Ground fault identification method based on zero-sequence voltage periodic differential energy

Abstract

The invention provides a ground fault identification method, device and equipment based on zero-sequence voltage periodic differential energy and a storage medium, and the method comprises the steps: collecting and processing a zero-sequence voltage signal of a bus in a power distribution network system, and obtaining a zero-sequence voltage data sequence; according to the zero-sequence voltage data sequence, calculating zero-sequence voltage periodic differential energy of the bus to obtain a zero-sequence voltage periodic differential energy sequence; determining a zero-sequence voltage period differential energy threshold according to a preset condition, and comparing the zero-sequence voltage period differential energy threshold with the zero-sequence voltage period differential energy sequence to obtain a comparison result; and generating a grounding fault identification result of the power distribution network system according to the comparison result and the zero-sequence voltage signal. Through periodic differential energy detection of the zero sequence voltage, voltage waveform change caused by a grounding fault is captured, an energy threshold value is set, and an instantaneous disturbance signal is effectively filtered. Periodic difference calculation is not sensitive to noise interference and data loss, and the stability and reliability of detection are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution networks, and in particular to a ground fault identification method based on zero-sequence voltage periodic differential energy. Background Art

[0002] In the distribution network, ground fault is an important issue that affects the safety and reliability of the system, especially the ground fault in the low-resistance grounding system, which often manifests as abnormal phenomena such as voltage fluctuation and harmonic distortion. The rapid identification and location of ground fault is of great significance to improve the safe operation of the distribution network. However, most of the existing ground fault identification methods rely on simple threshold judgments of current and voltage, which are difficult to deal with complex high-resistance ground faults or short-term disturbances. Summary of the invention

[0003] The main purpose of the present invention is to solve the technical problem that most existing ground fault identification methods rely on simple threshold judgments of current and voltage, and are difficult to cope with complex high-resistance ground faults or short-term disturbances; A first aspect of the present invention provides a ground fault identification method based on zero-sequence voltage periodic differential energy, the ground fault identification method based on zero-sequence voltage periodic differential energy comprising: Collect and process the zero-sequence voltage signal of the busbar in the distribution network system to obtain a zero-sequence voltage data sequence; Calculating the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage period differential energy sequence; Determine a zero-sequence voltage cycle differential energy threshold according to a preset condition, and compare the zero-sequence voltage cycle differential energy threshold with the zero-sequence voltage cycle differential energy sequence to obtain a comparison result; A ground fault is identified on the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result.

[0004] Optionally, in a first implementation of the first aspect of the present invention, collecting and processing the zero-sequence voltage signal of the bus in the distribution network system to obtain the zero-sequence voltage data sequence includes: Sampling the zero-sequence voltage signal of the busbar in the distribution network system to obtain original zero-sequence voltage sampling data; Performing DC offset removal and high-frequency interference filtering processing on the original zero-sequence voltage sampling data to obtain zero-sequence voltage data; Detecting missing sampling points in the zero-sequence voltage data, and performing zero-value filling on the missing sampling points in the zero-sequence voltage data to obtain continuous zero-sequence voltage data; The continuous zero-sequence voltage data is subjected to data formatting and time stamp alignment processing to obtain a zero-sequence voltage data sequence.

[0005] Optionally, in a second implementation of the first aspect of the present invention, calculating the zero-sequence voltage periodic differential energy of the bus according to the zero-sequence voltage data sequence to obtain the zero-sequence voltage periodic differential energy sequence includes: Performing period segmentation processing on the zero-sequence voltage data sequence to obtain a plurality of single-period zero-sequence voltage data sub-sequences; Calculating voltage differences of corresponding sampling points in the single-cycle zero-sequence voltage data subsequences of adjacent cycles to obtain a zero-sequence voltage cycle difference sequence; A square operation is performed on the zero-sequence voltage period difference sequence to obtain a zero-sequence voltage period difference energy sequence.

[0006] Optionally, in a third implementation of the first aspect of the present invention, the preset condition includes a bus zero-sequence voltage when the distribution network system is operating normally; The step of determining a zero-sequence voltage period differential energy threshold according to a preset condition, and comparing the zero-sequence voltage period differential energy threshold with the zero-sequence voltage period differential energy sequence to obtain a comparison result includes: Performing peak detection on the bus zero-sequence voltage during normal operation of the distribution network system to obtain the bus zero-sequence voltage peak value, and performing superposition influence analysis of noise signals on the zero-sequence voltage data sequence to obtain an analysis result; Determine a margin coefficient of a zero-sequence voltage period differential energy threshold value according to the analysis result, and calculate the zero-sequence voltage period differential energy threshold value according to the bus zero-sequence voltage peak value and the margin coefficient; Comparing each zero-sequence voltage cycle differential energy in the zero-sequence voltage cycle differential energy sequence with the zero-sequence voltage cycle differential energy threshold to obtain an element-level comparison result sequence; A statistical analysis is performed on the element-level comparison result sequence, and the number and duration of zero-sequence voltage period differential energy exceeding the zero-sequence voltage period differential energy threshold are calculated to obtain a comparison result.

[0007] Optionally, in a fourth implementation manner of the first aspect of the present invention, the performing superposition influence analysis of noise signals on the zero-sequence voltage data sequence to obtain the analysis result includes: Performing a joint analysis of the zero-sequence voltage data sequence in the time-frequency domain to obtain a time-frequency characteristic matrix, and separating and extracting noise components in the zero-sequence voltage data sequence according to the time-frequency characteristic matrix to obtain a noise signal sequence; Performing statistical characteristic analysis on the noise signal sequence, calculating probability distribution parameters of the noise signal sequence, and obtaining noise statistical characteristics; The noise statistical characteristics are quantitatively evaluated according to a preset noise impact evaluation index and the amplitude information of the zero-sequence voltage data sequence to obtain a quantitative evaluation result of the noise superposition effect; The quantitative evaluation results are normalized and compared with the preset noise impact level to obtain analysis results.

