with the strongest current signal strength and stable signal fluctuation; this node is designated as the convergence point. After determining the convergence point, the difference between the current signal strength of other nodes connected to it and the current signal strength of the convergence point is calculated to obtain the attenuation change value. The absolute value of the ratio of the attenuation change value to the current signal strength of the convergence point is the shunt attenuation rate.

[0092] For example, in a cabinet grounding system, the delay time between node A and node B is calculated to be 0.001 seconds with a synchronization strength of 90% using a cross-correlation function, and the delay time between node A and node C is 0.005 seconds with a synchronization strength of 30%. Node B is identified as the convergence point by identifying the signal concentration area. The current signal strength at node B is 9mA, and the current strength at node C is 3mA. The shunt attenuation rate was found to be 67%, and the current intensity at node A was 10mA. The shunt attenuation rate was found to be 10%.

[0093] In one implementation, the step of dividing the main path and branch paths according to the shunt attenuation rate includes:

[0094] The shunt attenuation rate is compared with a preset attenuation rate threshold, and the synchronization strength is compared with a preset synchronization strength threshold.

[0095] When the shunt attenuation rate is less than or equal to the attenuation rate threshold and the synchronization strength is greater than or equal to the synchronization strength threshold, the inter-node current signal transmission path is determined to be the mainstream path.

[0096] When the shunt attenuation rate is greater than the attenuation rate threshold or the synchronization strength is less than the synchronization strength threshold, the inter-node current signal transmission path is determined to be the branch path.

[0097] It should be noted that synchronization strength represents the synchronicity of current signals between nodes, reflecting the real-time nature of signal transmission and helping to identify the physical connection quality of the current path; shunt attenuation rate quantifies the current distribution ratio and is used to locate abnormal shunts. Current in the mainstream path propagates through low-impedance trunk lines, resulting in short paths, uniform impedance, small signal delay time, and high synchronization strength. Simultaneously, the current is concentrated, with few shunts and a low shunt attenuation rate. When current passes through branch paths, the delay time may increase due to path detours, contact impedance, or electromagnetic interference, resulting in low synchronization strength. Furthermore, the current may be distributed to multiple branches, or the line impedance may be higher, leading to high shunt attenuation. Therefore, paths with high synchronization strength and low shunt attenuation rate are identified as mainstream paths, while paths with low synchronization strength or high shunt attenuation rate are identified as branch paths.

[0098] For example, based on experimental data, the attenuation rate threshold is set to 10%, and the synchronization strength threshold is set to 80%. For instance, the shunt attenuation rate from node A to node B is 10%, and the synchronization strength is 90%. The shunt attenuation rate from node B to node C is 67%, and the synchronization strength is 30%. The shunt attenuation rate from node A to node B equals the attenuation rate threshold, and the synchronization strength is greater than the synchronization strength threshold. Therefore, node A to node B is the main path. The shunt attenuation rate from node B to node C is greater than the attenuation rate threshold, and the synchronization strength is less than the synchronization strength threshold. Therefore, node B to node C is a branch path.

[0099] In step S13, the step of extracting the peak frequencies of the main path and the branch paths through Fast Fourier Transform and obtaining the peak offset of each path by combining it with a preset frequency threshold includes:

[0100] Fast Fourier transform is performed on the current dynamic sequences of the main path and the branch path respectively to convert the time domain signal into the frequency domain signal and obtain the spectrum.

[0101] Calculate the amplitude of each frequency component in the spectrum;

[0102] The frequency point with the largest amplitude is identified as the peak frequency in the spectrum.

[0103] When the peak frequency exceeds a preset frequency threshold, the difference between the peak frequency and the frequency threshold is calculated to obtain the peak offset.

[0104] It should be noted that in this embodiment, a Fast Fourier Transform (FFT) is performed on the current dynamic sequence to convert the time-domain current signal into a spectrum. This decomposes the current signal of the current dynamic sequence into a superposition of sine waves of different frequencies, yielding a mainstream spectrum and a branch spectrum. Specifically, a FFT is performed on the current dynamic sequences of the mainstream path and the branch path, with a sampling frequency of 1000 Hz and 1024 sampling points, resulting in the mainstream spectrum and the branch spectrum. Normal current signals typically have stable frequency components, while abnormal conditions may lead to changes in frequency components or the emergence of new frequency components. By obtaining the spectrum through the FFT, these abnormal frequency components can be quickly identified.

