Synchronous phasor measurement type power distribution automation device and power distribution implementation method
By combining the synchronous phasor measurement distribution automation device with Beidou timing technology, the problems of complex setting and long fault isolation time of existing distribution network automation protection devices have been solved, and high-precision fault identification and second-level self-healing have been achieved.
Patent Information
- Application Number
- CN202511049487.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-29
- Publication Date
- 2025-10-03
- Estimated Expiration
- 2045-07-29
AI Technical Summary
Existing distribution network automation protection devices rely on centralized protection at the master station, which is complex to set up, prone to over-tripping, long fault isolation time, low single-phase grounding identification accuracy, and unable to achieve large-scale synchronous monitoring and strategic control.
The use of synchronized phasor measurement distribution automation equipment, combined with Beidou timing technology, enables high-precision, large-scale synchronous power grid monitoring. Utilizing improved transient power direction protection algorithms and neighborhood protection algorithms, the protection action thresholds are dynamically optimized to achieve precise fault location and self-healing within seconds.
It achieves high-precision fault identification and rapid isolation, reduces fault isolation time, improves the accuracy of single-phase grounding identification, and realizes distributed fault location and self-healing in seconds without relying on the master station.
Smart Images

Figure CN120750019A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power distribution automation, and in particular to a synchronous phasor measurement type power distribution automation device and a power distribution implementation method. Background Art
[0002] With the development of social productivity and industrial production, the requirements for power grid reliability are becoming increasingly higher. The extensive application of photovoltaic power generation has also changed the traditional single-source power supply structure of the distribution network. The development of technologies such as the Internet of Things, artificial intelligence, and big data has made automated fault handling in the power grid field possible. There is an urgent need for a new distribution automation device with high-precision, large-scale synchronous power grid monitoring, fast and accurate fault identification, isolation and self-healing to match the requirements of the new power grid, and at the same time provide data accumulation for fault handling based on artificial intelligence and big data automation.
[0003] The existing distribution network mainly uses the FTU feeder automation protection device of the distribution automation master station, which realizes the automatic protection of the distribution network based on the traditional three-stage overcurrent protection + reclosing mechanism. It has telesignaling, remote control, and telemetering functions, and has high requirements on the communication quality and processing capabilities of the automation master station. Its protection strategy relies on the master station timing and centralized analysis and judgment. The setting is complex, and it is easy to cause over-tripping. The fault isolation time is long, the single-phase grounding identification accuracy is low, and it is impossible to achieve large-scale synchronous monitoring and strategy control.
[0004] The existing FTU feeder automation protection device consists of a circuit breaker body, power supply voltage transformer, voltage transformer / sensor, current transformer / sensor, FTU controller, connecting cables, etc. It is mainly based on the master station centralized protection logic. Its system structure is as follows Figure 1 As shown in the figure, for typical overcurrent and ground fault types in the power grid, different overcurrent / ground fault protection values and delay levels are set for the upstream and downstream FTUs of the line to provide automated protection. When a fault occurs, multiple reclosing methods are used to isolate the fault.
[0005] The feeder automation protection process is as follows:
[0006] The first step is to initiate fault handling. Upon receiving the first fault current signal, the real-time server initiates the fault analysis preprocessing process. It then waits for a configurable delay to collect all fault signals and switch trip signals along the feeder, and waits for the circuit breakers along the line to reclose. After the signal collection period, it begins analyzing the collected signals, deleting erroneous signals, such as when a switch is closed but reports an overcurrent signal without tripping. The following conditions are evaluated to determine whether to initiate fault handling. If no corresponding trip position change information is available, the analysis and recovery process ceases.
[0007] Tripping must be accompanied by a corresponding fault signal, otherwise the fault analysis will be ignored.
[0008] The second step is to identify the last faulty element (switch) on the faulty feeder (along the power supply path before the fault). The area below this switch is the fault zone. If there is only one fault signal, the fault zone is definitely outside this point. If there are multiple fault signals, the last one in the power supply direction before the fault is searched for. To ensure processing safety, multi-branch faults (different branches of the same feeder reporting unrelated overcurrent signals, making it impossible to determine the last overcurrent switch) will directly exit the automatic analysis process, and a corresponding system event will be generated to notify the user.