[0008] Optionally, in a fifth implementation of the first aspect of the present invention, the performing ground fault identification on the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result includes: Analyzing whether there is zero-sequence voltage period differential energy exceeding a zero-sequence voltage period differential energy threshold value in the zero-sequence voltage period differential energy sequence according to the comparison result; If so, the zero-sequence voltage period differential energy exceeding the zero-sequence voltage period differential energy threshold is used as the target differential energy, and it is determined whether there is a zero-sequence current protection action state on the feeder of the distribution network system within the duration corresponding to the target differential energy; If there is no zero-sequence current protection action state, the ground fault identification result is determined to be a low-resistance ground fault; If there is a zero-sequence current protection action state, a statistical analysis is performed on the historical over-limit situation of the target differential energy to determine whether the over-limit situation of the target differential energy is the first over-limit situation within a preset time range; If the target differential energy exceeds the limit for the first time within a preset time range, determining that the ground fault identification result is a high-resistance ground fault; If the target differential energy is not exceeded for the first time within a preset time range, the ground fault identification result is determined to be intermittent isolated light grounding.

[0009] Optionally, in a sixth implementation of the first aspect of the present invention, after taking the zero-sequence voltage period differential energy exceeding the zero-sequence voltage period differential energy threshold as the target differential energy, the method further includes: Performing energy peak analysis on the target differential energy to obtain a differential energy peak, and identifying a start time of the differential energy peak; The starting time of the differential energy peak is used as the fault time of the ground fault, and the fault time is added to the ground fault identification result.

[0010] A second aspect of the present invention provides a ground fault identification device based on zero-sequence voltage periodic differential energy, the ground fault identification device based on zero-sequence voltage periodic differential energy comprising: The acquisition and processing module is used to acquire and process the zero-sequence voltage signal of the busbar in the distribution network system to obtain a zero-sequence voltage data sequence; A differential calculation module, used for calculating the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage period differential energy sequence; A comparison module, used for determining a zero-sequence voltage period differential energy threshold according to a preset condition, and comparing the zero-sequence voltage period differential energy threshold with the zero-sequence voltage period differential energy sequence to obtain a comparison result; The identification module is used to identify the ground fault of the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result.

[0011] A third aspect of the present invention provides a ground fault identification device based on zero-sequence voltage periodic differential energy, comprising: a memory and at least one processor, wherein instructions are stored in the memory, and the memory and the at least one processor are interconnected through a line; the at least one processor calls the instructions in the memory so that the ground fault identification device based on zero-sequence voltage periodic differential energy performs the steps of the above-mentioned ground fault identification method based on zero-sequence voltage periodic differential energy.

[0012] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the steps of the above-mentioned ground fault identification method based on zero-sequence voltage periodic differential energy.

[0013] The above-mentioned ground fault identification method, device, equipment and storage medium based on zero-sequence voltage periodic differential energy obtain a zero-sequence voltage data sequence by collecting and processing the zero-sequence voltage signal of the bus in the distribution network system; according to the zero-sequence voltage data sequence, the zero-sequence voltage periodic differential energy of the bus is calculated to obtain a zero-sequence voltage periodic differential energy sequence; the zero-sequence voltage periodic differential energy threshold is determined according to preset conditions, and the zero-sequence voltage periodic differential energy threshold is compared with the zero-sequence voltage periodic differential energy sequence to obtain a comparison result; the ground fault identification result of the distribution network system is generated according to the comparison result and the zero-sequence voltage signal. The present invention captures the voltage waveform changes caused by the ground fault through the periodic differential energy detection of the zero-sequence voltage, and sets the energy threshold to effectively filter the instantaneous disturbance signal. The periodic differential calculation is insensitive to noise interference and data loss, which ensures the stability and reliability of the detection.

[0014] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0015] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of a first embodiment of a ground fault identification method based on zero-sequence voltage periodic differential energy in an embodiment of the present invention; Figure 2 It is a schematic diagram of an embodiment of a ground fault identification device based on zero-sequence voltage periodic differential energy in an embodiment of the present invention; Figure 3 The figure is a schematic diagram of an embodiment of a ground fault identification device based on zero-sequence voltage periodic differential energy in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0019] To facilitate understanding of this embodiment, a ground fault identification method based on zero-sequence voltage period differential energy disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps: 101. Collect and process the zero-sequence voltage signal of the busbar in the distribution network system to obtain a zero-sequence voltage data sequence; In one embodiment of the present invention, the zero-sequence voltage signal of the bus in the distribution network system is collected and processed to obtain a zero-sequence voltage data sequence, including: sampling the zero-sequence voltage signal of the bus in the distribution network system to obtain original zero-sequence voltage sampling data; performing DC offset removal and high-frequency interference filtering on the original zero-sequence voltage sampling data to obtain zero-sequence voltage data; detecting missing sampling points in the zero-sequence voltage data, and filling the missing sampling points in the zero-sequence voltage data with zero values ​​to obtain continuous zero-sequence voltage data; performing data formatting and timestamp alignment on the continuous zero-sequence voltage data to obtain a zero-sequence voltage data sequence.

