Power distribution network small current single-phase ground fault protection method, system, device and medium

Abstract

The present application relates to the technical field of relay protection, and provides a power distribution network small current single-phase ground fault protection method, system, device and medium, comprising: when a suspected ground fault is monitored, starting a preset protection action criterion according to fault transient data; in response to the starting of the preset protection action criterion, performing gradient current limiting and starting fault self-healing monitoring; after entering the current limiting steady state period, generating a detection signal and actively injecting it based on the current power distribution network operating parameters; performing fault analysis according to the detection signal response characteristics and the current power distribution network operating parameters; performing protection delay analysis according to the fault level, fault distance and protection device downstream load priority to obtain a protection action delay; when the preset self-healing monitoring time is exceeded and the fault is not self-healed, triggering a circuit breaker to perform a protection action according to the protection action delay. The present application can improve the scientific rationality of small current ground fault protection in scenarios such as new energy weak feeding and weak fault characteristics, and effectively balance protection speed and power supply reliability.

Description

Technical Field

[0001] This invention relates to the field of relay protection technology, and in particular to a method, system, equipment and medium for protecting against small current single-phase grounding faults in distribution networks. Background Technology

[0002] With the construction of new power systems and the large-scale integration of distributed power sources, the distribution network exhibits the "three highs" characteristics: high proportion of imported power, high proportion of new energy sources, and high proportion of peak-valley difference. In the "three highs" distribution network, low-current grounding systems (including neutral point ungrounded and grounded through arc suppression coils) are widely used in 6kV~35kV medium-voltage distribution networks because they maintain symmetrical line voltage during faults and can operate with faults for short periods.

[0003] However, in high-efficiency, high-voltage, and high-power distribution networks, distributed photovoltaic and wind power are connected to the grid on a large scale through power electronic converters. During a fault, the converter only provides 1-3 times the rated current to the protection devices, exhibiting weak feedback characteristics. This can cause traditional overcurrent protection to fail to operate because the fault current does not reach the threshold. Furthermore, the dynamic adjustment of the converter makes the fault transient discrete and time-varying, undermining the linear analysis basis of the protection. At the same time, the large load fluctuation and flexible and varied topology of the high-efficiency, high-voltage, and high-power distribution network further exacerbate the problem of weakened fault characteristics. Therefore, when traditional active injection protection only monitors electrical parameters and the current limiting strategy mostly adopts fixed thresholds or traditional blocking methods, it is very easy to cause load shocks and increased power outage duration. This makes it face many technical bottlenecks in the application of high-efficiency, high-voltage, and high-power distribution networks, making it difficult to effectively balance protection speed and power supply reliability. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method, system, device, and medium for protecting against low-current single-phase grounding faults in distribution networks. This method employs a collaborative protection mechanism that uses adaptive generation of detection signal injection based on distribution network operating status data, fault analysis based on line physical state characteristics and signal injection timing characteristics, and protection delay design based on fault level, fault distance, and downstream load priority of the protection equipment. This mechanism effectively balances the speed of low-current grounding fault protection with power supply reliability.

[0005] In a first aspect, embodiments of the present invention provide a method for protecting against low-current single-phase grounding faults in a distribution network, the method comprising: When a suspected grounding fault is detected in the target distribution network, the preset protection action criteria are activated based on the acquired fault transient data. In response to the activation of the preset protection action criterion, a gradient current limiting strategy is executed and fault self-healing monitoring is initiated. After entering the current limiting steady state period, a detection signal generated based on the current distribution network operating parameters is actively injected into the target distribution network, and the detection signal response characteristics are acquired simultaneously. The detection signal response characteristics include the electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process. Fault analysis is performed based on the detection signal response characteristics and the current power distribution network operating parameters to obtain corresponding fault information; the fault information includes fault distance and fault level. Based on the fault level, the fault distance, and the downstream load priority of the protection device, a protection delay analysis is performed to obtain the protection action delay. If the preset self-healing monitoring time is exceeded and the fault has not self-healed, the corresponding circuit breaker is triggered to perform the protection action according to the protection action delay.

[0006] Furthermore, the steps for generating detection signals based on current distribution network operating parameters include: When the load data of the faulty line in the current distribution network operating parameters is less than the preset load threshold, a pulse signal with a preset duty cycle range is generated. When the load data of the faulty line is not less than the preset load threshold and the proportion of new energy output in the current distribution network operation parameters is greater than the preset output threshold, a wideband sweep frequency signal with a preset frequency range is generated. When the load data of the faulty line is not less than the preset load threshold and the output ratio of the new energy source is not greater than the preset output threshold, a single-frequency continuous wave signal is generated.

[0007] Furthermore, the step of performing fault analysis based on the detection signal response characteristics and the current distribution network operating parameters to obtain the corresponding fault information includes: Feature anomaly detection is performed on the electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process, respectively, to obtain corresponding feature detection results; the feature detection results include electrical feature detection results, physical feature detection results, and timing feature detection results; Based on the feature detection results and the current power distribution network operating parameters, feature fusion analysis is performed on the injected electrical features, the injected line physical state features, and the injection process timing features to obtain the corresponding fusion feature strengths. The fault level is determined by identifying the fault level based on the intensity of the fused feature and the abnormal feature type and number of abnormal features corresponding to the feature detection results. The fault distance is obtained by measuring the fault distance based on the signal frequency of the detection signal and the fault transient waveform data and line topology switching status data in the current distribution network operating parameters.

[0008] Further, the step of performing feature fusion analysis on the injected electrical features, the injected line physical state features, and the injection process timing features based on the feature detection results and the current distribution network operating parameters to obtain the corresponding fused feature strength includes: The electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process are extracted to obtain the corresponding electrical comprehensive characteristics, physical state comprehensive characteristics, and timing comprehensive characteristics. Based on the current power distribution network operating parameters and the physical feature detection results in the feature detection results, determine the feature weight vector corresponding to the detection signal response feature; The electrical integrated feature, the physical state integrated feature, and the temporal integrated feature are weighted and fused according to the feature weight vector to obtain the fused feature intensity.

[0009] Further, the step of determining the feature weight vector corresponding to the detection signal response feature based on the current distribution network operating parameters and the physical feature detection results in the feature detection results includes: When the proportion of new energy output in the current distribution network operation parameters is not greater than the preset output threshold and the physical feature detection result is normal, the feature weight vector is set as the basic weight vector. When the proportion of new energy output in the current power distribution network operation parameters is greater than the preset output threshold and the physical feature detection result is normal, the feature weight vector is set based on the preset time-series feature weight lower limit and the preset electrical feature weight upper limit; When the physical feature detection result is abnormal, the feature weight vector is set based on the preset physical state feature weight lower limit.

[0010] Further, the step of determining the fault distance based on the signal frequency of the detection signal and the fault transient waveform data and line topology switching status data in the current distribution network operating parameters includes: Based on the bus-side voltage waveform data and line-side current waveform data in the fault transient waveform data, the steady-state phasor at the signal frequency is obtained based on the discrete Fourier transform, and the probe voltage phasor and probe current phasor are obtained. Based on the line topology switching status data, obtain the line parameters of the currently faulty line, and based on the line parameters, obtain the detection unit length loop impedance of the currently faulty line at the signal frequency. Based on the circuit impedance per unit length of the detection, the detection voltage phasor, and the detection current phasor, a corresponding fault location equation is constructed; The fault distance is obtained by solving the fault location equation.

