A direct current system ground fault detection method and system

Abstract

The application discloses a direct current system grounding fault detection method and system, equal resistance is connected between the positive and negative bus of the direct current system and the ground to form a symmetrical resistance network, the positive / negative ground voltage and the branch ground current are collected in real time, the voltage deviation is calculated to generate a time domain fault mark, a high frequency component is extracted by wavelet transform and energy entropy analysis to generate a frequency domain mark, the ground current signal is subjected to wavelet transform and energy entropy calculation to extract a high frequency detail component, hierarchical collaborative decision is executed, double mark triggering immediately trips, single mark triggering starts a delay review mechanism, and after continuous triggering, tripping is executed and fault positioning and reporting are performed. The technical system improves the comprehensive identification speed of the direct current system grounding fault, forms a whole-process closed-loop protection of fast sensing and identification, reliable decision and accurate positioning, and improves the operation safety of the direct current distribution system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power detection, in particular to a direct current system grounding fault detection method and system BACKGROUND

[0002] Traditional direct current grounding protection methods mainly include resistance bridge method, harmonic injection method and transient analysis method, but these methods have certain limitations: the resistance bridge method relies on the symmetry of the balance bridge and cannot detect high resistance grounding faults, and is easily disturbed by the system ground capacitance; the harmonic injection method needs an additional signal source, which may affect the system stability. The transient analysis method has high requirements for sampling rate and algorithm real-time performance, and lacks accurate mathematical models. The deficiencies of traditional direct current grounding protection methods limit the ability of fast identification of direct current system grounding faults, high resistance fault detection and fault location, so a more efficient and reliable solution is needed to deal with the complex scenarios of direct current system grounding faults. SUMMARY

[0003] In order to overcome the deficiencies of the above-mentioned traditional direct current grounding protection methods, which limit the ability of fast identification of direct current system grounding faults, high resistance fault detection and fault location, the present application provides a direct current system grounding fault detection method and system.

[0004] In order to solve the above-mentioned problems, the present application is realized according to the following technical scheme:

[0005] The direct current system grounding fault detection method of the first aspect of the present application comprises the following steps: step one, connecting resistors with equal resistance between the positive and negative bus bars of the direct current system and the ground to form a symmetrical resistance network, the symmetrical resistance network is in a star type symmetrical topology structure; step two, collecting positive and negative voltage signals and branch ground current signals of the symmetrical resistance network; calculating the ground voltage deviation according to the positive and negative voltage signals, and comparing it with the preset threshold to generate a time domain fault flag; performing wavelet transform and energy entropy calculation on each branch ground current signal to extract high frequency detail components; calculating the energy entropy of each scale detail coefficient and comparing it with the dynamic threshold to generate a frequency domain fault flag; step three, performing hierarchical collaborative fault judgment and fault decision: when the time domain fault flag and the frequency domain fault flag are triggered at the same time, immediately execute the trip; if only a single flag is triggered, start the delay review mechanism, and only after the single flag continues to trigger, confirm the fault and trip; step four, instructing to cut off the fault branch and reporting the type and position of the fault.

[0006] Preferably, the symmetrical resistance network quantization ground voltage deviation relationship is: ; wherein, is the ground voltage difference, is the positive ground voltage of the symmetrical network, The positive-to-ground voltage of the symmetrical network, The DC bus voltage.

[0007] Preferably, the wavelet transform and energy entropy calculation are used to obtain the local features of the signal at different scales and times by convolving the ground current signal with a set of scaled and translated wavelet functions; the wavelet transform separates the high-frequency details and low-frequency approximations of the signal through multi-scale decomposition.

[0008] Preferably, in the wavelet transform, the energy entropy can be used to evaluate the complexity of the signal by calculating the distribution of the wavelet transform coefficients; the energy entropy calculation formula is: ; wherein, is the energy entropy, j is the layer number, is the energy of the jth layer of detail coefficients , is the total energy.

[0009] Preferably, the fault decision includes: the time-domain fault criterion is , and the duration is > 20 ms; wherein, is the ground voltage difference, is the first threshold value; the frequency-domain fault criterion is , and the high-frequency component energy ratio is > 30%; wherein, is the energy entropy, is the second threshold value; the delay review mechanism is to immediately trip when the time-domain fault criterion and the frequency-domain fault criterion are both satisfied, and to delay 10 ms and then perform secondary verification when only one of the time-domain fault criterion or the frequency-domain fault criterion is satisfied.