[0009] The third step is to determine the nature of the fault. The nature of the fault is determined based on whether the main fault location is directly connected to the busbar. If it is directly connected to the busbar (not through the disconnecting device), it can be determined that it is a (busbar) fault in the ring network cabinet or switch station. If the fault point is not directly connected to the busbar, determine whether it is a load fault or a line fault. The fault area should be a closed area surrounded by several switch knives (including the switch where the end overcurrent signal is located) and their points within the fault area. Starting from the determined end fault overcurrent position, search downward (in the power supply direction) for the switch knife with the collected overcurrent signal. If found, the point is a boundary point. If a switch or knife switch with a status of "off" is encountered, it is also added to the queue as a boundary point.
[0010] The fourth step is to isolate the fault. Fault isolation is based on the previously determined fault area. If the fault area remains unchanged, the disconnection steps are given in sequence, and the total power outage capacity in the fault area is calculated.
[0011] Step 5: Restore power to the non-faulty section. This operation is divided into two steps:
[0012] (1) Restore power to the main supply section (the fault has been isolated). At this time, close the relevant components in the trip list (if the trip component is a terminal overcurrent component, this step is not required);
[0013] (2) Restoring the restorable area below the fault point. This part requires first calculating whether the candidate points of each restorable area have sufficient capacity, then dividing the fault area into segments according to the situation, and repeating the calculation. Finally, a plan for closing the circuit breaker and restoring power supply is given, including the restoration of multiple sub-areas.
[0014] Traditional master station centralized feeder protection requires high-quality communication channels and computer master stations, which requires large investments and involves a wide range of complex projects. In particular, the setting is complex and prone to over-tripping. The line experiences multiple high-load shocks during the reclosing process, and the fault isolation time is long. Summary of the Invention
[0015] The technical problem to be solved by the present invention is to provide a synchronous phasor measurement type distribution automation device and a distribution implementation method to solve the deficiencies of the existing technology described in the background technology.
[0016] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:
[0017] A synchronous phasor measurement type power distribution automation device; including a circuit board, the circuit board is divided into a core board, a main board, an acquisition board, an analog signal sampling board, a key display board, a power module, a residual voltage board, a power supply voltage measurement switching board, an indicator light board and a PMU board;
[0018] The core board circuit includes a main control chip, a FLASH chip, an SRAM, an RTC circuit, an encryption chip, a 24C02 memory, a temperature measurement circuit and a parallel port drive circuit. The interface of the core board includes I2C communication, SPI communication, UART communication, parallel port communication and an I / O interface;
[0019] The mainboard includes a core board interface, a system power circuit, a switch remote control circuit, a switch remote signaling circuit, a power module remote control remote signaling circuit, an energy storage current, an operating current, a battery voltage, and an operating voltage measurement circuit, a communication network port, a maintenance network port, a 4G module serial port communication, a PMU communication serial port, a reserved 232 communication interface and a reserved serial port communication, an external indicator light drive circuit, and the mainboard is connected to the key display board through a soft connection;
[0020] The mainboard and the analog signal sampling board are connected by a flexible cable, which provides a 5V power supply to the analog signal sampling board. The analog signal sampling board sends the voltage and current measurement signals to the AD7616 interface of the mainboard through the flexible cable. The mainboard also includes a line loss module drive circuit, a frequency measurement circuit, a GPS, a Bluetooth circuit, and a communication interface circuit of the PMU board.
[0021] The acquisition board includes a primary voltage measurement circuit, a secondary voltage measurement circuit, a primary current measurement circuit, and a zero-voltage zero-current measurement circuit, wherein the primary voltage measurement circuit is compatible with the electromagnetic FTU voltage measurement input, and the primary current measurement circuit is compatible with the electromagnetic FTU current measurement input; the primary voltage measurement circuit, the secondary voltage measurement circuit, the primary current measurement circuit, and the zero-voltage zero-current measurement circuit are all electromagnetic acquisition transformers, and the voltage and current measurement signals output by the electromagnetic acquisition transformers are sent to the op amp follower circuit, and each measurement signal is divided into two mutually non-interfering measurement signals by the follower, and respectively sent to the main board and the PMU board through flexible cables;
[0022] The acquisition board also includes a ±15V power supply circuit. The 5V power from the mainboard is sent to the Goldensun A0515S-1WR3 power module, and the ±15V power output by the module provides working power for the on-board operational amplifier.