[0011] Further, the step of performing protection delay analysis based on the fault level, the fault distance, and the downstream load priority of the protection device to obtain the protection action delay includes: Based on the fault distance, the corresponding electrical distance base delay is obtained based on the preset inverse time protection action characteristic curve; The corresponding load priority correction coefficient is obtained based on the highest load priority among the downstream load priorities of the protection device; Based on the fault level, a fault level correction coefficient is determined, and based on the fault level correction coefficient and the load priority correction coefficient, the electrical distance base delay is corrected to obtain the protection action delay.

[0012] Secondly, embodiments of the present invention provide a low-current single-phase grounding fault protection system for distribution networks, the system comprising: The fault monitoring module is used to initiate preset protection action criteria based on the acquired fault transient data when a suspected grounding fault is detected in the target distribution network. The signal injection module is used to respond to the preset protection action criterion, execute the gradient current limiting strategy and start fault self-healing monitoring, and actively inject the detection signal generated based on the current distribution network operating parameters into the target distribution network after entering the current limiting steady state period, and simultaneously acquire the detection signal response characteristics; the detection signal response characteristics include the electrical characteristics after injection, the physical state characteristics of the line after injection and the timing characteristics of the injection process; The fault analysis module is used to perform fault analysis based on the response characteristics of the detection signal and the current power distribution network operating parameters to obtain corresponding fault information; the fault information includes fault distance and fault level. The protection delay module is used to perform protection delay analysis based on the fault level, the fault distance and the downstream load priority of the protection device to obtain the protection action delay, and to trigger the corresponding circuit breaker to perform protection action when the preset self-healing monitoring time is exceeded and the fault has not self-healed.

[0013] Thirdly, embodiments of the present invention also provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0014] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0015] This invention provides a method, system, device, and medium for protecting against single-phase grounding faults with low current in a distribution network. The method implements the following: when a suspected grounding fault is detected in the target distribution network, a preset protection action criterion is initiated based on acquired transient fault data; in response to the activation of the preset protection action criterion, a gradient current limiting strategy is executed and fault self-healing monitoring is initiated; after entering the current-limiting steady-state period, a detection signal generated based on the current distribution network operating parameters is actively injected into the target distribution network, simultaneously acquiring detection signal response characteristics including post-injection electrical characteristics, post-injection line physical state characteristics, and injection process timing characteristics; fault analysis is performed based on the detection signal response characteristics and current distribution network operating parameters to obtain fault information including fault distance and fault level; protection delay analysis is performed based on the fault level, fault distance, and downstream load priority of the protection equipment to obtain the protection action delay; and if the preset self-healing monitoring time is exceeded and the fault has not self-healed, the corresponding circuit breaker is triggered to perform protection action based on the protection action delay. Compared with existing technologies, this method for protecting small-current single-phase grounding faults in distribution networks improves the scientific rationality of protection against small-current grounding faults in scenarios with weak new energy feed and weak fault characteristics. It adopts a collaborative protection mechanism that uses adaptive generation of detection signal injection based on distribution network operating status data, fault analysis based on line physical state characteristics and signal injection timing characteristics, and protection delay design based on fault level, fault distance, and downstream load priority of protection equipment. This effectively balances protection speed and power supply reliability, thereby improving the safe and stable operation level of the distribution network. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the method for protecting a single-phase ground fault with low current in a power distribution network according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the distribution network low-current single-phase grounding fault protection system in an embodiment of the present invention; Figure 3 This is an internal structural diagram of the computer device in an embodiment of the present invention; The attached figures are labeled as follows: 1. Fault monitoring module; 2. Signal injection module; 3. Fault analysis module; 4. Protection delay module. Detailed Implementation

[0017] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. Obviously, the embodiments described below are only part of the embodiments of this invention and are used to illustrate the invention, but are not intended to limit the scope of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0018] The low-current single-phase grounding fault protection method for distribution networks provided by this invention is mainly applicable to low-current grounding systems in 6kV~35kV "three-high" distribution networks, including typical topologies such as ungrounded neutral point and grounded via arc suppression coil, and containing distributed photovoltaic grid-connected medium-voltage distribution network lines. In the "three-high" distribution network: distributed photovoltaic power stations adopt a topology combining Boost DC / DC converters and two-level voltage source converters (VSCs). The DC / DC converters use the disturbance observation method to achieve maximum power point tracking (MPPT). Tracking control: The VSC employs a dual closed-loop control of constant AC reactive power and constant DC voltage, ensuring stable output power during normal operation. In the medium-voltage distribution network, the dedicated photovoltaic grid-connected transmission line shares a busbar with conventional feeders. A Berylon model is used for modeling, and line parameters are adapted to the requirements of detection signal transmission, ensuring that the signal attenuation in the 200Hz-2kHz range is less than 50%. In the low-current grounding system, an arc-suppression coil is configured on the busbar side, with a compensation degree of 10%~15%, to suppress the capacitive current during single-phase grounding faults. During a fault, the system line voltage remains symmetrical, providing a short-time fault operation window for the protection scheme. Protection measurement points are set on both sides of the photovoltaic grid-connected transmission line, and voltage sensors (capacitive voltage divider type) and current sensors (capacitive voltage divider type) are configured on the busbar side and other feeders. The method employs a through-core structure to avoid three-phase current coupling interference and fiber optic vibration sensors (uniformly installed on line towers with spacing determined according to line length to ensure no blind spots in physical condition monitoring). The voltage and current sensors have a sampling frequency greater than 10kHz and an accuracy less than 0.2%, while the fiber optic vibration sensors have a monitoring accuracy of 0.1mm, meeting the requirements for detection signal acquisition. The response time of the protection equipment can be controlled within 50ms. The VSC and protection equipment are connected via a dual link combining hard-wired and fiber optic communication. Hard-wired transmission of emergency control commands (such as current-limiting start and trip commands) has a response time controlled within 1ms. Fiber optic communication can transmit large amounts of information such as detection signal reference values ​​and response data, ensuring a delay controlled within 10ms and guaranteeing the reliability of control commands and data transmission. The following embodiments will provide a detailed description of the distribution network low-current single-phase ground fault protection method of the present invention.

[0019] In one embodiment, such as Figure 1 As shown, a method for protecting a single-phase ground fault with low current in a power distribution network is provided, including the following steps: S11. When a suspected ground fault is detected in the target distribution network, a preset protection action criterion is activated based on the acquired fault transient data. Detecting a suspected ground fault in the target distribution network can be understood as detecting a zero-sequence voltage exceeding the limit or any phase voltage change rate exceeding the limit, indicating a suspected ground fault and the target distribution network system entering the ground fault transient stage. At this time, fault transient data for a period of time (0~5ms after the fault transient occurs) is immediately and continuously collected, and the detection algorithm of the preset protection action criterion is activated based on the obtained fault transient data to determine whether the protection action activation requirements are met. It should be noted that the preset protection action criterion can be the protection action judgment of an actual small current ground fault system, and the data items involved in the fault transient data only need to meet the analysis requirements of the preset protection action criterion; no restrictions are imposed here.