[0020] Specifically, first, high-precision voltage transformers (PTs) are installed on the busbars of the distribution network system. These PTs are usually connected in an open triangle or a star connection to accurately measure the zero-sequence voltage. Signal acquisition equipment, such as intelligent electronic devices (IEDs) or digital fault recorders (DFRs), are connected to the PTs to perform high-speed sampling of zero-sequence voltage signals. The sampling frequency is usually set to an integer multiple of the system power frequency, such as 10kHz (200 sampling points per cycle for a 50Hz system). This high sampling rate can capture rapidly changing transient phenomena, which is critical for accurately identifying ground faults. During the sampling process, the analog-to-digital converter (ADC) converts the analog signal into a digital signal to form the original zero-sequence voltage sampling data. For example, in a 10kV distribution system, an IED with a sampling frequency of 10kHz is used to sample the zero-sequence voltage, and 10,000 sampling points are obtained within 1 second, which constitute the original zero-sequence voltage sampling data sequence.

[0021] Specifically, after obtaining the original zero-sequence voltage sampling data, signal preprocessing is required to improve data quality. First, DC offset removal is performed. This step aims to eliminate the DC component due to imperfect measurement equipment or the system itself, ensuring that the signal fluctuates around zero. DC offset removal is usually achieved using a moving average method or a high-pass filter. Next, a digital filter is used to remove high-frequency interference. Common filtering methods include Butterworth filters, Chebyshev filters, or finite impulse response (FIR) filters. The cutoff frequency of the filter is usually set to several times the power frequency to retain useful harmonic information while removing high-frequency noise. For example, for a 50Hz system, a low-pass filter with a cutoff frequency of 1000Hz can be set.

[0022] Specifically, in actual operation, due to communication failure or temporary equipment failure, sampling points may be lost. In order to ensure the continuity and integrity of the data, it is necessary to detect and fill missing sampling points for zero-sequence voltage data. The detection method is usually based on the time interval analysis between sampling points. For example, if the time interval between two adjacent sampling points is significantly greater than the expected sampling period (such as greater than 1.5 times the normal sampling period), it is considered that there is a missing sampling point in the middle. For the detected missing sampling points, the zero-value filling method is used for processing. This method is simple and effective. Although it may introduce certain errors, it has little effect on short-term data loss. In some high-demand scenarios, more complex methods such as linear interpolation or spline interpolation can also be used. For example, in a system with a sampling rate of 10kHz, if the time interval between two adjacent sampling points is detected to be 300μs (normal should be 100μs), two zero values ​​are inserted between these two points to maintain data continuity.

[0023] Specifically, the last step is to format and timestamp-align the processed continuous zero-sequence voltage data for subsequent analysis. Data formatting includes unifying data types (such as converting all data to 32-bit floating point numbers), adjusting data precision, etc. Timestamp alignment is to ensure that each sampling point has accurate time information, which is crucial for subsequent fault moment location and various time-related analyses. Timestamp alignment can use interpolation techniques such as linear interpolation or cubic spline interpolation to obtain more accurate time information. Through this interpolation, a more accurate timestamp can be assigned to each sampling point, improving the accuracy of subsequent analysis. For example, when processing 1 second of data, if the original timestamp accuracy is only at the millisecond level, the timestamp accuracy can be increased to the microsecond level through interpolation, thereby assigning a unique, high-precision time identifier to each sampling point.

[0024] 102. Calculate the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage period differential energy sequence; In one embodiment of the present invention, the zero-sequence voltage periodic differential energy of the bus is calculated according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage periodic differential energy sequence, including: performing period segmentation processing on the zero-sequence voltage data sequence to obtain multiple single-cycle zero-sequence voltage data sub-sequences; calculating the voltage difference of corresponding sampling points in the single-cycle zero-sequence voltage data sub-sequences of adjacent cycles to obtain a zero-sequence voltage periodic differential sequence; and performing a square operation on the zero-sequence voltage periodic differential sequence to obtain a zero-sequence voltage periodic differential energy sequence.

[0025] Specifically, first, the obtained zero-sequence voltage data sequence is processed by period segmentation. The purpose of this step is to divide the continuous zero-sequence voltage data sequence into multiple single-cycle data subsequences to prepare for the subsequent period differential energy calculation. Period segmentation requires accurate identification of the start and end points of each power frequency cycle. Considering the possible frequency fluctuations in the actual system, it cannot be simply divided according to a fixed number of sampling points. Instead, the zero-crossing point detection method or the Fourier transform method can be used to achieve accurate period identification. The zero-crossing point detection method determines the period boundary by finding the moment when the signal crosses the zero point from positive to negative (or from negative to positive). The Fourier transform rule determines the main frequency component by analyzing the spectrum of the signal, thereby calculating the accurate period length. In practical applications, these two methods can be combined, first using the Fourier transform method to estimate the approximate period length, and then using the zero-crossing point detection method to accurately locate the period boundary to obtain the best segmentation effect. For example, for a 50Hz system with a sampling frequency of 10kHz, each cycle theoretically contains 200 sampling points. However, in actual operation, it may be found that some cycles contain 199 or 201 sampling points, which reflects the slight fluctuation of the system frequency. Through precise cycle segmentation, we can obtain a series of single-cycle zero-sequence voltage data subsequences, each of which accurately corresponds to a complete power frequency cycle.

