Active detection based active power distribution network phase-to-phase fault identification and reclosing method and system

Abstract

This invention relates to the field of power system relay protection technology, and particularly to a method and system for active distribution network phase-to-phase fault identification and reclosing based on active detection. The method involves injecting a three-phase sinusoidal high-frequency detection signal into the faulty line and filtering to extract the high-frequency components. Positive-sequence, negative-sequence, and zero-sequence current components are extracted from the filtered three-phase current. Based on the amplitude ratio of the negative-sequence current to the positive-sequence current, a threshold is set to determine the nature of transient and permanent faults. Reclosing logic is constructed based on the fault nature determination results to complete the reclosing decision. Compared with existing technologies, this invention, based on active high-frequency electrical detection, symmetrical component sequence feature extraction, and protection-new energy collaborative control, achieves phase-to-phase asymmetrical fault nature determination, adaptive reclosing, and islanding detection.

Description

Technical Field

[0001] This invention relates to the field of power system relay protection technology, and in particular to an active detection-based method and system for phase-to-phase fault identification and reclosing in distribution networks, applicable to fault handling and safe and stable control of medium and low voltage distribution networks containing distributed photovoltaic, wind power and other power electronic inverter power sources. Background Technology

[0002] The power distribution network is a critical link in the power system that directly faces users, and its power supply reliability is directly related to social production and residents' lives. Phase-to-phase faults are one of the common fault types in distribution networks, and the currently widely adopted handling method is automatic reclosing after the relay protection trips. However, phase-to-phase faults in distribution networks include transient and permanent faults. Blindly performing reclosing in the case of a permanent fault will lead to a secondary impact on the system, which in severe cases can damage power equipment, expand the scope of the accident, and even cause large-scale power outages.

[0003] In distribution networks with a high proportion of renewable energy penetration, the harm of this problem is further amplified. When a phase-to-phase fault occurs on an adjacent line of a distributed generation source (such as photovoltaic or wind power), the inverter of the distributed generation source will enter low-voltage ride-through (LVRT) mode to maintain grid connection due to the voltage drop at the grid connection point. If a permanent fault is not identified at this time and reclosing is performed blindly, the fault point will persist, and the grid connection point voltage will drop again. The operating time of the inverter in LVRT mode is strictly limited. If the voltage drop exceeds its tolerance threshold, the inverter will trigger the protection mechanism and be forced to disconnect from the grid. This will not only cause power generation losses of the distributed generation source, but also cause power imbalance in the distribution network, further deteriorate voltage stability, and even trigger cascading failures.

[0004] When a phase-to-phase fault occurs on the main distribution line connected to a distributed generation source, the relay protection system will quickly activate, triggering circuit breakers to trip and isolate the fault point. This action directly forces the downstream distributed generation inverter to temporarily disconnect from the grid and enter islanded operation mode. During the disconnection period, the system-side power supply and the distributed generation source lose electrical connection and synchronization control, and the voltage frequency and phase of the two power supplies will gradually deviate. When reclosing is performed after the fault is cleared, if the phase difference between the two voltages is not detected and adjusted before directly closing the circuit breaker, the voltage difference will generate a huge inrush current on the system impedance and line impedance. The inrush current will not only cause mechanical and thermal stress damage to power equipment such as circuit breakers and inverters, but may also cause severe fluctuations in system voltage, and even lead to malfunction of protection devices, further expanding the fault range and affecting the safe and stable operation of the distribution network.

[0005] Existing fault identification technologies mainly rely on the amplitude and phase changes or traveling wave characteristics of fault currents. However, under the complex topology and variable operating conditions of distribution networks, the accuracy and speed of these methods often fail to meet practical needs. Especially with the widespread integration of distributed power sources into distribution networks, traditional fault feature extraction and judgment logic is further interfered with, leading to an increased misjudgment rate for both transient and permanent faults. Therefore, there is an urgent need for a technical means to quickly and accurately identify the nature of a fault after it occurs. Summary of the Invention

[0006] Purpose of the invention: To address the pain points in handling phase-to-phase faults in high-proportion renewable energy distribution networks, this invention proposes an active detection-based method and system for phase-to-phase fault identification and reclosing in active distribution networks. Based on active high-frequency electrical detection, symmetrical component sequence feature extraction, and protection-renewable energy coordinated control, it realizes phase-to-phase asymmetrical fault nature discrimination, adaptive reclosing, and islanding detection.

