Online monitoring device for grounding faults in power distribution networks

Abstract

This invention relates to the field of power system distribution network operation and fault monitoring, specifically to an online monitoring device for distribution network grounding faults. It comprises an insulating base, an inner sensing magnetic ring, an outer compensation magnetic ring, and a processing terminal. The device utilizes a coaxial nested double-layer magnetic ring structure to construct a self-compensating loop where the outer layer reverse series winding is directly connected to the inner layer cancelling coil. Its core principle is that when external conductors asymmetrically generate leakage magnetic flux, an induced current drives the cancelling coil to generate an equal and opposite cancelling magnetic flux to counteract the interference. This invention eliminates parasitic leakage magnetic interference caused by asymmetrical spatial arrangement of three-phase conductors, and solves the measurement error problem under complex spatial magnetic fields using a self-compensating physical mechanism, effectively maintaining the zero magnetic field geometric accuracy of the core sensing axis.

Description

Technical Field

[0001] This invention relates to the field of power system distribution network operation and fault monitoring, specifically to an online monitoring device for distribution network grounding faults. Background Technology

[0002] In the distribution network operating environment, real-time monitoring of the zero-sequence transient signal of the three-phase conductors is the basis for judging the occurrence of grounding faults. To obtain the above fault characteristic data, existing solutions generally use sensing devices based on a single magnetic core structure for magnetic field measurement and control. Although this solution has the current sensing capability to meet the preset threshold under ideal balanced conditions, in actual installation and operation, due to the eccentricity or uneven spacing of the three-phase conductors, significant parasitic leakage magnetic field interference will be generated. In addition, the single magnetic core structure cannot meet the requirements of multiple operating conditions in terms of material properties. It has a high probability of magnetic saturation when facing a background of hundreds of amperes or large current disturbances. Furthermore, it faces the defect of insufficient sensitivity causing signal submersion when the fault arc is weak due to high resistance grounding.

[0003] Therefore, how to eliminate parasitic leakage flux interference caused by asymmetrical spatial arrangement of conductors, and stably extract weak zero-sequence transient signals under heavy load conditions, and solve the contradiction between sensitivity and anti-saturation capability of a single magnetic core, has become an urgent technical problem to be solved. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention provides an online monitoring device for grounding faults in power distribution networks. Specifically, the technical solution of the present invention includes: The system comprises an insulating base, an inner sensing magnetic ring, an outer compensation magnetic ring, and a processing terminal. The insulating base has a main through hole at its center for the three-phase conductors of the power distribution network to pass through. The insulating base has an inner ring groove and an outer ring groove arranged coaxially from the inside to the outside along the radial direction. An annular partition is provided between the inner ring groove and the outer ring groove. The inner sensing magnetic ring is installed in the inner ring groove, and the surface of the inner sensing magnetic ring is uniformly wound with a main sensing coil, a DC bias coil and a cancellation coil along the circumferential direction. The outer compensation magnetic ring is installed in the outer ring groove, and the outer compensation magnetic ring is coaxially nested outside the inner sensing magnetic ring. The surface of the outer compensation magnetic ring is wound with a reverse series winding. The two ends of the reverse series winding are directly connected to the canceling coil wound on the inner sensing magnetic ring through wires to form a self-generating compensation circuit. The output terminal of the main sensing coil and the input terminal of the DC bias coil are electrically connected to the processing terminal via shielded cables. The processing terminal is equipped with a micro-power constant current source circuit, and the output terminal of the micro-power constant current source circuit is connected to the DC bias coil.

[0005] As a further aspect of the present invention: the inner sensing magnetic ring is formed by winding an amorphous alloy strip, and the outer compensation magnetic ring is formed by stacking silicon steel sheets; the main through hole has a diameter of 60mm, the inner sensing magnetic ring has an outer diameter of 120mm, an inner diameter of 80mm, and a thickness of 20mm, the outer compensation magnetic ring has an outer diameter of 180mm, an inner diameter of 140mm, and a thickness of 20mm, and the annular partition has a thickness of 10mm.

[0006] As a further aspect of the present invention: the reverse series winding is composed of two sets of half-toroidal coils with opposite winding directions connected end to end; the main sensing coil, the DC bias coil and the canceling coil are interlayered insulated by a polyimide film.

[0007] As a further aspect of the present invention: the reverse series winding is configured to induce a leakage magnetic field and generate an induced current when the three-phase conductors of the power supply and distribution network pass through the main through hole and are spatially asymmetrically arranged to generate a leakage magnetic field; the canceling coil is connected to the reverse series winding and is configured to receive the induced current and generate a reverse canceling magnetic flux that is equal in magnitude and opposite in direction to the leakage magnetic field, so as to maintain the zero magnetic field geometric accuracy of the core sensing axis of the inner sensing magnetic ring.

[0008] As a further aspect of the present invention: the processing terminal continuously receives the induced current signal output by the main sensing coil, and the processing terminal is configured to extract the current spectrum distortion rate of the induced current signal in the high frequency band of 3000Hz to 5000Hz and the high frequency phase jitter variance of the waveform zero crossing point of the induced current signal.

[0009] As a further aspect of the present invention: the processing terminal is further configured to calculate the comprehensive fluctuation index and the attenuation rate of the induced signal; wherein, the comprehensive fluctuation index is used to quantify the disturbance in the time domain and the frequency domain, and its calculation formula is: comprehensive fluctuation index = high-frequency phase jitter variance × current spectrum distortion rate; The formula for calculating the attenuation rate of the induced signal is: Attenuation rate = Comprehensive fluctuation index × Attenuation conversion coefficient; The attenuation conversion coefficient is a pre-calibrated value related to the grid voltage level, and the dimension of the attenuation conversion coefficient is the reciprocal of the dimension of the high-frequency phase jitter variance.

[0010] As a further aspect of the present invention: the processing terminal is configured to invert the required relative permeability compensation amount of the inner sensing magnetic ring based on the attenuation rate, and to convert the nonlinear change of the arc impedance during a ground fault into a compensation curve of the required relative permeability. The higher the attenuation rate, the greater the required relative permeability compensation amount.

