Single-phase earth fault protection method for distribution network with distributed power sources

Abstract

The application belongs to the field of power distribution network relay protection and fault monitoring, and relates to a single-phase grounding fault protection method for a power distribution network containing a distributed power source. The method comprises the following steps: collecting bus voltage, feeder and branch current and DG operation information; calculating zero sequence voltage, zero sequence current, abrupt variable and transient energy, and obtaining an initial criterion through normalized weighted fusion; constructing a disturbance influence quantity correction criterion according to the DG operation state, output and access position; selecting the maximum after correction as a candidate, and using the criterion to distinguish the margin and branch space consistency check; constructing fault evidence and disturbance evidence cumulative quantity, and controlling the conversion and outputting the action signal through a state machine containing normal monitoring, disturbance blocking, fault starting, confirmation, action keeping, clearing confirmation and reset permission. The method reduces non-fault disturbance interference, reduces misselection, adapts to intermittent faults, and forms a continuous and stable protection process.

Description

Technical Field

[0001] This invention relates to the field of power distribution network relay protection and fault monitoring technology, specifically a method for protecting a single-phase grounding fault in a power distribution network containing distributed power sources. Background Technology

[0002] Rural power distribution lines in mountainous areas are typically characterized by long lines, many branches, dispersed loads, complex operating environments, and significant impact from tree obstructions and lightning strikes. Grounding faults are prone to exhibiting characteristics such as large transition resistance, weak fault current, and intermittent arcing process.

[0003] When a single-phase ground fault occurs in a low-current grounding system, the fault current amplitude is usually small, and the arc at the fault point may repeatedly extinguish and reignite, resulting in the zero-sequence voltage, zero-sequence current, and their sudden changes being characterized by weak amplitude, short duration, and unstable waveform. Existing protection and fault selection methods mainly determine whether a ground fault has occurred in the system and select the faulty feeder by means of zero-sequence voltage initiation, zero-sequence current amplitude comparison, phase comparison, transient energy comparison, and time-frequency characteristic analysis.

[0004] With the integration of new grid-connected entities such as distributed photovoltaics, energy storage, and micro gas turbines into rural power distribution networks, the power flow direction, current distribution, and transient response of feeders have become increasingly complex. Distributed generation (DG) connections to different feeders, as well as non-faulty operational processes such as grid connection, disconnection, or output changes, can all cause sudden changes in feeder current and short-term transient energy variations. This leads to problems such as false activation, unstable judgment, or temporary suppression of real faults in protection methods that rely solely on a single zero-sequence quantity or fixed threshold.

[0005] Most current research focuses on improving the accuracy of low-current grounding fault identification or fault feeder selection. However, the actual protection process still requires continuous processing, including non-fault disturbance identification, protection initiation, fault confirmation, operation maintenance, fault clearance confirmation, and reset permission. Without corresponding evidence accumulation and state constraints, even if the fault feeder judgment result can be obtained, problems such as protection false initiation, repeated resets, or insufficient operation maintenance may occur due to distributed power source operation disturbances, intermittent arcing, or instantaneous criterion fluctuations, making it difficult to form a stable and continuous protection operation process.

[0006] The existing technology still has the following shortcomings: (1) The transient characteristics of faults and the non-fault disturbance characteristics of distributed generation are easily confused. The zero-sequence current, zero-sequence current mutation and transient energy of low-current grounding faults are usually weak. When the transition resistance of the grounding point increases, the fault characteristics will be further reduced. However, the grid connection, disconnection or output step of distributed generation may also cause short-term changes in feeder current and transient quantities. If the existing method lowers the start-up threshold, it is easy to misjudge the operation changes of distributed generation as grounding faults. If the start-up threshold is raised, it may cause weak faults to fail to start reliably. At the same time, the degree of influence of different distributed generation connection locations on each feeder is different. Existing methods usually do not use the operation status of distributed generation, output changes and connection location together for fault criterion correction, making it difficult to take into account both disturbance resistance and fault detection sensitivity.

[0007] (2) Intermittent arcing ground faults have highly variable characteristics. Under conditions such as tree obstruction, damp insulation, conductor drop, and arc discharge, single-phase ground faults may repeatedly experience arcing, extinction, and reignition processes, causing discontinuous changes in zero-sequence voltage, zero-sequence current, and transient energy. Traditional instantaneous threshold criteria tend to activate during the arcing stage and then revert during the short-term extinction stage, resulting in the inability to continuously retain fault evidence, repeated changes in the judgment of candidate fault feeders, and unstable protection actions, making it difficult to reliably confirm and continuously maintain intermittent faults composed of multiple short-term arcing stages.

[0008] (3) Insufficient multi-measurement point collaborative judgment capability for the entire protection process. Some existing methods mainly focus on grounding anomaly identification or fault feeder selection, and have not yet uniformly processed the zero-sequence voltage of the bus, the zero-sequence current of the feeder, the zero-sequence current mutation, and the zero-sequence transient energy. They also lack a mechanism to perform spatial consistency verification of candidate fault feeders by utilizing the transient response relationship between feeder outlet and branch measurement points. At the same time, the distributed power source disturbance identification, disturbance blocking, blocking exit, fault initiation, fault confirmation, action holding, fault clearing confirmation, and reset permission are usually independent of each other and have not formed a unified state transition process. Therefore, it is difficult to simultaneously ensure the stability of fault feeder judgment and the continuity of protection action process under complex operating conditions. Summary of the Invention

[0009] The purpose of this invention is to provide a single-phase grounding fault protection method for a distribution network containing distributed power sources, addressing the problems of weak fault current, difficulty in fault identification due to high transition resistance, discontinuous characteristics of intermittent arcing faults, and the easy influence of distributed power source operation changes on protection criteria in 10kV rural power distribution network low current grounding systems in mountainous and rural areas.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a method for protecting a single-phase ground fault in a distribution network containing distributed power sources, comprising the following steps: The system collects the three-phase voltage of the busbar, the three-phase current at each feeder outlet, the three-phase current at each branch measuring point, and the operating information of each distributed power source; the operating information includes the commissioning status, active power output, and connection location. The zero-sequence voltage of the busbar is calculated based on the three-phase voltage of the busbar. The zero-sequence current, zero-sequence current mutation and zero-sequence transient energy of each feeder are calculated based on the three-phase current at the feeder outlet. The zero-sequence voltage of the busbar, the zero-sequence current, zero-sequence current mutation and zero-sequence transient energy of each feeder are normalized and then weighted and fused according to preset weights to obtain the initial fault criterion quantity of each feeder. Based on the operating information and access location of each distributed power source, the comprehensive disturbance impact of distributed power sources on each feeder is constructed; the initial fault criterion of each feeder is corrected using the comprehensive disturbance impact of distributed power sources to obtain the disturbance-corrected fault criterion of each feeder; the feeder with the largest disturbance-corrected fault criterion is selected as the candidate fault feeder, and the criterion discrimination margin between the candidate fault feeder and the other feeders is calculated; and the spatial consistency index of the candidate fault feeder is calculated using the zero-sequence current deviation at the outlet of the candidate fault feeder and the zero-sequence current deviation at the relevant branch measurement points. The cumulative fault evidence is constructed based on the disturbance correction fault criterion quantity, criterion discrimination margin, and spatial consistency index of the candidate fault feeder. The cumulative distributed power source disturbance evidence is constructed based on the maximum value of the comprehensive disturbance impact of each feeder. A protection state machine is constructed, which includes normal monitoring state, disturbance blocking state, fault initiation state, fault confirmation state, action holding state, fault clearing confirmation state, and reset permission state. Based on the comparison results of the cumulative fault evidence quantity, the cumulative distributed power source disturbance evidence quantity, and the preset threshold, the protection state machine is controlled to perform state transitions and output protection action signals.