[0026] Specifically, after obtaining a single-cycle zero-sequence voltage data subsequence, the next step is to calculate the voltage difference between the corresponding sampling points in adjacent cycles. This step is the core of the zero-sequence voltage cycle differential energy calculation. For each sampling point, we need to calculate the voltage difference between it and the sampling point at the corresponding moment in the next cycle. This difference calculation can effectively eliminate the periodic component under normal operating conditions and highlight the non-periodic changes caused by faults or disturbances. Under normal operating conditions, due to the periodic characteristics of the power system, Should be with is very close, so Δ should be close to zero. However, when a ground fault or other disturbance occurs in the system, this periodicity will be broken, resulting in Δ For example, suppose that at a certain time t, = 100V, and at the corresponding moment after one cycle, due to the occurrence of a ground fault, = 150V, then Δ = 50V. This large difference indicates an abnormal change in the system state. By performing such a difference calculation on the entire data sequence, we can obtain a zero-sequence voltage cycle difference sequence, and each element in this sequence reflects the change in the system state at the corresponding moment. Finally, the zero-sequence voltage cycle difference sequence is converted into a zero-sequence voltage cycle differential energy sequence. The purpose of this step is to further amplify the signal changes caused by faults or disturbances and improve the sensitivity of detection. The calculation method is to perform a square operation on each element in the zero-sequence voltage cycle difference sequence. The square operation has two important functions: first, it eliminates the positive and negative signs of the difference, so that all changes are converted into positive values, which is convenient for subsequent threshold comparisons; second, the square operation has an amplifying effect on large differences and has less effect on small differences, which helps to highlight significant changes in system states. The calculation of the cycle differential energy of the above process can be expressed as: ; in, is the zero-sequence voltage period differential energy at time t, and are the zero-sequence voltage values ​​at time t and the corresponding time after one cycle. In practical applications, the calculation results can be normalized as needed to facilitate comparison between different systems. For example, assuming that at a certain time t, the calculated zero-sequence voltage cycle difference is = 50V, then the corresponding cycle differential energy E(t) = = 2500V. This large energy value clearly indicates a significant change in the system state. In contrast, the cycle-to-cycle differential energy under normal operating conditions is usually much smaller. For example, if = 1V (indicating a very small fluctuation), then E(t) = 1^2 = 1V. This small energy value can usually be considered as a normal system fluctuation.

[0027] 103. Determine a zero-sequence voltage cycle differential energy threshold according to a preset condition, and compare the zero-sequence voltage cycle differential energy threshold with the zero-sequence voltage cycle differential energy sequence to obtain a comparison result; In one embodiment of the present invention, the preset condition includes the bus zero-sequence voltage when the distribution network system is operating normally; the zero-sequence voltage cycle differential energy threshold is determined according to the preset condition, and the zero-sequence voltage cycle differential energy threshold is compared with the zero-sequence voltage cycle differential energy sequence to obtain the comparison result, including: performing peak detection on the bus zero-sequence voltage when the distribution network system is operating normally to obtain the bus zero-sequence voltage peak value, and performing superposition influence analysis of noise signals on the zero-sequence voltage data sequence to obtain an analysis result; determining a margin coefficient of the zero-sequence voltage cycle differential energy threshold according to the analysis result, and calculating the zero-sequence voltage cycle differential energy threshold according to the bus zero-sequence voltage peak value and the margin coefficient; comparing each zero-sequence voltage cycle differential energy in the zero-sequence voltage cycle differential energy sequence with the zero-sequence voltage cycle differential energy threshold to obtain an element-level comparison result sequence; performing statistical analysis on the element-level comparison result sequence to calculate the number and duration of zero-sequence voltage cycle differential energy exceeding the zero-sequence voltage cycle differential energy threshold to obtain a comparison result.

[0028] Specifically, first, the peak value of the bus zero-sequence voltage is detected when the distribution network system is in normal operation. The purpose of this step is to obtain the reference value of the zero-sequence voltage under normal operation of the system, which provides a basis for the subsequent threshold calculation. The peak detection can use the sliding window method to find the maximum value within a certain time window, and perform statistical analysis through the results of multiple windows to eliminate the influence of accidental factors. At the same time, the superposition influence of noise signals is analyzed on the zero-sequence voltage data sequence. This analysis process can use a joint analysis method in the time and frequency domains, such as wavelet packet transform or Hilbert-Huang transform. These methods can provide information in the time domain and frequency domain at the same time, which helps to separate the noise components in the signal. By analyzing the wavelet packet coefficients of different scales and frequency bands, the energy distribution of the noise and the degree of influence on the signal can be estimated. For example, suppose that in a 10kV distribution system, the peak value of the bus zero-sequence voltage during normal operation is 100V through peak detection. However, through wavelet packet analysis, it is found that there is significant noise energy in the high frequency band (such as above 500Hz), accounting for 5% of the total signal energy. This information will be used for subsequent margin coefficient determination and threshold calculation.

[0029] Specifically, the margin coefficient is introduced to consider the noise level and possible normal fluctuations of the system to avoid false alarms due to minor disturbances. The margin coefficient can be determined by an adaptive method and dynamically adjusted according to the noise analysis results. For example, the margin coefficient is calculated using the nonlinear mapping relationship between the noise energy ratio and the margin coefficient, such as the following formula: ; Where k is the margin factor, is the estimated noise energy, is the total signal energy, and α and β are pre-set parameters based on system characteristics. This formula reflects the relationship that the higher the noise level, the larger the margin factor is required, and at the same time limits the growth rate of the margin factor through the logarithmic function. After determining the margin factor, the zero-sequence voltage period differential energy threshold can be calculated using the following formula: ; in, is the zero-sequence voltage cycle differential energy threshold, k is the margin coefficient, is the peak value of the zero-sequence voltage of the busbar. This formula takes into account the maximum zero-sequence voltage change that may occur under normal conditions and is appropriately amplified by the margin factor. After obtaining the zero-sequence voltage cycle differential energy threshold, the next step is to compare it with each element in the zero-sequence voltage cycle differential energy sequence. This process produces a Boolean element-level comparison result sequence, in which each element indicates whether the cycle differential energy at the corresponding moment exceeds the threshold. After obtaining the element-level comparison result sequence, it is necessary to perform statistical analysis on it to calculate the number of elements exceeding the threshold and the duration. This analysis process can use the sliding window method to count the number of "1"s and the longest continuous appearance time within a certain time window. For example, suppose that in a window containing 1000 sampling points (corresponding to 100ms), it is found that the cycle differential energy of 50 points exceeds the threshold, and the longest continuous exceeding time is 20 points (2ms). These statistical results constitute the final comparison result.