[0023] Power module, the structure is compatible with electromagnetic power module and capacitor power module;
[0024] Residual pressure board, the residual pressure board interface is compatible with electromagnetic FTU, electronic FTU and deep fusion FTU;
[0025] The PMU board, based on Beidou synchronization technology, uses 12.8 kHz high-frequency sampling and 0.5-level measurement accuracy to continuously sample and record the current and voltage at the measurement point, and time-tags each sampling point with a 1μs synchronization accuracy with the Universal Time Coordinated (UTC).
[0026] Optionally, the PMU board includes:
[0027] Beidou / GPS module, used to provide high-precision time synchronization and geographic location information;
[0028] AD conversion module, used to convert the analog quantity of the power grid into high-precision digital signals;
[0029] Vector measurement module, used to measure the synchronized phasors, frequency and frequency change rate of the power grid;
[0030] Data communication unit, used to transmit measurement data to the master station or other terminals in real time.
[0031] Optionally, the power supply voltage measurement switching board adopts an independent module design, and the circuit is divided into an AC 220V voltage measurement circuit and a two-way power supply switching circuit.
[0032] Optionally, the key-controlled display panel includes an LCD adapter board, a 4G module communication interface, a maintenance serial port interface, a maintenance network port interface, a remote local switch, a fault throw-in switch, a centralized local switch, and LCD function buttons.
[0033] A power distribution implementation method, using any of the above-mentioned synchronized phasor measurement type power distribution automation devices, is used to implement single-phase ground fault identification protection, comprising the following steps:
[0034] Step S1: Collect zero-sequence current data of each monitoring point on the line, and synchronize the timestamp based on Beidou / GPS timing alignment;
[0035] Step S2: According to the zero sequence current Amplitude classification fault type:
[0036] Strong characteristic fault: ,Adopt an improved transient power directional protection algorithm and combine the local voltage and current phase difference to determine the fault direction;
[0037] Weak feature failure: , obtain zero-sequence current waveform data of adjacent terminals through neighborhood protection, calculate waveform similarity and comprehensively judge the fault section;
[0038] Very weak characteristic fault: , based on the waveform similarity analysis of multi-point synchronous zero-sequence current data, the grounding section is determined;
[0039] Step S3: For strong characteristic faults, a zero-sequence overcurrent algorithm with adaptive threshold adjustment is used to dynamically optimize the protection action threshold;
[0040] Step S4: For weak and very weak characteristic faults, generate protection output action instructions, drive the circuit breaker to isolate the fault section, and upload the fault waveform and positioning results to the master station.
[0041] Optionally, the improved transient power direction protection algorithm in step S2 includes: extracting the zero-sequence voltage and current transient components within 1 / 4 cycle after the fault, calculating the instantaneous power direction; if the power direction is continuously reversed and the duration is ≥5ms, it is determined to be an internal fault, specifically:
[0042] Using db4 wavelet to analyze zero sequence voltage Perform 3-layer decomposition and calculate the maximum modulus of detail coefficients at each scale. When the maximum modulus of a certain scale exceeds 3 Determine the fault starting time ;in, is the instantaneous value of zero-sequence voltage at time t, is the standard deviation of the zero-sequence voltage noise during normal operation;
[0043] extract Zero-sequence voltage within the time window and zero sequence current , the transient component is separated by a 4th-order Butterworth high-pass filter and ,in: is the instantaneous value of zero-sequence current at time t, is the transient zero-sequence voltage component, is the transient zero-sequence current component;
[0044] Calculate instantaneous power , define the power direction function:
[0045]
[0046] represents the power direction determination result at time t;
[0047] Statistics within a 10ms sliding time window =-1 duration , if both ≥5ms and >0.5A, it is determined to be an internal fault; is the duration of power direction reversal.
[0048] Optionally, the waveform similarity calculation in step S2 includes:
[0049] Normalize the zero-sequence current waveforms of adjacent terminals; extract the zero-sequence current waveform data of adjacent terminals within one cycle after the fault occurs, and define:
[0050] represents the instantaneous value of zero-sequence current of the ith terminal at the nth sampling point;
[0051] represents the instantaneous value of zero-sequence current of the jth terminal adjacent to terminal i at the nth sampling point;
[0052] N is the total number of sampling points collected for each terminal, which is 128 points;
[0053] Preprocess the waveform and calculate the waveform mean , remove the DC component;
[0054] After normalization, we get , scale the amplitude to the interval [-1, 1];
[0055] Similarity is calculated using the mutual correlation coefficient method :
[0056]
[0057] in:
[0058] , is the mean of the normalized waveform;
[0059] When there is a sampling delay, the dynamic time warping (DTW) algorithm is used:
[0060]
[0061] in, is the end point value of the cumulative distance matrix, which is calculated by dynamic programming;
[0062] is the normalization factor, and the similarity range is [0, 1];
[0063] Calculate the waveform similarity between poles. For adjacent terminals m, n, k distributed along the line, if the similarity between pole m and pole n is ≥0.9, and the similarity between rod n and rod k ≤0.2, the fault is determined to be between rods n and k.