[0011] As a further aspect of the present invention: the processing terminal is configured to adjust the duty cycle of the pulse width modulation signal output from the micro-power constant current source circuit to the DC bias coil in real time according to the calculated compensation curve; the DC bias magnetization adjustment field generated by the DC bias coil dynamically changes the relative permeability and giant magnetoresistance effect of the inner layer sensing magnetic ring by changing the magnetic domain arrangement state inside the inner layer sensing magnetic ring.

[0012] As a further aspect of the present invention: when the processing terminal determines that the comprehensive fluctuation index and the attenuation rate increase and the amplitude of the induced current signal attenuates to less than or equal to a preset noise floor threshold, the processing terminal automatically reduces the duty cycle of the pulse width modulation signal, weakens the DC bias field strength, and causes the relative permeability of the inner sensing magnetic ring to rise back to the preset target relative permeability working area. When the processing terminal detects that a large load fluctuation in the power grid or an arc breakdown causes the amplitude of the induced current signal to suddenly increase to a value greater than the average amplitude of the previous statistical period and the distortion rate of the current spectrum decreases, the processing terminal instantly increases the duty cycle of the pulse width modulation signal, enhances the DC bias field strength, and reduces the relative permeability of the inner sensing magnetic ring to a preset anti-saturation relative permeability region.

[0013] The present invention has the following beneficial effects: 1. This device eliminates parasitic leakage magnetic interference caused by the asymmetrical spatial arrangement of three-phase conductors; by coaxially nesting an outer compensation magnetic ring outside the inner sensing magnetic ring, and directly connecting the reverse series winding on the outer compensation magnetic ring to the canceling coil on the inner sensing magnetic ring, a self-generating compensation circuit is formed; when the external conductors generate leakage magnetic field due to asymmetry, an induced current will be induced in the reverse series winding. This current drives the canceling coil to generate a reverse canceling magnetic flux of equal magnitude and opposite direction, thereby offsetting the interference and maintaining the zero magnetic field geometric accuracy of the core sensing axis of the inner sensing magnetic ring. 2. This device resolves the contradiction between sensitivity and anti-saturation capability in a single magnetic core; the processing terminal extracts the current spectrum distortion rate and high-frequency phase jitter variance of the induced current signal at the zero-crossing point of the waveform, and adjusts the duty cycle of the pulse width modulation signal output to the DC bias coil in real time; when the arc impedance changes and the signal attenuates, the DC bias field strength is reduced to restore the relative permeability of the magnetic ring to ensure high sensitivity; when a large load fluctuation causes a sudden increase in current, the bias field strength is increased to suppress the relative permeability, thereby effectively preventing bias saturation and achieving stable monitoring under complex working conditions. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is a schematic diagram of the overall structure of the device; Figure 2 This is a cross-sectional view of the insulating base of the device; Figure 3 This is a schematic diagram of the outer compensation magnetic ring structure of the device; Figure 4 This is a schematic diagram of the inner sensing magnetic ring structure of the device.

[0015] In the diagram: 1. Insulating base; 2. Main through hole; 3. Inner ring groove; 4. Outer ring groove; 5. Annular partition; 6. Inner sensing magnetic ring; 7. Main sensing coil; 8. DC bias coil; 9. Cancellation coil; 10. Outer compensation magnetic ring; 11. Reverse series winding; 12. Wire; 13. Shielded cable; 14. Processing terminal; 15. Micropower constant current source circuit; 16. Half-ring coil; 17. Polyimide film. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention. Example

[0017] Please see Figure 1 and Figure 2 An online monitoring device for grounding faults in a power distribution network, comprising: The insulating base 1, the inner sensing magnetic ring 6, the outer compensation magnetic ring 10 and the processing terminal 14 are provided. The insulating base 1 has a main through hole 2 for the three-phase conductors of the power distribution network to pass through. The inner ring groove 3 and the outer ring groove 4 are coaxially distributed in the radial direction from the inside to the outside of the insulating base 1. An annular partition 5 is provided between the inner ring groove 3 and the outer ring groove 4. The inner sensing magnetic ring 6 is installed in the inner ring groove 3. The main sensing coil 7, DC bias coil 8 and cancellation coil 9 are uniformly wound around the surface of the inner sensing magnetic ring 6 in the circumferential direction. The outer compensation magnetic ring 10 is installed in the outer ring groove 4. The outer compensation magnetic ring 10 is coaxially nested outside the inner sensing magnetic ring 6. The outer compensation magnetic ring 10 has a reverse series winding 11 wound on its surface. The two ends of the reverse series winding 11 are directly connected to the cancellation coil 9 wound on the inner sensing magnetic ring 6 through the wire 12 to form a self-generating compensation circuit. The output terminal of the main sensing coil 7 and the input terminal of the DC bias coil 8 are electrically connected to the processing terminal 14 via shielded cable 13, respectively. The processing terminal 14 is equipped with a low-power constant current source circuit 15, and the output terminal of the low-power constant current source circuit 15 is connected to the DC bias coil 8. The insulating base 1 is made of flame-retardant insulating material and is used to support the inner sensing magnetic ring 6, the outer compensation magnetic ring 10 and their coil assembly, and to ensure a stable insulation distance between the three-phase conductors and the magnetic components when the three-phase conductors pass through the main through hole 2; the main through hole 2 is used to allow the three-phase conductors of the distribution network to pass through as a whole, so that the three-phase conductors can jointly form the magnetic field excitation corresponding to the zero-sequence current. The inner ring groove 3 and the outer ring groove 4 are arranged coaxially along the radial direction of the insulating base 1, so that the inner sensing magnetic ring 6 and the outer compensation magnetic ring 10 are geometrically aligned. The annular partition 5 is used to structurally isolate the two magnetic circuits and limit the direct coupling between them. The inner sensing magnetic ring 6 serves as a fault information sensing component. Its main sensing coil 7 is used to output the induced current signal, the DC bias coil 8 is used to establish a DC bias magnetization adjustment field under the control of the processing terminal 14, and the cancellation coil 9 is used to receive the energy induced by the corresponding winding of the outer compensation magnetic ring 10 and generate a reverse cancellation magnetic flux. The outer compensation magnetic ring 10 is coaxially sleeved outside the inner sensing magnetic ring 6. The reverse series winding 11 on its surface directly forms a self-generated compensation circuit with the cancellation coil 9. This circuit does not rely on external power supply, but uses the leakage magnetic field generated by the asymmetrical spatial arrangement of the external three-phase conductors to induce current, thereby realizing real-time offsetting of parasitic leakage magnetic field. The processing terminal 14 receives the output signal of the main sensing coil 7 through the shielded cable 13, and provides an adjustable current to the DC bias coil 8 through the micro-power constant current source circuit 15, so that the inner sensing magnetic ring 6 switches between a high-sensitivity state and an anti-saturation state. This device, through the combined effects of coaxial nesting in the mechanical structure, a self-generating compensation circuit, and micro-power constant current bias adjustment in the electrical control, enables the stable extraction of weak zero-sequence transient signals under a hundred-ampere load background, solving the problem of the difficulty in balancing sensitivity and anti-saturation capability in a single magnetic core structure.

