A method and system for monitoring ground lines

Abstract

The application provides a grounding wire monitoring method and system, and relates to the technical field of grounding wire monitoring.The method comprises the following steps: collecting a voltage signal of an injection point and a current signal of a grounding loop; pre-processing the signals to obtain pre-processed signals; calculating impedance spectra of the grounding loop at different frequencies according to the pre-processed signals, and extracting key characteristic parameters in the impedance spectra; collecting environmental parameters and operating state data of power equipment, and fusing the key characteristic parameters to obtain fused data; adjusting a judgment threshold or a reference baseline corresponding to the key characteristic parameters by using the fused data; and judging the current grounding loop state by using the adjusted judgment threshold or reference baseline.The method of the application is based on impedance spectrum analysis, and overcomes electromagnetic interference and achieves the effects of adapting to dynamic baseline and distinguishing different electric leakage types through multi-stage signal processing, robust feature extraction and multi-dimensional data fusion analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ground wire monitoring, in particular to a ground wire monitoring method and system. BACKGROUND

[0002] In power substations, the safe operation of power equipment is of great importance, and reliable grounding connection is required. Traditional grounding wires rely on periodic offline detection and cannot be monitored in real time. Intelligent grounding wires integrate voltage detection and leakage monitoring functions, monitor leakage current through impedance spectrum analysis method, and evaluate the insulation state of equipment. This method injects an excitation signal and measures the response signal, calculates the impedance value of the grounding loop at different frequencies, analyzes the impedance spectrum to determine whether there is a leakage.

[0003] However, the operation environment of substations is complex, and there are strong power frequency electromagnetic fields, high-order harmonics and transient electromagnetic interferences. These interferences will reduce the signal-to-noise ratio of the monitoring signal, distort the impedance spectrum analysis results based on the original measurement data, and greatly reduce the accuracy of leakage judgment.

[0004] In addition, the power equipment connected by the intelligent grounding wire may be in different operating modes, and the environmental factors of the substation will also affect the impedance characteristics of the grounding grid and the grounding wire, resulting in the normal impedance spectrum baseline of the grounding loop not being fixed but dynamically changing with the equipment state and environmental conditions.

[0005] At the same time, the impedance spectrum analysis technology for leakage monitoring needs to inject an excitation signal into the grounding loop and collect the response signal. The frequency range, amplitude selection of the excitation signal, and the bandwidth, sampling rate of the measurement system directly affect the resolution and accuracy of the spectrum analysis. In a strong interference environment, in order to improve the measurement accuracy, specific excitation signal waveforms may need to be used, the energy of the excitation signal may need to be increased, or the data acquisition time may need to be extended. However, these measures may increase the power consumption of the monitoring equipment, affect the response speed of the equipment, or have a potential impact on the normal operation of the monitored equipment. At the same time, different types of leakage and fault locations may exhibit the most significant characteristics in different frequency ranges of the grounding loop impedance spectrum, and appropriate excitation frequency ranges need to be selected to effectively capture this information.

[0006] Therefore, in the specific scenario of strong electromagnetic interference, diverse equipment operating modes, changing environmental factors, and long-term operation in power substations, how to use impedance spectrum analysis-based technology to overcome the influence of electromagnetic interference on impedance spectrum measurement, effectively distinguish different types of leakage and equipment state changes, adapt to the dynamic changes of the grounding loop impedance spectrum baseline, and maintain the stability and reliability of the monitoring in the long-term operation, thereby avoiding false negatives and false positives, is a key problem that needs to be solved in the current intelligent grounding wire leakage monitoring technology. SUMMARY

[0007] The application aims to provide a grounding line monitoring method and system, which is based on impedance spectrum analysis, overcomes electromagnetic interference through multi-stage signal processing, robust feature extraction and multi-dimensional data fusion analysis, and achieves the effects of adapting to dynamic baseline and distinguishing different leakage types.

[0008] In a first aspect, the application provides a grounding line monitoring method applied to power equipment of a substation, including the following steps:

[0009] An excitation signal containing multiple discrete frequency points or sweep signals is injected into a grounding loop formed by the intelligent grounding line and the grounding grid, and the voltage signal at the injection point and the current signal of the grounding loop are synchronously collected;

[0010] The collected voltage signal and current signal are preprocessed to obtain preprocessed voltage signal and current signal;

[0011] According to the preprocessed voltage signal and current signal, the impedance spectrum of the grounding loop at different frequencies is calculated, and the key feature parameters in the impedance spectrum are extracted;

[0012] The environmental parameters and operating state data of the power equipment are collected and fused with the key feature parameters to obtain fusion data;

[0013] The judgment threshold or reference baseline corresponding to the key feature parameters is adjusted using the fusion data; the judgment threshold is a numerical limit for distinguishing between normal and abnormal states; the reference baseline is the normal value or range corresponding to the key feature parameters;

[0014] The current grounding loop state is judged using the adjusted judgment threshold or reference baseline.

[0015] The grounding line monitoring method provided by the application can improve the accuracy and reliability of the leakage monitoring of the intelligent grounding line based on impedance spectrum analysis in the environment of the substation, where there is strong electromagnetic interference and the impedance spectrum baseline of the grounding loop changes dynamically with the equipment state and environmental factors, so as to effectively distinguish between normal changes and different leakage states.

[0016] In a second aspect, the application provides a grounding line monitoring system applied to power equipment of a substation, including:

[0017] The collection module is used to inject an excitation signal containing multiple discrete frequency points or sweep signals into a grounding loop formed by the intelligent grounding line and the grounding grid, and synchronously collect the voltage signal at the injection point and the current signal of the grounding loop;

[0018] The preprocessing module is used to preprocess the collected voltage signal and current signal to obtain preprocessed voltage signal and current signal;

[0019] an extraction module configured to calculate an impedance spectrum of the grounding loop at different frequencies according to the preprocessed voltage signal and the current signal, and extract a key characteristic parameter in the impedance spectrum;

[0020] a fusion module configured to collect environmental parameters and operating state data of the power equipment, and fuse the key characteristic parameter to obtain fusion data;

[0021] an adjustment module configured to adjust a judgment threshold or a reference baseline corresponding to the key characteristic parameter by using the fusion data; the judgment threshold is a numerical limit for distinguishing between a normal state and an abnormal state; and the reference baseline is a normal value or range corresponding to the key characteristic parameter;

[0022] a judgment module configured to judge the current grounding loop state by using the adjusted judgment threshold or the reference baseline.

[0023] As can be seen from the above, the grounding line monitoring method provided by the application, in view of the electromagnetic interference specific to the transformer substation, adopts time-frequency joint analysis and other methods for signal preprocessing, effectively deals with the complex electromagnetic interference of the transformer substation, improves the quality of the original measurement data, extracts a key characteristic sensitive to electric leakage and relatively insensitive to interference and environmental changes from the processed impedance spectrum, and combines multi-dimensional data fusion and dynamic diagnosis, so that the system can adapt to the dynamic change characteristics of the grounding loop, avoids false positives, and significantly improves the accuracy and reliability of the intelligent grounding line in electric leakage monitoring in the complex environment of the transformer substation.

[0024] Other features and advantages of the application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the application as described in the written description and claims. The objects and other advantages of the application will be realized and attained by means of the structures particularly pointed out in the written description and claims. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A flowchart of the grounding line monitoring method provided by the embodiment of the application.

[0026] Figure 2 A structural schematic diagram of the grounding line monitoring system provided by the embodiment of the application.

[0027] Label explanation:

[0028] 100, acquisition module; 200, preprocessing module; 300, extraction module; 400, fusion module; 500, adjustment module; 600, judgment module; 700, prediction module. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0030] It should be noted that similar reference numerals and letters refer to similar items throughout the accompanying drawings, and therefore, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", and the like are only used to distinguish description, and cannot be understood as indicating or implying relative importance.

[0031] In a power substation, the safe operation of large high-voltage power equipment such as power transformers, high-voltage switch cabinets, etc. is of great importance. In order to ensure the electrical safety of the equipment during normal operation or failure, these devices need to be reliably connected to the grounding grid to provide a low-impedance discharge channel. The traditional grounding wire only provides a physical electrical connection path, and its state monitoring usually relies on periodic offline detection, which cannot achieve real-time monitoring.