[0064] Optionally, step S3 adopts a zero-sequence overcurrent algorithm with adaptive threshold adjustment to dynamically optimize the protection action threshold, specifically:
[0065] Real-time acquisition of zero-sequence current after a strong characteristic fault occurs Data, sampling frequency is not less than 1000Hz;
[0066] Calculate the maximum value of zero-sequence current in one power frequency cycle ,average value , and the zero-sequence current change rate ,in is the sampling interval;
[0067] Using the formula Calculate dynamic protection action threshold ,in, 、 、 is the weight coefficient, satisfying + + =1, and dynamically adjusted according to the time t after the fault occurs: 50ms before the fault occurs, = 0.6, = 0.3 , = 0.1; 50ms after the fault occurs, the weight coefficient is adjusted in real time according to the fluctuation of zero-sequence current.
[0068] Optionally, the method further includes implementing feeder automation protection, specifically the following steps:
[0069] Step A1: Each feeder terminal monitors the line current / voltage in real time. If a fault characteristic is detected and reaches the neighborhood protection activation threshold, the neighborhood protection logic is triggered;
[0070] Step A2: The feeder terminal where the fault point is located collects the fault status of the adjacent feeder terminals through neighborhood communication, and calculates the fault status of the adjacent feeder terminals according to the 4G communication delay T 4G-L and local protection action time T LPL , Neighborhood protection action time T NPL , reclosing time T ER Dynamically select the protection action mode:
[0071] Case 1: If T NPL <T 4G-L <T LPL , then the neighborhood protection starts, the upstream neighboring switch of the fault point is quickly opened, and the branch switch reserves time for action;
[0072] Case 2: If T LPL <T 4G-L < T ER , then the local protection trips and verifies the correctness of the trip through neighborhood communication before reclosing. If there is an error, the error correction closing is triggered;
[0073] Case 3: If T 4G-L ≥ T ER Or communication is interrupted, it switches to local protection logic and performs the opening operation independently;
[0074] Among them, before reclosing, the status of the adjacent feeder terminal is checked. If the switch is not upstream of the fault, the opening command is cancelled and the closing delay time is ≤100ms;
[0075] Step A3: When the non-fault point feeder terminal detects the neighborhood protection start threshold but there is no local fault, it immediately sends a no-fault status to the adjacent feeder terminal to assist in fault section analysis;
[0076] Step A4: Based on the fault section location results, drive the upstream switch to open and isolate the fault, correct the error and close the downstream switch to restore power supply, and upload the event record to the master station.
[0077] Optionally, the fault section analysis logic in step A2 includes: if feeder terminal m detects a fault and feeder terminal n has no fault, the fault is located between feeder terminal m and feeder terminal n; if feeder terminal m and feeder terminal n both report a fault, the fault is located in the downstream section of feeder terminal n.
[0078] The beneficial effects of the above technical solution of the present invention are as follows:
[0079] 1) Realize wide-area synchronization, high-frequency, and high-precision sampling functions:
[0080] The Beidou time pulse signal is used to achieve clock synchronization accuracy of 1μs for all FXU device measurement data. The terminal sampling frequency is no less than 12.8kHz, that is, no less than 256 sampling points are synchronously sampled per cycle. High-precision synchronous sampling data can effectively improve the accuracy of fault diagnosis.
[0081] 2) FXU grounding detection performance improvement:
[0082] Based on the zero-sequence current method and the zero-sequence power direction method, a multi-point fault waveform comparison algorithm based on neighborhood communication has been added. At the moment of a fault, the upstream and downstream adjacent terminals of the fault point exchange data. While comparing the state variables, a multi-point waveform similarity comparison is also performed simultaneously at the moment of the fault. Based on the comparison of multi-point current and voltage data and waveform similarity, a comprehensive ground fault analysis is conducted, which can accurately identify single-phase ground faults with a ground current ≥1A.