[0007] Technical solution: This invention discloses an active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network, comprising:

[0008] A three-phase sinusoidal high-frequency detection signal is injected into the faulty line, and the high-frequency component is filtered and extracted.

[0009] Extract the positive-sequence, negative-sequence, and zero-sequence current components from the filtered three-phase current;

[0010] Based on the ratio of the amplitude of negative sequence current to positive sequence current, a threshold is set to distinguish between transient and permanent faults.

[0011] Based on the fault nature determination results, reclosing logic is constructed to complete the reclosing determination.

[0012] Furthermore, the three-phase sinusoidal high-frequency detection signal is selected as 500Hz, with an amplitude of 0.1 times the rated voltage, that is:

[0013] ;

[0014] In the formula: U N This refers to the rated phase voltage amplitude of the line. , , These are the three-phase sinusoidal high-frequency detection signal voltages.

[0015] Furthermore, a 500Hz narrowband bandpass filter (BPF) is used to extract high-frequency components during filtering, while filtering out power frequency, harmonics, and noise.

[0016] Furthermore, when extracting the positive-sequence, negative-sequence, and zero-sequence current components from the filtered three-phase current, the symmetrical component method is used, and the transformation matrix is: , Let be the rotation factor, satisfying a 3 =1、1+a+a 2 =0.

[0017] Furthermore, by substituting the 500Hz three-phase current into the transformation matrix of the symmetrical component method, the formula for calculating the high-frequency sequence current is obtained:

[0018] ;

[0019] in, , , These are the positive-sequence, negative-sequence, and zero-sequence current components, respectively. These are the filtered currents for phases a, b, and c, respectively.

[0020] Furthermore, the distinction between transient and permanent faults is specifically as follows:

[0021] The fault characteristic coefficient k is defined as the ratio of the amplitude of the high-frequency negative sequence current to the amplitude of the positive sequence current.

[0022] Transient fault: The fault arc is extinguished, the line resumes three-phase symmetrical operation, and no negative sequence component is generated under high-frequency signal excitation, i.e. ;

[0023] Permanent fault: An asymmetrical low-impedance path persists at the fault point, generating a significant negative sequence current under high-frequency signal excitation, i.e. ;

[0024] Set threshold k set The fault diagnosis criteria are: .

[0025] Furthermore, the reclosing logic is as follows:

[0026] For permanent faults, the fault point persists, reclosing is blocked, and an alarm is uploaded.

[0027] For transient faults, once the fault disappears, a reclosing operation is performed.

[0028] Furthermore, after fault diagnosis is completed, the following steps are performed: fine-tuning the inverter output frequency, increasing the power angle, collecting the output power response, and calculating the power angle-power change rate, thereby completing the islanding detection.

[0029] Furthermore, island detection specifically includes:

[0030] The output active power satisfies the power angle characteristic equation:

[0031] ;

[0032] In the formula: U SU represents the system-side voltage amplitude. DG X represents the output voltage amplitude of the inverter. Σ δ represents the equivalent reactance of the system and the line, and δ represents the power angle of the inverter and the system.

[0033] A small adjustment Δf is applied to the inverter output frequency to increase the power angle by Δδ; the power angle-power change rate η is defined as follows: ,in, These represent the active power output of the inverter before and after frequency fine-tuning, respectively. These are the power angles of the inverter before and after frequency fine-tuning, respectively.

[0034] Set threshold η set The criteria for island detection are: .