[0018] The inner sensing magnetic ring 6 is formed by winding amorphous alloy strip, and the outer compensation magnetic ring 10 is formed by stacking silicon steel sheets; the main through hole 2 has a diameter of 60mm, the inner sensing magnetic ring 6 has an outer diameter of 120mm, an inner diameter of 80mm, and a thickness of 20mm, the outer compensation magnetic ring 10 has an outer diameter of 180mm, an inner diameter of 140mm, and a thickness of 20mm, and the annular partition 5 has a thickness of 10mm; Please see Figure 4 The inner sensing magnetic ring 6 is formed by winding an amorphous alloy strip, wherein the amorphous alloy strip refers to a soft magnetic material with high initial permeability and low coercivity, which is used in this invention to sensitively respond to the magnetic flux corresponding to zero-sequence transient currents in the milliampere to microampere ranges. Please see Figure 3 The outer compensation magnetic ring 10 is formed by stacking silicon steel sheets. The silicon steel sheet stacking structure has mechanical stability that meets the preset threshold and low-frequency magnetic flux carrying capacity that meets the rated requirements. It is suitable for receiving leakage magnetic flux changes caused by the spatial asymmetry of the three-phase conductors and driving the reverse series winding 11 to output compensation current. The diameter of the main through hole 2 is set to 60mm to meet the common requirements for the installation of three-phase cable bundles or insulated conductor combinations in medium-voltage distribution networks, and to ensure that there is an installation gap between the conductor and the inner diameter of the magnetic ring that meets the insulation and assembly requirements; the inner sensing magnetic ring 6 has an outer diameter of 120mm, an inner diameter of 80mm, and a thickness of 20mm, so that it forms a radial 10mm installation support area with the main through hole 2, and provides a cross-sectional area that meets the rated magnetic flux requirements to establish an internal magnetic circuit with adjustable permeability; The outer compensation magnetic ring 10 has an outer diameter of 180 mm, an inner diameter of 140 mm, and a thickness of 20 mm, forming a radial gap area occupied by the annular partition 5 between the outer compensation magnetic ring 10 and the inner sensing magnetic ring 6; the annular partition 5 has a thickness of 10 mm, which is used to limit the direct leakage magnetic coupling between the inner and outer magnetic rings and maintain coaxial positioning accuracy. This set of dimensions allows the inner sensing magnetic ring 6 to focus on extracting high-frequency weak fault features, while the outer compensation magnetic ring 10 focuses on offsetting low-frequency parasitic leakage flux. The two are arranged in layers in space, which can reduce the error propagation caused by structural coupling. Those skilled in the art can also make proportional adjustments to the above dimensions according to the requirements of conductor outer diameter, rated current level and insulation spacing without changing the coaxial nesting and dual magnetic ring division of labor principle.

[0019] The reverse series winding 11 is composed of two sets of half-toroidal coils 16 with opposite winding directions connected end to end; the main sensing coil 7, the DC bias coil 8 and the canceling coil 9 are interlayered insulated by a polyimide film 17. The reverse series winding 11 is composed of two sets of half-loop coils 16 with opposite winding directions connected end to end. The half-loop coils 16 are winding units that occupy half of the circumference area along the outer compensation magnetic ring 10. The two sets of half-loop coils 16 form opposite potential responses to spatial asymmetric leakage magnetic flux due to their opposite winding directions. The purpose of this structure is to convert parasitic leakage magnetic components from different directions into directionally selective induced currents on the outer compensation magnetic ring 10, and then transmit these induced currents to the cancellation coil 9 to form a reverse magnetic flux against the parasitic leakage magnetics in the inner sensing magnetic ring 6. Compared with the whole-ring unidirectional winding, this structure has a higher resolution of spatial bias leakage magnetic field and is easier to maintain the zero magnetic field condition of the inner sensing axis. The main sensing coil 7, DC bias coil 8 and cancellation coil 9 are all wound on the surface of the inner sensing magnetic ring 6, and the three sets of coils are interlayered with polyimide film 17. In this invention, the polyimide film 17 is used as a heat-resistant insulating medium. Its temperature resistance and dielectric strength can meet the requirements of long-term online operation and prevent inter-turn crosstalk between DC bias current, induced signal current and compensation current. In actual winding, the main sensing coil 7 can be arranged at the position closest to the inner sensing magnetic ring 6 to enhance the sensing efficiency of magnetic flux changes. The DC bias coil 8 and the canceling coil 9 are arranged in sequence on the outer layer, and a polyimide film 17 with a thickness of 25μm to 75μm is laid between each coil to balance insulation reliability and winding compactness.