[0030] Furthermore, the superposition influence analysis of the noise signal on the zero-sequence voltage data sequence to obtain the analysis result includes: performing a joint analysis of the zero-sequence voltage data sequence in the time-frequency domain to obtain a time-frequency feature matrix, and separating and extracting the noise components in the zero-sequence voltage data sequence according to the time-frequency feature matrix to obtain a noise signal sequence; performing a statistical characteristic analysis on the noise signal sequence to calculate the probability distribution parameters of the noise signal sequence to obtain noise statistical characteristics; performing a quantitative evaluation of the noise statistical characteristics according to a preset noise impact evaluation index and the amplitude information of the zero-sequence voltage data sequence to obtain a quantitative evaluation result of the noise superposition influence; and normalizing the quantitative evaluation result and comparing it with a preset noise impact level to obtain an analysis result.

[0031] Specifically, first, the zero-sequence voltage data sequence is jointly analyzed in the time and frequency domains to obtain the complete characteristics of the signal in the time and frequency dimensions. This step uses the Wavelet Packet Transform (WPT) method, which can provide multi-resolution analysis of the signal and is suitable for processing non-stationary signals. WPT decomposes the signal into sub-signals of different frequency bands and time scales to form a time-frequency feature matrix. The mathematical expression of WPT is as follows: ; in, are the wavelet packet coefficients, is the input zero-sequence voltage signal, is the wavelet packet basis function, Represent the scale, translation and oscillation parameters respectively. This formula describes the decomposition process of the signal at different times, frequencies and scales. After obtaining the time-frequency feature matrix, the noise component can be identified by analyzing the energy distribution of each frequency band. Usually, the noise is mainly concentrated in the high-frequency part, while the useful signal is mainly distributed in the low-frequency area. Therefore, a frequency threshold can be set, and the components above the threshold are regarded as noise. Through inverse transformation, the noise signal sequence can be reconstructed. For example, suppose a zero-sequence voltage signal with a sampling frequency of 10kHz is decomposed by a 6-layer wavelet packet. The analysis results show that the energy of the 5th and 6th layers (corresponding to frequencies above 250Hz) accounts for 5% of the total energy and is evenly distributed in time. This part can be regarded as a noise component, and the noise signal sequence is obtained through inverse transformation.

[0032] Specifically, after obtaining the noise signal sequence, the next step is to perform statistical characteristic analysis to quantify the characteristics of the noise. This step mainly includes calculating the probability distribution parameters of the noise signal. Assuming that the noise conforms to the generalized Gaussian distribution (GGD), its probability density function can be expressed as: ; Among them, α is the scale parameter, β is the shape parameter, μ is the location parameter, Γ(·) is the gamma function, and x represents the amplitude of each sampling point in the noise signal sequence. This distribution function can describe probability distributions of various shapes, from Laplace distribution (β=1) to Gaussian distribution (β=2). Through maximum likelihood estimation or moment estimation method, these parameters can be estimated from the noise signal sequence to obtain the statistical characteristics of the noise. For example, for the noise signal sequence obtained in the previous step, the parameters may be obtained by fitting: α=0.5, β=1.8, μ=0. This shows that the noise is close to the Gaussian distribution, but with a slight deviation, and there may be some impulsive interference.

[0033] Specifically, based on the noise statistical characteristics and the preset noise impact evaluation index, combined with the amplitude information of the zero-sequence voltage data sequence, a quantitative evaluation of the noise superposition effect is performed. The evaluation index may include signal-to-noise ratio (SNR), peak signal-to-noise ratio (PSNR), etc. Here, a comprehensive evaluation index I is introduced, and its calculation formula is: ; Among them, SNR is the signal-to-noise ratio, σs is the signal standard deviation, and w1, w2, and w3 are weight coefficients. This indicator comprehensively considers the signal-to-noise ratio, noise intensity (through reflection) and the shape of the noise distribution (through For example, if the calculated SNR=20dB, =0.1, β=1.8, weights w1=0.5, w2=0.3, w3=0.2, then the comprehensive index I = 0.520 + 0.30.1+ 0.2*(1 / 1.8) ≈ 10.11. This value reflects the comprehensive impact of noise on the signal.

[0034] Specifically, finally, the quantitative evaluation results are normalized to fall into the interval [0,1], which is convenient for comparison with the preset noise impact level. Normalization can use the min-max method: ; in, is the normalized index value, and are the expected minimum and maximum indicator values, respectively. After normalization, Compare with the preset noise impact level threshold to get the final analysis result. For example, suppose the preset noise impact level is: low impact (0-0.3), medium impact (0.3-0.6), high impact (0.6-1). If =0, =20, then the normalized value corresponding to I=10.11 calculated in the previous step is ≈ 0.51. This falls into the "moderate impact" range, indicating that the noise has some impact on the zero-sequence voltage signal, but has not yet reached the level of serious interference.