[0011] Furthermore, each feeder of the distribution network is equipped with a three-phase current measurement unit at its end to collect the three-phase current at the branch measurement point; the operation information of the distributed power source is obtained through the supporting control device or system monitoring interface of the distributed power source.

[0012] Furthermore, a limiting normalization function is used to map the zero-sequence voltage of the bus, the zero-sequence current of each feeder, the zero-sequence current mutation, and the zero-sequence transient energy to the interval [0,1] to obtain the corresponding normalized characteristic quantities; the normalized characteristic quantities are weighted and summed according to the preset weights to obtain the initial fault criterion quantity.

[0013] Furthermore, the method for constructing the comprehensive disturbance impact of the distributed power source includes: Calculate the first The degree of disturbance of each distributed power source: ; In the formula, express Time of the first The degree of disturbance of a distributed power source; express Time of the first The change in active power output of a distributed power source; Normalized reference value representing the change in active power output; express Time of the first The operational status value of a distributed power source; This represents the amplitude limiting normalization function; Within the preset disturbance hold time window The maximum value of the disturbance level is maintained to obtain the maintained disturbance amount: ; In the formula, express Time of the first The amount of disturbance to maintain the distributed power source; express Time of the first The degree of disturbance of a distributed power source; Represents a time variable; The sum of the holding disturbances of each distributed power source and their corresponding location influence coefficients is then calculated and normalized to obtain the comprehensive disturbance impact of the distributed power sources on the feeder: ; In the formula, express Time of the first The combined impact of distributed power source disturbances on each feeder; This represents the number of distributed power sources.

[0014] Furthermore, the calculation method for the disturbance correction fault criterion quantity is as follows: ; express Time of the first Disturbance correction fault criterion for each feeder; Indicates the disturbance correction factor; express Normalized zero-sequence voltage characteristics of the bus at any given time; The method for calculating the criterion discrimination margin is as follows: ; In the formula, express The criterion discrimination margin between the candidate faulty feeder and the other feeders at any given time; express The fault criterion quantity for candidate fault feeder disturbance correction at any time.

[0015] Furthermore, the method for calculating the spatial consistency index includes: Calculate the zero-sequence current deviation at the outlet of the candidate faulty feeder and the first Zero-sequence current deviation at each relevant branch measurement point; Within a preset transient time window, the distance between the candidate faulty feeder outlet and the first... The consistency of zero-sequence transient response among the relevant branch measurement points is the normalized cross-correlation value between the zero-sequence current deviation at the outlet of the candidate fault feeder and the zero-sequence current deviation at the relevant branch measurement points. The maximum value among the zero-sequence transient response consistency of all relevant branch measurement points corresponding to the candidate fault feeder is determined as the spatial consistency index.

[0016] Furthermore, the methods for constructing the accumulated amount of fault evidence and the accumulated amount of distributed power source disturbance evidence include: The cumulative amount of fault evidence is constructed using a recursive method: ; In the formula, express The cumulative amount of fault evidence at any given moment; Indicates the previous sampling time, i.e. The cumulative amount of fault evidence at any given moment; express The candidate faulty feeder number selected at any given time; Indicates the previous sampling time, i.e. The candidate faulty feeder number at any given time; Indicates the fault evidence memory coefficient; express The fault support quantity of the candidate fault feeder at any given time; when the candidate fault feeder changes, the calculation of the accumulated fault evidence quantity is reset; Based on the maximum value of the comprehensive disturbance impact of the feeder distributed power source, construct the cumulative amount of evidence for the distributed power source disturbance: ; In the formula, express The cumulative amount of evidence of distributed source disturbances at any given time; Indicates the previous sampling time, i.e. The cumulative amount of evidence of distributed source disturbances at any given time; Indicates the memory coefficient of perturbation evidence; express The maximum value of the combined disturbance impact of distributed power sources on each feeder at any given time.

[0017] Furthermore, the state transition conditions of the protection state machine include: when the accumulated amount of fault evidence reaches the fault initiation accumulation threshold and the disturbance correction fault criterion amount of the candidate fault feeder reaches the correction criterion initiation value, the system switches from the normal monitoring state to the fault initiation state; when the accumulated amount of distributed power source disturbance evidence reaches the disturbance blocking value and the accumulated amount of fault evidence does not reach the fault initiation accumulation threshold, the system switches from the normal monitoring state to the disturbance blocking state; in the fault initiation state, if the candidate fault feeder remains unchanged within the preset confirmation time window and the accumulated amount of fault evidence reaches the fault confirmation threshold, the system switches to the fault confirmation state; in the fault confirmation state, the current candidate fault feeder is latched as a fault feeder, and the system switches to the action holding state; in the action holding state, when the disturbance correction fault criterion amount of the fault feeder drops below the correction criterion release value, the accumulated amount of fault evidence drops below the fault evidence release value, and the bus zero-sequence voltage drops below the bus zero-sequence voltage release value, the system switches to the fault clearing confirmation state; in the fault clearing confirmation state, if the above conditions are continuously met for a preset time, the system switches to the reset permission state.

[0018] Furthermore, in the action holding state, the protection state machine continuously outputs protection action commands and maintains the latching results of the fault feeder; in the fault clearing confirmation state, if the fault characteristics reappear, the protection state machine returns to the action holding state and switches to the reset permission state after the conditions are met.

[0019] Furthermore, the protection action signals include at least one of the following: fault alarm signal, fault feeder indication signal, distributed power disturbance interlock signal, interlock exit signal, fault start signal, fault confirmation signal, protection action hold signal, fault clearing confirmation signal, and reset permission signal.