[0010] In another aspect, the present application provides a DC system ground fault detection system, which comprises: a symmetrical resistance network module for constructing a symmetrical resistance network; a sensor module for detecting the positive-to-ground voltage and the ground current of each branch; a signal adjustment module for suppressing high-frequency noise and eliminating AC coupling interference; a control module, which comprises a data acquisition unit, a time-domain processing module, and a frequency-domain processing module; and a relay module for fault isolation and alarm.

[0011] Preferably, the symmetrical resistance network satisfies the following conditions: the resistance value range is 10 kΩ-100 kΩ, and the power level is 1.5 times the rated value; a metal film resistor is used and packaged in a shielded box, and the temperature drift coefficient is < 50 ppm / ℃; the ground capacitance and the symmetrical resistance network form a low-pass filter.

[0012] Preferably, the sensor module comprises a voltage sensor and a current sensor; the signal adjustment module comprises a low-pass filter and a power frequency trap; and the sampling rate of the data acquisition unit is ≥ 200 kS / s.

[0013] Compared with the prior art, the present application has the beneficial effects that: by constructing a star-type symmetric monitoring topology through a symmetric resistance network, the positive and negative electrode voltage deviation detection sensitivity is improved by more than 40%, solving the problem of large detection blind area of the traditional voltage balance method in single-point grounding fault. The dual-domain detection technology combining wavelet transform and dynamic energy entropy analysis is adopted, the high-frequency detail components are accurately extracted through multi-scale decomposition, and the identification accuracy is improved from less than 70% of the traditional method to 95% by combining the dynamic threshold algorithm. The hierarchical collaborative decision mechanism innovatively combines time domain threshold triggering and frequency domain energy mutation characteristics, the dual flag verification mechanism reduces the misoperation rate to less than 0.5%, and at the same time, through the delay review strategy, the reliability when a single feature triggers is ensured. Combined with branch current synchronous monitoring and topology analysis algorithm, the fault branch can be accurately located in a short time, and the positioning error is less than 0.5 meters. The technology system improves the comprehensive identification speed of the DC system grounding fault, shortens more than 60% compared with the traditional method, forms a full-process closed-loop protection of rapid sensing and identification, reliable decision-making, and accurate positioning, and improves the operation safety of the DC distribution system. BRIEF DESCRIPTION OF DRAWINGS

[0014] The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings, in which:

[0015] Figure 1 is a step schematic diagram of a DC system grounding fault detection method of the present application; DETAILED DESCRIPTION

[0016] The preferred embodiments of the present application will be described below with reference to the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application.

[0017] As Figure 1As shown, the first aspect of the present invention provides a DC system grounding fault detection method, comprising the following steps: Step 1, connecting resistors of equal resistance between the positive and negative busbars of the DC system and ground to form a symmetrical resistor network, wherein the symmetrical resistor network has a star-shaped symmetrical topology; Step 2, acquiring the positive and negative voltage signals of the symmetrical resistor network and the grounding current signals of each branch; calculating the voltage deviation to ground based on the positive and negative voltage signals and comparing it with a preset threshold to generate a time-domain fault flag; performing wavelet transform and energy entropy calculation on the grounding current signals of each branch to extract high-frequency detail components; calculating the energy entropy of detail coefficients at each scale and comparing it with a dynamic threshold to generate a frequency-domain fault flag; Step 3, performing hierarchical collaborative fault judgment and fault decision: when the time-domain fault flag and the frequency-domain fault flag are triggered simultaneously, tripping is immediately executed; if only a single flag is triggered, a delayed verification mechanism is initiated, and the fault is confirmed and tripping is performed after only a single flag continues to trigger; Step 4, instructing the disconnection of the faulty branch and reporting the type and location of the fault.

[0018] In one embodiment, a preliminary fault detection is achieved by quantifying the voltage deviation to ground using a symmetrical resistor network.

[0019] Let the voltage of the positive terminal of the DC system relative to ground be... The voltage of the negative terminal of the DC system to ground is The system's capacitance to ground is C.

[0020] A symmetrical resistive network consists of the positive electrode resistance to ground. and negative electrode to ground resistance It constitutes and satisfies .