[0035] 104. According to the comparison result and the zero-sequence voltage signal, the ground fault of the distribution network system is identified to obtain a ground fault identification result.

[0036] In one embodiment of the present invention, the ground fault identification of the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain the ground fault identification result includes: analyzing whether there is a zero-sequence voltage cycle differential energy exceeding a zero-sequence voltage cycle differential energy threshold in the zero-sequence voltage cycle differential energy sequence according to the comparison result; if so, taking the zero-sequence voltage cycle differential energy exceeding the zero-sequence voltage cycle differential energy threshold as the target differential energy, and judging whether there is a zero-sequence current protection action state on the feeder of the distribution network system during the duration corresponding to the target differential energy; if there is no zero-sequence current protection action state, determining that the ground fault identification result is a low-resistance ground fault; if there is a zero-sequence current protection action state, performing statistical analysis on the historical limit crossing of the target differential energy to judge whether the limit crossing of the target differential energy is the first limit crossing within a preset time range; if the limit crossing of the target differential energy is the first limit crossing within a preset time range, determining that the ground fault identification result is a high-resistance ground fault; if the limit crossing of the target differential energy is not the first limit crossing within the preset time range, determining that the ground fault identification result is an intermittent solitary light grounding.

[0037] Specifically, first, the zero-sequence voltage cycle differential energy sequence is analyzed according to the comparison results to identify the energy value exceeding the threshold. The core of this step is to traverse the entire cycle differential energy sequence, compare each energy value with the preset threshold, and then identify the target differential energy. After identifying the target differential energy, the next step is to determine whether there is a zero-sequence current protection action state on the feeder of the distribution network system during the duration of the target differential energy. This requires access to the protection device status information of the distribution system. Usually, the action threshold of the zero-sequence current protection is set low so that low-resistance grounding faults can be detected. When the zero-sequence voltage cycle differential energy exceeds the threshold, but the zero-sequence current protection does not act, it is determined to be a low-resistance grounding fault. This is because low-resistance grounding faults usually cause large zero-sequence voltage changes, but due to the low fault impedance, the zero-sequence current may have triggered a faster protection action (such as quick-break protection), resulting in the zero-sequence current protection not being able to act in time to remove the fault. If the zero-sequence current protection does not act, it is necessary to further analyze the historical over-limit situation of the target differential energy. This step involves statistical analysis of the zero-sequence voltage cycle differential energy over a period of time (usually a few minutes to tens of minutes). This analysis can be implemented using the sliding time window method. When the zero-sequence voltage cycle differential energy exceeds the threshold, the zero-sequence current protection is activated, and it is the first time that the limit is exceeded, it is determined to be a high-resistance grounding fault. The characteristic of a high-resistance grounding fault is that the zero-sequence voltage changes significantly, but due to the presence of high impedance, the zero-sequence current is relatively small, and may just reach the action threshold of the zero-sequence current protection. The judgment of the first limit-exceeding is to distinguish between continuous high-resistance grounding and intermittent arc grounding. When the zero-sequence voltage cycle differential energy exceeds the threshold, the zero-sequence current protection is activated, and it is not the first time that the limit is exceeded, it is determined to be intermittent arc grounding. The characteristic of intermittent arc grounding is that the fault will appear and disappear periodically, so there will be multiple limit-exceeding situations within a certain time range. This fault is usually caused by insulation aging or contamination, and is self-extinguishing and repeatable.

[0038] Furthermore, after taking the zero-sequence voltage cycle differential energy exceeding the zero-sequence voltage cycle differential energy threshold as the target differential energy, it also includes: performing energy peak analysis on the target differential energy to obtain the differential energy peak, and identifying the starting time of the differential energy peak; taking the starting time of the differential energy peak as the fault time of the ground fault, and adding the fault time to the ground fault identification result.

[0039] Specifically, after identifying the target differential energy, it is first necessary to perform energy peak analysis on it. This process aims to accurately locate the critical moment of energy mutation from the target differential energy sequence, which usually corresponds to the moment of occurrence of the ground fault. Energy peak analysis can be achieved by calculating the first-order derivative of the target differential energy sequence. Specifically, the energy change rate can be calculated using the following formula: ; in, is the rate of change of energy at time t, is the zero-sequence voltage period differential energy at time t, is the sampling interval. This formula approximately calculates the instantaneous rate of change of the energy sequence. The value of can identify the moment when the energy rises rapidly, that is, the starting point of the energy peak. For example, assuming that in a 10kV distribution system, the sampling interval Δt is 1ms, at t=100ms, the calculated Suddenly jumping from the average value of 50V^2 / ms to 500V^2 / ms, this mutation point may be the starting moment of the energy peak.

[0040] Specifically, after obtaining the energy change rate sequence, the next step is to accurately identify the start time of the differential energy peak. This can be achieved by setting a change rate threshold and finding the first moment that exceeds the threshold. The following mathematical expression can be used to describe this process: ; in, is the identified peak start time, k is a preset multiple, is the standard deviation of the energy change rate. This expression looks for the first change rate point that is significantly higher than the normal fluctuation. The introduction of the standard deviation takes into account the normal fluctuation range of the system, making the threshold setting more adaptive. For example, if the standard deviation of the energy change rate is calculated The value of k=4 is 20V^2 / ms, and the change rate threshold is 80V^2 / ms. When the change rate at t=99.8ms is detected to exceed this threshold for the first time and reaches 85V^2 / ms, this moment is determined as the peak start time. Finally, after identifying the start time of the differential energy peak, it is directly used as the fault moment of the ground fault. This is based on the assumption that the beginning of the energy peak usually marks a sudden change in the system state, and in the case of a ground fault, this sudden change point is likely to be the moment when the fault occurs. After determining the fault moment, it needs to be added to the ground fault identification result.