[0035] This invention also discloses a system for active detection-based phase-to-phase fault identification and reclosing method in active distribution networks, comprising:

[0036] The signal injection module is used to inject a three-phase sinusoidal high-frequency detection signal into the faulty line and filter and extract the high-frequency components.

[0037] The current component acquisition module is used to extract the positive sequence, negative sequence, and zero sequence current components from the filtered three-phase current.

[0038] The fault nature discrimination module is used to distinguish between transient and permanent faults by setting a threshold based on the amplitude ratio of negative sequence current to positive sequence current.

[0039] The reclosing determination module is used to construct reclosing logic based on the fault nature judgment result to complete the reclosing determination.

[0040] Beneficial effects:

[0041] 1. This invention employs a collaborative control method combining active high-frequency detection and sequence component feature extraction: a 500Hz high-frequency signal is injected into the faulty line through a new energy inverter. After filtering and decomposition using the symmetrical component method, positive and negative sequence currents are obtained. The ratio of negative to positive sequence currents is used to identify permanent phase-to-phase faults. Fault nature identification can be completed according to a set threshold, and the identification results are consistent with theoretical criteria. This method is compatible with the operating characteristics of new energy inverters, reducing the risk of secondary impact from fault reclosing, equipment damage, and system instability. In this invention, there is no grounding path for asymmetrical phase-to-phase faults in the distribution network, and the vector sum of the three-phase currents is 0. Therefore, the zero-sequence component is always 0, and fault identification is only required through positive and negative sequence currents, simplifying the calculation logic. Based on the physical differences of the fault, it is not affected by DG control strategies or LVRT modes, resulting in high identification accuracy.

[0042] 2. The present invention also adopts an islanding detection method based on power angle characteristics. By fine-tuning the inverter output frequency and increasing the power angle, islanding detection is completed based on the response characteristics of output power to power angle. Ultimately, adaptive control is achieved to block reclosing in case of permanent faults, allow reclosing in case of transient faults and non-islanded states. Attached Figure Description

[0043] Figure 1 This is a framework diagram of the feature signal injection strategy of the present invention;

[0044] Figure 2 This is an overall flowchart of the active detection-based method for phase-to-phase fault identification and reclosing in a power distribution network according to the present invention.

[0045] Figure 3 This is a model of a distribution network high-proportion renewable energy access system according to an embodiment of the present invention;

[0046] Figure 4 The above are simulation waveforms under transient fault conditions according to an embodiment of the present invention.

[0047] Figure 5 The waveforms are simulation waveforms under permanent fault conditions in an embodiment of the present invention. Detailed Implementation

[0048] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0049] Example 1:

[0050] This invention discloses an active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network. See [link to relevant documentation]. Figure 2 The overall flowchart includes the following processes:

[0051] Step 1: After a phase-to-phase asymmetrical fault occurs in the distribution network, the relay protection device drives the circuit breaker to trip, completing the fault isolation; after tripping, a fixed delay of 0.5s is executed, and the fault identification process is started after the delay ends.

[0052] Step 2: After the delay ends, the DG inverter associated with the faulty line is coordinated and injected with a three-phase sinusoidal high-frequency detection signal. Since the system fundamental frequency is 50Hz, and the 3rd / 5th / 7th harmonics are 150Hz / 250Hz / 350Hz, 500Hz does not overlap with the power frequency and lower harmonics, satisfying the frequency domain orthogonal separation condition. 500Hz is lower than the switching frequency of conventional inverters (1kHz~10kHz), allowing the inverter power devices to output stably without hardware modification; therefore, the injected signal frequency is chosen to be 500Hz, with an amplitude of 0.1 times the rated voltage. The signal injection control strategy is as follows: Figure 1 As shown, the mathematical model of the injected signal is:

[0053] (1)

[0054] In the formula: U N This refers to the rated phase voltage amplitude of the line.