[0020] S12. In response to the activation of the preset protection action criterion, a gradient current limiting strategy is executed and fault self-healing monitoring is initiated. After entering the current limiting steady-state period, a detection signal generated based on the current distribution network operating parameters is actively injected into the target distribution network, and the detection signal response characteristics are acquired simultaneously. The gradient current limiting strategy is used to forcibly limit the fault current and maintain it in a stable state to avoid transient oscillations. In order to avoid converter shutdown, this embodiment preferably triggers the gradient current limiting strategy within a certain time range (5ms~20ms) after the activation of the preset protection action criterion. Based on the initial current limiting value (e.g., 80% of the rated current), the fault current is limited to 2-3 times the rated current by gradually reducing the current limiting step size to the lower limit of the current limit. This avoids the load impact caused by instantaneous power outage (such as motor shutdown or electronic equipment damage), and forces the fault transient to enter a stable state, so as to provide a stable data acquisition window for subsequent detection signal injection and protection action.

[0021] In practical applications, current limiting execution clamps the reference value of the power frequency current in the converter's inner current loop control to the required limited level. It should be noted that, to overcome the risk of grid disconnection and transient impact caused by traditional fixed current limiting values, the lower limit and current limiting step size of the gradient current limiting strategy in this embodiment can also be dynamically calculated based on the physical boundary constraints of the distribution network equipment. 1) The lower limit of the current limiting can be calculated based on the critical current condition for the self-extinguishing of the ground fault arc in the target distribution network and the power constraint condition for the minimum stable operation of the converter, so as to ensure that the current limiting depth satisfies both the physical environment for the self-healing of the arc at the fault point and maintains the converter's operation without disconnecting from the grid; that is, assuming the lower limit of the current limiting is The corresponding boundary constraint equations can be expressed as: In the formula, The minimum apparent power required for the converter to operate without disconnecting from the grid is determined by the hardware nameplate attributes of the converter in the target distribution network and the protection settings of the low voltage ride-through control strategy. It can be obtained in advance from the static database of the converter's local controller or the distribution network energy management system (EMS). and The rated line voltage and phase voltage of the target distribution network are the inherent reference parameters of the target distribution network, which can be directly retrieved from relevant ledgers or operation management systems. The equivalent comprehensive impedance of the fault circuit is a dynamic calculation parameter. It is obtained by retrieving the line impedance parameters (such as resistance per unit length, reactance, etc.) from the topology database of the target distribution network and combining them with the instantaneous voltage and current values ​​collected at the bus side during the initial stage of the fault transient. The equivalent impedance is calculated in real time based on Ohm's law or impedance measurement algorithm. The self-extinguishing reliability coefficient of the electric arc is an engineering experience setting value, which is mainly determined based on the voltage level of the target distribution network, the recovery characteristics of the equipment insulation medium, and the statistical law of historical faults. It is usually taken in the range of 0.6 to 0.8.

[0022] By substituting the real-time operating parameters of the target distribution network into the above formula, the lower limit of the current limit that can both induce the arc to self-extinguish and prevent the converter from shutting down can be calculated. Under typical "high-temperature, high-volume, and high-efficiency" distribution network conditions, the lower limit of the current limit is usually around 30% of the rated current, as calculated by the above boundary constraint equation.

[0023] 2) The current limiting step size can be dynamically adjusted based on the maximum allowable transient impact change rate of the equipment in the target distribution network and the preset self-healing monitoring time to ensure that secondary transient overvoltages and load shutdowns are not triggered during the gradient current reduction process; assuming the initial current limiting value is The current limiting step size is Therefore, the gradient descent process must be completed within the preset self-healing monitoring time. The process must be completed within 100ms (for example), and the rate of change of current caused by each current reduction step must not exceed the maximum allowable transient impact limit of the system. The corresponding set of step size tuning equations can be expressed as: In the formula, The execution cycle of a single rate limiting command can be set based on communication and control delays; N This represents the total number of steps for the downflow. This is a time margin for control.

[0024] By combining the above equations, the flow-limiting step size can be obtained. The safe range of values ​​is: Based on the step size tuning equations, when the initial current limiting value is set to 80% of the rated current and the lower current limiting limit is 30% of the rated current, the optimal current limiting step size is calculated. Approximately 10% of the rated current, this ensures a smooth reduction to the target value within the preset self-healing monitoring period, while also avoiding secondary overvoltage caused by excessive current change rate. Therefore, the preferred gradient current limiting strategy in this embodiment is to use 80% of the rated current as the initial current limiting value, and gradually reduce it to a lower limit of 30% with a gradient step size of 10%.

[0025] To avoid unnecessary protection actions due to short-term vibrations, this embodiment preferably monitors the fault self-healing status in real time during the execution of the gradient current limiting strategy. The longest monitoring time is the aforementioned preset self-healing monitoring time. If fault self-healing is detected within the preset self-healing monitoring time (e.g., instantaneous arc extinguishing, and electrical characteristic parameters such as impedance and phase returning to normal), the current limiting is immediately lifted and the target distribution network is restored to normal power supply.

[0026] After detecting the entry into the current-limiting steady-state period, within the current stabilization time window (20ms~120ms), a detection signal is adaptively injected based on the current distribution network operating parameters, and a complete set of detection signal response characteristics, including the electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process, are simultaneously collected for fault analysis. To ensure the reliability of the detection signal injection and improve the accuracy of state perception, this embodiment preferably adaptively generates different types of injection signals based on the current steady-state operating conditions of the target distribution network; specifically, the steps for generating detection signals based on the current distribution network operating parameters include: When the load data of the faulty line in the current distribution network operating parameters is less than a preset load threshold, a pulse signal with a preset duty cycle range is generated. The faulty line load data can be understood as the load rate of the faulty line, and the corresponding preset load threshold can be set based on actual application requirements to identify whether the faulty line is in an extremely light-load state. For example, it can be set to 20% of the line's rated capacity. That is, when the load data of the faulty line is lower than the preset load threshold, regardless of the proportion of renewable energy output, it is considered to be in an extremely light-load state. The equivalent damping of the distribution network is extremely small, and its sensitivity to externally injected energy increases sharply. If a steady-state detection signal is injected at this time, it is very easy to cause a rise in line voltage or induce low-frequency oscillations. Therefore, it is necessary to prioritize adaptive switching and inject a pulse signal with a duty cycle within the preset duty cycle range. This signal uses an extremely narrow pulse width to provide the transient high-frequency edge characteristics required for fault detection, while greatly limiting the total energy injected into the distribution network system, thereby minimizing the impact of signal injection on the voltage stability of the light-load grid (bus voltage fluctuation less than or equal to ±2%).

[0027] When the load data of the faulty line is not less than the preset load threshold and the proportion of renewable energy output in the current distribution network operating parameters is greater than the preset output threshold, a wideband sweep frequency signal within a preset frequency range is generated. The renewable energy output proportion can be understood as the percentage of renewable energy output relative to the total load. The corresponding preset output threshold can be determined based on actual application requirements; for example, the preset output threshold can be set to 50% of the total load for renewable energy output. That is, when the renewable energy output proportion is greater than the preset output threshold, the low-order harmonics (especially the 2nd and 3rd harmonics) generated by the corresponding renewable energy inverter group will be amplified and become the dominant component of background noise. If a single frequency signal is injected at this time, it is very easy to be overwhelmed by harmonics or experience frequency aliasing. Therefore, it is necessary to adaptively switch and inject a wideband sweep frequency signal within a preset frequency range. This wideband signal, through continuous frequency changes, can effectively penetrate specific harmonic frequency bands of the inverter, ensuring that even if some frequency bands are interfered with, the remaining frequency bands can still maintain a high signal-to-noise ratio, achieving the effect of resisting harmonic interference. It should be noted that the selection of the preset frequency range should avoid the power frequency and its integer multiples of harmonics, especially the low-order harmonic frequency band dominated by the inverter, so as to reduce signal distortion caused by harmonic interference. In this embodiment, the preset frequency range is preferably set to 100Hz~500Hz.