[0041] In this embodiment, the zero-sequence voltage signal of the bus in the distribution network system is collected and processed to obtain a zero-sequence voltage data sequence; the zero-sequence voltage periodic differential energy of the bus is calculated according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage periodic differential energy sequence; the zero-sequence voltage periodic differential energy threshold is determined according to a preset condition, and the zero-sequence voltage periodic differential energy threshold is compared with the zero-sequence voltage periodic differential energy sequence to obtain a comparison result; the ground fault identification result of the distribution network system is generated according to the comparison result and the zero-sequence voltage signal. The present invention captures the voltage waveform changes caused by the ground fault through periodic differential energy detection of the zero-sequence voltage, and sets the energy threshold to effectively filter the instantaneous disturbance signal. The periodic differential calculation is insensitive to noise interference and data loss, which ensures the stability and reliability of the detection.

[0042] The above describes the ground fault identification method based on zero-sequence voltage cycle differential energy in the embodiment of the present invention. The following describes the ground fault identification device based on zero-sequence voltage cycle differential energy in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a ground fault identification device based on zero-sequence voltage periodic differential energy includes: The acquisition and processing module 201 is used to acquire and process the zero-sequence voltage signal of the busbar in the distribution network system to obtain a zero-sequence voltage data sequence; The differential calculation module 202 is used to calculate the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage period differential energy sequence; A comparison module 203 is used to determine a zero-sequence voltage cycle differential energy threshold according to a preset condition, and compare the zero-sequence voltage cycle differential energy threshold with the zero-sequence voltage cycle differential energy sequence to obtain a comparison result; The identification module 204 is used to identify the ground fault of the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result.

[0043] In an embodiment of the present invention, the ground fault identification device based on zero-sequence voltage periodic differential energy runs the above-mentioned ground fault identification method based on zero-sequence voltage periodic differential energy, and the ground fault identification device based on zero-sequence voltage periodic differential energy obtains a zero-sequence voltage data sequence by collecting and processing the zero-sequence voltage signal of the bus in the distribution network system; calculates the zero-sequence voltage periodic differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage periodic differential energy sequence; determines the zero-sequence voltage periodic differential energy threshold according to preset conditions, and compares the zero-sequence voltage periodic differential energy threshold with the zero-sequence voltage periodic differential energy sequence to obtain a comparison result; generates a ground fault identification result of the distribution network system according to the comparison result and the zero-sequence voltage signal. The present invention captures the voltage waveform changes caused by the ground fault through the periodic differential energy detection of the zero-sequence voltage, and sets the energy threshold to effectively filter the instantaneous disturbance signal. The periodic differential calculation is insensitive to noise interference and data loss, which ensures the stability and reliability of the detection.

[0044] above Figure 2 The ground fault identification device based on zero-sequence voltage periodic differential energy in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The ground fault identification device based on zero-sequence voltage periodic differential energy in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0045] Figure 3 1 is a schematic diagram of the structure of a ground fault identification device based on zero-sequence voltage cycle differential energy provided by an embodiment of the present invention. The ground fault identification device 300 based on zero-sequence voltage cycle differential energy may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 may be temporary storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the ground fault identification device 300 based on zero-sequence voltage cycle differential energy. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the ground fault identification device 300 based on zero-sequence voltage cycle differential energy to implement the steps of the ground fault identification method based on zero-sequence voltage cycle differential energy.

[0046] The ground fault identification device 300 based on zero-sequence voltage period differential energy may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art may understand that Figure 3 The structure of the ground fault identification device based on zero-sequence voltage periodic differential energy shown does not constitute a limitation on the ground fault identification device based on zero-sequence voltage periodic differential energy provided by the present invention, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0047] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the ground fault identification method based on zero-sequence voltage periodic differential energy.

[0048] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0049] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0050] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A ground fault identification method based on zero-sequence voltage periodic differential energy, characterized in that: The ground fault identification method based on zero-sequence voltage period differential energy includes: Collect and process the zero-sequence voltage signal of the busbar in the distribution network system to obtain a zero-sequence voltage data sequence; Calculating the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage period differential energy sequence; Determine a zero-sequence voltage cycle differential energy threshold according to a preset condition, and compare the zero-sequence voltage cycle differential energy threshold with the zero-sequence voltage cycle differential energy sequence to obtain a comparison result; A ground fault is identified on the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result.

2. The ground fault identification method based on zero-sequence voltage periodic differential energy according to claim 1 is characterized in that: The collecting and processing of the zero-sequence voltage signal of the busbar in the distribution network system to obtain the zero-sequence voltage data sequence comprises: Sampling the zero-sequence voltage signal of the busbar in the distribution network system to obtain original zero-sequence voltage sampling data; Performing DC offset removal and high-frequency interference filtering processing on the original zero-sequence voltage sampling data to obtain zero-sequence voltage data; Detecting missing sampling points in the zero-sequence voltage data, and performing zero-value filling on the missing sampling points in the zero-sequence voltage data to obtain continuous zero-sequence voltage data; The continuous zero-sequence voltage data is subjected to data formatting and time stamp alignment processing to obtain a zero-sequence voltage data sequence.