[0105] It should be noted that in this embodiment, the amplitude of each frequency component in the spectrum is calculated. Specifically, the amplitude is calculated for the complex value of each frequency point. The amplitude can identify high-frequency noise or low-frequency fluctuations in the current signal and capture dynamic anomaly characteristics.

[0106] It should be noted that the frequency point with the largest amplitude identified in the spectrum is designated as the peak frequency. The peak frequency includes the mainstream peak frequency and branch peak frequencies. The mainstream peak frequency is used to determine whether the current is within the normal power frequency range. The peak frequency reflects the core mode of current oscillation in the system under normal or abnormal conditions. Specifically, the peak frequency of normal grounding current is usually stable near the power frequency (50 Hz), while leakage current caused by insulation degradation will generate high-frequency harmonic components.

[0107] For example, the amplitude of each frequency component in the mainstream spectrum and the branch spectrum are calculated respectively. The frequency point with the largest amplitude in the mainstream spectrum is identified as 50 Hz as the mainstream peak frequency, and the frequency point with the largest amplitude in the branch spectrum is identified as 60 Hz as the branch peak frequency.

[0108] It should be noted that in this embodiment, when the peak frequency exceeds the preset frequency threshold, the difference between the peak frequency and the frequency threshold is calculated to obtain the peak offset. The peak offset includes the mainstream peak offset and the branch peak offset. The peak offset quantifies the degree of frequency anomaly and provides a quantitative basis for fault classification and location.

[0109] For example, the frequency threshold can be set to 50 Hz. The mainstream peak frequency in the mainstream spectrum is 50 Hz, and the branch peak frequency in the branch spectrum is 60 Hz. The mainstream peak frequency is equal to the frequency threshold. There is no peak offset in the mainstream path. The branch peak frequency is greater than the frequency threshold. The difference between the branch peak frequency and the frequency threshold is calculated to obtain a branch peak offset of 10 Hz.

[0110] In step S14, the path current of the main path and the branch path is first monitored by the peak offset. Specifically, monitoring resources are dynamically allocated according to the peak offset, high-risk paths are prioritized, and detection efficiency is improved. This is especially suitable for complex multi-branch cabinet grounding systems, significantly improving the fault detection speed and location accuracy of the cabinet grounding system, and reducing operation and maintenance costs.

[0111] In one implementation, when the path current exceeds the abnormal trigger threshold, the timing data within the reconstruction time window of the path is extracted to obtain the abnormal time period, including:

[0112] When the current in the path exceeds the abnormal trigger threshold, the timing data of all nodes of the path within the reconstruction time window are retrieved from memory.

[0113] The standard deviation of the time series data is calculated. When the standard deviation of a continuous time interval exceeds the fluctuation threshold, the continuous intervals are merged to obtain the abnormal time period.

[0114] It should be noted that when the path current exceeds the abnormal trigger threshold, the timing data of all nodes of the path within the reconstruction time window is retrieved from memory. The reconstruction time window includes a preset fixed window and an adaptive window. The preset fixed window is a time range pre-set according to the system design, while the adaptive window is dynamically adjusted based on the dynamic characteristics of the signal. Specifically, timing data is retrieved only when an abnormality is triggered, rather than continuously storing all data. This provides a high-quality data foundation for subsequent source tracing, saves memory resources, and is suitable for long-term operating monitoring systems.

[0115] For example, based on experimental data, an anomaly trigger threshold is set to 1 Hz, and the path current of the main path and the branch path is monitored. The branch peak offset is 10 Hz. When the anomaly trigger threshold is exceeded, the timing data within the reconstruction time window is extracted. The fixed window is 1 second, and the adaptive window is dynamically adjusted by 0.5 seconds according to the signal fluctuation frequency to obtain the path current from 0 to 1.4 seconds.

[0116] It should be noted that in this embodiment, the standard deviation of the time-series data is calculated according to time intervals to measure the current fluctuation amplitude. When the standard deviation exceeds the fluctuation threshold, the interval is marked as an abnormal candidate interval. Multiple consecutive abnormal candidate intervals are merged to obtain the abnormal time period. Specifically, the abnormal time period is shortened from the hour level to the minute level, which significantly improves operation and maintenance efficiency and can accurately capture abnormal time periods. Through dynamic windowing and standard deviation analysis, misjudgments caused by instantaneous interference are avoided, and only the truly abnormal and continuous time periods are extracted.