[0028] When the load data of the faulty line is not less than the preset load threshold and the output ratio of the new energy source is not greater than the preset output threshold, a single-frequency continuous wave signal is generated; that is, when the current distribution network operating parameters are under normal operating conditions, it is only necessary to inject a single-frequency continuous wave signal that avoids the power frequency and its integer multiples (e.g., 235Hz) to balance detection accuracy and calculation efficiency.

[0029] In addition, the amplitude of the detection signal is preferably controlled within 2% to 5% of the rated voltage. The lower limit of 2% is to meet the minimum detection sensitivity of the 0.2-level sensor, that is, the minimum detectable voltage signal is 0.2% of the rated voltage. The upper limit of 5% is to ensure that the bus voltage distortion rate after injection is less than or equal to 2% and does not affect the power quality. At the same time, the detection signal adopts the form of three-phase high-order harmonic voltage signal, which not only meets the detection accuracy requirements of the sensor, but also avoids affecting the normal operation of the user's electrical equipment. In practical applications, after generating the detection signal through the above steps, the active injection of the detection signal is achieved through the dual closed-loop control of the converter. That is, the detection signal reference value is superimposed on the modulation wave to generate a comprehensive modulation wave that is asynchronous with the system power frequency to drive the converter. This simplifies the control logic and filters out the influence of the detection frequency component on the phase-locked loop, ensuring that the amplitude of the power frequency fault current is limited while ensuring the stable and distortion-free injection of the high-frequency detection signal and avoiding system oscillation. It should be noted that a bandpass filter of the corresponding frequency can be added at the output of the converter that performs the detection signal injection to filter out the high-frequency noise generated during the injection process. At the same time, a 50Hz notch filter is added at the signal acquisition end of each sensor to eliminate the influence of the power frequency component on the analysis of the detection signal. Through the above hardware-level processing, it is ensured that the signal-to-noise ratio of the finally extracted feature signal is not less than 30dB.

[0030] The dynamic detection signal generation mechanism provided in this embodiment can maintain a high signal-to-noise ratio in scenarios with high penetration of new energy sources, low load, and topology switching, solving the adaptation problem of complex operating conditions in "three-high" distribution networks. It does not require additional signal injection equipment and can realize detection signal injection using only the existing converter of the system, which can effectively reduce the cost of engineering applications.

[0031] To avoid the limitations of relying solely on electrical parameters for fault analysis and protection delay when fault characteristics are weakened, this embodiment preferably incorporates the analysis of line physical state and signal injection timing changes in addition to electrical parameters. Specifically, the detection signal response characteristics include post-injection electrical characteristics, post-injection line physical state characteristics, and injection process timing characteristics. Post-injection electrical characteristics can be understood as the equivalent injection impedance calculated from the instantaneous voltage and current values ​​collected on the fault line bus side after the detection signal injection (this can be obtained by performing discrete Fourier transform or wavelet transform on the voltage and current instantaneous value sequences to extract voltage and current phasors with the same frequency as the detection signal, and dividing the voltage phasor by the current phasor to obtain the equivalent injection impedance), and transient response parameters of the renewable energy inverter (such as inverter output current distortion rate and DC-side voltage fluctuation). This addresses the problem of indistinct fault characteristics caused by weak renewable energy feedwater, and also considers the inverter's transient response parameters. The output current distortion rate is the ratio of the effective value of the current harmonics to the effective value of the fundamental current within the frequency band of the detection signal. The DC-side voltage fluctuation is the percentage of the peak-to-peak value of the DC-side voltage to the rated DC voltage during the fault transient period. The physical state characteristics of the line after injection include the line vibration frequency and amplitude of the faulty line. These vibration frequencies and amplitudes can be obtained by analyzing the Rayleigh scattering light signal acquired by the fiber optic vibration sensor installed on the faulty line tower using a phase-generated carrier demodulation algorithm, and then analyzing the reconstructed low-frequency vibration time-domain waveform reflecting the microscopic deformation of the faulty line. The timing characteristics of the detection signal response during the injection process can include the rate of change of voltage drop and the rate of change of current distortion over three power frequency cycles in the three stages before, during, and after the injection. This can be obtained by using the injection time of the detection signal as the zero point and employing a sliding time window technique to calculate the derivatives of the voltage drop rate and current distortion rate within adjacent cycles. It should be noted that the acquisition of each characteristic data in the above-mentioned detection signal response characteristics can be achieved with reference to relevant existing technologies, and will not be detailed here.

[0032] This embodiment adopts a real-time operating status based on the output of new energy sources and the load of lines within the target distribution network. It adaptively switches between wideband sweep frequency signals, pulse signals, or single-frequency continuous wave signals to achieve dynamic matching between the detection signal and the operating conditions of the distribution network. This enhances the anti-interference capability and fault feature extraction effect of the fault detection signal under different scenarios such as high new energy penetration, light load, and heavy load, improves the accuracy, reliability, and environmental adaptability of distribution network fault detection, and reduces the impact of detection signal injection on power grid operation. It has strong engineering practicality and reliability.

[0033] S13. Perform fault analysis based on the detection signal response characteristics and the current distribution network operating parameters to obtain corresponding fault information; the fault information includes fault distance and fault level; wherein, fault analysis can be understood as performing cross-validation calculations based on the detection signal response characteristics, including post-injection electrical characteristics, post-injection line physical state characteristics, and injection process timing characteristics, to complete the accurate analysis process of fault level and fault distance; specifically, the step of performing fault analysis based on the detection signal response characteristics and the current distribution network operating parameters to obtain corresponding fault information includes: The injected electrical characteristics, the injected line physical state characteristics, and the injection process timing characteristics are respectively subjected to feature anomaly detection to obtain corresponding feature detection results. These results may include electrical feature detection results, physical feature detection results, and timing feature detection results. The acquisition process may include: detecting feature anomalies in the injected electrical characteristics by comparing each electrical characteristic with a corresponding preset steady-state reference value. For example, if the equivalent injected impedance of the line decreases by more than a preset decrease percentage (e.g., 20%) compared to the corresponding steady-state impedance reference value, or if the inverter output current distortion rate jumps by more than a predicted jump threshold (e.g., 5%), then the electrical characteristic is determined to be abnormal, and the electrical feature detection result is set as abnormal. Conversely, if no electrical feature anomalies are found, the electrical feature detection result is set as normal. Detecting feature anomalies in the injected line physical state characteristics can be performed based on a preset normal micro-... The safe frequency range and safe amplitude upper limit under wind vibration and mechanical resonance are used to detect anomalies in the real-time acquired line vibration frequency and line vibration amplitude. For example, if the line vibration amplitude exceeds the safe amplitude upper limit for three consecutive sampling periods, or if the line vibration frequency shows anomalies in the infrasound band outside the safe frequency range, it indicates that the line is experiencing abnormal galloping, strand breakage, or external force damage, etc., and its physical state characteristics are determined to be abnormal, and the physical characteristic detection result is set as abnormal. Conversely, if there are no abnormal physical state characteristics, the physical characteristic detection result is set as normal. The time sequence characteristics of the injection process are used to detect anomalies by judging whether the rate of change of voltage drop rate or the rate of change of current distortion rate exceeds the extreme envelope generated by the maximum allowable load sudden switching of the target distribution network. If it exceeds, it is determined to be a time sequence characteristic anomaly, and the time sequence characteristic detection result is set as abnormal. Conversely, if there are no abnormal time sequence characteristics, the time sequence characteristic detection result is set as normal.