[0055] Step 3: Synchronously acquire the three-phase voltage u using the line PT and CT at a sampling frequency of 20kHz. abc (n), current i abc (n); A 500Hz narrowband bandpass filter (BPF) is used to extract high-frequency components and filter out power frequency, harmonics and noise.

[0056] The bandpass filter operator is a linear time-invariant system, and the high-frequency components after filtering are:

[0057] (2)

[0058] Expanded into three-phase electrical quantities:

[0059] (3)

[0060] Step 4: Substitute the filtered 500Hz three-phase current into the symmetrical component method to calculate the positive-sequence, negative-sequence, and zero-sequence current components. The transformation matrix is:

[0061] (4)

[0062] In the formula: Let be the rotation factor, satisfying a 3 =1、1+a+a 2 =0.

[0063] Substituting the 500Hz three-phase current into the formula, we obtain the formula for calculating the high-frequency sequence current:

[0064] (5)

[0065] In a distribution network with phase-to-phase asymmetrical faults, there is no grounding path and the vector sum of the three-phase currents is 0. Therefore, the zero-sequence component is always 0. Fault identification can be achieved only through positive and negative sequence currents, simplifying the calculation logic.

[0066] Step 5: Calculate the ratio of the magnitude of the negative sequence current to the positive sequence current, and set a threshold to complete the fault nature determination.

[0067] The fault characteristic coefficient k is defined as the ratio of the amplitude of the high-frequency negative-sequence current to the amplitude of the positive-sequence current.

[0068] (6)

[0069] Transient fault: The fault arc is extinguished, the line resumes three-phase symmetrical operation, and no negative sequence component is generated under high-frequency signal excitation, that is:

[0070] (7)

[0071] Permanent fault: An asymmetrical low-impedance path persists at the fault point, generating a significant negative sequence current under high-frequency signal excitation, i.e.:

[0072] (8)

[0073] By combining distribution network line parameters and simulation iterations, a threshold k is set. set =0.15, the fault judgment criterion is:

[0074] (9)

[0075] This criterion is based on the difference in the physical nature of the fault and is not affected by the DG control strategy or LVRT mode, resulting in a high accuracy rate.

[0076] Step 6: Construct the reclosing logic:

[0077] Permanent fault: The fault point persists, and reclosing will inevitably cause a secondary impact, blocking reclosing and uploading alarms;

[0078] Transient fault: Once the fault disappears, reclosing can be performed;

[0079] The logical expression is:

[0080] (10)

[0081] In the formula: Reclose=1 means overlap is allowed, and Reclose=0 means overlap is blocked.

[0082] Step 7: After fault diagnosis, fine-tune the inverter output frequency, increase the power angle, collect the output power response, and calculate the power change rate to complete the islanding detection.

[0083] When a power electronic inverter is connected to the grid, its output active power satisfies the power angle characteristic equation:

[0084] (11)

[0085] In the formula: U S U represents the system-side voltage amplitude. DG X represents the output voltage amplitude of the inverter. Σ δ represents the system and line equivalent reactance, and δ represents the power angle of the inverter and the system.

[0086] A small adjustment Δf = 0.2 Hz is applied to the inverter output frequency to make the power angle increment Δδ = 2°;

[0087] Power response calculation: Define the power angle-power change rate η:

[0088] (12)

[0089] Island Criterion:

[0090] Grid-connected state: The inverter is constrained by system synchronization, δ is stable, P out As δ changes significantly, η becomes relatively large;

[0091] Islanding state: The inverter loses system synchronization, delta drifts freely, and P... out There is no stable response, and η is extremely small.

[0092] Set threshold η set =0.5kW / °, the island detection criterion is:

[0093] (13)

[0094] This method is highly reliable, with a response time of <0.1s, and can be reliably integrated with the fault diagnosis process.