[0020] The reverse series winding 11 is configured to induce a leakage magnetic field and generate an induced current when the three-phase conductors of the power supply and distribution network pass through the main through hole 2 and are spatially asymmetrically arranged to generate a leakage magnetic field; the canceling coil 9 is connected to the reverse series winding 11 and is configured to receive the induced current and generate a reverse canceling magnetic flux that is equal in magnitude and opposite in direction to the leakage magnetic field, so as to maintain the zero magnetic field geometric accuracy of the core sensing axis of the inner sensing magnetic ring 6. After the external three-phase conductors are inserted into the main through hole 2, if the spatial arrangement of the three-phase conductors is eccentric, the spacing is unequal, or the installation posture is different, additional parasitic leakage flux will be superimposed on the zero-sequence current. The parasitic leakage flux preferentially couples to the outer compensation magnetic ring 10 located on the outside and cuts the reverse series winding 11 on its surface, forming an induced current in the winding. The induced current flows directly into the cancellation coil 9 on the surface of the inner sensing magnetic ring 6 through the conductor 12, and a reverse magnetomotive force is established in the cancellation coil 9. Since the reverse series winding 11 and the cancellation coil 9 form a self-generated compensation circuit, the larger the parasitic leakage flux, the larger the induced current, and the larger the reverse cancellation flux established by the cancellation coil 9, thus forming an automatic compensation relationship corresponding to the amplitude of the parasitic leakage flux. Equal in magnitude and opposite in direction means that when the design turns ratio, conductor 12 impedance and magnetic ring cross-sectional area meet the matching conditions, the magnetic flux generated by the cancellation coil 9 has a 1:1 ratio to the magnetic flux corresponding to the parasitic leakage flux, so that the residual bias magnetic flux in the core area of ​​the inner sensing magnetic ring 6 is reduced to within the predetermined allowable error range. The zero magnetic field geometric accuracy of the core sensing axis refers to the fact that the average value of the non-fault bias magnetic flux of the main sensing path, based on the geometric center circumference of the inner sensing magnetic ring 6, is controlled within a predetermined threshold, thereby ensuring that the output of the main sensing coil 7 mainly reflects the true zero-sequence fault information rather than installation asymmetry noise. To achieve the above compensation effect, the total resistance of the reverse series winding 11 and the offset coil 9 should be controlled within a range that makes the time constant of the compensation circuit lower than the period of high-frequency characteristic change of the fault, so as to ensure that the compensation response has sufficient following capability.

[0021] The processing terminal 14 continuously receives the induced current signal output by the main sensing coil 7. The processing terminal 14 is configured to extract the current spectrum distortion rate of the induced current signal in the high frequency band of 3000Hz to 5000Hz and the high frequency phase jitter variance of the waveform zero crossing point of the induced current signal. The processing terminal 14 includes a signal conditioning circuit, an analog-to-digital converter circuit, and a data processing unit. The induced current signal output by the main sensing coil 7 is transmitted to the processing terminal 14 via the shielded cable 13. It is first converted into a voltage signal by transimpedance amplification or sampling resistor, and then input to the analog-to-digital converter circuit after bandpass filtering and anti-aliasing processing. The data processing unit performs discrete acquisition of the signal at a sampling rate of not less than 20 kS / s and extracts fault characteristics in the high-frequency band of 3000 Hz to 5000 Hz. Since the nonlinear characteristics of ground fault arcs can induce a variety of irregular harmonic components in the high frequency band, the degree of spectral irregularity caused by ground fault arcs can be effectively characterized by calculating the ratio between the sum of squares of the amplitudes of non-reference harmonic components in the target high frequency band and the sum of squares of the total amplitudes in that band. Similarly, since arc reignition, extinction, and contact instability can generate high-frequency signal phase shifts that exceed the normal tolerance range near the zero-crossing point, the dispersion of the high-frequency phase shift obtained by calculating the waveform zero-crossing time within a continuous sampling window can accurately characterize the above-mentioned time jitter characteristics. The selection of the 3000Hz to 5000Hz frequency band is based on the fact that this frequency band can take into account the frequency interval between the high-frequency transient components generated by the fault arc and the conventional power frequency and low-order harmonic components of the distribution network, which is conducive to improving the separation between effective fault characteristics and background load components. The processing terminal 14 continuously receives and calculates the above two parameters, so that the subsequent attenuation rate calculation and permeability compensation have a direct data basis. The extraction process of current spectrum distortion rate and high frequency phase jitter variance is performed in a fixed data flow order, and its input source is the continuous discrete sampling sequence output by the analog-to-digital conversion circuit and the sampling clock reference inside the processing terminal 14. For the current spectrum distortion rate, the data processing unit first extracts an analysis window of a predetermined length from the continuous sampling sequence, and performs amplitude normalization and window function weighting on the analysis window data. Then, it performs spectrum decomposition on the 3000Hz to 5000Hz frequency band, identifies the stable component with the largest amplitude as the reference harmonic component in the analysis window, and the remaining components as non-reference harmonic components. It calculates the ratio between the sum of squares of the amplitudes of the non-reference harmonic components and the sum of squares of the total amplitudes of the frequency band to obtain parameter B. Stable components refer to components whose frequency position changes less than the preset frequency resolution unit and whose amplitude maintains the local maximum characteristic in multiple consecutive analysis windows, so as to avoid misjudging transient spikes as reference harmonics; For the high-frequency phase jitter variance at the zero-crossing point of the waveform, the data processing unit first extracts the zero-crossing time sequence of the main power frequency wave from the original induced current signal, and then takes each zero-crossing time as a time reference point, extracts the high-frequency local waveform in its neighborhood, calculates the phase offset of the high-frequency local waveform relative to the reference zero-crossing time, and forms a phase offset sequence P1, P2, P3... corresponding to multiple consecutive zero-crossing points. Then, the discreteness of the phase offset sequence is statistically analyzed to obtain the parameter A. Parameter A characterizes the temporal dispersion of high-frequency responses near multiple consecutive zero crossings. The larger the parameter A, the more unstable the arc conduction state. The larger the parameter B, the higher the proportion of abnormal spectral components in the target high-frequency band. After calculation, parameters A and B are output to the attenuation rate calculation module, and are used as time-domain disturbance input and frequency-domain distortion input, respectively, to participate in the subsequent comprehensive fluctuation index calculation. In actual implementation, the analysis window length can be selected from 2ms to 10ms, and adjacent analysis windows can be updated by sliding at a fixed step size, so as to balance the ability to track transient signals and the stability of statistical calculation. To ensure the programmable implementation of the algorithm and the clarity of data flow, the specific processing logic for calculating the phase offset is as follows: The data processing unit uses a bandpass digital filter to extract the intercepted high-frequency local waveform, determines the first peak time of the local waveform through a peak detection algorithm, calculates the time difference between the peak time and the reference zero-crossing time, and multiplies the time difference by the high-frequency center frequency and a constant. Or, it can be converted into a phase offset of 360°; To ensure dimensional consistency in the subsequent calculation of the comprehensive volatility index, this embodiment uniformly adopts the radian constant 2π for conversion, so that the calculated high-frequency phase jitter variance is in square radians. Correspondingly, the dimension of the attenuation conversion factor K is strictly the reciprocal of the square radians. The high-frequency center frequency is the arithmetic center frequency of the target high-frequency band, specifically 4000Hz; the discreteness statistics specifically involve calculating the mathematical variance of multiple consecutive phase offsets, which is used as parameter A. The specific processing logic for calculating the ratio to obtain parameter B is as follows: take the sum of the squares of the amplitudes of all discrete frequency points in the 3000Hz to 5000Hz frequency band as the total amplitude energy, take the sum of the squares of the amplitudes of the frequency points after removing the reference harmonic components as the non-reference harmonic energy, and divide the two to obtain parameter B.