[0034] Based on the feature detection results and the current distribution network operating parameters, feature fusion analysis is performed on the injected electrical features, the injected line physical state features, and the injection process timing features to obtain the corresponding fusion feature strength. The feature fusion analysis can be understood as a weighted fusion of the comprehensive features corresponding to the three types of features: injected electrical features, injected line physical state features, and injection process timing features. Specifically, the step of performing feature fusion analysis on the injected electrical features, injected line physical state features, and injection process timing features based on the feature detection results and the current distribution network operating parameters to obtain the corresponding fusion feature strength includes: The post-injection electrical characteristics, post-injection line physical state characteristics, and injection process timing characteristics are respectively subjected to comprehensive feature extraction to obtain corresponding comprehensive electrical features, comprehensive physical state characteristics, and comprehensive timing characteristics. Comprehensive feature extraction can be understood as performing intra-class normalization and weighting on various features such as post-injection electrical characteristics, post-injection line physical state characteristics, and injection process timing characteristics to obtain comprehensive feature indices that represent these features. This addresses the inconsistency in multi-source feature dimensions and provides a reliable data foundation for subsequent multi-source feature fusion analysis. For example, in practical applications, if the post-injection electrical characteristics include equivalent injection impedance and inverter output current distortion rate, the process of obtaining the corresponding comprehensive electrical characteristics may include: allocating weight coefficients for equivalent injection impedance and inverter output current distortion rate based on principal component analysis (PCA); first, normalizing the equivalent injection impedance and inverter output current distortion rate using the Min-Max normalization method to obtain corresponding dimensionless values; and then weighting and summing the normalized dimensionless values ​​based on the corresponding weight coefficients to obtain the required comprehensive electrical characteristics. The acquisition of the corresponding physical state comprehensive characteristics and timing comprehensive characteristics can also be achieved by referring to the above description of the acquisition of electrical comprehensive characteristics, which will not be elaborated here.

[0035] Based on the current distribution network operating parameters and the physical feature detection results in the feature detection results, a feature weight vector corresponding to the detection signal response feature is determined. The feature weight vector can be understood as a vector that simultaneously includes the fused weight values ​​corresponding to electrical comprehensive features, physical state comprehensive features, and time-series comprehensive features. In principle, a fixed weight vector can be set based on experience. However, considering the differences in the contribution of various detection signal response features to fault analysis under different distribution network operating conditions, in order to focus on the most discriminative features under different distribution network operating conditions and to maximize the comprehensiveness and reliability of fault analysis, this embodiment preferably dynamically allocates the feature weight vector by combining the distribution network's new energy output conditions and the line physical state detection results. This is used for adaptive weighted fusion of the detection signal response features. Specifically, the step of determining the feature weight vector corresponding to the detection signal response feature based on the current distribution network operating parameters and the physical feature detection results in the feature detection results includes: When the proportion of renewable energy output in the current distribution network operating parameters is not greater than the preset output threshold and the physical characteristic detection result is normal, the feature weight vector is set as the basic weight vector. The description of the preset output threshold is given above and will not be repeated here. In practical applications, if the proportion of renewable energy output in the current distribution network operating parameters is less than or equal to the preset output threshold and the physical characteristic detection result is normal, it is considered that the renewable energy fluctuation is small, the electrical characteristic reliability is high, and the line physical condition is good, thus determining it as a normal operating condition. In this case, the pre-set basic weight vector is directly used as the required feature weight vector. This not only reduces the computational overhead caused by dynamic weight adjustment and ensures the efficiency and reliability of fault analysis under normal operating conditions, but also avoids excessive dominance of certain features, making the fault judgment basis comprehensive and stable. It should be noted that the basic weight vector can be set according to actual application requirements. In this example, the fusion weight values ​​corresponding to the electrical comprehensive feature, physical state comprehensive feature, and time-series comprehensive feature are preferably set to 0.6, 0.2, and 0.2 respectively.

[0036] When the proportion of renewable energy output in the current distribution network operating parameters is greater than a preset output threshold and the physical characteristic detection result is normal, the feature weight vector is set based on a preset lower limit of time-series feature weight and a preset upper limit of electrical feature weight. Both the preset lower limit of time-series feature weight and the preset upper limit of electrical feature weight can be adjusted based on actual needs. In this embodiment, the preset lower limit of time-series feature weight is preferably set to 0.5, and the preset upper limit of electrical feature weight is set to 0.2. In practical applications, if the proportion of renewable energy output in the current distribution network operating parameters is detected to be greater than the preset output threshold and the physical characteristic detection result is normal, it is considered that weak renewable energy feed will cause electrical characteristic distortion. The fusion weight value corresponding to the electrical comprehensive feature should be reduced, and the reduced weight should be increased to the fusion weight value corresponding to the time-series comprehensive feature. At this time, the fusion weight value corresponding to the physical state comprehensive feature remains unchanged. It should be noted that the dynamic reduction formula for the fusion weight value corresponding to the electrical comprehensive feature in this embodiment is expressed as: In the formula, and These represent the fusion weight values ​​of the electrical integrated characteristics before and after dynamic reduction, respectively; This is the attenuation adjustment coefficient, which can be adjusted according to actual application requirements; This represents the proportion of new energy output in the current distribution network operating parameters, and ; The preset output threshold is preferably set to 50%.

[0037] When the physical feature detection result is abnormal, the feature weight vector is set based on the preset lower limit of the physical state feature weight. The preset lower limit of the physical state feature weight can be adjusted based on actual needs; in this embodiment, it is preferably set to 0.5. In practical applications, a physical state veto mechanism is introduced. When the physical feature detection result is abnormal, it can be considered that the corresponding line has an early hidden danger, requiring mandatory intervention and adjustment. The fusion weight value of the physical state comprehensive feature is adjusted to be greater than or equal to the preset lower limit of the physical state feature weight, thereby cutting off the fault evolution at the physical source in advance, achieving accurate anti-interference judgment, and improving fault analysis capabilities. It should be noted that the weight increase of the fusion weight value of the physical state comprehensive feature is equivalent to the total weight reduction of the other two fusion weight values, and needs to be proportionally allocated to the reduction of the fusion weight values ​​corresponding to the electrical comprehensive feature and the timing comprehensive feature. The specific allocation ratio can be determined based on the ratio of the other two fusion weight values, which will not be detailed here.