[0020] Compared with existing technologies, the present invention has the following advantages: Reduce the impact of non-fault disturbances on the criteria: By constructing a comprehensive disturbance impact quantity that includes the commissioning status, active power output and access location of distributed power sources, the initial fault criteria after fusion are differentiated and corrected. This can reduce the interference of non-fault disturbances on the fault criteria calculation during the grid connection, shutdown or output change of distributed power sources. Reduce the misselection of faulty feeders: By calculating the criterion discrimination margin between the candidate faulty feeder and the other feeders, and combining it with the consistency of the zero-sequence transient response between the candidate faulty feeder outlet and the relevant branch measurement points, it is helpful to reduce the possible misselection of faulty feeders in scenarios where multiple feeder criteria are close or there is local disturbance. To deal with intermittent grounding faults: By constructing the cumulative amount of fault evidence in a recursive manner and combining it with the state transition logic in the state machine, fault evidence can be retained during the short-term arc extinction phase of intermittent single-phase grounding faults, thus avoiding repeated activation or deactivation of the protection device due to discontinuous changes in fault characteristics. Forming a continuous protection action process: By constructing a state machine that includes normal monitoring, disturbance blocking, fault initiation, fault confirmation, action holding, fault clearing confirmation, and reset permission state, and controlling state transitions based on the cumulative amount of fault evidence and disturbance evidence, it is possible to achieve continuous processing of the entire process from disturbance identification, fault initiation and confirmation to action holding and reset. Comprehensive utilization of multi-source characteristics for fault diagnosis: By normalizing and weighting the zero-sequence voltage of the bus, the zero-sequence current of each feeder, the zero-sequence current mutation and the zero-sequence transient energy, the electrical characteristics at the system level and feeder level can be comprehensively utilized, providing a basis for subsequent disturbance correction and fault diagnosis. Adaptable to high-resistance grounding and multi-branch structures: By integrating multiple features, performing disturbance correction, implementing spatial consistency verification and evidence accumulation, it can adapt to the attenuation and instability of fault features under high transition resistance and multi-branch conditions, which helps to maintain the stability of fault feeder judgment. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall process of the present invention; Figure 2 This is a schematic diagram of the measurement point arrangement of the present invention; Figure 3 The timing diagram for the lockout exit and operation hold of ground faults under grid-connected disturbances of distributed power sources; Figure 4 This is a timing diagram of the accumulation of fault evidence and the operation maintenance process under intermittent grounding faults. Detailed Implementation

[0022] This invention proposes a single-phase grounding fault protection method for 10kV rural distribution networks with distributed generation (DG) access and low current grounding systems. The method integrates the zero-sequence characteristics of busbar, feeder, and branch measurement points to form an initial fault criterion. It then incorporates the DG's operational status, output changes, and access location for disturbance correction, and uses the criterion to distinguish margins and spatial consistency to verify candidate faulty feeders. Based on this, fault evidence and DG disturbance evidence are constructed, and a state machine is used to complete the protection process, including disturbance blocking, fault initiation, fault confirmation, action holding, fault clearing confirmation, and reset permission.

[0023] like Figure 1As shown, the ground fault protection method proposed in this invention generally includes four stages: multi-source information measurement and basic feature construction, multi-feature fusion initial fault criterion construction, distributed power source disturbance correction and candidate fault feeder verification, evidence accumulation state machine and protection action output.

[0024] I. Multi-source information measurement and basic feature construction

[0025] This invention first collects multi-source measurement information required for protection judgment, including the three-phase voltage of the bus, the three-phase current at each feeder outlet, the three-phase current at branch measuring points, and operational information such as the commissioning status, connection location, and active power output of distributed generation sources. Specifically, the three-phase voltage of the bus is used to reflect the degree of system grounding anomaly; the three-phase current at the feeder outlet is used to extract the zero-sequence current, zero-sequence current mutation, and zero-sequence transient energy of each feeder; the three-phase current at branch measuring points is used for spatial consistency verification of subsequent candidate fault feeders; and the commissioning status information of distributed generation sources is used to identify non-fault disturbances caused by grid connection, decommissioning, and changes in output.

[0026] To clarify the sources of multi-source measurement information, corresponding measurement units are set up at the busbar, feeder outlets, and main branch inlets; the operational status, connection location, and active power output information of distributed power sources are obtained from the corresponding distributed power source control devices or monitoring interfaces; such as Figure 2 As shown, the measurement units in this scheme are arranged according to a three-level electrical location: "busbar - feeder outlet - branch inlet". Distributed power source (DG) operation information is obtained through the existing control interface, and the specific locations are as follows: 1. Busbar three-phase voltage measurement points; A set of three-phase voltage measurement units (the measurement point at the top of the busbar in the figure) is set up on the common busbar on the low-voltage side of the transformer to collect the three-phase voltage data of the busbar, which is used to reflect the degree of abnormal grounding in the system.

[0027] 2. Three-phase current measuring point at the feeder outlet; At the beginning of each of the three outgoing feeders (between the busbar and the first section of the feeder), one set of three-phase current measurement units is installed, for a total of three sets, to extract the zero-sequence current, zero-sequence current abrupt change, and zero-sequence transient energy of the corresponding feeder: Feeder 1 (including line segment) , Arranged at the busbar outlet and line Measurement points between; Second feeder (including line segment) , Arranged at the busbar outlet and line Measurement points between; Feeder 3 (including line segment) , Arranged at the busbar outlet and line Measurement points between.

[0028] 3. Three-phase current measuring points at branch inlets; Seven sets of three-phase current measurement units are installed at the load branch inlet of each feeder end for spatial consistency verification of candidate faulty feeders. Feeder 1 (line) (After the end node): Set up one set of measurement points at the entrance of each of the three mountain and rural load branches; Second feeder (line) (After the end node): Set up one set of measurement points at the entrance of each of the two mountain and rural load branches; Feeder 3 (line) (After the end node): Set up one set of measurement points at the entrance of each of the two mountain and rural load branches.

[0029] 4. Distributed power supply operation information acquisition interface; Operational information such as the commissioning status, connection location, and active power output of each grid-connected distributed power source is directly obtained through the distributed power source's supporting control device or system monitoring interface, without adding any new electrical measurement points. This information is used to identify non-fault disturbances caused by the grid connection, disconnection, and output changes of distributed power sources.

[0030] In addition, the fault locations marked in the diagram are divided into two categories: one category is distributed in each feeder trunk line section ( and between, and between, and One type is distributed on the middle section of each load branch line, used to simulate fault scenarios at different locations.

[0031] Let the three-phase voltages of the bus be respectively , , Then the zero-sequence voltage of the bus is: (1); In the formula, express Zero-sequence voltage of the bus at any given moment; , , They represent The instantaneous voltage values ​​of phases A, B, and C of the bus at any given time.

[0032] Assume a 10kV rural power distribution network low current grounding system has a total of feeder line, number The three-phase currents at the feeder outlet measuring points are respectively , , Then the first The zero-sequence current of the feeder is: (2); In the formula, express Time of the first The zero-sequence current of each feeder; different feeders have different zero-sequence responses under single-phase ground faults, so this quantity is one of the main inputs for forming feeder fault criteria. , , They represent Time of the first Instantaneous values ​​of phase A, phase B, and phase C currents at the feeder outlet.