[0032] With the development of technology and the increasing demand for equipment state monitoring, intelligent grounding wires with electrostatic voltage detection function have emerged. This intelligent grounding wire integrates an electrostatic voltage detection module while providing basic grounding path function, which is used to confirm whether the equipment is in a non-electric state before the equipment is shut down or overhauled, thereby improving the operation safety. On this basis, in order to further improve the operation reliability and predictive maintenance capability of the equipment, the intelligent grounding wire is endowed with a leakage current monitoring function. This function aims to monitor the abnormal current flowing into the grounding grid through the grounding wire, i.e. the leakage current, and assess the insulation state of the equipment or whether there is a grounding fault by analyzing the characteristics of the leakage current.

[0033] Currently, a technical solution for intelligent grounding line leakage monitoring uses a method based on impedance spectrum analysis. The basic principle of this method is: by injecting excitation signals of a specific frequency range into the grounding loop (including the monitored equipment, intelligent grounding line and grounding net), and measuring the response signals (such as voltage and current) of the loop to these excitation signals, the impedance values of the grounding loop at different frequencies are calculated. The curve or data set of these impedance values changing with frequency is analyzed, that is, impedance spectrum analysis is performed. Under normal circumstances, the insulation of the monitored equipment is good, and the equivalent circuit model of the grounding loop and its impedance spectrum present specific baseline characteristics. When the insulation of the equipment deteriorates or leakage occurs, the equivalent circuit model of the grounding loop will change, for example, the equivalent parallel resistance decreases, the equivalent capacitance increases, etc., which in turn causes the impedance spectrum characteristics to change. By detecting and analyzing these spectrum changes, it is theoretically possible to determine whether there is leakage and the nature or degree of the leakage. This method has the advantages of non-invasiveness (relative to direct measurement of leakage current, the excitation signal is usually small) and providing rich information (spectrum characteristics).

[0034] However, as a typical place where high-voltage power facilities are concentrated, the operation environment of a power substation is extremely complex and harsh. First, the equipment in the substation generates strong power frequency (50 / 60 Hz) electric and magnetic fields during operation, as well as high-order harmonics. Second, high-voltage switch operation, lightning arrester action, short-circuit fault, etc. will produce high-amplitude, wide-band transient electromagnetic interference. In addition, there may be wireless communication equipment, power line carrier communication, control signal lines, etc. in the substation and its surroundings, which will generate high-frequency signals of different frequencies. These interference signals will be coupled into the monitoring loop of the intelligent grounding line through capacitive coupling, inductive coupling or conductive coupling, superimposed on the injected excitation signals and the collected response signals, significantly reducing the signal-to-noise ratio of the monitoring signals. Especially when monitoring the weak early-stage insulation deterioration leakage current, the impedance spectrum change amplitude of the grounding loop caused by the leakage current may be very weak and easily overwhelmed by strong electromagnetic interference, making the impedance spectrum analysis results based on the original measurement data severely distorted, resulting in a significant decrease in the accuracy of leakage detection, and possibly false negatives (failure to detect leakage) or false positives (mistaking interference or normal changes for leakage).

[0035] Due to the presence of the above-mentioned strong electromagnetic interference, the directly collected voltage and current signals contain a large amount of noise components. If these noises are not effectively suppressed or filtered out, subsequent spectrum analysis and feature extraction will face great difficulties. Traditional narrowband or broadband filtering methods may filter out not only the noise but also the effective spectrum information of the leakage signal at certain frequencies, especially when the frequency of the interference signal is close to the leakage characteristic frequency, it is difficult to effectively separate. This makes it extremely challenging to extract accurate and robust features related to the leakage state from the noise-polluted impedance spectrum data.

[0036] In addition, the insulation of power equipment in a substation is a slow aging process. In the early stage of insulation degradation, only very weak leakage current can be generated, and the change in ground loop impedance spectrum caused by such weak leakage current is relatively subtle and can be a slow evolution process, which is difficult to identify by simple threshold judgment. Unlike this, a sudden insulation breakdown or external line short circuit fault can cause a larger instantaneous leakage current, and the change in impedance spectrum caused by this can be more dramatic and rapid. When using the method based on impedance spectrum analysis for leakage monitoring, it is necessary to effectively distinguish between the characteristics of slow-evolving insulation degradation and the characteristics of sudden faults, to avoid misjudgment of spectrum fluctuations caused by transient interference, while being able to timely discover early signs of insulation degradation. Simple impedance amplitude or phase change threshold judgment cannot effectively distinguish between different types of leakage or insulation degradation stages.

[0037] At the same time, power equipment connected to the smart grounding line can be in different operating modes, such as normal live operation, power-off maintenance, standby state, etc. In different operating modes, the electrical connection state and equivalent impedance characteristics inside the equipment can change, which can affect the total impedance spectrum of the entire grounding loop. For example, the equivalent capacitance characteristics of the equipment can be different when it is live than when it is powered off. At the same time, environmental factors such as ambient temperature, humidity changes, seasonal changes in soil resistivity, etc. of the substation can also affect the impedance characteristics of the grounding grid and grounding line. These factors cause the normal impedance spectrum baseline of the grounding loop to be not fixed but to change dynamically with the device state and environmental conditions. If impedance spectrum analysis is performed, using a fixed model or baseline for comparison will not be able to adapt to such dynamic changes, and can misjudge the impedance spectrum changes caused by device state or environmental changes as leakage, or fail to accurately identify real leakage when it occurs due to baseline drift.

[0038] Impedance spectrum analysis technology for leakage monitoring requires injecting an excitation signal into the grounding loop and collecting the response signal. The frequency range, amplitude selection of the excitation signal, and the bandwidth, sampling rate of the measurement system directly affect the resolution and accuracy of the spectrum analysis. In a strong interference environment, in order to improve the measurement accuracy, specific excitation signal waveforms, increased excitation signal energy, or extended data acquisition time can be used. However, these measures can increase the power consumption of the monitoring equipment, affect the response speed of the equipment, or have potential impact on the normal operation of the monitored equipment. At the same time, different types of leakage and fault locations can exhibit the most significant characteristics in different frequency ranges of the grounding loop impedance spectrum, and appropriate excitation frequency ranges need to be selected to effectively capture this information.

[0039] Finally, as a device deployed in outdoor or indoor high-voltage areas for a long time, the smart grounding line needs to have high reliability and stability. During long-time operation, the performance of the monitoring circuit components may drift, the interface connected to the grounding line or grounding net may be oxidized, poorly contacted or loose, which may affect the accuracy and stability of the impedance spectrum measurement, cause the monitoring data to be distorted, and further affect the reliability of the leakage judgment. It is necessary to ensure the stability and anti-interference ability of the entire monitoring system under harsh environments and long-time operation while ensuring the monitoring accuracy.

[0040] Therefore, in the specific scenario of a power substation with strong electromagnetic interference, various device operation modes, changing environmental factors and long-time operation, how to use the technology based on impedance spectrum analysis to overcome the influence of electromagnetic interference on impedance spectrum measurement, effectively distinguish different types of leakage and device state changes, adapt to the dynamic changes of the grounding loop impedance spectrum baseline, and maintain the stability and reliability of the monitoring in long-time operation, thereby avoiding false negatives and false positives, is a key problem to be solved for the current smart grounding line leakage monitoring technology.

[0041] To this end, with reference to the accompanying drawings Figure 1 The application provides a grounding line monitoring method applied to power equipment of a substation, including the following steps:

[0042] An excitation signal containing multiple discrete frequency points or a sweep signal is injected into a grounding loop formed by the smart grounding line and the grounding net, and a voltage signal at the injection point and a current signal of the grounding loop are synchronously collected;

[0043] The collected voltage signal and current signal are preprocessed to obtain preprocessed voltage signal and current signal;