[0038] This embodiment adaptively sets the weight vector of the detection signal response features based on the power output conditions of new energy sources in the distribution network and the detection results of the physical status of the lines. It can ensure balanced and reliable fault judgment under normal operating conditions, enhance the contribution of time-series features to compensate for electrical feature distortion in scenarios with a high proportion of new energy access, and highlight the weight of fault precursor features when the physical status of the lines is abnormal. This effectively improves the accuracy, stability and adaptability of multi-source feature fusion analysis under complex operating conditions, and is especially beneficial for improving the fault analysis effect under weak faults, hidden faults and new energy fluctuation scenarios.

[0039] The electrical comprehensive feature, the physical state comprehensive feature, and the time sequence comprehensive feature are weighted and fused according to the feature weight vector to obtain the fused feature strength. The fused feature strength can be understood as a multi-dimensional comprehensive fault feature strength quantification index, which can be used for the quantitative determination of fault level. It can perform multi-dimensional comprehensive analysis from electrical features, line physical state features, and injection time sequence features, thereby improving the accuracy of fault analysis and high-resistance fault identification and reducing the fault missed rate.

[0040] The fault level is identified based on the intensity of the fused feature and the abnormal feature type and number corresponding to the feature detection results. The fault level classification can be determined based on actual application requirements. In practical applications, the fault level acquisition process may include: when only the electrical feature detection result is abnormal and the fused feature intensity is less than the first intensity threshold, the fault level is set to Class I fault (minor fault); when two features are abnormal and the fused feature intensity is greater than or equal to the first intensity threshold and less than or equal to the second intensity threshold, the fault level is set to Class II fault (general fault); when three features are abnormal and the fused feature intensity is greater than the second intensity threshold, the fault level is set to Class III fault (serious fault). It should be noted that both the first and second intensity thresholds can be set according to actual application requirements. In this embodiment, the first intensity threshold is preferably set to 30% and the second intensity threshold to 60%, and the required fault level is obtained based on this and the above classification criteria, for subsequent adjustment of protection action delay.

[0041] The fault distance is obtained by measuring the fault distance based on the signal frequency of the detection signal and the fault transient waveform data and line topology switching status data in the current distribution network operating parameters; specifically, the step of measuring the fault distance based on the signal frequency of the detection signal and the fault transient waveform data and line topology switching status data in the current distribution network operating parameters includes: Based on the bus-side voltage waveform data and line-side current waveform data in the fault transient waveform data, the steady-state phasor at the signal frequency is obtained using Discrete Fourier Transform (DFT), resulting in the probe voltage phasor and probe current phasor. The bus-side voltage waveform data and line-side current waveform data can be understood as discrete-time sequences formed by synchronously acquiring the instantaneous values ​​of voltage and current for one or more integer periods after the probe signal has stabilized, using a bus-side voltage sensor and a line-side current sensor at a fixed sampling frequency, respectively. Performing DFT on the bus-side voltage waveform data and line-side current waveform data respectively yields the probe voltage phasor and probe current phasor at the signal frequency. The specific process of extracting the steady-state phasor using DFT can be referenced from existing technologies. Under single-phase ground fault conditions, the probe voltage phasor... For the voltage phasor of the faulty phase, probe the current phasor. The injected frequency current phasor is the current phasor flowing out of the faulty line.

[0042] Based on the line topology switching status data, the line parameters of the currently faulty line are obtained, and based on the line parameters, the detection unit length loop impedance of the currently faulty line at the signal frequency is obtained. The line topology switching status data can be understood as data describing the network connectivity of the current target distribution network. Based on this, the line parameters of the currently faulty line (including unit length loop resistance and unit length loop inductance) can be retrieved from the topology database of the target distribution network, and combined with the angular frequency of the detection signal. Taking into account both the phase conductor parameters and the ground return current characteristics, the detection loop impedance per unit length of the currently faulty line at the signal frequency is calculated. , can be represented as: In the formula, in, and These are the signal frequency and the corresponding angular frequency of the probe signal, respectively. The imaginary unit; and These are the unit length loop resistance and loop inductance at the probe signal frequency, respectively, and their values ​​are retrieved from a pre-stored line parameter library based on the real-time topology.

[0043] Based on the detection unit length loop impedance, the detection voltage phasor, and the detection current phasor, a corresponding fault location equation is constructed; wherein, the fault location equation is the detection frequency loop equation for a single-phase ground fault, which can be expressed as: In the formula, The probe voltage phasor at the frequency of the signal monitored on the bus side; The probe current phasor at the signal frequency monitored on the line side; The distance to the fault is to be determined. The probe current phasor at the signal frequency flowing through the fault point; For ground fault transition resistance; For signal frequency The impedance of the detection unit length loop.

[0044] The fault location equation is solved to obtain the fault distance. In practical applications, considering that in high-voltage power distribution networks, distributed power sources connected through converters will generate boosted currents at injected frequencies, leading to increased current at line-side monitoring points... and the point of failure Due to phase deflection, this embodiment preferably employs the method of extracting the imaginary part (pure reactance component) to eliminate the ground fault transition resistance. The effect of the phase of the boost current: Divide both sides of the above fault location equation by the phasor of the detection current at the line-side monitoring point. The equivalent injection impedance is obtained. : Then take the imaginary part of both sides of the above equation: In the formula, This is an operation to extract the imaginary part (pure reactance component).

[0045] Because the frequency of the actively injected detection signal (such as 235Hz or a wideband signal) is much higher than the power frequency, and in non-power frequency bands, the ground fault transition resistance is... It exhibits purely resistive characteristics, and the inductive reactance component excited in the loop by the injected detection signal is much greater than the resistive component; at this time, the imaginary part of the second term on the rightmost side of the above equation regarding the ground fault transition resistance and current deviation approaches zero. Therefore, the fault distance... The exact result can be obtained through the following formula: The fault location equation provided in this embodiment is constructed entirely based on the signal frequency of the actively injected detection signal, and the method of extracting the pure reactance component to accurately calculate the fault distance is adopted. It can effectively eliminate the interference of high-resistance transition resistance and distributed power supply boosting current on the location accuracy in single-phase grounding faults by utilizing the physical characteristics of inductive reactance dominance under high-frequency detection signals. It overcomes the application defects of traditional power frequency location measurement which are greatly affected by transition resistance and multi-terminal weak feeder power supply boosting current, thereby effectively reducing the location error.

[0046] S14. Perform protection delay analysis based on the fault level, the fault distance, and the downstream load priority of the protection device to obtain the protection action delay. If the preset self-healing monitoring time is exceeded and the fault has not self-healed, trigger the corresponding circuit breaker to perform the protection action according to the protection action delay. The downstream load priority of the protection device includes the priority of all downstream feeder loads corresponding to the protection device that needs to perform the protection action. It can be obtained based on a pre-built load priority database. The load priority database stores the load priority classification information corresponding to industrial loads, residential loads, and important users (such as hospitals, data centers, and other users with extremely high power supply reliability requirements). It can support remote updates and local real-time calls.

[0047] In this embodiment, the protection action delay can be understood as the duration for controlling the delayed execution of protection actions; specifically, the step of performing protection delay analysis based on the fault level, the fault distance, and the downstream load priority of the protection equipment to obtain the protection action delay includes: Based on the fault distance, the corresponding electrical distance base delay is obtained based on the preset inverse time protection action characteristic curve; wherein, the process of obtaining the preset inverse time protection action characteristic curve includes: obtaining the total length of the current protected line, the preset shortest base action time and the preset longest base action time corresponding to the fault in the zone; establishing a mapping model between fault distance and base action time based on a power function to obtain the preset inverse time protection action characteristic curve.