3. The ground fault identification method based on zero-sequence voltage periodic differential energy according to claim 1, characterized in that: The step of calculating the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain the zero-sequence voltage period differential energy sequence comprises: Performing period segmentation processing on the zero-sequence voltage data sequence to obtain a plurality of single-period zero-sequence voltage data sub-sequences; Calculating voltage differences of corresponding sampling points in the single-cycle zero-sequence voltage data subsequences of adjacent cycles to obtain a zero-sequence voltage cycle difference sequence; A square operation is performed on the zero-sequence voltage period difference sequence to obtain a zero-sequence voltage period difference energy sequence.

4. The ground fault identification method based on zero-sequence voltage periodic differential energy according to claim 1, characterized in that: The preset conditions include the zero-sequence voltage of the busbar when the distribution network system is operating normally; The step of determining a zero-sequence voltage period differential energy threshold according to a preset condition, and comparing the zero-sequence voltage period differential energy threshold with the zero-sequence voltage period differential energy sequence to obtain a comparison result includes: Performing peak detection on the bus zero-sequence voltage during normal operation of the distribution network system to obtain the bus zero-sequence voltage peak value, and performing superposition influence analysis of noise signals on the zero-sequence voltage data sequence to obtain an analysis result; Determine a margin coefficient of a zero-sequence voltage period differential energy threshold value according to the analysis result, and calculate the zero-sequence voltage period differential energy threshold value according to the bus zero-sequence voltage peak value and the margin coefficient; Comparing each zero-sequence voltage cycle differential energy in the zero-sequence voltage cycle differential energy sequence with the zero-sequence voltage cycle differential energy threshold to obtain an element-level comparison result sequence; A statistical analysis is performed on the element-level comparison result sequence, and the number and duration of zero-sequence voltage period differential energy exceeding the zero-sequence voltage period differential energy threshold are calculated to obtain a comparison result.

5. The ground fault identification method based on zero-sequence voltage periodic differential energy according to claim 4 is characterized in that: The performing of the superposition influence analysis of the noise signal on the zero-sequence voltage data sequence to obtain the analysis result includes: Performing a joint analysis of the zero-sequence voltage data sequence in the time-frequency domain to obtain a time-frequency characteristic matrix, and separating and extracting noise components in the zero-sequence voltage data sequence according to the time-frequency characteristic matrix to obtain a noise signal sequence; Performing statistical characteristic analysis on the noise signal sequence, calculating probability distribution parameters of the noise signal sequence, and obtaining noise statistical characteristics; The noise statistical characteristics are quantitatively evaluated according to a preset noise impact evaluation index and the amplitude information of the zero-sequence voltage data sequence to obtain a quantitative evaluation result of the noise superposition effect; The quantitative evaluation results are normalized and compared with the preset noise impact level to obtain analysis results.

6. The ground fault identification method based on zero-sequence voltage periodic differential energy according to claim 1, characterized in that: The performing ground fault identification on the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result comprises: Analyzing whether there is zero-sequence voltage period differential energy exceeding a zero-sequence voltage period differential energy threshold value in the zero-sequence voltage period differential energy sequence according to the comparison result; If so, the zero-sequence voltage period differential energy exceeding the zero-sequence voltage period differential energy threshold is used as the target differential energy, and it is determined whether there is a zero-sequence current protection action state on the feeder of the distribution network system within the duration corresponding to the target differential energy; If there is no zero-sequence current protection action state, the ground fault identification result is determined to be a low-resistance ground fault; If there is a zero-sequence current protection action state, a statistical analysis is performed on the historical over-limit situation of the target differential energy to determine whether the over-limit situation of the target differential energy is the first over-limit situation within a preset time range; If the target differential energy exceeds the limit for the first time within a preset time range, determining that the ground fault identification result is a high-resistance ground fault; If the target differential energy is not exceeded for the first time within a preset time range, the ground fault identification result is determined to be intermittent isolated light grounding.

7. The ground fault identification method based on zero-sequence voltage period differential energy according to claim 6, characterized in that: After the zero-sequence voltage period differential energy exceeding the zero-sequence voltage period differential energy threshold is used as the target differential energy, the method further includes: Performing energy peak analysis on the target differential energy to obtain a differential energy peak, and identifying a start time of the differential energy peak; The starting time of the differential energy peak is used as the fault time of the ground fault, and the fault time is added to the ground fault identification result.

8. A ground fault identification device based on zero-sequence voltage periodic differential energy, characterized in that: The ground fault identification device based on zero-sequence voltage period differential energy comprises: The acquisition and processing module is used to acquire and process the zero-sequence voltage signal of the busbar in the distribution network system to obtain a zero-sequence voltage data sequence; A differential calculation module, used for calculating the zero-sequence voltage period differential energy of the bus according to the zero-sequence voltage data sequence to obtain a zero-sequence voltage period differential energy sequence; A comparison module, used for determining a zero-sequence voltage period differential energy threshold according to a preset condition, and comparing the zero-sequence voltage period differential energy threshold with the zero-sequence voltage period differential energy sequence to obtain a comparison result; The identification module is used to identify the ground fault of the distribution network system according to the comparison result and the zero-sequence voltage signal to obtain a ground fault identification result.

9. A ground fault identification device based on zero-sequence voltage periodic differential energy, characterized in that: The ground fault identification device based on zero-sequence voltage period differential energy comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the ground fault identification device based on zero-sequence voltage period differential energy to perform the steps of the ground fault identification method based on zero-sequence voltage period differential energy as claimed in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the ground fault identification method based on zero-sequence voltage periodic differential energy as described in any one of claims 1 to 7 are implemented.

Images