[0044] According to the pre-processed voltage signal and current signal, the impedance spectrum of the grounding circuit at different frequencies is calculated, and key characteristic parameters in the impedance spectrum are extracted; the key characteristic parameters include impedance amplitude and phase in a specific frequency point or frequency band, slope and curvature of the impedance spectrum curve in a specific frequency range, frequency, amplitude and bandwidth of the resonance peak or anti-resonance peak in the spectrum, and spectrum energy distribution of a specific frequency band (it should be noted that "specific frequency point or frequency band" is usually selected based on analysis of the equivalent circuit model of the grounding circuit and the performance of different types of insulation degradation or leakage fault in the frequency domain response. For example, the dielectric loss characteristics of certain insulation materials may exhibit significant changes in a specific frequency range, or structural defects of the grounding grid may cause abnormal impedance at a specific frequency point. A person skilled in the art can identify frequency points or frequency bands sensitive to device state changes through historical operation data analysis, theoretical calculation or experimental verification. The impedance amplitude and phase of these sensitive frequency points or frequency bands can directly reflect the changes of the resistance, inductance, capacitance and other parameters of the grounding circuit, thereby indicating potential abnormalities. The "specific frequency range" is also determined based on experience and analysis. The slope reflects the speed of change of impedance with frequency, and the curvature describes the change of this rate. In some frequency intervals, the impedance spectrum curve of a normal grounding circuit may exhibit a gentle or linear change, while when insulation degradation or leakage occurs, the slope or curvature of the curve may change significantly, such as appearing inflection points or sharp rises / falls. Therefore, selecting these frequency ranges sensitive to state changes for slope and curvature calculation helps to capture subtle abnormalities in the spectrum shape. The "specific frequency band" refers to one or more frequency intervals with specific diagnostic value selected from the entire frequency range covered by the excitation signal injected into the grounding circuit formed by the smart grounding wire and the grounding grid according to actual diagnostic needs. The selection of these frequency bands is based on the analysis of the impedance spectrum of the grounding circuit. Specifically, different types of leakage and fault locations may exhibit the most significant characteristics in different frequency ranges of the grounding circuit impedance spectrum. Therefore, the "specific frequency band" is selected to effectively capture these key information related to insulation degradation or leakage fault of the device. By identifying these frequency intervals that can most significantly reflect abnormal states, the effectiveness of extracting key characteristic parameters from the impedance spectrum can be improved, thereby improving the accuracy of judging the state of the grounding circuit.

[0045] Collecting environmental parameters and operating state data of the power equipment, and fusing with the key characteristic parameters to obtain fused data;

[0046] According to the pre-established dynamic reference model or diagnostic rule set, the judgment threshold corresponding to the key characteristic parameter or the reference baseline is adjusted by using the fusion data; the judgment threshold is a numerical limit for distinguishing between normal and abnormal states (such as early degradation or leakage fault); the reference baseline is the normal value or range corresponding to the key characteristic parameter (the reference baseline is the value or range that the impedance spectrum characteristic parameter should have under the specific environmental conditions and equipment state in the normal operating state); the adjustment of these judgment thresholds or reference baselines is to adapt the judgment process to the normal fluctuations of the grounding loop impedance spectrum caused by changes in environmental factors (such as temperature, humidity) and equipment operating state (such as live, power off);

[0047] Using the adjusted judgment threshold or reference baseline, the current grounding loop state is judged to be normal, early degradation, or different degrees of leakage fault.

[0048] The present application aims to solve the problem of accurately determining the state of the grounding loop in a complex substation environment. First, an excitation signal is injected into the grounding loop and voltage and current signals are collected, which is the basic step of obtaining the impedance information of the grounding loop. By injecting a known signal and measuring the loop response, the impedance characteristics of the loop can be calculated. Thus, the original data reflecting the electrical characteristics of the grounding loop can be obtained. Next, the collected voltage and current signals are preprocessed. The substation environment has strong electromagnetic interference, and the directly collected signals contain a lot of noise. The preprocessing step removes or suppresses the interference components through signal processing techniques, improves the signal quality, provides clean data for subsequent accurate impedance spectrum calculation, and thus reduces the influence of interference on the measurement results. Then, the impedance spectrum of the grounding loop at different frequencies is calculated according to the preprocessed voltage and current signals, and key feature parameters are extracted from it. The impedance spectrum reflects the electrical characteristics of the grounding loop at different frequencies. Extracting key feature parameters is to convert complex spectrum curves into representative numerical values or patterns, which are associated with changes in the state of the grounding loop, such as insulation degradation or leakage that may cause impedance changes at specific frequencies. At the same time, environmental parameters and operating state data of power equipment are collected and fused with the extracted key feature parameters. The impedance characteristics of the grounding loop are affected by environmental factors and equipment operating state, which will cause the baseline of the impedance spectrum to change. Collecting these data and fusing them with the impedance characteristics is to obtain more comprehensive information reflecting the actual working conditions of the grounding loop, so that the state of the grounding loop can be understood more comprehensively. Using the fused data, according to the pre-established dynamic reference model or diagnostic rule set, adjust the judgment threshold or reference baseline corresponding to the key feature parameters. Traditional fixed threshold or baseline is difficult to adapt to changes in environment and equipment state. By fusing environmental and operating data and combining dynamic models or rules, the judgment threshold or reference baseline can be adjusted according to the actual working conditions, so that it is more consistent with the current normal state range, thereby improving the adaptability of the judgment standard. Finally, the adjusted judgment threshold or reference baseline is used to determine the current state of the grounding loop. Based on the adjusted and more accurate judgment standard, the key feature parameters currently extracted are evaluated to distinguish whether the grounding loop is in a normal state, an early degradation stage, or has different degrees of leakage fault, thereby improving the accuracy of determining the state of the grounding loop in a complex and variable environment and reducing false positives and false negatives.

[0049] In particular, the method works as follows: First, by exciting signal injection and response signal collection, the electrical response data of the grounding circuit at different frequencies are obtained. These raw data may be affected by strong electromagnetic interference in the substation. Therefore, it is necessary to preprocess the collected voltage and current signals. The preprocessing aims to filter out power frequency interference, transient high amplitude interference and random noise to obtain purer signals. In this way, the accuracy of subsequent impedance spectrum calculation can be improved. Based on the preprocessed signals, the impedance spectrum of the grounding circuit is calculated. The impedance spectrum is a function of frequency, reflecting the comprehensive characteristics of resistance, inductance, capacitance, etc. of the circuit. Key feature parameters are extracted from the impedance spectrum, such as impedance values at specific frequency points, shape features of the spectrum curve or resonance peak information, which are considered to be related to the health status of the grounding circuit. However, these impedance features will be affected by environmental factors (such as temperature, humidity) and equipment operating conditions (such as voltage, current), resulting in impedance feature values that are not fixed in normal state. To solve this problem, the method further collects environmental parameters and operating state data of power equipment, and fuses these data with the extracted impedance key feature parameters. The fused data contains the electrical characteristic information of the grounding circuit and the working condition information it is in. Using these fused data, the method dynamically adjusts the judgment threshold or reference baseline for judging the state of the grounding circuit according to the pre-established dynamic reference model or diagnostic rule set. For example, when the ambient temperature rises, the resistance of the grounding grid may increase, causing the overall impedance to rise, at which time the dynamic reference model will adjust the impedance reference baseline in the normal state accordingly. Finally, the current extracted impedance key feature parameters are compared with the adjusted judgment threshold or reference baseline. If the key feature parameters exceed the adjusted normal range, it is judged that the grounding circuit is in an abnormal state, such as early degradation or electric leakage fault. By this way of dynamically adjusting the judgment standard, the method can adapt to the complex and variable environment and equipment operating state of the substation, improving the accuracy and reliability of electric leakage monitoring.

[0050] In some embodiments, a sweep excitation signal is injected into the grounding loop, with a frequency range set to 100 Hz to 10 kHz. The voltage signal at the injection point and the current signal of the grounding loop are simultaneously collected with a sampling rate set to 50 kHz. The collected voltage and current signals are first filtered by a digital notch filter to remove the 50 Hz power frequency and its third harmonic (150 Hz) interference. Further, the filtered signals are subjected to short-time Fourier transform to identify and remove transient components with amplitudes exceeding a set threshold. In this way, the influence of transient interference is suppressed. The processed voltage and current signals are measured multiple times and averaged to reduce random noise. In this way, the pre-processed voltage and current signals are obtained. According to the pre-processed signals, the impedance spectrum of the grounding loop in the range of 100 Hz to 10 kHz is calculated. Key feature parameters are extracted from the impedance spectrum, such as the impedance amplitude and phase at 1 kHz, and the impedance amplitude curve slope in the frequency band of 2 kHz to 5 kHz. At the same time, the ambient temperature of the substation and the operating current of the monitored equipment are collected. The 1 kHz impedance amplitude, phase, and spectral slope are weighted and fused with the ambient temperature and operating current to obtain the fusion data. The fusion weights are determined by historical data analysis. According to the pre-established dynamic reference model based on historical data, which takes into account the influence of temperature and current on impedance characteristics, the normal range reference baseline of 1 kHz impedance amplitude is adjusted using the current fusion data. Finally, the current calculated 1 kHz impedance amplitude is compared with the adjusted reference baseline. If the current amplitude is lower than the lower limit of the reference baseline, it is judged that there may be a leakage fault.