[0048] Specifically, if the fault distance is less than or equal to the total length of the line within the zone (i.e., Then, the analytical expression of the mapping relationship curve is: In the formula, The fault distance is the one calculated above; To protect the entire length of the line; and These are the preset minimum basic action time (rapid response to near-end faults) and the maximum basic action time (coordination to far-end faults), both of which can be set based on actual application requirements; The preset inverse time-limit curvature coefficient (usually 1 or 2); The calculated electrical distance base delay is shown by the formula, which shows that the base action time decreases as the fault distance shortens.

[0049] If the fault distance is greater than the total length of the line, then it is an out-of-area fault (i.e.) If the action time exceeds the coordination threshold, a locking inaction command will be output directly.

[0050] Based on the highest load priority among the downstream load priorities of the protection equipment, the corresponding load priority correction coefficient is obtained. The highest load priority can be understood as the highest priority among all downstream feeders in the downstream load priority of the protection equipment. In practical applications, load priorities from highest to lowest are: important users, residential loads, and industrial loads. When multiple types of mixed loads exist on the same downstream feeder of the protection equipment, the highest load priority is determined according to the highest priority load type. After determining the highest load priority, the required load priority correction coefficient is obtained based on a pre-built mapping table between load priorities and correction coefficients. The mapping table between load priorities and correction coefficients can be constructed based on actual application requirements. For example, the load priority correction coefficient for important user load priorities can be set to 0.7 (accelerated disconnection), the load priority correction coefficient for residential user load priorities can be set to 1.0 (normal disconnection), and the load priority correction coefficient for industrial user load priorities can be set to 1.15 (allowing short-term withstand).

[0051] Based on the fault level, a fault level correction coefficient is determined, and the electrical distance base delay is corrected according to the fault level correction coefficient and the load priority correction coefficient to obtain the protection action delay. The fault level correction coefficient can be obtained by querying a pre-built mapping table between fault levels and correction coefficients. This mapping table can also be constructed based on actual application requirements; for example, the fault level correction coefficient for Class I faults can be set to 1.5 (delay to prevent false tripping), the fault level correction coefficient for Class II faults can be set to 1 (normal disconnection), and the fault level correction coefficient for Class III faults can be set to 0.6 (fast disconnection). The final protection action delay can be expressed as: in, For electrical distance basis delay; and These are the load priority correction factor and the fault level correction factor, respectively. To protect the action delay.

[0052] This embodiment combines the three dimensions of fault distance, fault level, and downstream load priority to jointly set the protection action delay, thereby achieving adaptive delay optimization of distribution network protection. While ensuring rapid fault clearing, it also takes into account protection selectivity, power supply continuity, and power supply reliability of important loads. It effectively solves the problems of traditional fixed delay or simple inverse time delay in complex distribution networks, such as easy false tripping, failure to tripping, and long power outage time of important loads.

[0053] In practical applications, during the aforementioned fault analysis and protection action delay calculation phase (e.g., 120ms~200ms after the fault), the fault self-healing status is still being monitored synchronously (assuming a preset self-healing monitoring duration of 100ms): if the fault self-heals, the current limiting is immediately lifted, the probe signal injection is stopped, and the system resumes normal operation control mode (e.g., MPPT control, dual closed-loop control); furthermore, when the aforementioned fault analysis and protection action delay calculation phase is completed, the preset self-healing monitoring duration has actually been exceeded (assuming a preset self-healing monitoring duration of 100ms). At this time, action decisions will be made based on the monitoring results: if the fault has self-healed, the current limiting is immediately lifted, the probe signal injection is stopped, and the system resumes normal operation control mode (e.g., MPPT control, dual closed-loop control); if it is confirmed that the preset self-healing monitoring duration has been exceeded and the fault has not self-healed, then based on the obtained protection action delay, the corresponding circuit breaker is selectively controlled to trip and disconnect the faulty line based on the fault type (intra-zone fault) to avoid power outages in non-faulty sections. After the fault is cleared, the probe signal injection is stopped, and normal operation is restored.

[0054] When a suspected grounding fault is detected in a target distribution network, the present invention provides a method for initiating a preset protection action criterion based on the acquired fault transient data. After the preset protection action criterion is initiated, a gradient current limiting strategy is executed and fault self-healing monitoring is activated. After entering the current limiting steady-state period, a detection signal generated based on the current distribution network operating parameters is actively injected into the target distribution network. Simultaneously, detection signal response characteristics, including electrical characteristics after injection, physical state characteristics of the lines after injection, and timing characteristics of the injection process, are acquired. Fault analysis is performed based on the detection signal response characteristics and the current distribution network operating parameters to obtain fault information including fault distance and fault level, as well as fault information based on the fault level, fault distance, and protection equipment. The technical solution involves analyzing the protection delay based on the load priority to obtain the protection action delay. If the preset self-healing monitoring time is exceeded and the fault has not self-healed, the corresponding circuit breaker is triggered to perform the protection action according to the protection action delay. By adopting a collaborative protection mechanism that uses adaptive generation of detection signal injection based on distribution network operation status data, fault analysis based on line physical state characteristics and signal injection timing characteristics, and protection delay design based on fault level, fault distance, and downstream load priority of protection equipment, the scientific rationality of low current grounding fault protection in scenarios such as weak new energy feed and weak fault characteristics can be improved. This effectively balances protection speed and power supply reliability, thereby improving the safe and stable operation level of the distribution network.

[0055] It should be noted that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise explicitly stated in this document, there is no strict order requirement for the execution of these steps, and they can be executed in other orders.

[0056] In one embodiment, such as Figure 2 As shown, a low-current single-phase ground fault protection system for a power distribution network is provided, the system comprising: Fault monitoring module 1 is used to initiate preset protection action criteria based on the acquired fault transient data when a suspected grounding fault is detected in the target distribution network. The signal injection module 2 is used to respond to the preset protection action criterion, execute the gradient current limiting strategy and start fault self-healing monitoring, and actively inject the detection signal generated based on the current distribution network operating parameters into the target distribution network after entering the current limiting steady state period, and simultaneously acquire the detection signal response characteristics; the detection signal response characteristics include the electrical characteristics after injection, the physical state characteristics of the line after injection and the timing characteristics of the injection process; Fault analysis module 3 is used to perform fault analysis based on the detection signal response characteristics and the current power distribution network operating parameters to obtain corresponding fault information; the fault information includes fault distance and fault level; The protection delay module 4 is used to perform protection delay analysis based on the fault level, the fault distance and the downstream load priority of the protection device to obtain the protection action delay, and when the preset self-healing monitoring time is exceeded and the fault has not self-healed, the corresponding circuit breaker is triggered to perform the protection action according to the protection action delay.

[0057] Specific limitations regarding the protection system for single-phase ground faults with low current in distribution networks can be found in the limitations of the protection method for single-phase ground faults with low current in distribution networks described above. The corresponding technical effects are equivalent and will not be repeated here. Each module in the aforementioned protection system for single-phase ground faults with low current in distribution networks can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0058] Figure 3 An internal structural diagram of a computer device is shown in one embodiment. This computer device may specifically be a terminal or a server. Figure 3As shown, the computer device includes a processor, memory, network interface, display, camera, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it can implement a method for protecting against small-current single-phase grounding faults in a power distribution network. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0059] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. Specific computing devices may include more or fewer components than those shown in the figure, or combine certain components, or have the same component arrangement.