[0021] In one embodiment, under normal conditions: when the system is symmetrical, the voltages of the positive and negative terminals relative to ground satisfy:

[0022] ;

[0023] At this time, the voltage difference to ground However, the total leakage current of the system is zero.

[0024] Under fault conditions: Assuming the positive terminal is grounded through the fault resistor, the equivalent circuit equation is: ;

[0025] in This is the fault resistor.

[0026] Substituting the above equivalent circuit equation into and Solving for:

[0027] .

[0028] The simplified formula above yields: Symmetrical network positive pole-to-ground voltage: Symmetrical network negative pole-to-ground voltage: Ground voltage difference: Wherein, is the DC bus voltage.

[0029] In a preferred embodiment, the symmetrical resistance network quantifies the ground voltage deviation relationship as: Wherein, is the ground voltage difference, is the symmetrical network positive pole-to-ground voltage, is the symmetrical network negative pole-to-ground voltage, is the DC bus voltage, the symmetrical resistance network quantifies the ground voltage deviation relationship as: always holds true.

[0030] The parameter design and sensitivity analysis when constructing the star-shaped symmetrical topology are as follows: fault resistance The absolute values of the ground voltage and are changed, not their difference.

[0031] When in a low resistance state: Symmetrical network positive pole-to-ground voltage: .

[0032] When in a high resistance state: Symmetrical network negative pole-to-ground voltage: .

[0033] Calculate the power loss: symmetrical resistance needs to satisfy Wherein, is the power loss, is the rated power.

[0034] The symmetrical resistance network and the ground capacitance C form a low-pass filter, and the anti-capacitance interference calculation formula is: needs to ensure Wherein, is the system noise frequency, is the cut-off frequency.

[0035] In a preferred embodiment, the wavelet transform and energy entropy calculation convolve the ground current signal through a set of scaled and translated wavelet functions to obtain the local features of the signal at different scales and times; the wavelet transform separates the high-frequency details and low-frequency approximations of the signal through multi-scale decomposition.

[0036] It can be understood that the wavelet transform is a time-frequency analysis tool, which can decompose a signal into frequency components at multiple scales for localized analysis in time and frequency. Compared with the traditional Fourier transform, the wavelet transform includes continuous wavelet transform and discrete wavelet transform. The wavelet transform can provide the instantaneous frequency characteristics of the signal, and thus has an advantage in processing signals with mutations or transients such as ground fault signals. Energy entropy is an index for quantifying the complexity of a signal, and is particularly suitable for detecting transient changes in a signal. In wavelet transform, energy entropy can be used to evaluate the complexity of a signal by calculating the distribution of wavelet transform coefficients. The application of energy entropy in fault detection can determine whether abnormal changes have occurred in a signal, and improve the accuracy and efficiency of the judgment.

[0037] In a preferred embodiment, in the wavelet transform, the energy entropy can be used to evaluate the complexity of a signal by calculating the distribution of the wavelet transform coefficients; the energy entropy calculation formula is: ; wherein, is the energy entropy, j is the layer number, is the energy of the jth layer of detail coefficients , is the total energy.

[0038] By convolution with a set of scaled and translated wavelet functions, the local features of a signal at different scales and time positions are obtained, which is a continuous wavelet transform. The mathematical expression of the continuous wavelet transform is as follows:

[0039] ; wherein, is a set of scaled and translated wavelet functions (a, b) in the continuous wavelet transform, is the signal.

[0040] Wavelet basis selection and multi-resolution analysis, the ground fault transient signal contains high frequency components (such as the rapid rising edge of arc discharge), while normal noise is mostly low frequency or stationary signal. The wavelet transform separates the high frequency details (Detail Coefficients) and the low frequency approximation (Approximation Coefficients) of the signal through multi-scale decomposition.

[0041] In an embodiment, the preferred wavelet basis is db4, which has tight support and orthogonality, and the waveform matches the rising edge of the fault transient signal; the preferred decomposition layer number J=5: taking into account the high frequency resolution and computational efficiency.

[0042] Discrete wavelet transform decomposes a signal into a set of discrete high and low frequency subbands. Through multi-level decomposition, the discrete wavelet transform can extract multi-scale information of the signal, thereby effectively processing and analyzing the mutation characteristics in the signal. Decomposed into approximation coefficients and the detail coefficients D1, D2,..., DJ: the calculation formula is .