[0051] It should be noted that the present application is applicable to power equipment arranged with intelligent grounding wires, which can be connected with external communication devices. Users can send control signals to the external communication devices through user terminals to control the intelligent grounding wires to perform electric detection and other operations. At the same time, the intelligent grounding wires can also send feedback information to the user terminals through the external communication devices. Therefore, in actual application, the information output by the present application (such as the judgment result of the current grounding loop state, the fault occurred, and the early warning information below) can also be sent to the user terminals, so that users can remotely communicate through the user terminals. Further, users can also design special APPs as needed to display functions such as electric detection, grounding end resistance detection, abnormal feedback, fault point positioning detection, etc. on the visual user interface.

[0052] In some embodiments, the method further comprises the steps of:

[0053] Recording and storing historical impedance spectrum feature parameters, environmental parameters, and equipment state information;

[0054] By long-term trend analysis of historical data, the change rate and direction of key feature parameters over time are obtained.

[0055] According to the rate and direction of the change of the key characteristic parameters over time, the insulation deterioration speed of the power equipment is evaluated, and early warning is given before the power equipment fails according to the evaluation result.

[0056] The historical impedance spectrum characteristic parameters, environmental parameters and equipment state information are recorded and stored, and a data basis for subsequent analysis is established. By accumulating these data for a long time, the changing trajectory of the grounding loop state over time can be observed. By long-term trend analysis of the historical data, the rate and direction of the change of the key characteristic parameters over time are obtained, and the accumulated historical data are utilized. By analyzing the numerical change of the key characteristic parameters at different time points, it can be identified whether these parameters remain stable, slowly increase, slowly decrease or present an accelerated change trend. The change rate and direction quantify this trend. According to the rate and direction of the change of the key characteristic parameters over time, the insulation deterioration speed of the power equipment is evaluated, and the result of the trend analysis is associated with the equipment insulation state. Insulation deterioration is usually a gradual process, and its manifestation on the impedance spectrum characteristic may also be a slow change. By analyzing the rate and direction of the change of the key characteristic parameters, the speed of insulation deterioration can be judged. According to the evaluation result, early warning is given before the power equipment fails, which is the purpose of the whole scheme. Through the evaluation of the insulation deterioration speed, the time point when the insulation state may reach the failure threshold can be predicted. Before the predicted failure time, the system can give an early warning signal according to the evaluation result of the deterioration speed, prompting the operation and maintenance personnel to pay attention to the equipment state or arrange maintenance.

[0057] Specifically, the technical solution increases the use of historical data on the basis of judging the current grounding loop state. First, the system continuously records and stores the impedance spectrum key feature parameters obtained each time the monitoring is performed, such as the impedance amplitude, phase at a specific frequency, and the like, and records the environmental parameters at the time of monitoring, such as temperature, humidity, and the running state information of the power equipment. These historical data are stored in a database. Subsequently, the system processes the stored historical data, for example, for a certain key feature parameter, the value sequence thereof in the past period of time is extracted. By applying time series analysis methods, such as linear regression, exponential smoothing, or moving average, the average rate and change direction of the parameter over time are calculated. For example, if a certain impedance amplitude parameter has been continuously decreasing in the past year, trend analysis can calculate the average monthly decrease value and determine the direction of the decrease. According to the calculated change rate and direction of the key feature parameters, the system evaluates the insulation deterioration speed of the power equipment. For example, if the impedance amplitude decrease rate exceeds a preset threshold, it is judged that the insulation deterioration speed is fast; if the change rate is low, it is judged that the deterioration speed is slow. Thus, the insulation deterioration speed information can be obtained. Finally, according to the evaluated insulation deterioration speed, the system predicts the time when the insulation state reaches the failure threshold. If the system issues an early warning signal according to the evaluation result of the deterioration speed before the predicted failure time, for example, if it is predicted that a failure may occur in the next three months, the system will issue a warning. This improves the predictive maintenance capability. These added steps extend the monitoring method from simply judging the current state to predicting and warning the future state, thereby comprehensively improving the accuracy and reliability of the intelligent grounding line leakage monitoring.

[0058] In some embodiments, the step of pre-processing the collected voltage signal and current signal to obtain a pre-processed voltage signal and a pre-processed current signal includes:

[0059] The collected voltage signal and current signal are cleaned by using a digital notch filter or by a Fourier analysis method to suppress the power frequency and its harmonics to obtain a first-processed voltage signal and a first-processed current signal.

[0060] For the first-processed voltage signal and the first-processed current signal, a time-frequency analysis method is used to identify and remove transient high-amplitude components to obtain a second-processed voltage signal and a second-processed current signal.

[0061] For the voltage signal after the second processing, a plurality of voltage values are continuously collected from the voltage signal after the second processing according to a preset first measurement period, and all voltage values in the same first measurement period are averaged to obtain an average voltage value corresponding to the first measurement period. The average voltage values obtained in all different first measurement periods are taken as the voltage signal after the third processing, so as to reduce the random noise of the voltage signal and serve as the final preprocessed voltage signal, or the voltage signal after the second processing is directly subjected to Kalman filtering to reduce the random noise of the voltage signal, so as to obtain the voltage signal after the third processing and serve as the final preprocessed voltage signal.

[0062] For the current signal after the second processing, a plurality of current values are continuously collected from the current signal after the second processing according to a preset second measurement period, and all current values in the same second measurement period are averaged to obtain an average current value corresponding to the second measurement period. The average current values obtained in all different second measurement periods are taken as the current signal after the third processing, so as to reduce the random noise of the current signal and serve as the final preprocessed current signal, or the current signal after the second processing is directly subjected to Kalman filtering to reduce the random noise of the current signal, so as to obtain the current signal after the third processing and serve as the final preprocessed current signal.

[0063] The collected voltage signal and current signal are cleaned to suppress the power frequency and its harmonics. This can be achieved by designing a digital notch filter, which has high attenuation characteristics at the power frequency (e.g. 50 Hz or 60 Hz) and its integer multiple frequencies. Alternatively, the collected signal can be subjected to a fast Fourier transform (FFT), and the amplitude of the frequency points or frequency bands corresponding to the power frequency and its harmonic components in the frequency domain can be set to zero or greatly attenuated, and then an inverse Fourier transform (IFFT) is performed to obtain the time domain signal. Further, transient high amplitude components can be identified and removed using time-frequency analysis methods, such as wavelet analysis. By selecting a suitable wavelet basis function, the transient high amplitude components are usually represented by a few large coefficients in the wavelet domain. By setting a threshold, wavelet coefficients less than the threshold can be set to zero or shrunk, and then wavelet reconstruction is performed to remove the transient components. In this way, the random noise of the voltage signal and the current signal is reduced. The signal can be smoothed by repeatedly collecting signal samples under the same state and then calculating the average of these samples. As another implementation, a state space model describing the dynamic characteristics of the signal can be constructed, and a Kalman filtering algorithm is applied to optimally estimate the true value of the signal according to real-time measurement data, so as to filter out random noise.

[0064] Specifically, in the substation environment, the collected grounding loop voltage and current signals are susceptible to various disturbances. First, strong power frequency electromagnetic fields can induce power frequency and harmonic components in the signals. Through digital notch filtering or Fourier analysis methods, these periodic disturbances of fixed frequency can be effectively attenuated, obtaining a primary processed signal and improving the purity of the signal. Second, transient events such as switching operations can produce short-time high-amplitude pulse disturbances. These disturbances are widely distributed in time and frequency domains, and it is difficult to remove them with traditional filters. Using time-frequency analysis methods such as wavelet analysis, the location of these transient events in time and frequency can be accurately located, and they can be separated and removed from the signal through threshold processing, obtaining a secondary processed signal, and avoiding the influence of transient disturbances on subsequent analysis. Finally, there are still random noises in the signals processed by the previous two steps, which will affect the accuracy of impedance calculation. By averaging multiple measurements, the average of random noise tends to zero, thereby reducing the influence of noise. Alternatively, Kalman filtering can be used to optimally estimate and suppress noise based on the dynamic characteristics of the signal, further improving the signal-to-noise ratio and accuracy of the signal, obtaining the final preprocessed voltage signal and current signal. These clean signals are used to calculate the impedance spectrum of the grounding loop, which can significantly improve the accuracy and reliability of the spectrum analysis, providing basic data for accurately determining the state of the grounding loop.