[0033] To reduce the impact of measurement noise and steady-state unbalanced current on transient characteristic calculations, a recursive method is used to calculate the first... Zero-sequence current reference value for each feeder: (3); In the formula, express Time of the first Reference value for zero-sequence current of each feeder; express Time of the first Reference value for zero-sequence current of each feeder; Indicates the sampling time interval; Indicates the recursive coefficient; express Time of the first Zero-sequence current of the feeder; This represents the reference value update status quantity. A value of 0 indicates that the reference value update is allowed, while a value of 1 indicates that the reference value update is paused (usually when the transient quantity exceeds a set threshold to maintain the stability of the baseline background value).

[0034] recursion coefficient The calculation formula is: (4); In the formula, This represents the update time constant, used to control the rate at which the reference value tracks the background of normal fluctuations. When the transient quantity exceeds the update pause threshold, the reference value stops updating. After the relevant quantity recovers, the recursive update continues.

[0035] Based on the difference between the current zero-sequence current and the background reference value, construct the first... The zero-sequence current deviation and zero-sequence current abrupt change of the feeder are expressed as follows: (5); In the formula, express Time of the first Zero-sequence current deviation of the feeder The direction of change of the zero-sequence response is retained for subsequent spatial consistency calculations between the feeder outlet and the branch measurement points; express Time of the first The zero-sequence current mutation of a feeder represents the magnitude of the change in zero-sequence current relative to the recent normal operating level, and is used for multi-feature fusion and zero-sequence transient energy calculation.

[0036] To reflect the duration and cumulative intensity of the zero-sequence current abrupt change over a certain period of time, a sliding time window is used. Internal calculation Zero-sequence transient energy of a feeder: (6); In the formula, express Time of the first Zero-sequence transient energy of a single feeder; This represents the sliding time window for transient energy calculation; Indicates time No. The sudden change in zero-sequence current of a feeder; Represents the integral variable, with a range of values ​​of . .

[0037] For feeders with multi-branch structures, the measuring points of each branch are numbered uniformly. Mark; let the first The three-phase currents at each branch measuring point are respectively , , Then the zero-sequence current at the measuring point of this branch is: (7); In the formula, , , They represent Time of the first Instantaneous values ​​of phase A, phase B, and phase C currents at each branch measuring point; express Time of the first The zero-sequence current at each branch measuring point is not used in the calculation of the initial fault criterion for multi-feature fusion. Instead, it is used after the candidate fault feeder is formed to calculate the consistency of the zero-sequence transient response between the feeder outlet and the relevant branch measuring points, providing a spatial response basis for the verification of the candidate fault feeder.

[0038] Assume a 10kV rural power distribution network low current grounding system has a total of The first distributed power source, the... The operational status of a distributed power source is represented as follows: (8); In the formula, express Time of the first The operational status value of a distributed power source. Indicates the first One distributed power source is in grid-connected and operational status. Indicates the first One distributed power source is in a decommissioned state.

[0039] The grid connection or decommissioning of distributed power sources can be represented by changes in their operational status: (9); In the formula, express Time of the first The operational status value of a distributed power source; express Time of the first The change in the operational status of a distributed power source, when When, it indicates that a distributed power source has been connected to the grid or disconnected within the sampling interval. When the state remains unchanged; This represents the sampling time interval.

[0040] Let the first The active power output of each distributed power source is Then the change in its active power output is: (10); In the formula, express Time of the first The change in active power output of a distributed power source is used to characterize the absolute degree of increase, decrease or step change in the active power output of the distributed power source, so that the output change in different directions can form a non-negative disturbance input of the distributed power source. express Time of the first Active power output of each distributed power source; express Time of the first Active power output of each distributed power source; The first The location of the distributed power source is as follows express, It can correspond to the feeder number, branch number, or network node number, and is used to subsequently determine the electrical relationship between the distributed power source and each feeder.

[0041] Operational status Changes in state Contributing merit and effort Change in output and access location It is not directly used as a single-phase ground fault operation condition, but rather as an input for subsequent distributed power source disturbance identification and fault criterion correction.

[0042] The result obtained in this step Zero-sequence voltage of the bus at any time , Time of the first Zero-sequence current of the feeder , Time of the first Zero-sequence current mutation of a feeder , Time of the first Zero-sequence transient energy of a feeder Used to construct initial fault criteria for multi-feature fusion; Time of the first Zero-sequence current deviation of the feeder and Time of the first Zero-sequence current at each branch measuring point Used for subsequent spatial consistency verification; Time of the first The operational status value of a distributed power source , Time of the first Changes in the operational status of a distributed power source , Time of the first Active power output of each distributed power source , Time of the first Changes in active power output of distributed power sources , No. The location of the distributed power source Used to construct the impact of distributed power source disturbances.

[0043] II. Construction of Initial Fault Criterion through Multi-Feature Fusion

[0044] After completing the multi-source information measurement and basic feature construction, the zero-sequence voltage of the bus, the zero-sequence current of the feeder, the abrupt change in the zero-sequence current of the feeder, and the zero-sequence transient energy of the feeder are normalized, and the initial fault criterion quantity of each feeder is constructed by weighted fusion. This criterion quantity is used to characterize the comprehensive strength of the grounding anomaly characteristics of each feeder, providing a basis for subsequent distributed power source disturbance correction and candidate fault feeder selection.

[0045] To enable features from different amplitude ranges to participate in unified calculations, this invention employs an amplitude-limiting normalization function: (11); In the formula, This represents the amplitude limiting normalization function, used to normalize the input variables. Limiting the amplitude to a closed interval of 0 to 1 avoids excessive influence of a single feature amplitude on the fusion result. This represents the input variable, i.e., the original feature quantity to be normalized.

[0046] Let the normalized reference values ​​of the zero-sequence voltage of the bus, the zero-sequence current of the feeder, the sudden change in the zero-sequence current of the feeder, and the zero-sequence transient energy of the feeder be respectively... , , , Then the first The normalized features corresponding to the feed lines are as follows: ; ; ; ; In the formula, express Normalized zero-sequence voltage characteristics of the bus at any given time; express Normalized zero-sequence current characteristics of the feeder at any given time; express Characteristics of the normalized zero-sequence current abrupt change in the feeder at any given moment; express The normalized zero-sequence transient energy characteristics of the feeder at any time; the zero-sequence voltage of the bus is a system grounding anomaly characteristic shared by all feeders, and the differences between feeders are mainly reflected by the zero-sequence current, the zero-sequence current mutation, and the zero-sequence transient energy.