[0117] For example, the fluctuation threshold can be set to 0.5mA. The standard deviation of the current within each 0.5-second time interval is calculated, resulting in a standard deviation of 0.1414mA for the time interval 0.0-0.5 seconds, 2.5mA for the time interval 0.5-1.0 seconds, and 0.1414mA for the time interval 1.0-1.5 seconds. The standard deviation of 2.5mA in the time interval 0.5-1.0 seconds exceeds the fluctuation threshold, indicating an abnormal time period of 0.5-1.0 seconds.

[0118] In step S15, the step of generating a dynamic flow direction vector by backtracking historical current data in memory based on the abnormal time period to obtain the complete flow trajectory of the abnormal current includes:

[0119] Based on the abnormal time period, extract the historical current value corresponding to the time period from memory;

[0120] Calculate the current flow direction and intensity between each node to generate a dynamic flow direction vector;

[0121] By analyzing the changing trend of the dynamic flow direction vector, the complete flow trajectory of the abnormal current is obtained.

[0122] It should be noted that, based on the abnormal time period, historical current values ​​for the corresponding time period are extracted from memory. The current values ​​must be aligned by timestamps to ensure the correlation between current values ​​from different nodes at the same moment. This, combined with timing correlation, eliminates transient interference (such as random noise) and improves fault location accuracy.

[0123] For example, in a cabinet grounding system, there are four nodes A, B, C, and D. The connections between the nodes are as follows: node A is connected to nodes B and C, node B is connected to node D, and node C is connected to node D. Based on the abnormal time period of 0.5-1.0 seconds, the historical current values ​​for the corresponding time period are extracted from memory.

[0124] It should be noted that the current flow direction is determined by the temporal relationship of current changes between adjacent nodes. The integral value of the current change within a sliding window is used as the flow intensity. A dynamic flow direction vector set is generated for the abnormal time period window, with vector elements being (source node, target node, flow intensity). This dynamic flow direction vector represents the current direction and magnitude. In cabinet grounding current detection, the dynamic flow direction vector quantifies the changes in current flow direction and intensity, intuitively reflecting the actual current distribution. Its function is to clarify the propagation path of abnormal current and pinpoint the origin of the anomaly through spatial and temporal analysis.

[0125] In the cabinet grounding system, there are four nodes A, B, C, and D. During the abnormal time window of 0.5-1.0 seconds, the flow intensity from node A to node B is 1 mA, and the flow direction is from node A to node B. The dynamic flow direction vector is obtained. =(A,B,1A), the flow intensity from node A to node C is 2mA, and the flow direction is from node A to node C, thus obtaining the dynamic flow direction vector. =(A,C,1A), where the flow intensity from node D to node B is 1mA, and the flow direction is from node B to node D, thus obtaining the dynamic flow direction vector. Given the flow vector = (B, D, 1A), the flow intensity from node C to node D is 1 mA, and the flow direction is from node C to node D. =(C,D,1A).

[0126] It should be noted that by analyzing the changing trend of the dynamic flow vector, and connecting the dynamic flow vectors in chronological order, a current propagation chain is formed, yielding the complete flow trajectory of the abnormal current. This enables real-time tracking of the propagation path of the abnormal source, avoiding cascading faults caused by grounding failures.

[0127] For example, in a cabinet grounding system there are four nodes A, B, C, and D. Analyze the dynamic flow vector. =(A,B,1A) =(A,C,1A) =(B,D,1A) and =(C,D,1A), starting from node A, the current flows from node A to node B, then from node B to node D, and then from node A to node C, and then from node C to node D, thus obtaining the complete flow trajectory of the abnormal current. The main path is A → B → D, and the branch path is A → C → D.

[0128] In step S16, the step of locating the current change position between nodes at the point of abrupt change in flow direction based on the complete flow trajectory, as the preliminary coordinate position of the anomaly source, includes:

[0129] Based on the complete flow trajectory, calculate the rate of change of current between the node and its connected adjacent nodes. When the rate of change of current exceeds a preset current change threshold, it is marked as a point of sudden change in flow direction.

[0130] When the flow direction change point is detected, the location of the current change is determined by combining the dynamic flow direction vector corresponding to the flow direction change point, which serves as the preliminary coordinate location of the anomaly source.