[0065] In some embodiments, the step of collecting environmental parameters and operating state data of the power equipment and fusing with key characteristic parameters to obtain fused data includes:

[0066] Collecting environmental parameters of the power equipment, including temperature, humidity, electromagnetic interference intensity, and operating state data, including voltage, current, active power and reactive power;

[0067] Using wavelet threshold denoising method to filter out noise interference of environmental parameters and operating state data, and using interpolation algorithm based on historical data to fill in missing values, obtaining clean and complete data set;

[0068] The mutual information method is used to evaluate the correlation of each environmental parameter and each operating state data with respect to each key characteristic parameter, obtain an evaluation result about the correlation of each environmental parameter and each operating state data with respect to each key characteristic parameter, and select environmental parameters and operating state data with a correlation greater than a preset threshold according to the evaluation result to obtain a filtered parameter set (it should be noted that the selection of the preset threshold is usually based on a comprehensive consideration of analysis of historical data, statistical principles, and domain knowledge. Specifically, in actual applications, the preset threshold can be determined in the following way: first, historical data analysis. During normal operation of the equipment, a large amount of environmental parameters, operating state data, and corresponding impedance spectrum key characteristic parameters are collected. Then, the mutual information values between each environmental parameter and operating state data and the key characteristic parameters in these historical data are calculated. By observing the distribution and range of these mutual information values under normal working conditions, it can be identified which parameters usually show strong correlation and which show weak correlation. For example, the average level and fluctuation range of the mutual information values can be counted, and an initial threshold can be set based on experience so that most parameters that are considered to have an impact under normal conditions can be retained. Second, combined with statistical principles. Mutual information quantifies the degree of dependence between two variables. Although there is no direct statistical significance test standard for mutual information, the numerical value reflects the amount of information sharing. Generally, the higher the mutual information value, the stronger the correlation between the two parameters. Therefore, the selection of the preset threshold can be based on the understanding of the mutual information value, and a threshold that can distinguish between "meaningful correlation" and "random fluctuation or weak correlation" can be set. For example, an empirical value such as 0.1, 0.2, or higher can be set as the minimum mutual information requirement. Third, incorporate domain knowledge and expert experience. The operation of power equipment and the characteristics of grounding systems are complex, and some environmental or operating parameters are theoretically or based on long-term operation and maintenance experience known to have a significant impact on the grounding circuit impedance. These prior knowledge can guide the initial setting of the preset threshold. For example, if it is known that temperature has a large impact on the grounding net impedance, even if the initial calculated mutual information value is slightly lower than a certain general threshold, the threshold can be adjusted appropriately to retain the temperature parameter. Finally, through system performance verification and optimization. Before actual deployment, the monitoring system under different preset thresholds can be tested offline or online. By evaluating the false positive rate, false negative rate, and diagnostic accuracy of the system under different thresholds, a threshold that can balance these performance indicators is selected. This is an iterative optimization process aimed at finding a threshold that can effectively filter out irrelevant interference while retaining enough useful information to support accurate judgment.

[0069] The Min-Max scaling method is used to normalize the environmental parameters and operating state data in the filtered parameter set to scale the environmental parameters and operating state data to the interval [0, 1] and eliminate dimensional differences.

[0070] The normalized environment parameters and operating state data are fused with the key feature parameters through a weighted fusion algorithm to obtain a fused multi-dimensional feature vector, and the fused multi-dimensional feature vector is used as fusion data for subsequent dynamic reference model or diagnostic rule set adjustment; the weight coefficients in the weighted fusion algorithm are determined by the corresponding mutual information values.

[0071] Specifically, the technical scheme first acquires original data reflecting the external environment and internal operating state of the power equipment. These original data can be affected by various factors and contain noise or have missing values. By applying wavelet threshold denoising and historical data interpolation techniques, the original environmental parameters and operating state data are purified and completed, improving the reliability and integrity of the data and laying a foundation for subsequent analysis. Next, the mutual information method is used to quantify the correlation strength between the processed environmental parameters and operating state data and the key feature parameters of the grounding loop impedance spectrum, and data filtering is performed based on a pre-set correlation threshold, removing redundant information with low correlation to the key feature parameters, reducing the data dimension, and focusing on factors that actually affect the grounding loop state judgment. Subsequently, the filtered environmental parameters and operating state data are normalized to eliminate the influence of different physical dimensions and numerical ranges, ensuring the fairness of various data in the fusion process. Finally, a weighted fusion algorithm is used to combine the normalized environmental parameters and operating state data with the key feature parameters. In the weighted fusion process, the weights are determined based on the previously calculated mutual information values, so that environmental or operating parameters with higher correlation to the key feature parameters have a greater proportion in the fusion result. The multi-dimensional fusion data generated in this way comprehensively reflects the complex relationship between the grounding loop state, environmental conditions, and equipment operating state. This fusion data is used as input to dynamically adjust the threshold or reference baseline for grounding loop state judgment, improving the accuracy of grounding loop state judgment under different environmental and operating conditions (it should be noted that the calculation of mutual information value is usually based on the estimation of the probability distribution of the involved variables. Specifically, for any two variables X (such as environmental temperature) and Y (such as impedance amplitude), the following steps are needed to calculate the mutual information value: First, a large number of historical data samples are collected, which contain paired observations of variables X and Y. If these variables are continuous, in order to facilitate calculation, it is usually necessary to divide their value range into several discrete intervals, and convert continuous data into discrete categories. Second, based on these discretized data samples, estimate their probability distribution. This includes estimating the marginal probability P(X) of variable X taking each discrete interval, estimating the marginal probability P(Y) of variable Y taking each discrete interval, and estimating the joint probability P(X, Y) of variables X and Y taking a particular discrete interval. These probabilities can be approximated by counting the frequency of samples in each interval or each pair of intervals. Then, based on these estimated probabilities, the mutual information value is calculated. The principle of calculating mutual information is to measure the amount of information about one variable obtained by observing another variable. If two variables are independent of each other, the mutual information value is zero; if there is a dependency between them, the mutual information value will be greater than zero, and the greater the value, the stronger the dependency).

[0072] In some embodiments, for example, the condition of the transformer grounding line is monitored. The temperature, humidity, load current and winding temperature around the transformer are collected. At the same time, the impedance amplitude of the grounding loop at 1 kHz and 10 kHz frequencies is measured as a key characteristic parameter. The collected temperature and current data may contain transient fluctuations, which are filtered out using a wavelet threshold method. If the temperature sensor data is missing at a certain time point, the average value of the previous stable time points is used for filling. The mutual information of temperature and 1 kHz impedance amplitude, and the mutual information of load current and 1 kHz impedance amplitude are calculated. It is assumed that the correlation between temperature and impedance is high, and the correlation between current and impedance is low. For example, if the preset threshold is 0.4, only data greater than the preset threshold is selected. For example, if the correlation of temperature data is 0.7 and the correlation of current data is 0.3, only temperature data is selected for fusion. The historical temperature data is Min-Max normalized to map the current temperature value to 0 to 1. The normalized temperature data is fused with the impedance amplitude at 1 kHz and 10 kHz, and the weight is determined according to the mutual information value (it should be noted that the specific way of determining the weight coefficient according to the mutual information value can be realized as follows: first, the mutual information value between each pre-processed, screened and normalized environmental parameter and operating state data and the key characteristic parameters of the grounding loop impedance spectrum is calculated. For example, if temperature, humidity and operating current are selected as environmental and operating parameters, and impedance amplitude and phase are selected as key characteristic parameters, the mutual information of temperature and impedance amplitude, the mutual information of temperature and impedance phase, and the mutual information of humidity, operating current and impedance amplitude and phase are calculated. After obtaining these mutual information values, normalization processing can be used to determine the weight coefficient. For example, for a certain key characteristic parameter, the mutual information values of each environmental parameter and operating state data related to it are summed up, and then each mutual information value is divided by the sum, thereby obtaining the weight of the parameter in fusion. In this way, the greater the mutual information value of a parameter, the greater its weight in the fusion data, and the greater its influence on the final fusion result. This method ensures that in the fusion process, external factors that are more closely related to the change of the grounding loop state can obtain a higher weight, so that the fusion data more accurately reflects the influence of the actual working condition on the impedance characteristics of the grounding loop).

[0073] In certain embodiments, using the fusion data, the step of adjusting the judgment threshold or reference baseline corresponding to the key characteristic parameter includes:

[0074] According to the pre-established dynamic reference model, the judgment threshold or reference baseline corresponding to the key characteristic parameter is adjusted using the fusion data.

