Search

CN-122017669-A - Transient waveform-based pole-mounted breaker ground fault detection method

CN122017669ACN 122017669 ACN122017669 ACN 122017669ACN-122017669-A

Abstract

The invention relates to the field of breaker fault detection, in particular to a method for detecting a pole-mounted breaker ground fault based on transient waveforms, which adopts an FPGA and DSP heterogeneous architecture to collect high-frequency zero sequence transient signals, and calculating the waveform time-frequency local fluctuation rate by analyzing the extremum distribution characteristic, and converting the waveform time-frequency local fluctuation rate into local inertia weight by utilizing a nonlinear function. In signal decomposition iteration, the system dynamically adjusts the weighted fusion proportion of the standard screening path and the inertia maintaining path by using the inertia weight, so as to construct an adaptive screening model to extract pure natural modal components. Finally, hilbert transform is performed on the extracted components, and the fault direction is determined based on the instantaneous phase difference of the voltage and the current. The technical scheme effectively solves the problem of modal aliasing in the traditional empirical mode decomposition, and remarkably improves the feature extraction capability and identification accuracy of the system for single-phase grounding, disconnection and high-resistance faults in a strong noise environment.

Inventors

  • WEN ZHONGHAI
  • DING BAOQUAN

Assignees

  • 保定市全达电力设备有限公司

Dates

Publication Date
20260512
Application Date
20260203

Claims (10)

  1. 1. The method for detecting the grounding fault of the pole-mounted circuit breaker based on the transient waveform is characterized by comprising the steps of collecting transient zero-sequence voltage and zero-sequence current signals of a power distribution network line to obtain a transient signal sequence; performing adaptive signal decomposition on the transient signal sequence to extract a first-order natural mode component, performing Hilbert transformation on the first-order natural mode component, and judging a fault direction according to the transformed transient phase difference; the self-adaptive signal decomposition comprises iterative screening of the current sequence to be processed, and specifically comprises the following steps: the method comprises the steps of constructing an upper envelope line and a lower envelope line of a sequence to be processed, obtaining a mean line of the upper envelope line and the lower envelope line, subtracting the sequence to be processed from the mean line to obtain a standard residual sequence; and acquiring a plurality of extreme points of the sequence to be processed, acquiring inertial weights representing local waveform fluctuation according to the distribution characteristics of the extreme points, and carrying out weighted fusion on the standard residual sequence and the sequence to be processed to obtain the sequence to be processed of the next iteration, wherein the inertial weights are fusion weights of the standard residual sequence in the fusion process.
  2. 2. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 1, wherein the method for calculating the inertia weight is specifically as follows: ; Wherein the method comprises the steps of Representing the inertial weight; is a preset critical threshold value; the sensitivity factor is preset and is used for controlling the steepness degree of weight change; the extreme value interval fluctuation rate is used for representing the distribution characteristic of the extreme value points.
  3. 3. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 2, wherein the method for calculating the extreme value interval fluctuation rate is specifically as follows: ; Wherein the method comprises the steps of Representing the extreme value interval fluctuation rate; An index representing an extreme point; Represent the first Extreme point (P) Time differences between the extreme points; Represent the first Extreme point (P) Time differences between the extreme points; Representing the amplitude of the current extreme point; representing a global maximum amplitude within the entire sampling window; indicating the set minimum positive number.
  4. 4. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 2, wherein the inertia weight is forcedly set to be 1 if the calculated inertia weight value is greater than a preset confidence threshold value.
  5. 5. The method for detecting the grounding fault of the pole-mounted circuit breaker based on the transient waveform according to claim 1, wherein the calculation method of weighted fusion is as follows: ; Wherein the method comprises the steps of Representing a sequence to be processed for a next round of iteration; representing the inertial weight; representing the standard residual sequence; representing the sequence to be processed for the current round.
  6. 6. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 1, wherein constructing an upper envelope and a lower envelope of the sequence to be processed comprises: Connecting all local maximum points of the sequence to be processed based on cubic spline interpolation to obtain an upper envelope; And connecting all local minimum value points of the sequence to be processed based on cubic spline interpolation to obtain a lower envelope curve.
  7. 7. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 6, further comprising taking the average value of the upper envelope line and the lower envelope line as the average value line.
  8. 8. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 1, wherein the step of judging the fault direction according to the transformed transient phase difference comprises the following steps: if the instantaneous phase difference of the first-order natural mode components of the transient zero-sequence voltage and the zero-sequence current is within a preset tolerance range of 180 degrees, determining that the fault is located on the load side of the circuit breaker; And if the instantaneous phase difference is within a preset tolerance range of 0 degrees, judging that the fault is positioned on the power supply side of the circuit breaker.
  9. 9. The method for detecting the ground fault of the pole-mounted circuit breaker based on the transient waveform according to claim 8, wherein the preset tolerance range is 30 degrees to 60 degrees.
  10. 10. The method for detecting the grounding fault of the on-pole circuit breaker based on the transient waveform according to claim 1, wherein the detection method is executed based on an intelligent controller of a dual-core processing unit comprising an FPGA and a DSP; The FPGA is responsible for collecting and caching the transient signal sequence; the DSP is responsible for performing iterative screening, hilbert transformation, and fault determination.