[0047] After completing feature normalization, construct the first... Initial fault criterion quantity for multi-feature fusion of feeders: (13); In the formula, express Time of the first Multi-feature fusion initial fault criterion quantity for a single feeder; , , , These are the fusion weights for the zero-sequence voltage of the bus, the zero-sequence current of the feeder, the abrupt change in the zero-sequence current of the feeder, and the zero-sequence transient energy of the feeder, respectively. Each fusion weight satisfies the constraints of non-negativity and normalization. (14); In the formula, Indicates the first The fusion weights corresponding to each feature; Each fusion weight was determined using an offline sample calibration method. Typical samples were selected, including normal operation, distributed generation grid connection or disconnection, distributed generation output changes, continuous grounding faults, and intermittent grounding faults. Under the condition of satisfying non-negativity and normalization constraints, the initial fault criteria corresponding to different fusion weight combinations were compared. The criterion discrimination margin between the faulty feeder and other feeders, the criterion response under non-fault disturbances of the distributed generation, and the correctness of the faulty feeder judgment were used as calibration criteria to select a weight combination that balances fault detection sensitivity and disturbance rejection capability. Once the weights were determined, they were fixed and used within the same protection area, and no online adjustments were made during protection operation.

[0048] As one implementation method, in this embodiment , , , The values ​​are 0.25, 0.20, 0.30, and 0.25, respectively. These values ​​are the calibration results of this embodiment and do not constitute a limitation on the fusion weights of other protected areas.

[0049] The larger the value, the higher the value. The stronger the comprehensive characteristics of the grounding anomaly of a feeder at the current moment, the more likely it is to be considered. This criterion is only used as the basis for subsequent distributed power source disturbance correction and candidate fault feeder selection, and is not directly used to confirm the faulty feeder or trigger protection actions.

[0050] III. Distributed power source disturbance correction and candidate fault feeder verification

[0051] Initial fault criteria for each feeder are obtained through multi-feature fusion. Subsequently, based on the operational status, output changes, and connection location of the distributed power source, the disturbance impact was constructed, and then... Differential corrections are performed; then candidate faulty feeders are selected, and verification is performed by combining criteria for discrimination margin and spatial consistency.

[0052] Distributed power source grid connection, disconnection, or power output step change may cause a short-term increase in zero-sequence current, zero-sequence current surge, and zero-sequence transient energy in the feeder. If directly based on... Making a judgment might misjudge changes in the operation of distributed generation sources as single-phase grounding faults; therefore, the disturbance impact of a single distributed generation source is first constructed based on the changes in the operating status and active power output of the distributed generation source: (15); In the formula, express Time of the first The degree of disturbance of a distributed power source; express Time of the first The change in active power output of a distributed power source; Normalized reference value representing the change in active power output; express Time of the first The operational status value of a distributed power source; This represents the limiting normalization function. When distributed power sources experience grid connection, disconnection, or significant output changes... Increase; when its operating state and output remain stable. Keep it at a low level.

[0053] To prevent false alarms caused by the instantaneous disappearance of transient events in distributed generation sources, a disturbance hold-up window for distributed generation sources is set. By holding a single disturbance within a time window, the amount of disturbance held is obtained: (16); In the formula, express Time of the first The amount of disturbance to maintain the distributed power source; express Time of the first The degree of disturbance of a distributed power source; Represents a time variable.

[0054] As one implementation method, distributed power source disturbance holding window Take 40ms. Through... Maintaining a time window ensures that transient disturbances caused by grid connection, disconnection, or sudden output changes of distributed power sources can be effectively characterized in subsequent criterion corrections and state assessments.

[0055] set up For the first The distributed power supply pair The location influence coefficient of the initial fault criterion for the feeder line satisfies Its value is determined based on the location of the distributed power source and its electrical relationship with the feeder measurement point.

[0056] No. The total disturbance impact of distributed generation on the feeder is: (17); In the formula, express Time of the first The combined impact of distributed power source disturbances on each feeder; This refers to the number of distributed power sources in a 10kV rural power distribution network with low current grounding.

[0057] The initial fault criterion of the feeder is corrected by using the combined disturbance of distributed power sources on the feeder, resulting in the disturbance-corrected fault criterion of the feeder: (18); In the formula, express Time of the first Disturbance correction fault criterion for each feeder; Indicates the disturbance correction factor; express Normalized zero-sequence voltage characteristics of the bus at any given time.

[0058] When the operation of distributed power sources changes significantly and the zero-sequence voltage support is weak, the disturbance correction effect of distributed power sources is strong; when a real ground fault causes a significant increase in the zero-sequence voltage, the correction amount automatically decreases to avoid the fault criterion being overly suppressed when the disturbance of distributed power sources and the real fault occur at the same time.

[0059] Based on the disturbance correction fault criterion value of each feeder, the feeder with the largest current criterion value is selected as the candidate faulty feeder: (19); In the formula, express The candidate faulty feeder number selected at any given time; This indicates the total number of feeders.

[0060] To avoid misselection when the criteria values ​​of multiple feeders are close, calculate the criterion discrimination margin between the candidate faulty feeder and the other feeders: (20); In the formula, express The criterion discrimination margin between the candidate faulty feeder and the other feeders at any given time; express The disturbance correction fault criterion quantity of the candidate fault feeder at any time (the disturbance correction fault criterion quantity of the candidate fault feeder is calculated by equation (18)). The larger the value, the more prominent the abnormal characteristics of the candidate faulty feeder.

[0061] For candidate faulty feeders with branch structures, spatial consistency verification is performed using their zero-sequence current and the zero-sequence current at the branch measurement points. The same background reference processing method as the feeder outlet is used (when calculating the reference value of the zero-sequence current at the branch measurement points, the "recursive update" algorithm, coefficients, and logic control established earlier for the zero-sequence current at the feeder outlet are fully reused) to calculate the first... Zero-sequence current reference values ​​for each branch measuring point: (twenty one); In the formula, express Time of the first Zero-sequence current reference value at each branch measuring point; express Time of the first Zero-sequence current at each branch measuring point.

[0062] No. The zero-sequence current deviation at each branch measuring point is: (twenty two); In the formula, express Time of the first The zero-sequence current deviation of each branch measurement point is used for spatial response calculation.

[0063] Let the zero-sequence current deviation at the outlet of the candidate faulty feeder be... Then in the time window Inside, the candidate faulty feeder outlet and the first The consistency of the zero-sequence transient response among the branch measurement points is: (twenty three); In the formula, express Time Candidate Fault Feeder Output and the First Consistency of zero-sequence transient response among branch measurement points; Indicates time Zero-sequence current deviation at the outlet of the candidate faulty feeder; This represents a pre-set positive number used to prevent mathematical calculation overflow caused by the denominator (energy product) being zero within the integration time window.

[0064] When the transient response energy at the feeder outlet or branch measurement point of a candidate fault is lower than the effective response threshold, an effective spatial correlation cannot be formed. In this case, let... : (twenty four); In the formula, This represents the minimum effective response threshold.