[0075] In some embodiments, the step of establishing the dynamic reference model comprises:

[0076] obtaining historical operation data, the historical operation data comprising historical impedance spectrum feature parameters, historical environmental parameters and historical equipment state information;

[0077] based on the historical operation data, performing clustering analysis on the data under normal operating conditions using a clustering algorithm to obtain a plurality of normal operating condition clustering centers, and constructing a dynamic reference model under normal operating conditions by calculating the covariance matrix of each clustering center.

[0078] The obtaining of the historical operation data provides a basic data set for model training, which records the performance of the equipment under different times and conditions. The application of the clustering algorithm aims to identify a plurality of typical modes or sub-states under normal operating conditions, such as the normal performance of the equipment under different seasons or different load levels. From this, clustering centers representing these typical normal modes can be obtained. Further, the covariance matrix of each clustering center is calculated, which quantifies the normal fluctuation range of each feature parameter and their mutual relationship under this particular normal mode, thereby defining the normal data distribution under this mode. By collecting these clustering centers and their corresponding covariance matrices, a reference model is constructed that can reflect the diversity and dynamic changes of normal operating conditions.

[0079] Specifically, the technical solution establishes a reference model that can adapt to environmental and operating state changes by analyzing historical data. First, historical data generated by the device during long-term operation is collected, which contains the characteristic parameters of the ground loop impedance spectrum measured under various environmental conditions (such as temperature, humidity) and operating states (such as voltage, current). Then, data samples of the device under normal operating conditions are selected from these historical data. Next, a clustering algorithm is applied to the multi-dimensional data (including characteristic parameters, environmental parameters, and operating state data) under normal operating conditions. The clustering process groups similar data points into multiple clusters, each representing a specific normal operating mode. For example, one cluster may correspond to the normal state of the device when it is running under high temperature and humidity in summer and high load, while another cluster may correspond to the normal state of the device when it is running under low temperature and dryness in winter and low load. The output of the clustering algorithm is the center point of each cluster, which represents the typical value of each normal mode. Subsequently, for each data point in each cluster, its covariance matrix is calculated. The covariance matrix describes the dispersion of data points in each dimension and the correlation between dimensions, thus defining the data distribution range and shape under a specific normal mode. By obtaining the coordinates of all normal operating clustering centers and their corresponding covariance matrices, a dynamic reference model is constructed. This model no longer relies on a single fixed baseline, but can find the most matching normal mode in the model according to the current device environment and operating state, and use the covariance matrix corresponding to the mode to determine the normal range or confidence interval under the current state. Thus, in subsequent judgment of the current ground loop state, the current measured fusion data (including real-time characteristic parameters, environmental parameters, and operating state data) can be compared with the most matching normal mode in the dynamic reference model, so that the current state is more accurately judged whether it deviates from the normal range under a specific condition, improving the accuracy of leakage judgment and reducing false positives and false negatives.

[0080] In some embodiments, the historical operation data is assumed to contain the impedance amplitude of the grounding loop at a specific frequency, the ambient temperature and the load current of the device. First, historical data for months or years is collected, and data points when the device is confirmed to be in a normal state are marked. For example, 10,000 data samples in a normal state are collected, and each sample is a three-dimensional vector [impedance amplitude, temperature, load current]. Then, a K-means clustering algorithm is used to perform clustering analysis on the 10,000 samples, and the number of clustering clusters K is set to 3. After the clustering algorithm is run, three cluster centers C1, C2, and C3 are obtained, which respectively represent three typical normal operating modes. For example, C1 may correspond to [low impedance amplitude, low temperature, low load], C2 corresponds to [medium impedance amplitude, medium temperature, medium load], and C3 corresponds to [high impedance amplitude, high temperature, high load]. Then, for the data points belonging to each clustering cluster, the three-dimensional covariance matrix Σ1, Σ2, and Σ3 is calculated. For example, Σ1 describes the fluctuation range of the impedance amplitude, temperature, and load current and the correlation between them under the normal operating condition corresponding to C1. Thus, the dynamic reference model is composed of {C1, Σ1}, {C2, Σ2}, and {C3, Σ3}. In real-time monitoring, the current fusion data [current impedance amplitude, current temperature, current load current] is obtained, the Mahalanobis distance of the data point from C1, C2, and C3 is calculated, the clustering center with the smallest Mahalanobis distance is selected, for example, C2. Then, using the covariance matrix Σ2 corresponding to C2, the confidence level or Mahalanobis distance threshold (i.e., the confidence interval) of the current data point under the normal mode is calculated, and the confidence interval is used as a reference baseline or threshold for judging whether the current state is normal.

[0081] Further, according to the pre-established dynamic reference model, the step of adjusting the judgment threshold or reference baseline corresponding to the key feature parameter using the fusion data includes:

[0082] Real-time acquisition of current fusion data, calculation of the Mahalanobis distance of the current fusion data from each normal operating condition cluster center, and selection of the cluster center corresponding to the smallest Mahalanobis distance;

[0083] Obtaining the covariance matrix corresponding to the selected cluster center through the dynamic reference model, and calculating the confidence interval of the current fusion data according to the covariance matrix;

[0084] The confidence interval is adaptively adjusted as the judgment threshold or reference baseline of the key characteristic parameter. If the current key characteristic parameter is out of the confidence interval, it is determined as an abnormal state, otherwise it is determined as a normal state. The adjustment of the confidence interval includes: when the environmental parameter is detected to mutate, the confidence interval is increased to reduce the false positive rate; when the device running state is stable, the confidence interval is reduced to improve the detection sensitivity (it needs to be noted that the confidence interval is a statistical interval used to estimate the range of a parameter true value, which is determined according to the distribution characteristics of the data (such as mean and covariance matrix). When applied to the key characteristic parameter, the confidence interval defines the statistical range of the key characteristic parameter that is considered to be a normal value under the current specific environment and running condition. The boundary of the confidence interval or the range defined by it directly constitutes or determines the judgment threshold or reference baseline of the key characteristic parameter under the current working condition. In other words, the confidence interval itself is the specific manifestation of the dynamically adjusted reference baseline, or its upper and lower limits are the dynamically adjusted judgment threshold. For example, if the adjusted reference baseline is a range [A, B], then [A, B] is calculated according to the confidence interval. Whether the current key characteristic parameter is within [A, B] is to judge by using the adjusted reference baseline).

[0085] In the complex environment and equipment operating state of power substation, the normal impedance spectrum baseline of grounding circuit is not fixed, and it is difficult to accurately identify abnormal state by using fixed judgment standard, which may lead to misjudgment or missed judgment. The technical scheme realizes adaptive adjustment of judgment threshold or reference baseline by introducing dynamic reference model and combining real-time environment parameters and equipment operating state data. Firstly, the system collects fusion data including impedance spectrum key feature parameters, environment parameters and equipment operating state data in real time. Then, the current fusion data is compared with the dynamic reference model established based on historical normal data. The dynamic reference model includes multiple cluster centers representing different normal working condition modes and their corresponding covariance matrices. By calculating the Mahalanobis distance between the current fusion data and each cluster center, it can be determined which normal working condition mode the current state is closest to. The cluster center with the smallest distance is selected, and its corresponding covariance matrix is obtained. The covariance matrix reflects the distribution characteristics of the fusion data dimensions (including impedance features, environment parameters and operating state) under the specific normal working condition. Using the covariance matrix, the confidence interval of the current fusion data belonging to the normal working condition mode can be calculated. This confidence interval defines the normal fluctuation range of the key feature parameters under the current environment and operating state. When judging, the current key feature parameters are compared with the calculated confidence interval. If the key feature parameters fall within the confidence interval, the grounding circuit state is considered normal; if it exceeds the confidence interval, it is determined as abnormal state. In order to further optimize the judgment performance, the system dynamically adjusts the width of the confidence interval according to the change trend of the environment parameters and equipment operating state. For example, when detecting that the environment temperature, humidity and other parameters change rapidly and greatly, the grounding circuit impedance may fluctuate temporarily, at this time, appropriately increasing the confidence interval can improve the system's tolerance to environmental changes, reduce the misjudgment of normal fluctuation as abnormal, and reduce the false alarm rate. On the contrary, when the equipment is in stable operating state for a long time, and the environment parameters are also relatively stable, the grounding circuit impedance characteristics should be more stable, at this time, reducing the confidence interval can improve the system's detection sensitivity to small abnormal changes, which is helpful to early detection of potential insulation deterioration signs. Through this adaptive adjustment mechanism based on dynamic reference model and real-time data, the system can more accurately reflect the normal state range under the current working condition, thereby improving the accuracy and reliability of grounding line state judgment in complex environment, effectively distinguishing normal change from abnormal state, and avoiding false alarm and missed alarm.