[0043] The energy entropy is calculated as the energy of the jth layer of detail coefficients , the total energy . The energy entropy S reflects the degree of disorder of the energy distribution, and the calculation formula is: ; when a fault occurs, the proportion of high-frequency energy increases, and S increases significantly.

[0044] Under normal conditions, the energy entropy of the signal is low because the signal changes smoothly and regularly. When a ground fault occurs, the current or voltage of the system will suddenly change or fluctuate, which will be manifested as a higher energy concentration in the wavelet transform coefficients. The calculation result of the energy entropy will also rise rapidly, indicating that the system has an abnormal fault.

[0045] In a preferred embodiment, the fault decision includes: the time-domain fault criterion is , and the duration is > 20 ms; wherein, is the voltage difference to ground, is the first threshold value; the frequency-domain fault criterion is , and the proportion of high-frequency component energy is > 30%; wherein, is the energy entropy, is the second threshold value; the delay review mechanism is that when the time-domain fault criterion and the frequency-domain fault criterion are met at the same time, the circuit is immediately tripped, and when only the time-domain fault criterion or the frequency-domain fault criterion is met, a second verification is performed after a delay of 10 ms.

[0046] It can be understood that under normal conditions, the mean value μ and the standard deviation σ of the entropy value of the historical data under normal conditions are statistically analyzed.

[0047] Set = μ - 4σ to ensure that the false alarm rate is < 0.1%.

[0048] The flow of generating the time-domain fault flag and triggering is:

[0049] S10, read the positive pole-to-ground voltage of the DC system as and the negative pole-to-ground voltage of the DC system as .

[0050] S20, obtain and by the formulas and , the voltage difference to ground, is the positive pole-to-ground voltage difference, is the negative pole-to-ground voltage difference.

[0051] S30, comparison .

[0052] Primary fault judgment is performed, and the criterion is that when or a primary fault warning is triggered; secondary fault judgment is performed again, and the criterion is that the energy entropy is confirmed as a fault.

[0053] The processing time is less than 20 ms, wherein is a voltage offset threshold value; is an energy entropy threshold value.

[0054] S40, if the limit is exceeded, a time domain fault flag is triggered.

[0055] The ground voltage deviation can be calculated in real time, and the primary criterion is executed.

[0056] The flow of generating a frequency domain fault flag and triggering is

[0057] S11, buffer 1 ms window current signal.

[0058] S12, execute 5-layer db4 wavelet decomposition.

[0059] S13, extract 5-layer detail coefficients.

[0060] S14, calculate energy entropy S.

[0061] S15, compare S with threshold value .

[0062] S16, if the limit is exceeded, a frequency domain fault flag is triggered.

[0063] The wavelet transform and energy entropy calculation are executed to realize high resistance fault identification.

[0064] On the other hand, the application provides a DC system ground fault detection system, the fault detection system comprising: a symmetric resistance network module for building a symmetric resistance network; a sensor module for detecting positive and negative ground voltage, and branch ground line current; a signal adjustment module for suppressing high-frequency noise and eliminating AC coupling interference; a control module, the control module comprising a data acquisition unit, a time domain processing module and a frequency domain processing module; a relay module, the relay module for fault isolation and alarm.

[0065] It can be understood that the symmetric resistance network module is packaged in a shielded box to avoid electromagnetic interference. Low-temperature drift resistors (such as metal film resistors, temperature drift coefficient < 50ppm / ℃) are selected.

[0066] Further, the symmetrical resistance network satisfies the following conditions: the resistance value ranges from 10kΩ to 100kΩ, the power level is 1.5 times of the rated value; the metal film resistance is adopted and is packaged in a shielding box, the temperature drift coefficient is less than 50ppm / ℃; the ground capacitance and the symmetrical resistance network form a low-pass filter.

[0067] Further, the sensor module comprises a voltage sensor and a current sensor; the signal adjustment module comprises a low-pass filter and a power frequency trap; the sampling rate of the data acquisition unit is greater than or equal to 200kS / s. The voltage sensor and the current sensor are arranged in the sensor module for detecting the voltage of the positive and negative poles to ground and the current of each branch ground wire, the voltage sensor is installed at the positive pole to ground and the negative pole to ground, and the current sensor is installed at each branch ground wire for locating the fault branch. In an embodiment, the data acquisition unit in the control module is an AD converter, the resolution of the AD converter is 16 bits (such as AD7606), and the sampling rate is 200kS / s, which can meet the requirements of wavelet transform analysis. Synchronous sampling is performed: the positive and negative voltage signals are synchronously collected to reduce the phase error.