[0083] 3) Achieve neighborhood protection independent of the master station, automatically completing accurate fault location, isolation, and self-healing within seconds:
[0084] The neighborhood protection function, based on 4G wireless communication, utilizes synchronized line status, current, and voltage data from adjacent terminals to identify faults within a specific area (including the protection zone between the switch at the current level and the adjacent downstream switch, and the FA zone between the switch at the current level and the adjacent upstream switch), locate and isolate the fault, and implement intelligent distributed feeder automation (FA). This distributed intelligent FA function, independent of master stations and trunk lines, achieves absolute selectivity, accurately minimizing isolation of faulty sections and restoring power to non-faulty sections within seconds. Neighborhood protection compensates for the selectivity limitations of overcurrent protection, enabling the nearest tripping of upstream and downstream switches at the fault point. BRIEF DESCRIPTION OF THE DRAWINGS
[0085] Figure 1 This is a structural diagram of a fully automatic FA system in the prior art;
[0086] Figure 2 This is a circuit diagram of the power distribution automation device of the present invention;
[0087] Figure 3 This is a basic functional block diagram of the wide area synchronous measurement unit of the present invention;
[0088] Figure 4 This is an example diagram of the single-phase grounding protection identification method of the present invention;
[0089] Figure 5 This is an example effect diagram of the single-phase grounding protection identification method of the present invention. DETAILED DESCRIPTION
[0090] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.
[0091] The device assembly structure of the present invention is as follows:
[0092] The device adopts a double-door structure. The first door seals the outside of the chassis, with a sealing strip on the inside and a lightning logo, product name and factory logo printed on the outside.
[0093] The second door houses the operating buttons and display. It features a 2.4-inch OLED display, LCD function buttons, a power start / stop button, status indicators, a protective pressure plate, a power switch, and manual opening and closing controls. An acrylic baffle is installed on the inside of the second door where the display is located, allowing the display to be viewed from the outside. This design effectively prevents static electricity and damage to the display from hard objects.
[0094] The aviation plugs (network port aviation plug, power aviation plug, current aviation plug, 14-core aviation plug) are designed with independent mounting plates. The aviation plug plates can be replaced separately to increase or decrease functional requirements according to different project requirements.
[0095] Antenna Installation: The FXU includes a GPS / BDS antenna on the mainboard, a 4G antenna on the master communication module, and a 4G antenna on the PMU board. The antennas are attached to the outside of the chassis through the waterproof locks at the bottom. The 4G antenna is secured to the bottom of the chassis with a bracket. The GPS antenna is magnetic. Once the chassis is installed on site, attach the GPS antenna to the angle iron securing the chassis, with the receiving surface facing upward.
[0096] Indicator light installation: The indicator light board is installed on the inside of the bottom of the chassis and connected to the mainboard through a flexible cable; a lampshade is installed at the indicator light hole position on the outside of the bottom of the chassis for waterproofing and anti-static purposes.
[0097] The mainboard, signal measurement board, power supply voltage measurement switching board, power module, and PMU board are designed as independently installed components. This circuitry can be pre-assembled and tested before being installed inside the chassis, improving assembly and testing efficiency.
[0098] The chassis has a reserved mounting area for the BDCU board, which can be installed according to project requirements. When installing the BDCU board, the aviation plug-in baseplate must be replaced with one with waterproof locks and antenna mounting brackets. The BDCU's two 4G antennas are connected to the chassis exterior through waterproof locks and secured to the aviation plug-in baseplate at the bottom of the chassis.
[0099] like Figure 2 As shown, the circuit structure of the synchronous phasor measurement type distribution automation device of the present invention is as follows:
[0100] Circuit board: The circuit board is divided into core board, main board, acquisition board, key display board, power module, residual pressure board, power supply voltage measurement switching board, indicator board and PMU board (BDCU board can also be expanded).
[0101] Core board: The core board circuit includes the main control chip, FLASH chip, SRAM, RTC circuit, encryption chip, 24C02 memory, temperature measurement circuit and parallel port driver circuit. The core board's interfaces include I2C communication, SPI communication, UART communication, parallel port communication and I / O interface.
[0102] Mainboard: The mainboard circuit includes the core board interface, system power circuit, switch remote control circuit, switch remote signaling circuit, power module remote control and signaling circuit, energy storage current, operating current, battery voltage, and operating voltage measurement circuits, two network communication circuits (communication network port and maintenance network port), 4G module serial communication port, PMU communication serial port, one reserved 232 communication interface and one reserved serial communication port (the reserved communication interface is obtained by expanding the core board's SPI3 through the WK2132 chip), and external indicator light drive circuit. The mainboard is connected to the key display board via a flexible connector.