Description

Transient waveform-based pole-mounted breaker ground fault detection method Technical Field The invention relates to the field of breaker fault detection, in particular to a pole-mounted breaker ground fault detection method based on transient waveforms. Background The Feeder Terminal (FTU) and the on-column circuit breaker of the distribution network serve as key equipment of the intelligent distribution network, and are widely applied to monitoring and control of distribution lines. When a single-phase earth fault occurs in the power distribution network, transient signals of zero-sequence voltage and zero-sequence current are generated in the circuit. These signals are often of non-linear, non-stationary nature and often are accompanied by high frequency oscillations or arc noise. In order to achieve accurate fault detection and localization, advanced signal processing techniques are often required to perform time-frequency analysis on these complex transient signals to extract effective fault feature components. At present, an Empirical Mode Decomposition (EMD) algorithm is a mainstream method for processing such non-stationary transient signals due to the characteristics of being capable of adaptively decomposing a complex signal into a plurality of eigenmode functions (IMFs) without a preset basis function. The core of the algorithm is that the layer-by-layer screening and decomposition are carried out by fitting an upper envelope curve and a lower envelope curve through extreme points. However, in an actual power distribution network fault detection scenario, the extreme point distribution of the transient signal tends to be very uneven due to intermittent fault arcs or strong noise interference. When the envelope is constructed by using the traditional EMD algorithm to conduct cubic spline interpolation, the uneven distribution of extreme points is extremely easy to cause overshoot or fitting distortion of the envelope. This can directly cause severe modal aliasing, i.e. fluctuating components of different scales are mixed in the same modality, or components of the same scale are dispersed into different modalities. The physical reality of the decomposition result is seriously damaged by modal aliasing, so that the extracted fault characteristics are inaccurate, and further misjudgment or refusal of the breaker to the fault direction can be caused. Disclosure of Invention Aiming at the problem that the conventional EMD algorithm possibly causes misjudgment of the breaker to the fault direction during modal aliasing, the invention provides a pole-mounted breaker grounding fault detection method based on transient waveforms, which comprises the steps of collecting transient zero-sequence voltage and zero-sequence current signals of a power distribution network line to obtain a transient signal sequence; the method comprises the steps of performing adaptive signal decomposition on a transient signal sequence to extract a first-order inherent mode component, performing Hilbert transformation on the first-order inherent mode component, judging a fault direction according to a transformed instantaneous phase difference, performing iterative screening on a current sequence to be processed, and specifically, constructing an upper envelope and a lower envelope of the sequence to be processed, acquiring a mean value line of the upper envelope and the lower envelope, subtracting the mean value line from the sequence to be processed to obtain a standard residual sequence, acquiring a plurality of extreme points of the sequence to be processed, acquiring inertia weights representing local waveform fluctuation according to distribution characteristics of the extreme points, and performing weighted fusion on the standard residual sequence and the sequence to be processed to obtain a sequence to be processed of the next round of iteration, wherein the inertia weights are fusion weights of the standard residual sequence in a fusion process. According to the invention, the screening process of the traditional self-adaptive signal decomposition is improved by introducing the inertial weight based on the extreme point distribution characteristic, and compared with the prior art that standard residual errors are directly used for single iteration, the invention can adaptively adjust the sequence retention proportion according to the fluctuation intensity of local waveforms in the screening process. The mechanism effectively inhibits the modal aliasing phenomenon and the end-point effect, improves the accuracy and purity of the first-order inherent modal component extraction, thereby ensuring more accurate instantaneous phase difference obtained by the follow-up Hilbert transformation, and remarkably improving the reliability of the discrimination of the grounding fault direction of the circuit breaker on the distribution network column. Further, the method for calculating the inertia weight specifically comprises the fo