[0086] In some embodiments, the dynamic reference model can be based on the historical collected fusion data, and a K-means clustering algorithm is used to cluster the multi-dimensional fusion data under normal conditions to obtain K cluster centers. For each cluster center, the covariance matrix of the data within the cluster is calculated. For example, in a certain substation, historical data shows that the fusion data of the grounding loop impedance spectrum characteristics and the environmental / operating parameters under high temperature and high humidity in summer, low temperature and dryness in winter, and different load levels presents different distribution patterns. Through clustering analysis, the cluster centers and covariance matrices representing these typical normal conditions can be obtained. During real-time monitoring, the fusion data vector composed of the current temperature 25℃, humidity 70%, equipment load 80%, and the corresponding impedance spectrum key feature parameters is collected. The Mahalanobis distance of this vector from all cluster centers is calculated, and it is assumed that the distance from the cluster center representing the "high temperature and high humidity in summer, high load" condition is the smallest. The covariance matrix Σ corresponding to this cluster center is obtained. Using this Σ, the Mahalanobis distance square of the current fusion data vector is calculated, and a 95% confidence interval is determined according to the chi-square distribution. If the Mahalanobis distance square value of the current fusion data vector is less than the upper limit of the confidence interval, it is judged to be normal (it should be noted that when the fusion data vector is considered to follow a multivariate normal distribution, the square of its Mahalanobis distance follows a chi-square (x^2) distribution. The degrees of freedom of the chi-square distribution are equal to the number of dimensions of the fusion data vector, i.e. the total number of feature parameters contained in the fusion data. Therefore, the upper limit of the confidence interval is determined by looking up the critical value corresponding to a certain confidence level (such as 95% or 99%) on the chi-square distribution. This critical value represents that under normal conditions, a certain percentage (such as 95%) of the Mahalanobis distance square values will fall below this critical value. Any real-time calculated Mahalanobis distance square value that exceeds this critical value is considered to be outside the normal fluctuation range and is judged to be an abnormal state. For example, if the fusion data vector contains 3 feature parameters, the degrees of freedom of the chi-square distribution are 3. If a 95% confidence level is chosen, the system will look up the chi-square value with degrees of freedom 3 and cumulative probability 0.95 in the chi-square distribution table, which is the upper limit of the confidence interval). If the system detects that the environmental temperature suddenly rises from 25℃ to 35℃ in a short time, the judgment module can expand the upper limit of the confidence interval by 10% according to the pre-set rules or model, to adapt to the impedance fluctuation caused by the sudden temperature change and reduce the risk of false positives. Conversely, if the equipment load and environmental parameters remain stable for several hours, the judgment module can reduce the upper limit of the confidence interval by 5% to improve the detection ability of weak abnormal signals.

[0087] Reference is made to the accompanying drawings that Figure 2 The present application provides a grounding line monitoring system applied to power equipment of a substation, comprising:

[0088] The acquisition module 100 is configured to inject an excitation signal containing a plurality of discrete frequency points or a sweep signal into a grounding loop formed by the smart grounding wire and the grounding net, and synchronously acquire a voltage signal at the injection point and a current signal of the grounding loop;

[0089] The preprocessing module 200 is configured to preprocess the acquired voltage signal and current signal to obtain preprocessed voltage signal and current signal;

[0090] The extraction module 300 is configured to calculate the impedance spectrum of the grounding loop at different frequencies according to the preprocessed voltage signal and current signal, and extract key characteristic parameters in the impedance spectrum; the key characteristic parameters include impedance amplitude and phase in a specific frequency point or frequency band, slope and curvature of the impedance spectrum curve in a specific frequency range, frequency, amplitude and bandwidth of the resonance peak or anti-resonance peak in the spectrum, and spectrum energy distribution in a specific frequency band;

[0091] The fusion module 400 is configured to acquire environmental parameters and operating state data of the power equipment, and fuse the key characteristic parameters to obtain fusion data;

[0092] The adjustment module 500 is configured to adjust the judgment threshold or reference baseline corresponding to the key characteristic parameters according to the pre-established dynamic reference model or diagnostic rule set, using the fusion data; the judgment threshold is a numerical limit for distinguishing between normal state and abnormal state; the reference baseline is a normal value or range corresponding to the key characteristic parameters;

[0093] The judgment module 600 is configured to judge whether the current grounding loop state belongs to normal, early degradation or different degrees of electric leakage fault, using the adjusted judgment threshold or reference baseline.

[0094] In some embodiments, the prediction module 700 is further included, and the prediction module 700 is configured to perform the following steps:

[0095] Record and store historical impedance spectrum characteristic parameters, environmental parameters and equipment state information;

[0096] Obtain the change rate and direction of the key characteristic parameters over time by long-term trend analysis on the historical data;

[0097] According to the change rate and direction of the key characteristic parameters over time, evaluate the insulation degradation speed of the power equipment, and according to the evaluation result, issue an early warning before the power equipment fails.

[0098] In some embodiments, the preprocessing module 200 performs the following steps when preprocessing the acquired voltage signal and current signal to obtain preprocessed voltage signal and current signal:

[0099] The collected voltage signal and current signal are cleaned by using a digital notch filter or by a Fourier analysis method to suppress the power frequency and its harmonics, to obtain a first-processed voltage signal and a first-processed current signal;

[0100] For the first-processed voltage signal and the first-processed current signal, a time-frequency analysis method is used to identify and remove transient high-amplitude components, to obtain a second-processed voltage signal and a second-processed current signal;

[0101] For the second-processed voltage signal, a plurality of voltage values are continuously collected from the second-processed voltage signal according to a preset first measurement period, and all voltage values of the same first measurement period are averaged to obtain an average voltage value corresponding to the first measurement period, all average voltage values obtained for different first measurement periods are taken as a third-processed voltage signal, to reduce the random noise of the voltage signal, and are taken as a final pre-processed voltage signal, or the second-processed voltage signal is directly subjected to Kalman filtering to reduce the random noise of the voltage signal, to obtain a third-processed voltage signal, which is taken as a final pre-processed voltage signal;

[0102] For the second-processed current signal, a plurality of current values are continuously collected from the second-processed current signal according to a preset second measurement period, and all current values of the same second measurement period are averaged to obtain an average current value corresponding to the second measurement period, all average current values obtained for different second measurement periods are taken as a third-processed current signal, to reduce the random noise of the current signal, and are taken as a final pre-processed current signal, or the second-processed current signal is directly subjected to Kalman filtering to reduce the random noise of the current signal, to obtain a third-processed current signal, which is taken as a final pre-processed current signal.

[0103] In some embodiments, the fusion module 400 performs the following when collecting environmental parameters and operating state data of the power equipment and fusing with key feature parameters to obtain fusion data:

[0104] The environmental parameters of the power equipment include temperature, humidity, electromagnetic interference intensity, and the operating state data include voltage, current, active power and reactive power;

[0105] A wavelet threshold denoising method is used to filter out noise interference of the environmental parameters and the operating state data, and an interpolation algorithm based on historical data is used to fill in missing values, to obtain a clean and complete data set;

[0106] The mutual information method is used to evaluate the correlation of each environmental parameter and each running state data with respect to each key characteristic parameter, to obtain an evaluation result about the correlation of each environmental parameter and each running state data with respect to each key characteristic parameter, and to filter out environmental parameters and running state data with a correlation greater than a preset threshold according to the evaluation result, to obtain a filtered parameter set;

[0107] The Min-Max scaling method is used to normalize the environmental parameters and running state data in the filtered parameter set, so as to scale the environmental parameters and running state data to the interval [0, 1] and eliminate dimensional differences.

[0108] The normalized environmental parameters and running state data are fused with the key characteristic parameters by a weighted fusion algorithm to obtain a fused multi-dimensional feature vector, which is used as fusion data for subsequent adjustment of a dynamic reference model or a diagnosis rule set. The weight coefficient in the weighted fusion algorithm is determined by the corresponding mutual information value.

[0109] In some embodiments, the adjustment module 500 is configured to adjust the judgment threshold or reference baseline corresponding to the key characteristic parameter using the fusion data.

[0110] The judgment threshold or reference baseline corresponding to the key characteristic parameter is adjusted using the fusion data according to a pre-established dynamic reference model.

[0111] In some embodiments, the adjustment module 500 is configured to adjust the judgment threshold or reference baseline corresponding to the key characteristic parameter using the fusion data according to a pre-established dynamic reference model.