[0103] A flexible cable connects the mainboard to the analog signal sampling board, providing 5V power to the board. The sampling board also transmits voltage and current measurement signals to the mainboard's AD7616 interface via the flexible cable. The mainboard also includes circuits for the line loss module driver, frequency measurement, GPS, Bluetooth, and the PMU board's communication interface.
[0104] The acquisition board includes a primary voltage measurement circuit, a secondary voltage measurement circuit, a primary current measurement circuit, and a zero-voltage, zero-current measurement circuit. The primary voltage measurement circuit is compatible with the voltage measurement input of an electromagnetic FTU, and the primary current measurement circuit is compatible with the current measurement input of an electromagnetic FTU. The PCB packaging design is compatible with electromagnetic acquisition transformers and electronic and deep-integrated acquisition transformers.
[0105] The voltage and current measurement signals output by the transformer are sent to the op amp follower circuit. The follower divides each measurement signal into two mutually independent measurement signals, which are then sent to the main board and PMU board respectively through flexible cables.
[0106] The acquisition board also includes a ±15V power supply circuit. The 5V power from the mainboard is sent to the Jinshengyang A0515S-1WR3 power module, and the ±15V power output by the module provides working power for the on-board operational amplifier.
[0107] Key-operated display panel: The key-operated display panel includes an LCD adapter board, a 4G module communication interface (DB9 port), a maintenance serial port, a maintenance network port, a remote local switch, a fault drop / drop switch, a centralized local switch, and LCD function buttons. As needed, in addition to the six standard indicators ("power," "communication status," "self-test fault," "conventional protection," "local FA," and "centralized FA"), eight new indicators are added: "open position remote signaling," "closed position remote signaling," "energy storage remote signaling," "lockout," "run," "overcurrent alarm," "zero current alarm," and "ground alarm."
[0108] Power Module: The structure is compatible with electromagnetic power modules and capacitor-type power modules. The power modules are purchased externally, and the mainboard power interface and control interface are designed to be functionally compatible. Different models and manufacturers of power modules can be selected according to project requirements.
[0109] Residual pressure board: The residual pressure board interface is compatible with electromagnetic FTU, electronic FTU and deep fusion FTU.
[0110] Power supply voltage measurement switching board: adopts independent module design, the circuit is divided into AC 220V voltage measurement circuit and two-way power supply switching circuit (electromagnetic type).
[0111] The circuit board package adopts a structural compatibility form of the capacitance measurement board (deep fusion) and the power supply voltage measurement switching board (electromagnetic);
[0112] After the power supply voltage passes through a measurement transformer (the capacitor power supply of the Deep Fusion FTU uses an isolated op amp to measure the power supply voltage), it is sent to a voltage follower. The follower splits the signal into two independent measurement signals, which are then sent to the signal interfaces of the mainboard and the PMU. Simultaneously, the mainboard sends 5V power via a flexible cable to the power supply voltage measurement switching board. A DC chip converts +5V to -5V, providing ±5V power to the voltage follower.
[0113] PMU board: Based on Beidou synchronization technology, it adopts 12.8 kHz high-frequency sampling and 0.5-level measurement accuracy to continuously sample and record the current and voltage at the measurement point, and marks each sampling point with a time tag synchronized with the Universal Time (UTC) with an accuracy of 1 μs.
[0114] In this embodiment, the PMU board includes:
[0115] Beidou / GPS module, used to provide high-precision time synchronization and geographic location information;
[0116] AD conversion module, used to convert the analog quantity of the power grid into high-precision digital signals;
[0117] Vector measurement module, used to measure the synchronized phasors, frequency and frequency change rate of the power grid;
[0118] Data communication unit, used to transmit measurement data to the master station or other terminals in real time.
[0119] The PMU performs high-precision, large-scale synchronous sampling, as follows: the Beidou high-precision synchronization module and wireless lateral communication module are integrated inside the distribution network feeder automation terminal, so that devices installed at any position on the line can be accurately synchronized to the Beidou satellite time. The time difference between the Beidou second pulse and the sampling pulse is ≤1μs, the error of the fundamental voltage phase angle measurement is ≤0.2°, and the error of the fundamental current phase measurement is ≤0.5°. It has a recording function, and the recording time scale accuracy adopts GPS or Beidou time scale. Under the rated value, the recording error of zero-sequence voltage and zero-sequence current is no more than 10us, and the recording error of the same measurement channel on different terminals is no more than 10us.