[0065] The relevant branch measurement points described in this invention refer to branch measurement points that have a direct electrical connection with the current candidate fault feeder and are pre-included in the protection device's measurement point configuration. Spatial consistency verification only uses the branch measurement points corresponding to the candidate fault feeder, excluding branch measurement points of other feeders. When a candidate fault feeder corresponds to multiple relevant branch measurement points, the zero-sequence transient response consistency between each branch measurement point and the candidate fault feeder outlet is calculated, and the maximum value is taken as the spatial consistency index of the candidate fault feeder. Branches without configured measurement units do not participate in spatial consistency calculations, and it is not required that all terminal branches of the candidate fault feeder have measurement points.

[0066] The spatial consistency of a candidate faulty feeder is taken as the maximum value among the consistency values ​​of its relevant branch measurement points: (25); In the formula, express Spatial consistency index of candidate fault feeders at any given time; This indicates the total number of branch measuring points with electrical connections configured for the candidate faulty feeder; if the candidate faulty feeder does not have relevant branch measuring points configured (i.e. ), then let This is to keep the factor in a neutral state in subsequent calculations.

[0067] The disturbance correction fault criterion quantity for the candidate faulty feeder and the maximum combined disturbance impact quantity of the distributed generation are denoted as follows: (26); In the formula, express Disturbance correction fault criterion quantity for candidate fault feeder at any time; express The maximum value of the combined disturbance impact of distributed power sources on each feeder at any given time is used to characterize the overall non-fault disturbance intensity of the system. express Time of the first The combined impact of distributed power source disturbances on each feeder.

[0068] This step yields... Disturbance correction fault criterion quantity of candidate fault feeder at any time , The maximum value of the combined disturbance impact of distributed power sources on each feeder at any given time. , Spatial consistency index of candidate fault feeders at any time , Criterion discrimination margin between the candidate faulty feeder and the other feeders at any given time As input for subsequent evidence accumulation and state judgment.

[0069] IV. Evidence Accumulation State Machine and Protection Action Output

[0070] In obtaining , , , Then, fault evidence and distributed power source disturbance evidence are constructed, and disturbance interlocking, fault initiation, fault confirmation, action holding, fault clearing confirmation and reset permission are completed through a state machine.

[0071] First, the fault support quantity of the candidate faulty feeder at the current moment is constructed based on the disturbance-corrected fault criterion quantity of the candidate faulty feeder, the criterion discrimination margin between the candidate faulty feeder and other feeders, and the spatial consistency index of the candidate faulty feeder: (27); In the formula, express Fault support quantity of candidate fault feeder at any time; Indicates the activation value of the modification criterion; This represents the minimum criterion discrimination margin.

[0072] The cumulative amount of fault evidence is constructed using a recursive method: (28); In the formula, express The cumulative amount of fault evidence at any given moment; Indicates the previous sampling time, i.e. The cumulative amount of fault evidence at any given moment; Indicates the previous sampling time, i.e. The candidate faulty feeder number at any given time; This represents the fault evidence memory coefficient, used to control the degree to which historical fault evidence is retained in the current calculation result. This memory coefficient is calibrated offline based on the sampling time interval, the arc extinction interval of intermittent faults, and the fault confirmation speed. When the candidate fault feeder remains unchanged, fault evidence continues to accumulate; when the candidate fault feeder changes, it is recalculated according to the new candidate fault feeder. As one implementation method, the memory time parameter corresponding to the fault evidence is set to 12ms.

[0073] Based on the maximum value of the comprehensive disturbance impact of the feeder distributed power source, construct the cumulative amount of evidence for the distributed power source disturbance: (29); In the formula, express The cumulative amount of evidence of distributed source disturbances at any given time; Indicates the previous sampling time, i.e. The cumulative amount of evidence of distributed source disturbances at any given time; This represents the memory coefficient of the disturbance evidence, used to control the decay rate of the disturbance evidence from distributed generation sources. This memory coefficient is calibrated offline based on the sampling time interval and the continuous characteristics of distributed generation source grid connection, disconnection, and power output abrupt changes. As one implementation method, the memory time parameter corresponding to the distributed generation disturbance evidence is set to 6 ms.

[0074] The state machine of this invention includes a normal monitoring state, a disturbance lockout state, a fault initiation state, a fault confirmation state, an action holding state, a fault clearing confirmation state, and a reset permission state, and its state set is represented as follows: (30); In the formula, This indicates a normal monitoring status; Indicates a disturbance-locked state; Indicates a fault start-up status; Indicates fault confirmation status; Indicates that the action is in a state of being held; Indicates that the fault has been cleared and confirmed. This indicates a return to a licensed state.

[0075] Under normal monitoring conditions Below, when the accumulated amount of fault evidence Reaching the cumulative threshold for fault start Furthermore, the disturbance correction fault criterion quantity of the candidate fault feeder It also simultaneously satisfies the modification criterion activation value. When required, the state machine enters the fault start state. : (31); When the accumulated amount of evidence of distributed power source disturbance Strong enough to reach the distributed power source disturbance blocking value. The amount of accumulated evidence of failure Failed to reach the cumulative threshold for fault start When this happens, the state machine enters the disturbance lockout state. To avoid non-fault disturbances triggering erroneous protection activation: (32); It should be noted that during the disturbance lockout period, the state machine will continue to update and calculate the fault criteria and fault evidence, and will not stop detecting real faults.

[0076] In the disturbance-locked state Below, if the amount of fault evidence accumulates during the lockout period If the start-up requirements are met (i.e., equation (31) is satisfied), then the interlocking is immediately disengaged and the system enters the fault start-up state. If the distributed power source disturbance disappears and the fault evidence is released simultaneously, the state machine returns to the normal monitoring state. The conditions for returning to the normal monitoring state are: (33); in, and These represent the distributed power source disturbance release value and the fault evidence release value, respectively.

[0077] Enter fault start-up state Subsequently, if the candidate faulty feeder is within the confirmation time window (As one implementation method, The value remains unchanged within 8ms, and the accumulated amount of fault evidence is... Fault confirmation threshold reached Then it enters the fault confirmation state. : (34); (35); If the accumulated amount of fault evidence drops below the release threshold before confirmation, the startup will be canceled, and the system will either return to normal monitoring or enter disturbance blocking state depending on the current distributed power source disturbance status.

[0078] Entering fault confirmation state Then, the candidate faulty feeder at this moment is latched as the faulty feeder: (36); In the formula, Indicates the moment when the fault confirmation state is entered; This indicates the latched fault feeder number. After fault confirmation, the state machine... The corresponding correction criteria and spatial consistency results are used for subsequent judgment, and the latching results are no longer modified by changes in real-time candidate fault feeders. Subsequently, the state machine transitions to the action hold state and outputs the action hold signal; In the state of motion Below, when the faulty feeder Disturbance correction fault criterion quantity (Calculated by equation (18)) Reduced to the modified criterion release value The following is the cumulative amount of fault evidence. Reduced to fault evidence release value Below, and the absolute value of the zero-sequence voltage of the bus. Reduced to bus zero-sequence voltage release value The following steps will trigger a fault clearing confirmation state. : (37).