[0131] It should be noted that the calculation of the current change rate between a node and its connected adjacent nodes is used. When the current change rate exceeds a preset current change threshold, it is marked as a flow direction abrupt change point. Specifically, the absolute value of the current difference between adjacent nodes is calculated and divided by a fixed time interval to obtain the current change rate of adjacent nodes. The current change rate reflects the energy attenuation or enhancement characteristics of the current in the flow. When the current change rate exceeds the preset current change threshold, it is marked as a flow direction abrupt change point. Specifically, the preset current change threshold can be adjusted according to system characteristics to balance sensitivity and false alarm rate. After marking the abrupt change point, the investigation scope can be quickly narrowed down, guiding subsequent feature analysis. This method is suitable for complex topologies and solves the problem of tracing the source of multi-path currents.

[0132] For example, the current change threshold can be set to 3A / s. In the cabinet grounding system, there are two adjacent nodes A and B. The current difference between node A and node B is 5.0mA. The fixed time interval is 1 second. 5.0 / 1 gives the current change rate per unit distance between node A and node B as 5A / s, which exceeds the current change threshold. Node B is marked as a point of sudden change in current direction.

[0133] It should be noted that, in this embodiment, when the abrupt change in flow direction is detected, the location of the current change is determined by combining the dynamic flow direction vector corresponding to the abrupt change point, serving as the preliminary coordinate location of the anomaly source. This approach is applicable to complex topologies and solves the problem of tracing the source of multi-path currents.

[0134] For example, in a cabinet grounding system containing nodes A (0,0,0) and B (1,0,0), the detected flow direction change point is at nodes A and B. Taking the center of AB (0.5,0,0), and combining it with the dynamic flow direction vector... =(A,B,1A), determine the location of the current change as node B, and determine the physical location coordinates of node B (1,0,0) as the initial coordinate position of the anomaly source.

[0135] In step S17, the step of extracting sensor current data of the corresponding node based on the preliminary coordinate position, removing noise interference, and performing feature analysis to obtain the precise current characteristics of the anomaly source includes:

[0136] Extract the sensor current data of the corresponding node based on the preliminary coordinate position;

[0137] By combining the reconstruction error range, noise interference is removed from the sensor current data to obtain effective current data;

[0138] The current time-domain characteristics of the anomaly source are obtained by calculating the mean and variance of the effective current data.

[0139] The effective current data is analyzed in the frequency domain using the Fast Fourier Transform algorithm to obtain the current frequency domain characteristics of the anomaly source.

[0140] The amplitude, frequency, and phase information in the current time-domain features and the current frequency-domain features are used as precise current features of the anomaly source.

[0141] It should be noted that in this embodiment, the sensor current data of the corresponding node is extracted based on the preliminary coordinate position. The sensor current data includes time domain signals and frequency domain signals, which can improve the accuracy of anomaly source location, narrow down the range of anomaly sources, and provide detailed data support for accurate current feature analysis.

[0142] It should be noted that a wavelet thresholding denoising algorithm is employed, using the db4 wavelet basis function and a decomposition layer of 5 to separate high-frequency noise and low-frequency features in the signal. By setting the wavelet threshold to 0.15, the system performs soft thresholding on the high-frequency components, eliminating noise interference and retaining the valid signal. A reconstruction error range is set; when the reconstructed signal of a frequency band exceeds the reconstruction error range, that frequency band is determined to be noise and set to zero, yielding valid current data. The reconstruction error range can be set to ±0.01mA. Electromagnetic interference is common in cabinet grounding systems, and direct analysis of the raw data may lead to misjudgment. Combining the reconstruction error range with noise interference removal from the sensor current data preserves the true fault characteristics.

[0143] For example, the sensor current data of node B exhibits drastic fluctuations of [0.5A, 1.8A, 0.6A, 2.1A,...] within 1 second. After denoising, the low-frequency trend component [0.5A, 0.7A, 0.6A, 0.8A, ...] is retained, eliminating the spikes caused by high-frequency noise.

[0144] It should be noted that in this embodiment, the effective current data is calculated using the mean and variance to obtain the current time-domain characteristics of the anomaly source. The mean reflects the average amplitude level of the current and is used to detect whether there is persistent leakage or abnormal grounding resistance. The variance reflects the intensity of current fluctuations and is used to capture intermittent faults. By combining these two parameters, the stability and fluctuation patterns of the anomaly source can be quickly identified.

[0145] It should be noted that the effective current data is analyzed in the frequency domain using a Fast Fourier Transform algorithm to obtain the current frequency domain characteristics of the anomaly source. By converting the current signal from the time domain to the frequency domain, harmonic components exceeding the normal frequency range, such as high-frequency noise and low-frequency oscillations, are detected. Fault types are distinguished by differences in spectral distribution. Current frequency domain characteristics are an important supplement to current time domain analysis, and are particularly suitable for detecting latent, periodic, or high-frequency transient anomalies.