[0022] Example 2: The processing terminal 14 is also configured to calculate the comprehensive fluctuation index and the attenuation rate of the induced signal; wherein, the comprehensive fluctuation index is used to quantify the disturbance in the time domain and the frequency domain, and its calculation formula is: comprehensive fluctuation index = high-frequency phase jitter variance × current spectrum distortion rate. The formula for calculating the attenuation rate of the induced signal is: Attenuation rate = Comprehensive fluctuation index × Attenuation conversion factor; The attenuation conversion factor is a pre-calibrated value related to the grid voltage level, and the dimension of the attenuation conversion factor is the reciprocal of the dimension of the high-frequency phase jitter variance. After obtaining the high-frequency phase jitter variance A and the current spectrum distortion rate B, the processing terminal 14 performs a product operation on the two to obtain the comprehensive fluctuation index. This comprehensive fluctuation index is used to uniformly quantify the degree of phase instability in the time domain and the degree of spectrum distortion in the frequency domain, so as to avoid misjudgment of a single index under the conditions of arc intermittency or load disturbance. Parameter A can be obtained by calculating the variance of the high-frequency phase offset sequence corresponding to multiple consecutive zero crossings, and parameter B can be obtained by calculating the spectral energy distribution in the 3000Hz to 5000Hz frequency band; the attenuation conversion coefficient K related to the grid voltage level is a pre-calibrated coefficient, and its value can be obtained by experimental samples or field calibration based on the insulation level, fault arc sustaining conditions and line distribution parameters of 10kV, 20kV or 35kV distribution networks. Specifically, the attenuation conversion factor K has a negative correlation with the distribution network voltage level. Since the higher the distribution network voltage level, the higher the corresponding line insulation level and arc sustaining voltage, the high-frequency disturbance energy excited under the same fault grounding impedance is relatively weakened. Therefore, the value of the pre-calibrated attenuation conversion factor K decreases accordingly as the voltage level increases, so as to ensure that the normalization benchmark of the final calculated attenuation rate remains consistent under different voltage levels. Multiplying the overall fluctuation index by K yields the inductive signal attenuation rate, which describes the degree of attenuation of the current fault signal relative to the noise floor of the sensing link. For example, when A is 0.12, B is 0.35, and K is 1.8, the overall fluctuation index is 0.042, and the inductive signal attenuation rate is 0.0756. The processing terminal 14 determines whether the current fault signal has entered the weak induction zone based on the result, and transmits the attenuation rate to the subsequent compensation calculation module; the method of multiplying A and B corresponds to the coupled description of the dual characteristics of arc impedance fluctuation in this invention, which aims to enhance the identification of real grounding arcs.