[0079] In the fault clearing confirmation state If the above fault clearing conditions (i.e., equation (37)) continue to be met for a preset time, then the system enters the reset permission state. As one implementation method, the continuous confirmation time for fault clearing conditions is set to 20ms. If the fault characteristic reappears, the system returns to the action hold state. After entering the reset permission state, a reset permission signal is output, and the system waits for an external reset command.

[0080] Each start, confirmation, and release threshold satisfies the following hysteresis relationship to avoid repeated state switching around the threshold: (38).

[0081] The threshold values ​​for each state are calibrated offline based on samples from normal operation, non-fault disturbances of distributed power sources, typical ground faults, and fault clearing. Among these, the correction criterion activation value... The minimum criterion discrimination margin is determined based on the statistical upper limit of the modified criterion under normal operation and distributed power source disturbance conditions, and the effective response under weak fault conditions; Determined based on the criterion difference between the faulty feeder and the other feeders; fault start-up cumulative threshold. Fault confirmation threshold and fault evidence release value The values ​​for distributed power source disturbance blocking are determined based on the rise, confirmation, and return processes of fault evidence. and distributed power source disturbance release value The release value of the correction criterion is determined based on the range of changes in disturbance evidence during the grid connection, decommissioning, and output changes of distributed power sources; and bus zero-sequence voltage release value The value is determined based on the steady-state recovery range after fault clearance. The starting value is higher than the corresponding release value to create state transition hysteresis and avoid repeated state switching near the threshold.

[0082] As one implementation method, the criterion activation value is modified. The minimum criterion discriminant margin is set at 0.329334. Set the value to 0.190630 and adjust the criterion release value accordingly. Set the threshold to 0.214067; cumulative threshold for fault start-up. Set the threshold to 0.180000. Take 0.694641, the fault evidence release value. Set to 0.072000; Distributed source disturbance blocking value. Take 0.450000 as the distributed generation disturbance release value. Take 0.150000, the zero-sequence voltage release value of the bus. The value is set to 24.4895V. The above parameters are the offline calibration results of this embodiment and do not constitute a limitation on the parameter values ​​under other system structures and operating conditions.

[0083] This invention outputs at least one of the following signals: fault alarm, fault feeder indication, distributed power source disturbance lockout, lockout exit, fault initiation, fault confirmation, protection action hold, fault clearance confirmation, and reset permission signal. Through evidence accumulation and state transitions, it achieves continuous judgment of distributed power source disturbances, single-phase ground faults, and the fault clearance process.

[0084] To illustrate the operation of the protection method under the superposition of distributed power source operation disturbances and actual grounding faults and intermittent grounding faults, the following two typical operating conditions are selected for verification.

[0085] Figure 3 In the illustrated operating condition, the distributed power source connects to feeder F1 at 0.080s; at 0.115s, a single-phase ground fault with a transition resistance of 500Ω occurs on feeder F1. The fault lasts until 0.190s (corresponding to...). Figure 3 (1) DG status signal and fault input signal; in the initial stage of distributed power grid connection, by Figure 3 As can be seen in (2), the evidence of disturbance (Accumulated evidence of distributed power source disturbances) increases and fault evidence increases. The accumulated fault evidence has not yet reached the activation condition, and the state machine enters the disturbance lockout state at 0.083s. Figure 3 In step (3), the Block signal changes from 0 to 1; after a real ground fault occurs in 0.115s, as... Figure 3 As shown in (2), evidence of the fault Gradually rising, while perturbing evidence The decay begins. At 0.129s, the disturbance evidence drops below the latch release threshold, and the fault evidence meets the activation requirements. The state machine immediately exits the latch, corresponding to... Figure 3 The Block signal in (3) is cleared, and the BlockExit signal generates a pulse, entering the fault start-up process; thereafter, by Figure 3 (2) It can be seen that fault evidence continues to accumulate. At 0.155s, the fault evidence reaches the confirmation threshold, and the state machine confirms that feeder F1 is a faulty feeder. Figure 3 The Trip signal in (3) flips from 0 to 1, completing the output of the protection action command.

[0086] Figure 4In the operating condition shown, feeder F3 experienced an intermittent single-phase ground fault comprising four short-duration arcing events (stages 1 to 4); Figure 4 As can be seen from (1) and (2), the zero-sequence voltage of the busbar exhibits a significant oscillating response during each arcing process; at the same time, Figure 4 The intermittent fault input signal in (2) exhibits a high-frequency alternating state of arc ignition and arc extinction; by Figure 4 As can be seen in (3), the F3 correction criterion (orange dashed line) of the F3 feeder increases significantly during the arcing stage and decreases rapidly during the arc extinction gap. Although the real-time value of the criterion fluctuates, fault evidence... (The blue solid line) can continuously accumulate across subsequent arc-burning stages, maintaining a stable rise within a short arc-extinguishing interval, effectively overcoming the problem of intermittent characteristic discretization; by Figure 4 As can be seen from (4), based on the above accumulated evidence, the state machine reaches the start condition at 0.116s, the Start signal flips from 0 to 1 (corresponding to the "Start F3: 0.116s" mark in the figure), and enters the fault start state; and reaches the confirmation threshold at 0.128s, the Trip signal flips from 0 to 1 (corresponding to the "Confirm F3: 0.128s" mark in the figure), confirming that the F3 feeder is a fault feeder, and at the same time the action holding signal Hold signal is synchronously set to 1, entering the action holding state. In the subsequent short-term arc extinguishing process, the Hold signal remains at a high level, avoiding repeated return of the protection device or loss of action, and ensuring the reliability of fault clearing; F3 correction criterion refers to the fault criterion quantity for feeder disturbance correction; The Start signal is the fault start signal, the BlockExit signal is the disturbance lockout exit signal, the Block signal is the disturbance lockout (start) signal, the Trip signal is the protection action command signal, and the Hold signal is the action hold status signal; Trip=1 indicates that a protection action command has been formed, and Hold=1 indicates that the action hold status is valid.