[0120] At the same time, high bandwidth, high precision, and high frequency sampling are also key technologies for wide-area synchronous measurement devices in distribution networks. Grounding algorithms and neighborhood protection algorithms are deployed in wide-area synchronous measurement units in distribution networks to achieve fast and accurate switch action. Figure 3 shown.
[0121] Based on Beidou's high-precision synchronization technology, the synchronized phasor measurement distribution automation device supports 12.8 kHz high-frequency sampling and 0.5-level measurement accuracy, continuously sampling and recording current and voltage at measurement points. Furthermore, utilizing the built-in Beidou satellite navigation module and its own unique algorithm, it time-tags each sampling point with 1μs synchronization accuracy to UTC (Universal Time Coordinated), enabling high-precision synchronized data collection in the neighborhood and ensuring accurate and reliable fault analysis results.
[0122] Through the PMU and Beidou synchronized clock, comprehensive and accurate monitoring of the power grid can be carried out over a large area, and comprehensive visual management of the distribution network can be achieved.
[0123] In this embodiment, the key-controlled display panel includes an LCD adapter board, a 4G module communication interface, a maintenance serial port interface, a maintenance network port interface, a remote local switch, a fault drop-in / drop-out switch, a centralized local switch, and LCD function buttons.
[0124] A power distribution implementation method, using any of the above-mentioned synchronized phasor measurement type power distribution automation devices, is used to implement single-phase ground fault identification protection, comprising the following steps:
[0125] Step S1: Collect zero-sequence current data of each monitoring point on the line, and synchronize the timestamp based on Beidou / GPS timing alignment;
[0126] Step S2: According to the zero sequence current Amplitude classification fault type:
[0127] Strong characteristic fault: ,Adopt an improved transient power directional protection algorithm and combine the local voltage and current phase difference to determine the fault direction;
[0128] Weak feature failure: , obtain zero-sequence current waveform data of adjacent terminals through neighborhood protection, calculate waveform similarity and comprehensively judge the fault section;
[0129] Very weak characteristic fault: , based on the waveform similarity analysis of multi-point synchronous zero-sequence current data, the grounding section is determined;
[0130] Step S3: For strong characteristic faults, a zero-sequence overcurrent algorithm with adaptive threshold adjustment is used to dynamically optimize the protection action threshold;
[0131] Step S4: For weak and very weak characteristic faults, generate protection output action instructions, drive the circuit breaker to isolate the fault section, and upload the fault waveform and positioning results to the master station.
[0132] In this embodiment, the improved transient power direction protection algorithm in step S2 includes: extracting the zero-sequence voltage and current transient components within 1 / 4 cycle after the fault, calculating the instantaneous power direction; if the power direction is continuously reversed and the duration is ≥5ms, it is determined to be an internal fault, specifically:
[0133] Using db4 wavelet to analyze zero sequence voltage Perform 3-layer decomposition and calculate the maximum modulus of detail coefficients at each scale. When the maximum modulus of a certain scale exceeds 3 Determine the fault starting time ;in, is the instantaneous value of zero-sequence voltage at time t, is the standard deviation of the zero-sequence voltage noise during normal operation;
[0134] extract Zero-sequence voltage within the time window and zero sequence current , the transient component is separated by a 4th-order Butterworth high-pass filter and ,in: is the instantaneous value of zero-sequence current at time t, is the transient zero-sequence voltage component, is the transient zero-sequence current component;
[0135] Calculate instantaneous power , define the power direction function:
[0136]
[0137] represents the power direction determination result at time t;
[0138] Statistics within a 10ms sliding time window =-1 duration , if both ≥5ms and >0.5A, it is determined to be an internal fault; is the duration of power direction reversal.