[0023] The processing terminal 14 is configured to invert the relative permeability compensation amount required for the inner sensing magnetic ring 6 based on the attenuation rate, and to convert the nonlinear change of the arc impedance when a ground fault occurs into a compensation curve of the required relative permeability. The higher the attenuation rate, the greater the required relative permeability compensation amount. The processing terminal 14 is equipped with a relative permeability inversion estimation logic. The purpose of this inversion estimation logic is to accurately calculate the amount of compensation required for the inner sensing magnetic ring 6 to restore sensing sensitivity or anti-saturation capability when the nonlinear change of arc impedance causes the sensing signal to attenuate. The estimation logic comprises two processing stages in the data flow: the pre-matching stage receives the currently calculated attenuation rate as input, combines it with the pre-stored magnetic response characteristic calibration data of the inner layer sensing magnetic ring 6, and outputs the target relative permeability value; the post-comparison stage receives the target relative permeability value and the current estimated value of the actual relative permeability, and uses it to calculate the relative permeability compensation amount; the inversion estimation logic as a whole characterizes the physical causal relationship between the unstable behavior on the arc side and the controllable quantity on the sensor magnetic circuit side; Specifically, the processing terminal 14 inverts the relative permeability compensation amount required for the inner sensing magnetic ring 6 based on the pre-stored calibration relationship between the attenuation rate and the magnetic response characteristics of the inner sensing magnetic ring 6. In this invention, the relative permeability compensation amount refers to the range of permeability change that needs to be adjusted to restore the inner sensing magnetic ring 6 to a predetermined sensing sensitivity or a predetermined anti-saturation capability. This calibration relationship can be obtained through prototype testing. Specifically, the amplitude of the sensing magnetic ring 6 of the inner layer is measured under different DC bias currents to detect standard fault signals, and a correspondence table or function between the attenuation rate and the target permeability is established. After receiving the current attenuation rate, the processing terminal 14 looks up the required compensation amount in the correspondence table or calculates it in the function, and outputs the compensation curve that changes with time. The nonlinear change of the arc impedance is mapped to the permeability compensation requirement through the attenuation rate formed by A, B, and K, thereby converting the unstable behavior on the arc side into a controllable quantity that can be executed on the sensor magnetic circuit side. The higher the attenuation rate, the greater the relative permeability compensation required. This means that in order to offset the decrease in detection sensitivity caused by the attenuation of the induced signal, the processing terminal 14 will calculate a larger permeability adjustment range so that the inner sensing magnetic ring 6 is closer to the working area that is more conducive to the detection of weak signals. For inner sensing magnetic rings 6 with different material batches or different number of turns, the compensation curve can be recalibrated, but the control principle of determining the compensation amount based on the attenuation rate remains unchanged. The process of inverting the compensation amount based on the attenuation rate can be broken down into the following steps: Step 1, the processing terminal 14 reads the attenuation rate obtained from the current sampling window and compares the attenuation rate with the attenuation rate range in the calibration table; Step 2, if the current attenuation rate falls exactly on a calibrated node, the target relative permeability value corresponding to that node is directly read; if the current attenuation rate is between two adjacent calibrated nodes, interpolation estimation is performed according to the changing trend between adjacent nodes to obtain the intermediate target relative permeability value. The interpolation estimation preferably adopts a first-order linear interpolation algorithm, that is, the corresponding target relative permeability value is calculated according to the distance ratio between two adjacent upper and lower attenuation rate nodes in the calibration table based on the current attenuation rate. The data structure of the calibration table is a two-dimensional mapping array containing multiple attenuation rate nodes and corresponding target relative permeability values. This array is stored as a static constant in the non-volatile memory of the processing terminal 14 so that the program can quickly address and call it. Step 3: The processing terminal 14 compares the target relative permeability value with the estimated value of the actual relative permeability corresponding to the current operating point to obtain the relative permeability compensation amount; Step 4: The compensation amounts obtained from multiple consecutive sampling windows are arranged in chronological order and smoothly connected to form a compensation curve; The estimated value of the actual relative permeability corresponding to the current operating point can be determined by combining the current DC bias current level, the corresponding material calibration data and the most recent standard pulse response result. The purpose is to avoid compensation mismatch caused by directly using fixed empirical values. The specific processing logic determined by the comprehensive analysis is as follows: using the theoretical permeability of the current DC bias current range in the corresponding material calibration data as the reference value, the ratio of the measured amplitude of the most recent standard pulse response result to the initial calibration reference amplitude is calculated as the correction coefficient. Multiplying the reference value by the correction coefficient will yield the estimated value of the actual relative permeability corresponding to the current operating point. The predetermined sensing sensitivity refers to the state in which the output amplitude of the main sensing coil 7 to the standard fault injection signal reaches the predetermined reference value. The predetermined anti-saturation capability refers to the state in which the output of the main sensing coil 7 still remains within the linear tolerance range under the rated large load disturbance condition. The processing terminal 14 can select one of them as the compensation target according to the current operating scenario. When the attenuation rate continues to increase, the processing terminal 14 prioritizes increasing the compensation amount according to the sensitivity recovery target; when the attenuation rate decreases and the large current disturbance increases, the processing terminal 14 prioritizes limiting the compensation amount according to the anti-saturation target; in order to prevent the compensation amount from exceeding the reversible adjustment range of the magnetic ring material, the target relative permeability value is also constrained by an upper and lower limit, which are determined by the pre-test results of the inner sensing magnetic ring 6 material, and the boundary value is output when the limit is exceeded. Therefore, the compensation curve is not an abstract empirical quantity, but a control result obtained by combining attenuation rate input, calibration table lookup, interpolation conversion, comparison of actual operating points and boundary constraints, and is further transmitted to the duty cycle adjustment module of the micropower constant current source circuit 15.

[0024] The processing terminal 14 is configured to adjust the duty cycle of the pulse width modulation signal output to the DC bias coil 8 in real time according to the calculated compensation curve; the DC bias magnetization adjustment field generated by the DC bias coil 8 dynamically changes the relative permeability and giant magnetoresistance effect of the inner layer sensing magnetic ring 6 by changing the magnetic domain arrangement state inside the inner layer sensing magnetic ring 6. The low-power constant current source circuit 15 includes a reference source, a power switching device, a current sampling unit, and a pulse width modulation control unit. After the processing terminal 14 outputs the control command corresponding to the compensation curve, the pulse width modulation control unit adjusts the duty cycle of the power switching device according to the set carrier frequency, so that the average current output by the constant current source to the DC bias coil 8 changes. After the DC bias coil 8 is energized, a DC bias magnetization adjustment field is established around the inner sensing magnetic ring 6. This magnetization adjustment field changes the equivalent relative permeability of the inner sensing magnetic ring 6 in a controllable manner by changing the arrangement of magnetic domains and the motion state of domain walls inside the amorphous alloy. In this invention, the giant magnetoresistance effect refers to the phenomenon that the magnetic material changes its high-frequency electromagnetic response under the action of a bias magnetic field. This effect is related to the sensing ability of the main sensing coil 7 to detect high-frequency fault components. The processing terminal 14 adjusts the duty cycle of the pulse width modulation signal in real time, which means that the bias current can be corrected according to the latest attenuation rate after the continuous sampling window is updated, without relying on manually setting a fixed bias value. In actual implementation, the pulse width modulation frequency can be set from 10kHz to 100kHz, which ensures the smoothness of constant current output and reduces the coupling effect on the main sensing signal frequency band. Through this electrical control method, the inner sensing magnetic ring 6 can maintain the magnetic operating point corresponding to the current sensing requirements in different fault stages.