[0095] Simulation verification

[0096] To fully verify the correctness, reliability, and practicality of the proposed permanent fault identification method in distribution networks with a high proportion of renewable energy integration, a distribution network simulation model closely resembling actual engineering scenarios was built based on the PSCAD / EMTDC electromagnetic transient simulation platform. By setting typical fault conditions, the consistency between the fault identification results and theoretical criteria was compared and analyzed to verify the effectiveness of the method. To verify the correctness of the proposed permanent fault identification method, a simulation model of a distribution network closely resembling actual engineering scenarios was built in PSCAD. Figure 3 The model shown is a distribution network system with a high proportion of renewable energy integration. The model parameters are as follows:

[0097] Positive sequence resistance: 0.03206 Ω / km;

[0098] Positive sequence reactance: 0.36 Ω / km;

[0099] Positive sequence susceptance: 3.7 × 10−6 S / km;

[0100] Zero-sequence resistance: 0.096 Ω / km;

[0101] Zero-sequence reactance: 1.5 Ω / km;

[0102] Zero-sequence susceptance: 3 × 10−6 S / km;

[0103] The lengths of L1, L2, and L3 are 10km, 10km, and 5km, respectively.

[0104] This invention utilizes the Matlab / Simulink electromagnetic transient simulation platform to conduct simulation experiments on transient and permanent line faults. The simulation scenario is set as a two-phase fault (A and B) occurring in the middle section of line L2. After the relay protection completes the fault trip, a specific excitation signal is injected into the fault line port after a 0.5s delay.

[0105] The simulated waveforms for transient fault conditions and permanent fault conditions are as follows: Figure 4 and Figure 5 As shown in the simulation results, after injecting a 500Hz positive sequence voltage excitation at the fault end, the induced response current of the line is dominated by the positive sequence component, while the negative sequence component has an extremely low amplitude and can be almost ignored. The ratio of the negative sequence component to the positive sequence component of the current after the fault approaches zero.

[0106] Simulation waveforms under permanent fault conditions are as follows: Figure 4 As shown, under the same 500Hz positive sequence voltage injection condition, the amplitude content of the positive sequence component and the negative sequence component in the fault response current are basically the same, and the amplitude ratio of the negative sequence component to the positive sequence component is approximately 1.

[0107] The distance between fault f1 and bus A and the fault type were continuously adjusted. The simulation results are shown in Table 1.

[0108] Table 1 Simulation results for different fault distances and fault types

[0109] 3km from busbar A AB phase transient fault 0.000032 Transient 5km from busbar A AB phase transient fault 0.000032 Transient 8km from busbar A AB phase transient fault 0.000032 Transient 3km from busbar A AB phase-to-phase permanent fault 1.044802 Permanent 5km from busbar A Permanent fault between phases AB 1.062842 Permanent 8km from busbar A Permanent fault between phases AB 1.096878 Permanent

[0110] This invention addresses the problems of traditional reclosing methods, such as the inability to accurately distinguish between transient and permanent faults, the potential for secondary impacts after a high proportion of renewable energy integration, inverter grid disconnection, and phase mismatch in islanded grid connection. It employs active high-frequency detection for fault type identification and islanding detection. After a fault trip delay, a 500Hz high-frequency signal is injected by the distributed power inverter. After filtering and sequence component decomposition, the fault nature is determined by the high-frequency negative-sequence / positive-sequence current ratio. The inverter frequency is then fine-tuned, and islanding detection is completed based on the power angle-power change rate. Ultimately, this achieves adaptive control that blocks reclosing for permanent faults and allows reclosing for transient faults in non-islanded states.

[0111] Verified by electromagnetic transient simulation in Matlab / Simulink, this method can identify the nature of faults under phase-to-phase transient and permanent fault conditions at different line locations, according to a set threshold. The identification results are consistent with the theoretical criteria and can be adapted to the operating characteristics of new energy inverters, reducing the risk of secondary impact from fault reclosing, equipment damage, and system instability.