[0060] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0061] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0062] In summary, the present invention provides a method, system, computer equipment, and storage medium for protecting small-current single-phase grounding faults in distribution networks. By employing a collaborative protection mechanism that adaptively generates detection signal injection based on distribution network operating status data, performs fault analysis based on line physical state characteristics and signal injection timing characteristics, and designs protection delays based on fault level, fault distance, and downstream load priority of protection equipment, the invention can improve the scientific rationality of small-current grounding fault protection in scenarios such as weak feeder power from new energy sources and weak fault characteristics. It effectively balances protection speed and power supply reliability, thereby improving the safe and stable operation level of the distribution network.

[0063] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0064] The above-described embodiments are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for protecting against low-current single-phase grounding faults in a power distribution network, characterized in that, The method includes: When a suspected grounding fault is detected in the target distribution network, the preset protection action criteria are activated based on the acquired fault transient data. In response to the activation of the preset protection action criterion, a gradient current limiting strategy is executed and fault self-healing monitoring is initiated. After entering the current limiting steady state period, a detection signal generated based on the current distribution network operating parameters is actively injected into the target distribution network, and the detection signal response characteristics are acquired simultaneously. The detection signal response characteristics include the electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process. Fault analysis is performed based on the detection signal response characteristics and the current power distribution network operating parameters to obtain corresponding fault information; the fault information includes fault distance and fault level. Based on the fault level, the fault distance and the downstream load priority of the protection device, a protection delay analysis is performed to obtain the protection action delay. If the preset self-healing monitoring time is exceeded and the fault has not healed itself, the corresponding circuit breaker is triggered to perform the protection action according to the protection action delay. The steps for generating detection signals based on current power distribution network operating parameters include: When the load data of the faulty line in the current distribution network operating parameters is less than the preset load threshold, a pulse signal with a preset duty cycle range is generated. When the load data of the faulty line is not less than the preset load threshold and the proportion of new energy output in the current distribution network operation parameters is greater than the preset output threshold, a wideband sweep frequency signal with a preset frequency range is generated. When the load data of the faulty line is not less than the preset load threshold and the output ratio of the new energy source is not greater than the preset output threshold, a single-frequency continuous wave signal is generated.

2. The method for protecting against single-phase grounding faults with low current in a distribution network as described in claim 1, characterized in that, The step of performing fault analysis based on the detection signal response characteristics and the current distribution network operating parameters to obtain the corresponding fault information includes: Feature anomaly detection is performed on the electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process, respectively, to obtain corresponding feature detection results; the feature detection results include electrical feature detection results, physical feature detection results, and timing feature detection results; Based on the feature detection results and the current power distribution network operating parameters, feature fusion analysis is performed on the injected electrical features, the injected line physical state features, and the injection process timing features to obtain the corresponding fusion feature strengths. The fault level is determined by identifying the fault level based on the intensity of the fused feature and the abnormal feature type and number of abnormal features corresponding to the feature detection results. The fault distance is obtained by measuring the fault distance based on the signal frequency of the detection signal and the fault transient waveform data and line topology switching status data in the current distribution network operating parameters.

3. The method for protecting against single-phase grounding faults with low current in a distribution network as described in claim 2, characterized in that, The step of performing feature fusion analysis on the injected electrical features, the injected line physical state features, and the injection process timing features based on the feature detection results and the current distribution network operating parameters to obtain the corresponding fusion feature strength includes: The electrical characteristics after injection, the physical state characteristics of the line after injection, and the timing characteristics of the injection process are extracted to obtain the corresponding electrical comprehensive characteristics, physical state comprehensive characteristics, and timing comprehensive characteristics. Based on the current power distribution network operating parameters and the physical feature detection results in the feature detection results, determine the feature weight vector corresponding to the detection signal response feature; The electrical integrated feature, the physical state integrated feature, and the temporal integrated feature are weighted and fused according to the feature weight vector to obtain the fused feature intensity.

4. The method for protecting a single-phase ground fault in a distribution network with low current as described in claim 3, characterized in that, The step of determining the feature weight vector corresponding to the detection signal response feature based on the current distribution network operating parameters and the physical feature detection results in the feature detection results includes: When the proportion of new energy output in the current distribution network operation parameters is not greater than the preset output threshold and the physical feature detection result is normal, the feature weight vector is set as the basic weight vector. When the proportion of new energy output in the current power distribution network operation parameters is greater than the preset output threshold and the physical feature detection result is normal, the feature weight vector is set based on the preset time-series feature weight lower limit and the preset electrical feature weight upper limit; When the physical feature detection result is abnormal, the feature weight vector is set based on the preset physical state feature weight lower limit.

5. The method for protecting against single-phase grounding faults with low current in a distribution network as described in claim 2, characterized in that, The step of determining the fault distance based on the signal frequency of the detection signal and the fault transient waveform data and line topology switching status data in the current distribution network operating parameters includes: Based on the bus-side voltage waveform data and line-side current waveform data in the fault transient waveform data, the steady-state phasor at the signal frequency is obtained based on the discrete Fourier transform, and the probe voltage phasor and probe current phasor are obtained. Based on the line topology switching status data, obtain the line parameters of the currently faulty line, and based on the line parameters, obtain the detection unit length loop impedance of the currently faulty line at the signal frequency. Based on the circuit impedance per unit length of the detection, the detection voltage phasor, and the detection current phasor, a corresponding fault location equation is constructed; The fault distance is obtained by solving the fault location equation.

6. The method for protecting a single-phase ground fault in a distribution network with low current as described in claim 1, characterized in that, The step of performing protection delay analysis based on the fault level, the fault distance, and the downstream load priority of the protection device to obtain the protection action delay includes: Based on the fault distance, the corresponding electrical distance base delay is obtained based on the preset inverse time protection action characteristic curve; The corresponding load priority correction coefficient is obtained based on the highest load priority among the downstream load priorities of the protection device; Based on the fault level, a fault level correction coefficient is determined, and based on the fault level correction coefficient and the load priority correction coefficient, the electrical distance base delay is corrected to obtain the protection action delay.

7. A low-current single-phase grounding fault protection system for power distribution networks, characterized in that, The system, employing the low-current single-phase grounding fault protection method for distribution networks as described in claim 1, comprises: The fault monitoring module is used to initiate preset protection action criteria based on the acquired fault transient data when a suspected grounding fault is detected in the target distribution network. The signal injection module is used to respond to the preset protection action criterion, execute the gradient current limiting strategy and start fault self-healing monitoring, and actively inject the detection signal generated based on the current distribution network operating parameters into the target distribution network after entering the current limiting steady state period, and simultaneously acquire the detection signal response characteristics; the detection signal response characteristics include the electrical characteristics after injection, the physical state characteristics of the line after injection and the timing characteristics of the injection process; The fault analysis module is used to perform fault analysis based on the response characteristics of the detection signal and the current power distribution network operating parameters to obtain corresponding fault information; the fault information includes fault distance and fault level. The protection delay module is used to perform protection delay analysis based on the fault level, the fault distance and the downstream load priority of the protection device to obtain the protection action delay, and to trigger the corresponding circuit breaker to perform protection action when the preset self-healing monitoring time is exceeded and the fault has not self-healed.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images