[0025] When the processing terminal 14 determines that the comprehensive fluctuation index and attenuation rate have increased and the amplitude of the induced current signal has attenuated to less than or equal to the preset noise floor threshold, the processing terminal 14 automatically reduces the duty cycle of the pulse width modulation signal and weakens the DC bias field strength, so that the relative permeability of the inner sensing magnetic ring 6 rises back to the preset target relative permeability working area. When the processing terminal 14 detects that the amplitude of the induced current signal suddenly increases to a level greater than the average amplitude of the previous statistical period due to large load fluctuations or arc breakdown in the power grid and the distortion rate of the current spectrum decreases, the processing terminal 14 immediately increases the duty cycle of the pulse width modulation signal, enhances the DC bias field strength, and reduces the relative permeability of the inner sensing magnetic ring 6 to the preset anti-saturation relative permeability region. During operation, the processing terminal 14 continuously compares the changing trends of the current attenuation rate, the comprehensive fluctuation index, and the current spectrum distortion rate. When the comprehensive fluctuation index increases, the attenuation rate increases, and the output amplitude of the main sensing coil 7 approaches the preset noise floor threshold, the processing terminal 14 determines that the arc impedance increases and the fault signal tends to be weak. As the arc impedance increases, the fault signal tends to be weak and approaches the noise floor of the sensing link. Therefore, at this time, the processing terminal 14 reduces the duty cycle of the pulse width modulation signal, which reduces the output current of the micro-power constant current source, thereby weakening the DC bias field strength and returning the inner sensing magnetic ring 6 to the preset target relative permeability working area, thereby improving the sensing capability of weak high-frequency fault components. The specific method for reducing the duty cycle is as follows: the duty cycle is gradually reduced in each control update cycle with a preset fixed step size until the output amplitude of the main sensing coil 7 rises back to a level greater than the noise floor threshold or reaches the lower limit of the duty cycle; the noise floor threshold can be determined by multiplying the root mean square value of the statistical noise during the no-load or healthy operation of the equipment by a predetermined multiple. Conversely, when the processing terminal 14 detects a short-term surge in the output amplitude of the main sensing coil 7 and a decrease in the current spectrum distortion rate, it indicates that the proportion of large current components in the signal has increased, which may be caused by large load fluctuations in the power grid or arc breakdown. Because large load fluctuations or arc breakdowns in the power grid can cause a short-term surge in the output amplitude of the main sensing coil 7, which can easily lead to bias saturation of the inner sensing magnetic ring 6, the processing terminal 14 increases the duty cycle of the pulse width modulation signal to generate a bias field greater than the current operating state in the DC bias coil 8. This pushes the operating point of the inner sensing magnetic ring 6 toward the preset anti-saturation relative permeability region, thereby improving the anti-bias saturation capability and preventing distortion of the output of the main sensing coil 7. The specific method for increasing the duty cycle is as follows: based on the proportion by which the current output amplitude exceeds the average amplitude of the previous statistical period, the duty cycle is increased in a single step, using the system's preset maximum adjustment rate to suppress the relative permeability and prevent deep saturation of the magnetic ring; the specific formula for this operation is: Target duty cycle = Current duty cycle + Base adjustment step size × [(Current output amplitude / Average amplitude of the previous statistical period) - 1] The calculated target duty cycle must be limited to the maximum duty cycle safety threshold set by the system; where the target duty cycle is the adjusted pulse width modulation signal duty cycle; the current duty cycle is the pulse width modulation signal duty cycle of the current control cycle; the basic adjustment step size is the preset duty cycle step increase reference value; the current output amplitude is the output amplitude of the main sensor coil 7 in the current control cycle; and the average amplitude of the previous statistical cycle is the average output amplitude of the main sensor coil 7 in the previous statistical cycle. This adjustment method allows the inner sensing magnetic ring 6 to switch between a high-sensitivity state and an anti-saturation state, and the switching is based directly on the results of feature extraction, attenuation rate calculation, and compensation curve generation, ensuring that there is a consistent implementation logic between the mechanical structure, signal processing, and bias control. In this invention, the preset noise floor threshold is used to characterize the upper boundary of the background noise that the sensing link can accept under fault-free or healthy operating conditions. Its logical function is to serve as the triggering basis for weak fault identification and bias downshift control. The process of determining this threshold can be as follows: First, collect multiple sets of output data from the main sensor coil 7 under the conditions of no-load, normal load and acceptable environmental electromagnetic interference. Calculate the root mean square value of each set of data, then take its statistical mean and multiply it by a safety factor of 1.5 to 3 to obtain the noise floor threshold. When the rate of change of the on-site environmental parameters exceeds the preset environmental fluctuation threshold, the threshold can also be updated in segments over time. To avoid accidental triggering caused by occasional spikes in a single sampling window, the processing terminal 14 preferably adopts a joint judgment method for multiple consecutive sampling windows: when the output amplitude of the main sensing coil 7 is lower than or equal to the noise floor threshold for at least two consecutive sampling windows, and the synchronous attenuation rate and the comprehensive fluctuation index continue to rise, the duty cycle reduction command is output. Correspondingly, the criterion for a sudden increase in current caused by large load fluctuations or arcing is not determined solely by the instantaneous amplitude, but by the increment of the output amplitude of the main sensing coil 7 relative to the previous statistical period, the decreasing trend of the current spectrum distortion rate, and the duration. Among these, if the current output amplitude is higher than the average amplitude of the previous statistical period and remains so for at least one control update period, while the current spectrum distortion rate decreases relative to the previous statistical period, the processing terminal 14 determines that it has entered the anti-saturation priority state and outputs an instruction to increase the duty cycle. The previous statistical period can take 5 to 20 consecutive sampling windows to provide a reference background for the current working condition. In order to prevent the duty cycle from switching back and forth between the two states frequently, the processing terminal 14 can also set a hysteresis interval, that is, the exit condition of reducing the duty cycle is higher than its entry condition, and the exit condition of increasing the duty cycle is lower than its entry condition, thereby forming a stable dual threshold switching logic. Therefore, the control actions of automatically reducing or immediately increasing the duty cycle of the pulse width modulation signal all have clear sources of input parameters, judgment order and triggering conditions, enabling those skilled in the art to implement the specific control process of the processing terminal 14 accordingly.