[0112] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent transformations or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for active detection-based phase-to-phase fault identification and reclosing in an active distribution network, characterized in that, include: A three-phase sinusoidal high-frequency detection signal is injected into the faulty line, and the high-frequency component is filtered and extracted. Extract the positive-sequence, negative-sequence, and zero-sequence current components from the filtered three-phase current; Based on the ratio of the amplitude of negative sequence current to positive sequence current, a threshold is set to distinguish between transient and permanent faults. Based on the fault nature determination results, reclosing logic is constructed to complete the reclosing determination.

2. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 1, characterized in that, The three-phase sinusoidal high-frequency detection signal is selected at 500Hz, with an amplitude of 0.1 times the rated voltage, that is: ； In the formula: U N This refers to the rated phase voltage amplitude of the line. , , These are the three-phase sinusoidal high-frequency detection signal voltages.

3. The method for phase-to-phase fault identification and reclosing in an active distribution network based on active detection, as described in claim 1, is characterized in that... When filtering and extracting high-frequency components, a 500Hz narrowband bandpass filter (BPF) is used to extract high-frequency components and filter out power frequency, harmonics and noise.

4. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 1, characterized in that, When extracting the positive-sequence, negative-sequence, and zero-sequence current components from the filtered three-phase current, the symmetrical component method is used, and the transformation matrix is: , Let be the rotation factor, satisfying a 3 =1、1+a+a 2 =0.

5. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 4, characterized in that, Substituting the 500Hz three-phase current into the transformation matrix of the symmetrical component method, we obtain the formula for calculating the high-frequency sequence current: ； in, , , These are the positive-sequence, negative-sequence, and zero-sequence current components, respectively. These are the filtered currents for phases a, b, and c, respectively.

6. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 1, characterized in that, The fault nature distinction between transient and permanent faults is as follows: The fault characteristic coefficient k is defined as the ratio of the amplitude of the high-frequency negative sequence current to the amplitude of the positive sequence current. Transient fault: The fault arc is extinguished, the line resumes three-phase symmetrical operation, and no negative sequence component is generated under high-frequency signal excitation, i.e. ; Permanent fault: An asymmetrical low-impedance path persists at the fault point, generating a significant negative sequence current under high-frequency signal excitation, i.e. ; Set threshold k set The fault diagnosis criteria are: .

7. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 1, characterized in that, The reclosing logic is as follows: For permanent faults, the fault point persists, reclosing is blocked, and an alarm is uploaded. For transient faults, once the fault disappears, a reclosing operation is performed.

8. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 1, characterized in that, After fault diagnosis is completed, the following steps are performed: fine-tuning the inverter output frequency, increasing the power angle, collecting the output power response, and calculating the power angle-power change rate, thereby completing the islanding detection.

9. The active detection-based method for phase-to-phase fault identification and reclosing in an active distribution network according to claim 8, characterized in that, Island detection specifically refers to: The output active power satisfies the power angle characteristic equation: ； In the formula: U S U represents the system-side voltage amplitude. DG X represents the output voltage amplitude of the inverter. Σ δ represents the equivalent reactance of the system and the line, and δ represents the power angle of the inverter and the system. A small adjustment Δf is applied to the inverter output frequency to increase the power angle by Δδ; the power angle-power change rate η is defined as follows: ,in, These represent the active power output of the inverter before and after frequency fine-tuning, respectively. These are the power angles of the inverter before and after frequency fine-tuning, respectively. Set threshold η set The criteria for island detection are: .

10. A system based on the active detection-based phase-to-phase fault identification and reclosing method for active distribution networks as described in any one of claims 1 to 9, characterized in that, include: The signal injection module is used to inject a three-phase sinusoidal high-frequency detection signal into the faulty line and filter and extract the high-frequency components. The current component acquisition module is used to extract the positive sequence, negative sequence, and zero sequence current components from the filtered three-phase current. The fault nature discrimination module is used to distinguish between transient and permanent faults by setting a threshold based on the amplitude ratio of negative sequence current to positive sequence current. The reclosing determination module is used to construct reclosing logic based on the fault nature judgment result to complete the reclosing determination.

Images