[0146] It should be noted that, in this embodiment, the amplitude, frequency, and phase information in the current time-domain features and the current frequency-domain features are used as the precise current features of the anomaly source. Specifically, by combining and analyzing the current time-domain features and the current frequency-domain features, the stability and fluctuation patterns of the anomaly source can be quickly identified.

[0147] For example, the initial location of the anomaly source is node B. Current data from node B is extracted, and noise interference is removed from the sensor current data using a reconstruction error range of ±0.01mA to obtain valid current data. The mean and variance of the valid current data are calculated to obtain the time-domain characteristics of the anomaly source's current, with a mean of 4.25mA and a variance of 0.0292. A fast Fourier transform algorithm is then used to perform frequency domain analysis on the valid current data to obtain the frequency-domain characteristics of the anomaly source's current, with an amplitude of 4.8dB, a frequency of 55Hz, and a phase of 20°, thus obtaining the precise current characteristics of the anomaly source.

[0148] In step S18, the precise current characteristic is compared with a preset normal current characteristic to obtain a comparison result. Based on the comparison result, the current vector is projected onto a three-dimensional mesh model of the cabinet grounding system, and the diffusion path of the leakage current is simulated to determine the micro-leakage point, including:

[0149] The precise current characteristics are compared with the preset normal current characteristics. When the amplitude exceeds the amplitude threshold and the frequency offset exceeds the offset threshold, the corresponding dynamic flow vector is extracted to determine the three-dimensional coordinates of each node in the cabinet grounding system and obtain the current vector.

[0150] The current vector is projected onto the three-dimensional mesh model;

[0151] By combining the Euler-Lagrange particle tracking method, the diffusion path of leakage current is simulated. When the current density exceeds a critical value at a certain coordinate, that coordinate point is determined to be a micro-leak point.

[0152] It should be noted that the precise current characteristics are compared with preset normal current characteristics. The comparison results include amplitude exceeding an amplitude threshold and frequency offset exceeding an offset threshold, amplitude not exceeding an amplitude threshold, or frequency offset not exceeding an offset threshold. When the amplitude exceeds the amplitude threshold and the frequency offset exceeds the offset threshold, the corresponding dynamic flow vector is extracted. When the amplitude does not exceed the amplitude threshold or the frequency offset does not exceed the offset threshold, it is determined to be a normal fluctuation, and the current characteristics of that path are marked as "normal state," without triggering the leak point location process, thus avoiding misjudgment due to instantaneous interference or normal fluctuations. Comparing the precise current characteristics with preset normal characteristics can more comprehensively identify abnormal situations and avoid misdiagnosis caused by misjudgment of a single feature. The amplitude threshold is set according to the fluctuation range of the current amplitude during normal operation and is used to distinguish between normal fluctuations and abnormal situations. The frequency offset threshold is used to identify abnormal changes in the current frequency.

[0153] Specifically, when the amplitude exceeds the amplitude threshold and the frequency offset exceeds the offset threshold, the corresponding dynamic flow vector is extracted to determine the three-dimensional coordinates of each node in the rack grounding system, thus obtaining the current vector. The current vector contains information about the direction and magnitude of the current, describing its flow in three-dimensional space. This transforms the abstract current characteristics into a concrete three-dimensional spatial vector, accurately locating the source and flow path of abnormal currents, which helps to quickly determine the fault location and reduce troubleshooting time. Specifically, based on the structure of the rack grounding system and the sensor placement, the three-dimensional coordinates of each node are pre-established. These three-dimensional coordinates are determined based on the rack's physical layout and the actual sensor installation positions, with the rack's geometric center as the origin, and are used to accurately locate each node in three-dimensional space.

[0154] For example, the normal current characteristics can be set to an amplitude range of [3.5, 4.5] dB, a frequency range of [49, 51] Hz, a phase range of [0, 10]°, and an offset threshold of 2 Hz. The precise current characteristics of node B are extracted, with a mean of 4.25 mA, a variance of 0.0292, an amplitude of 4.8 dB, a frequency of 55 Hz, and a phase of 20°. Comparing these precise current characteristics with the preset normal current characteristics, if the amplitude exceeds the amplitude threshold and the frequency offset exceeds the offset threshold, the corresponding dynamic flow vector is extracted. =(A,B,1A / s), Based on the layout of the cabinet grounding system, determine the three-dimensional coordinates (1, 2, 3) of node B, and obtain the current vector with a current intensity of 4.8mA. The current flows from node A to B with a direction vector (0,0,1).