[0068] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Any modification, equivalent change and modification of the above embodiment based on the technical essence of the present application, without departing from the technical solution of the present application, still belongs to the scope of the technical solution of the present application.

Claims

1. A direct current system ground fault detection method characterized by, The method comprises the following steps: Step 1: connecting resistors with equal resistance between the positive and negative bus bars and the ground to form a symmetric resistance network, wherein the symmetric resistance network is in a star symmetric topology; Step 2: collecting positive and negative voltage signals and branch ground current signals of the symmetric resistance network; calculating the ground voltage deviation according to the positive and negative voltage signals and comparing the ground voltage deviation with a preset threshold to generate a time-domain fault flag; performing wavelet transform on the branch ground current signals and calculating the energy entropy to extract high-frequency detail components; calculating the energy entropy of each scale detail coefficient and comparing the energy entropy with a dynamic threshold to generate a frequency-domain fault flag; Step 3: performing hierarchical collaborative fault judgment and fault decision: when the time-domain fault flag and the frequency-domain fault flag are triggered at the same time, immediately execute tripping; if only a single flag is triggered, start a delay review mechanism, and confirm the fault and trip after the single flag continues to trigger; Step 4: instructing to cut off the fault branch and reporting the type and position of the fault.

2. The DC system ground fault detection method according to claim 1, characterized in that: The symmetric resistance network quantization to ground voltage deviation relationship is: ​ wherein, is the voltage difference to ground, is the positive pole to ground voltage of the symmetrical network, is the negative pole to ground voltage of the symmetrical network, is the DC bus voltage.

3. The DC system ground fault detection method according to claim 1, characterized in that: the wavelet transform and energy entropy calculation perform convolution on the ground current signals through a group of scaled and translated wavelet functions to obtain local features of the signals at different scales and times; the wavelet transform separates the high-frequency details and low-frequency approximations of the signals through multi-scale decomposition.

4. The DC system ground fault detection method according to claim 1, characterized in that: in the wavelet transform, the energy entropy is calculated by the distribution of the wavelet transform coefficients to evaluate the complexity of the signals; The energy entropy calculation formula is: ; wherein, is the energy entropy, j is the layer number, is the energy of the jth layer of detail coefficients , is the total energy.

5. The DC system ground fault detection method according to claim 1, characterized in that: the fault decision includes: The time domain fault criterion is , and the duration > 20 ms; wherein, is a voltage difference to ground, is a first threshold value; The frequency domain fault criterion is , and the high-frequency component energy proportion is >30%. wherein, is an energy entropy, is a second threshold value; the delay review mechanism immediately trips when the time-domain fault criterion and the frequency-domain fault criterion are met at the same time, and performs secondary verification after a delay of 10 ms when only the time-domain fault criterion or the frequency-domain fault criterion is met.

6. A DC system ground fault detection system based on the DC system ground fault detection method according to any one of claims 1 to 5, characterized by, The fault detection system comprises: a symmetric resistance network module for constructing a symmetric resistance network; a sensor module for detecting positive and negative ground voltages and branch ground currents; a signal adjustment module for suppressing high-frequency noise and eliminating AC coupling interference; a control module including a data acquisition unit, a time-domain processing module and a frequency-domain processing module; a relay module for fault isolation and alarm.

7. A direct current system ground fault detection system as recited in claim 6, wherein, The symmetric resistance network satisfies the following conditions: the resistance value ranges from 10 kΩ to 100 kΩ, and the power level is 1.5 times the rated value; metal film resistors are used and packaged in a shielded box, and the temperature drift coefficient is less than 50 ppm / ℃; the ground capacitance and the symmetric resistance network form a low-pass filter.

8. The DC system ground fault detection system according to claim 6, characterized in that: the sensor module includes voltage sensors and current sensors; the signal adjustment module includes a low-pass filter and a power frequency trap; the sampling rate of the data acquisition unit is greater than or equal to 200 kS / s.

Images