[0087] Table 1. Protection Action Results under Typical Fault Conditions

[0088] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources, characterized in that, Includes the following steps: The system collects the three-phase voltage of the busbar, the three-phase current at each feeder outlet, the three-phase current at each branch measuring point, and the operating information of each distributed power source; the operating information includes the commissioning status, active power output, and connection location. The zero-sequence voltage of the busbar is calculated based on the three-phase voltage of the busbar. The zero-sequence current, zero-sequence current mutation and zero-sequence transient energy of each feeder are calculated based on the three-phase current at the feeder outlet. The zero-sequence voltage of the busbar, the zero-sequence current, zero-sequence current mutation and zero-sequence transient energy of each feeder are normalized and then weighted and fused according to preset weights to obtain the initial fault criterion quantity of each feeder. Based on the operating information and connection location of each distributed power source, the comprehensive disturbance impact of distributed power sources on each feeder is constructed. The initial fault criteria of each feeder are corrected by using the comprehensive disturbance impact of distributed power sources to obtain the disturbance-corrected fault criteria of each feeder; the feeder with the largest disturbance-corrected fault criteria is selected as the candidate fault feeder, and the criterion discrimination margin between the candidate fault feeder and the other feeders is calculated. The spatial consistency index of the candidate fault feeder is calculated by using the zero-sequence current deviation at the outlet of the candidate fault feeder and the zero-sequence current deviation at the relevant branch measurement points. The cumulative amount of fault evidence is constructed based on the disturbance correction fault criterion quantity, criterion discrimination margin and spatial consistency index of the candidate fault feeder. The cumulative amount of distributed power source disturbance evidence is constructed based on the maximum value of the comprehensive disturbance impact of each feeder. A protection state machine is constructed that includes normal monitoring state, disturbance blocking state, fault initiation state, fault confirmation state, action holding state, fault clearing confirmation state and reset permission state. Based on the comparison results of the accumulated amount of fault evidence, the accumulated amount of distributed power source disturbance evidence, and the preset threshold, the control protection state machine performs state transitions and outputs protection action signals.

2. The method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 1, characterized in that: Each feeder of the distribution network is equipped with a three-phase current measurement unit at its end to collect the three-phase current at the branch measurement point; the operation information of the distributed power source is obtained through the supporting control device or system monitoring interface of the distributed power source.

3. The method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 2, characterized in that: By using a limiting normalization function, the zero-sequence voltage of the bus, the zero-sequence current of each feeder, the zero-sequence current mutation, and the zero-sequence transient energy are mapped to the interval [0,1] to obtain the corresponding normalized characteristic quantities. The normalized feature quantities are weighted and summed according to the preset weights to obtain the initial fault criterion quantity.

4. The method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 3, characterized in that: The methods for constructing the comprehensive disturbance impact of the distributed power source include: Calculate the first The degree of disturbance of each distributed power source: ； In the formula, express Time of the first The degree of disturbance of a distributed power source; express Time of the first The change in active power output of a distributed power source; Normalized reference value representing the change in active power output; express Time of the first The operational status value of a distributed power source; This represents the amplitude limiting normalization function; Within the preset disturbance hold time window The maximum value of the disturbance level is maintained to obtain the maintained disturbance amount: ； In the formula, express Time of the first The amount of disturbance to maintain the distributed power source; express Time of the first The degree of disturbance of a distributed power source; Represents a time variable; The sum of the holding disturbances of each distributed power source and their corresponding location influence coefficients is then calculated and normalized to obtain the comprehensive disturbance impact of the distributed power sources on the feeder: ； In the formula, express Time of the first The combined impact of distributed power source disturbances on each feeder; This represents the number of distributed power sources.

5. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 4, characterized in that: The calculation method for the disturbance correction fault criterion quantity is as follows: ； express Time of the first Disturbance correction fault criterion for each feeder; Indicates the disturbance correction factor; express Normalized zero-sequence voltage characteristics of the bus at any given time; The method for calculating the criterion discrimination margin is as follows: ； In the formula, express The criterion discrimination margin between the candidate faulty feeder and the other feeders at any given time; express The fault criterion quantity for candidate fault feeder disturbance correction at any time.

6. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 5, characterized in that: The methods for calculating the spatial consistency index include: Calculate the zero-sequence current deviation at the outlet of the candidate faulty feeder and the first Zero-sequence current deviation at each relevant branch measurement point; Within a preset transient time window, the distance between the candidate faulty feeder outlet and the first... The consistency of zero-sequence transient response among the relevant branch measurement points is the normalized cross-correlation value between the zero-sequence current deviation at the outlet of the candidate fault feeder and the zero-sequence current deviation at the relevant branch measurement points. The maximum value among the zero-sequence transient response consistency of all relevant branch measurement points corresponding to the candidate fault feeder is determined as the spatial consistency index.

7. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 6, characterized in that: The methods for constructing the accumulated amount of fault evidence and the accumulated amount of distributed power source disturbance evidence include: The cumulative amount of fault evidence is constructed using a recursive method: ； In the formula, express The cumulative amount of fault evidence at any given moment; Indicates the previous sampling time, i.e. The cumulative amount of fault evidence at any given moment; express The candidate faulty feeder number selected at any given time; Indicates the previous sampling time, i.e. The candidate faulty feeder number at any given time; Indicates the fault evidence memory coefficient; express The fault support quantity of the candidate fault feeder at any given time; when the candidate fault feeder changes, the calculation of the accumulated fault evidence quantity is reset; Based on the maximum value of the comprehensive disturbance impact of the feeder distributed power source, construct the cumulative amount of evidence for the distributed power source disturbance: ； In the formula, express The cumulative amount of evidence of distributed source disturbances at any given time; Indicates the previous sampling time, i.e. The cumulative amount of evidence of distributed source disturbances at any given time; Indicates the perturbation evidence memory coefficient; express The maximum value of the combined disturbance impact of distributed power sources on each feeder at any given time.

8. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 7, characterized in that: The state transition conditions of the protection state machine include: when the accumulated amount of fault evidence reaches the fault initiation accumulation threshold and the disturbance correction fault criterion amount of the candidate fault feeder reaches the correction criterion initiation value, the system switches from the normal monitoring state to the fault initiation state; when the accumulated amount of disturbance evidence of the distributed power source reaches the disturbance blocking value and the accumulated amount of fault evidence does not reach the fault initiation accumulation threshold, the system switches from the normal monitoring state to the disturbance blocking state; in the fault initiation state, if the candidate fault feeder remains unchanged within the preset confirmation time window and the accumulated amount of fault evidence reaches the fault confirmation threshold, the system switches to the fault confirmation state; in the fault confirmation state, the current candidate fault feeder is latched as a fault feeder, and the system switches to the action holding state; in the action holding state, when the disturbance correction fault criterion amount of the fault feeder drops below the correction criterion release value, the accumulated amount of fault evidence drops below the fault evidence release value, and the bus zero-sequence voltage drops below the bus zero-sequence voltage release value, the system switches to the fault clearing confirmation state; in the fault clearing confirmation state, if the above conditions are continuously met for a preset time, the system switches to the reset permission state.

9. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 8, characterized in that: In the action holding state, the protection state machine continuously outputs protection action commands and maintains the latching results of the fault feeder; in the fault clearing confirmation state, if the fault characteristics reappear, the protection state machine returns to the action holding state and switches to the reset permission state after the conditions are met.

10. A method for protecting a single-phase grounding fault in a distribution network containing distributed power sources according to claim 9, characterized in that: The protection action signals include at least one of the following: fault alarm signal, fault feeder indication signal, distributed power source disturbance interlock signal, interlock exit signal, fault start signal, fault confirmation signal, protection action hold signal, fault clearing confirmation signal, and reset permission signal.

Images