[0026] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An online monitoring device for grounding faults in a power distribution network, comprising: An insulating base (1), an inner sensing magnetic ring (6), an outer compensation magnetic ring (10), and a processing terminal (14) are characterized in that: the insulating base (1) has a main through hole (2) for the three-phase conductors of the power distribution network to pass through, and the insulating base (1) has an inner ring groove (3) and an outer ring groove (4) arranged coaxially from the inside to the outside in a radial direction, and an annular partition (5) is provided between the inner ring groove (3) and the outer ring groove (4); The inner sensing magnetic ring (6) is installed in the inner ring groove (3), and the surface of the inner sensing magnetic ring (6) is uniformly wound with a main sensing coil (7), a DC bias coil (8) and a cancellation coil (9) in the circumferential direction. The outer compensation magnetic ring (10) is installed in the outer ring groove (4). The outer compensation magnetic ring (10) is coaxially nested outside the inner sensing magnetic ring (6). The outer compensation magnetic ring (10) has a reverse series winding (11) wound on its surface. The two ends of the reverse series winding (11) are directly connected to the canceling coil (9) wound on the inner sensing magnetic ring (6) through wires (12) to form a self-generating compensation circuit; The output terminal of the main sensing coil (7) and the input terminal of the DC bias coil (8) are electrically connected to the processing terminal (14) via shielded cables (13); The processing terminal (14) is equipped with a micro-power constant current source circuit (15), and the output terminal of the micro-power constant current source circuit (15) is connected to the DC bias coil (8).

2. The online monitoring device for grounding faults in power distribution networks according to claim 1, characterized in that: The inner sensing magnetic ring (6) is formed by winding amorphous alloy strip, and the outer compensation magnetic ring (10) is formed by stacking silicon steel sheets. The main through hole (2) has a diameter of 60 mm, the inner sensing magnetic ring (6) has an outer diameter of 120 mm, an inner diameter of 80 mm, and a thickness of 20 mm, the outer compensation magnetic ring (10) has an outer diameter of 180 mm, an inner diameter of 140 mm, and a thickness of 20 mm, and the annular partition (5) has a thickness of 10 mm.

3. The online monitoring device for grounding faults in power distribution networks according to claim 1, characterized in that: The reverse series winding (11) is composed of two sets of half-loop coils (16) with opposite directions connected end to end; the main sensing coil (7), the DC bias coil (8) and the canceling coil (9) are interlayered by a polyimide film (17).

4. The online monitoring device for grounding faults in power distribution networks according to claim 1, characterized in that: The reverse series winding (11) is configured to induce a leakage magnetic field and generate an induced current when the three-phase conductors of the power supply and distribution network pass through the main through hole (2) and the spatial arrangement is asymmetrical to generate a leakage magnetic field. The cancelling coil (9) is connected to the reverse series winding (11) and is configured to receive the induced current and generate a reverse cancelling magnetic flux that is equal in magnitude and opposite in direction to the leakage magnetic field, so as to maintain the zero magnetic field geometric accuracy of the core sensing axis of the inner sensing magnetic ring (6).

5. The online monitoring device for grounding faults in power distribution networks according to claim 1, characterized in that: The processing terminal (14) continuously receives the induced current signal output by the main sensing coil (7). The processing terminal (14) is configured to extract the current spectrum distortion rate of the induced current signal in the high frequency band of 3000Hz to 5000Hz and the high frequency phase jitter variance of the waveform zero crossing point of the induced current signal.

6. The online monitoring device for grounding faults in power distribution networks according to claim 5, characterized in that: The processing terminal (14) is also configured to calculate the comprehensive fluctuation index and the attenuation rate of the induced signal; wherein the comprehensive fluctuation index is used to quantify the disturbance in the time domain and the frequency domain, and its calculation formula is: comprehensive fluctuation index = high frequency phase jitter variance × current spectrum distortion rate; The formula for calculating the attenuation rate of the induced signal is: Attenuation rate = Comprehensive fluctuation index × Attenuation conversion coefficient; The attenuation conversion coefficient is a pre-calibrated value related to the grid voltage level, and the dimension of the attenuation conversion coefficient is the reciprocal of the dimension of the high-frequency phase jitter variance.

7. The online monitoring device for grounding faults in power distribution networks according to claim 6, characterized in that: The processing terminal (14) is configured to invert the required relative permeability compensation amount of the inner sensing magnetic ring (6) based on the attenuation rate, and to convert the nonlinear change of the arc impedance when a ground fault occurs into a compensation curve of the required relative permeability. The higher the attenuation rate, the greater the required relative permeability compensation amount.

8. The online monitoring device for grounding faults in power distribution networks according to claim 7, characterized in that: The processing terminal (14) is configured to adjust the duty cycle of the pulse width modulation signal output to the DC bias coil (8) of the micro-power constant current source circuit (15) in real time according to the calculated compensation curve. The DC bias magnetization adjustment field generated by the DC bias coil (8) dynamically changes the relative permeability and giant magnetoresistance effect of the inner sensing magnetic ring (6) by changing the magnetic domain arrangement state inside the inner sensing magnetic ring (6).

9. The online monitoring device for grounding faults in power distribution networks according to claim 8, characterized in that: When the processing terminal (14) determines that the comprehensive fluctuation index and the attenuation rate increase and the amplitude of the induced current signal attenuates to less than or equal to the preset noise floor threshold, the processing terminal (14) automatically reduces the duty cycle of the pulse width modulation signal, weakens the DC bias field strength, and makes the relative permeability of the inner sensing magnetic ring (6) rise back to the preset target relative permeability working area. When the processing terminal (14) detects that the amplitude of the induced current signal suddenly increases to a value greater than the average amplitude of the previous statistical period and the distortion rate of the current spectrum decreases due to large load fluctuations or arc breakdown of the power grid, the processing terminal (14) instantly increases the duty cycle of the pulse width modulation signal, enhances the DC bias field strength, and reduces the relative permeability of the inner sensing magnetic ring (6) to a preset anti-saturation relative permeability region.

Images