[0155] It should be noted that the current vector is projected onto the three-dimensional mesh model. Specifically, the three-dimensional mesh model is a pre-built three-dimensional model of the cabinet grounding system, used for visualizing and analyzing current flow. This model divides the cabinet grounding system into multiple mesh cells, each corresponding to a set of nodes. The generated current vectors are mapped onto the three-dimensional mesh model, with each current vector corresponding to a mesh cell. In this way, the current flow path and intensity can be visually displayed in the three-dimensional model, and interference signals from adjacent nodes can be distinguished.

[0156] It should be noted that, by combining the Eulerian-Lagrange particle tracking method to simulate the diffusion path of leakage current, in the three-dimensional mesh model, the initial coordinates of the anomaly source are used as the current source, and a preset leakage current value is injected. These particles move according to the direction and magnitude of the current vector. When the current density exceeds a critical value at a certain coordinate, that coordinate point is determined to be a micro-leakage point. The critical value is determined based on the statistical significance principle. Criteria, setting a critical value ,in, This represents the average current density under normal conditions. This represents the standard deviation of the current density under normal conditions. By simulating the diffusion path of leakage current, the location of minute leak points can be determined more accurately, improving the detection sensitivity of minute leak points by more than 10 times and enhancing the accuracy of fault diagnosis.

[0157] For example, in the normal region, the particle density is uniformly distributed with a mean of 50 particles per unit volume and a standard deviation of 5. Therefore, the critical value can be set to 65. The three-dimensional coordinates of node B are (1, 2, 3), and the current vector is a current intensity of 4.8 mA. The current flows from node A to B with a direction vector of (0,0,1). The current vector is projected onto the three-dimensional mesh model from (1,2, 3) to (1, 2, 4). Using the Euler-Lagrange particle tracking method, a preset leakage current value of 0.1 mA is injected at node B. Based on the current vector and the current field in the three-dimensional mesh model, the trajectory of each particle is observed, and the current density at each mesh point is calculated. If the current density near node B (1, 2, 4) significantly increases to 70 particles per unit volume, exceeding the critical value, then this coordinate point is determined to be a micro-leakage point.

[0158] In summary, this invention discloses a cabinet grounding current detection method and system based on a current sensor. By monitoring the current values ​​of each node in the cabinet grounding system in real time, it can promptly capture minute changes in current, solving the problem of insufficient sensitivity to dynamic changes in existing technologies. By dividing the current into main and branch paths, it provides a more accurate understanding of the current distribution in the cabinet grounding system, prioritizing the analysis of high-energy, low-attenuation main paths to reduce the impact of branch path noise on source tracing. Furthermore, it combines time-domain (dynamic sequence), frequency-domain (peak offset), and spatial-domain (3D grid) analysis to avoid the limitations of single-dimensional information. Simultaneously, the dynamic flow vector, by associating historical data with real-time anomalies, constructs a causal chain, avoiding misjudgments based solely on instantaneous data. The spatial projection of the 3D model further binds electrical characteristics with physical location, improving positioning accuracy and thus enhancing the accuracy of anomaly source tracing.

[0159] Reference Figure 2 The second embodiment of the present invention provides a cabinet grounding current detection system based on a current sensor, comprising:

[0160] The current value acquisition module is used to acquire the current value of each node in the cabinet grounding system in real time through sensors, and record the current dynamic sequence by collecting the current value per second.

[0161] The path division module is used to calculate the arrival delay time of current signals between different nodes based on the current dynamic sequence, and to determine the synchronization strength and shunt attenuation rate of current signals between nodes based on the delay time, and to divide the main path and branch paths based on the shunt attenuation rate.

[0162] The peak module is used to extract the peak frequencies of the main path and the branch path through fast Fourier transform, and to obtain the peak offset of each path by combining it with a preset frequency threshold.

[0163] The abnormal time period acquisition module is used to monitor the path current of the main path and the branch path through the peak offset. When the path current exceeds the abnormal trigger threshold, the time series data within the reconstruction time window of the path is extracted to obtain the abnormal time period.

[0164] The backtracking module is used to backtrack historical current data in memory based on the abnormal time period, generate a dynamic flow direction vector, and obtain the complete flow trajectory of the abnormal current.

[0165] The positioning module is used to locate the current change position between nodes at the point of abrupt change in flow direction based on the complete flow trajectory, as the preliminary coordinate position of the anomaly source.