[0112] The current fusion data is obtained in real time, the Mahalanobis distance between the current fusion data and each normal operating condition cluster center is calculated, and the cluster center corresponding to the smallest Mahalanobis distance is selected.

[0113] The covariance matrix corresponding to the selected cluster center is obtained through the dynamic reference model, and the confidence interval of the current fusion data is calculated according to the covariance matrix.

[0114] The confidence interval is used as the judgment threshold or reference baseline of the key characteristic parameter for adaptive adjustment. If the current key characteristic parameter exceeds the confidence interval, it is determined to be an abnormal state, otherwise it is determined to be a normal state. The adjustment of the confidence interval includes: when a sudden change in the environmental parameter is detected, the confidence interval is increased to reduce the false alarm rate; when the equipment running state is stable, the confidence interval is reduced to improve the detection sensitivity.

[0115] In this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

[0116] The above description is merely illustrative of the application and not intended to be limiting. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described or shown herein, embody the principles of the application and are thus within its spirit and scope. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application should be included in the scope of the application.

Claims

1. A grounding wire monitoring method, applied to power equipment in a substation, characterized in that, Includes the following steps: An excitation signal containing multiple discrete frequency points or sweep frequency signals is injected into the grounding loop formed by the smart grounding wire and the grounding grid, and the voltage signal at the injection point and the current signal of the grounding loop are collected simultaneously. The acquired voltage and current signals are preprocessed to obtain preprocessed voltage and current signals; Based on the preprocessed voltage and current signals, the impedance spectrum of the grounding loop at different frequencies is calculated, and key characteristic parameters in the impedance spectrum are extracted. Collect environmental parameters and operating status data of power equipment, and fuse them with key characteristic parameters to obtain fused data; By utilizing fused data, the judgment thresholds or reference baselines corresponding to key feature parameters are adjusted; the judgment thresholds are numerical limits used to distinguish between normal and abnormal states; the reference baselines are the normal values ​​or ranges of the corresponding key feature parameters. The current grounding loop status is determined using the adjusted judgment threshold or reference baseline.

2. The grounding wire monitoring method according to claim 1, characterized in that, It also includes the following steps: Record and store historical impedance spectrum characteristic parameters, environmental parameters, and equipment status information; By conducting long-term trend analysis on historical data, the rate and direction of change of key characteristic parameters over time can be obtained. The rate and direction of change of key characteristic parameters over time are used to assess the insulation degradation rate of power equipment, and early warnings are issued before power equipment failures occur based on the assessment results.

3. The grounding wire monitoring method according to claim 1, characterized in that, The steps for preprocessing the acquired voltage and current signals to obtain preprocessed voltage and current signals include: The acquired voltage and current signals are cleaned by using a digital notch filter or by Fourier analysis to suppress power frequency and its harmonics, resulting in a single-processed voltage and current signal. For the voltage and current signals after primary processing, time-frequency analysis is used to identify and remove transient high-amplitude components to obtain the voltage and current signals after secondary processing. For the voltage signal after secondary processing, multiple voltage values ​​are continuously collected from the voltage signal after secondary processing according to a preset first measurement cycle, and the average of all voltage values ​​in the same first measurement cycle is calculated to obtain the average voltage value of the corresponding first measurement cycle. The average voltage values ​​obtained from all different first measurement cycles are used as the voltage signal after tertiary processing to reduce the random noise of the voltage signal and as the final pre-processed voltage signal. Alternatively, Kalman filtering can be directly applied to the voltage signal after secondary processing to reduce the random noise of the voltage signal and obtain the voltage signal after tertiary processing, which is also used as the final pre-processed voltage signal. For the current signal after secondary processing, multiple current values ​​are continuously collected from the current signal according to a preset second measurement cycle, and the average of all current values ​​in the same second measurement cycle is calculated to obtain the average current value of the corresponding second measurement cycle. The average current values ​​obtained from all different second measurement cycles are used as the current signal after tertiary processing to reduce the random noise of the current signal and as the final pre-processed current signal. Alternatively, Kalman filtering can be directly applied to the current signal after secondary processing to reduce the random noise of the current signal and obtain the current signal after tertiary processing, which is also used as the final pre-processed current signal.

4. The grounding wire monitoring method according to claim 1, characterized in that, Key characteristic parameters include impedance amplitude and phase at a specific frequency point or band, slope and curvature of the impedance spectrum curve within a specific frequency range, frequency, amplitude and bandwidth of resonance peaks or anti-resonance peaks in the spectrum, and spectral energy distribution in a specific frequency band.

5. The grounding wire monitoring method according to claim 1, characterized in that, The current grounding loop status includes normal, early deterioration, and varying degrees of leakage.

6. The grounding wire monitoring method according to claim 4, characterized in that, The steps for collecting environmental parameters and operating status data of power equipment and fusing them with key characteristic parameters to obtain fused data include: Collect environmental parameters of power equipment, including temperature, humidity, electromagnetic interference intensity, and operating status data, including voltage, current, active power and reactive power; Wavelet thresholding denoising method is used to filter out noise interference from environmental parameters and operational status data, and interpolation algorithm based on historical data is used to fill in missing values ​​to obtain a clean and complete dataset. The mutual information method is adopted to evaluate the correlation between each environmental parameter and each operating status data and each key characteristic parameter. The evaluation results of the correlation between each environmental parameter and each operating status data and each key characteristic parameter are obtained. Based on the evaluation results, environmental parameters and operating status data with a correlation greater than a preset threshold are selected to obtain the filtered parameter set. The Min-Max scaling method is used to normalize the environmental parameters and operational status data in the filtered parameter set. The normalized environmental parameters and operating status data are fused with key feature parameters through a weighted fusion algorithm to obtain a fused multidimensional feature vector, which is then used as the fused data.

7. The grounding wire monitoring method according to claim 1, characterized in that, The steps for adjusting the judgment thresholds or reference baselines corresponding to key feature parameters using fused data include: Based on a pre-established dynamic reference model, the judgment thresholds or reference baselines corresponding to key feature parameters are adjusted using fused data.

8. The grounding wire monitoring method according to claim 7, characterized in that, The steps for establishing a dynamic reference model include: Acquire historical operating data, which includes historical impedance spectrum characteristic parameters, historical environmental parameters, and historical equipment status information; Based on the historical operating data, a clustering algorithm is used to perform cluster analysis on the data under normal operating conditions, resulting in multiple normal operating condition cluster centers. By calculating the covariance matrix of each cluster center, a dynamic reference model under normal operating conditions is constructed.

9. The grounding wire monitoring method according to claim 8, characterized in that, The steps of adjusting the judgment thresholds or reference baselines corresponding to key feature parameters based on a pre-established dynamic reference model and using fused data include: Real-time acquisition of current fused data, calculation of Mahalanobis distance between current fused data and each normal operating condition cluster center, and selection of the cluster center corresponding to the smallest Mahalanobis distance; The covariance matrix corresponding to the selected cluster centers is obtained through a dynamic reference model, and the confidence interval of the current fused data is calculated based on the covariance matrix. The confidence interval is adaptively adjusted as a judgment threshold or reference baseline for key feature parameters. If the current key feature parameter exceeds the confidence interval, it is determined to be an abnormal state; otherwise, it is determined to be a normal state. The adjustment of the confidence interval includes: increasing the confidence interval when a sudden change in environmental parameters is detected to reduce the false alarm rate; and decreasing the confidence interval when the device is operating stably to improve detection sensitivity.

10. A grounding wire monitoring system, applied to power equipment in a substation, characterized in that, include: The acquisition module is used to inject an excitation signal containing multiple discrete frequency points or sweep frequency signals into the grounding loop formed by the smart grounding wire and the grounding grid, and simultaneously acquire the voltage signal at the injection point and the current signal of the grounding loop. The preprocessing module is used to preprocess the acquired voltage and current signals to obtain preprocessed voltage and current signals. The extraction module is used to calculate the impedance spectrum of the grounding loop at different frequencies based on the preprocessed voltage and current signals, and extract the key characteristic parameters in the impedance spectrum. The fusion module is used to collect environmental parameters and operating status data of power equipment and fuse them with key characteristic parameters to obtain fused data; The adjustment module is used to adjust the judgment threshold or reference baseline corresponding to the key feature parameters using the fused data; the judgment threshold is the numerical boundary used to distinguish between normal and abnormal states; the reference baseline is the normal value or range of the corresponding key feature parameter. The judgment module is used to determine the current grounding loop status using the adjusted judgment threshold or reference baseline.

Images