[0166] The feature analysis module is used to extract the sensor current data of the corresponding node based on the preliminary coordinate position, remove noise interference and perform feature analysis to obtain the accurate current characteristics of the abnormal source.

[0167] The leakage point output module is used to compare the precise current characteristics with the preset normal current characteristics to obtain the comparison result. Based on the comparison result, the current vector is projected onto the three-dimensional mesh model of the cabinet grounding system, and the diffusion path of the leakage current is simulated to determine the tiny leakage point.

[0168] It should be noted that the cabinet grounding current detection system based on a current sensor provided in this embodiment of the invention is used to execute all the process steps of the cabinet grounding current detection method based on a current sensor in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0169] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the various embodiments of the cabinet grounding current detection method based on current sensors described above, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module in the above-described device embodiments, such as the peak module.

[0170] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0171] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0172] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0173] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0174] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0175] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0176] 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. In particular, it should be noted that 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 for those skilled in the art.

Claims

1. A method for detecting cabinet grounding current based on a current sensor, characterized in that, include: The current values ​​of each node in the cabinet grounding system are acquired in real time by sensors, and the current dynamic sequence is obtained by recording the current values ​​collected per second. The arrival delay time of current signals between different nodes is calculated based on the current dynamic sequence, and the synchronization strength and shunt attenuation rate of current signals between nodes are determined based on the delay time. The main path and branch path are divided based on the shunt attenuation rate. The peak frequencies of the main path and the branch paths are extracted by fast Fourier transform, and the peak offset of each path is obtained by combining the preset frequency threshold. The path current of the main path and the branch path is monitored by the peak offset. When the path current exceeds the abnormal trigger threshold, the time series data within the reconstruction time window of the path is extracted to obtain the abnormal time period. Based on the abnormal time period, the historical current data in memory is traced back to generate a dynamic flow direction vector, and the complete flow trajectory of the abnormal current is obtained. Based on the complete flow trajectory, the current change locations between nodes at the points of abrupt change in flow direction are used as the preliminary coordinate locations of the anomaly source. Based on the preliminary coordinate positions, the sensor current data of the corresponding nodes are extracted, noise interference is removed, and feature analysis is performed to obtain the accurate current characteristics of the anomaly source. The precise current characteristics are compared with the preset normal current characteristics to obtain the comparison results. Based on the comparison results, the current vector is projected onto the three-dimensional mesh model of the cabinet grounding system, and the diffusion path of the leakage current is simulated to determine the micro-leakage point. The step of determining the synchronization strength and shunt attenuation rate of the inter-node current signal based on the delay time includes: The reciprocal of the delay time is calculated and normalized to obtain the synchronization strength of the inter-node current signal; The convergence point of the path is determined by identifying the concentrated signal area. The attenuation change value is calculated based on the current signal strength of the convergence point and the current signal strength of other nodes connected to it, and the shunt attenuation rate is obtained.

2. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, The method of dividing the main path and branch paths according to the shunt attenuation rate includes: The shunt attenuation rate is compared with a preset attenuation rate threshold, and the synchronization strength is compared with a preset synchronization strength threshold. When the shunt attenuation rate is less than or equal to the attenuation rate threshold and the synchronization strength is greater than or equal to the synchronization strength threshold, the inter-node current signal transmission path is determined to be the mainstream path. When the shunt attenuation rate is greater than the attenuation rate threshold or the synchronization strength is less than the synchronization strength threshold, the inter-node current signal transmission path is determined to be the branch path.

3. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, The step of extracting the peak frequencies of the main path and the branch paths using Fast Fourier Transform, and then combining this with a preset frequency threshold to obtain the peak offset of each path, includes: Fast Fourier transform is performed on the current dynamic sequences of the main path and the branch path respectively to convert the time domain signal into the frequency domain signal and obtain the spectrum. Calculate the amplitude of each frequency component in the spectrum; The frequency point with the largest amplitude is identified as the peak frequency in the spectrum. When the peak frequency exceeds a preset frequency threshold, the difference between the peak frequency and the frequency threshold is calculated to obtain the peak offset.

4. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, When the path current exceeds the abnormal trigger threshold, the timing data within the reconstruction time window of that path is extracted to obtain the abnormal time period, including: When the path current exceeds the abnormal trigger threshold, the timing data of all nodes of the path within the reconstruction time window are retrieved from memory, wherein the reconstruction time window includes a preset fixed window and an adaptive window. The standard deviation of the time series data is calculated. When the standard deviation of a continuous time interval exceeds the fluctuation threshold, the continuous time intervals are merged to obtain the abnormal time period.

5. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, The step of backtracking historical current data in memory based on the abnormal time period to generate a dynamic flow direction vector and obtain the complete flow trajectory of the abnormal current includes: Based on the abnormal time period, extract the historical current value corresponding to the time period from memory; Calculate the current flow direction and intensity between each node to generate a dynamic flow direction vector; By analyzing the changing trend of the dynamic flow direction vector, the complete flow trajectory of the abnormal current can be obtained.

6. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, The step of locating the current change position between nodes at the point of abrupt change in flow direction based on the complete flow trajectory, as the preliminary coordinate position of the anomaly source, includes: Based on the complete flow trajectory, calculate the rate of change of current between the node and its connected adjacent nodes. When the rate of change of current exceeds a preset current change threshold, it is marked as a point of sudden change in flow direction. When the flow direction change point is detected, the location of the current change is determined by combining the dynamic flow direction vector corresponding to the flow direction change point, which serves as the preliminary coordinate location of the anomaly source.

7. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, The step of extracting sensor current data of the corresponding node based on the preliminary coordinate position, removing noise interference, and performing feature analysis to obtain the precise current characteristics of the anomaly source includes: Extract the sensor current data of the corresponding node based on the preliminary coordinate position; By combining the reconstruction error range, noise interference is removed from the sensor current data to obtain effective current data; The current time-domain characteristics of the anomaly source are obtained by calculating the mean and variance of the effective current data. The effective current data is analyzed in the frequency domain using the Fast Fourier Transform algorithm to obtain the current frequency domain characteristics of the anomaly source. The amplitude, frequency, and phase information in the current time-domain features and the current frequency-domain features are used as precise current features of the anomaly source.

8. The cabinet grounding current detection method based on a current sensor according to claim 1, characterized in that, The step of comparing the precise current characteristics with preset normal current characteristics, projecting the current vector onto a three-dimensional mesh model of the cabinet grounding system based on the comparison result, and simulating the diffusion path of leakage current to determine minute leakage points includes: The precise current characteristics are compared with the preset normal current characteristics. When the amplitude exceeds the amplitude threshold and the frequency offset exceeds the offset threshold, the corresponding dynamic flow vector is extracted to determine the three-dimensional coordinates of each node in the cabinet grounding system and obtain the current vector. The current vector is projected onto the three-dimensional mesh model; By combining the Euler-Lagrange particle tracking method, the diffusion path of leakage current is simulated. When the current density exceeds a critical value at a certain coordinate point, that coordinate point is determined to be a micro-leak point.

9. A cabinet grounding current detection system based on a current sensor, characterized in that, The method for detecting cabinet grounding current based on a current sensor as described in any one of claims 1 to 8 includes: The current value acquisition module is used to acquire the current value of each node in the cabinet grounding system in real time through sensors, and record the current dynamic sequence by collecting the current value per second. The path division module is used to calculate the arrival delay time of current signals between different nodes based on the current dynamic sequence, and to determine the synchronization strength and shunt attenuation rate of current signals between nodes based on the delay time, and to divide the main path and branch paths based on the shunt attenuation rate. The peak module is used to extract the peak frequencies of the main path and the branch path through fast Fourier transform, and to obtain the peak offset of each path by combining it with a preset frequency threshold. The abnormal time period acquisition module is used to monitor the path current of the main path and the branch path through the peak offset. When the path current exceeds the abnormal trigger threshold, the time series data within the reconstruction time window of the path is extracted to obtain the abnormal time period. The backtracking module is used to backtrack historical current data in memory based on the abnormal time period, generate a dynamic flow direction vector, and obtain the complete flow trajectory of the abnormal current. The positioning module is used to locate the current change position between nodes at the point of abrupt change in flow direction based on the complete flow trajectory, as the preliminary coordinate position of the anomaly source. The feature analysis module is used to extract the sensor current data of the corresponding node based on the preliminary coordinate position, remove noise interference and perform feature analysis to obtain the accurate current characteristics of the abnormal source. The leakage point output module is used to compare the precise current characteristics with the preset normal current characteristics to obtain the comparison result. Based on the comparison result, the current vector is projected onto the three-dimensional mesh model of the cabinet grounding system, and the diffusion path of the leakage current is simulated to determine the tiny leakage point.

Images