[0139] In this embodiment, the waveform similarity calculation in step S2 includes:
[0140] Normalize the zero-sequence current waveforms of adjacent terminals; extract the zero-sequence current waveform data of adjacent terminals within one cycle after the fault occurs, and define:
[0141] represents the instantaneous value of zero-sequence current of the ith terminal at the nth sampling point;
[0142] represents the instantaneous value of zero-sequence current of the jth terminal adjacent to terminal i at the nth sampling point;
[0143] N is the total number of sampling points collected for each terminal, which is 128 points;
[0144] Preprocess the waveform and calculate the waveform mean , remove the DC component;
[0145] After normalization, we get , scale the amplitude to the interval [-1, 1];
[0146] Similarity is calculated using the mutual correlation coefficient method :
[0147]
[0148] in:
[0149] , is the mean of the normalized waveform;
[0150] When there is a sampling delay, the dynamic time warping (DTW) algorithm is used:
[0151]
[0152] in, is the end point value of the cumulative distance matrix, which is calculated by dynamic programming;
[0153] is the normalization factor, and the similarity range is [0, 1];
[0154] Calculate the waveform similarity between poles. For adjacent terminals m, n, k distributed along the line, if the similarity between pole m and pole n is ≥0.9, and the similarity between rod n and rod k ≤0.2, the fault is determined to be between rods n and k.
[0155] In this embodiment, step S3 adopts a zero-sequence overcurrent algorithm with adaptive threshold adjustment to dynamically optimize the protection action threshold, specifically:
[0156] Real-time acquisition of zero-sequence current after a strong characteristic fault occurs Data, sampling frequency is not less than 1000Hz;
[0157] Calculate the maximum value of zero-sequence current in one power frequency cycle ,average value , and the zero-sequence current change rate ,in is the sampling interval;
[0158] Using the formula Calculate dynamic protection action threshold ,in, 、 、 is the weight coefficient, satisfying + + =1, and dynamically adjusted according to the time t after the fault occurs: 50ms before the fault occurs, = 0.6, = 0.3 , = 0.1; 50ms after the fault occurs, the weight coefficient is adjusted in real time according to the fluctuation of zero-sequence current.
[0159] The single-phase grounding identification protection method is as follows:
[0160] Based on the zero-sequence overcurrent, phase current asymmetry method, and zero-sequence power direction method, a multi-point zero-sequence current waveform comparison method based on domain communication is introduced to effectively improve the accuracy of ground fault identification with weak characteristics, and achieve a 100% false alarm-free identification rate for ground faults with zero-sequence current above 1A.
[0161] For single-phase ground faults with strong characteristics (zero current > 3A, adjustable parameters), local power directional protection utilizes local measurement data. A transient power directional protection algorithm, improved based on massive field data, accurately reflects single-phase ground faults. Furthermore, a zero-sequence overcurrent algorithm based on adaptive threshold adjustment overcomes the threshold setting challenges of traditional solutions and is designed to address the increasingly frequent complex ground faults caused by out-of-phase ground shorts between lines.
[0162] For single-phase grounding faults with weak characteristics (zero current <3A, parameters can be adjusted), neighborhood protection can improve protection sensitivity by making comprehensive judgments based on the obtained adjacent terminal voltage and current data, accurately determine grounding faults with zero-sequence current not less than 1A in the protection section, and complete the protection output action.
[0163] When the grounding current is small enough to approach the inherent measurement error of the equipment, the error influence cannot be eliminated based on single-point data, and analysis can be performed based on multi-point synchronous zero-sequence current data.
[0164] like Figure 4 、 Figure 5 As shown, in a "12,000Ω" grounding test at a certain location, 3I_0 < 1A, making it impossible to locate the fault based on single-point data. The zero-sequence current similarity between poles 1 and 2 is close to 1, while the similarity between poles 2 and 3 is between -1 and 0. Based on this, it can be determined that the grounding section is between poles 2 and 3.
[0165] This embodiment also includes implementing feeder automatic protection, and the specific steps are as follows:
[0166] Step A1: Each feeder terminal monitors the line current / voltage in real time. If a fault characteristic is detected and reaches the neighborhood protection activation threshold, the neighborhood protection logic is triggered;
[0167] Step A2: The feeder terminal where the fault point is located collects the fault status of the adjacent feeder terminals through neighborhood communication, and calculates the fault status of the adjacent feeder terminals according to the 4G communication delay T 4G-L and local protection action time T LPL , Neighborhood protection action time T NPL , reclosing time T ER Dynamically select the protection action mode:
[0168] Case 1: If T NPL <T 4G-L <T LPL , then the neighborhood protection starts, the upstream neighboring switch of the fault point is quickly opened, and the branch switch reserves time for action;
[0169] Case 2: If T LPL <T 4G-L < T ER , then the local protection trips and verifies the correctness of the trip through neighborhood communication before reclosing. If there is an error, the error correction closing is triggered;
[0170] Case 3: If T 4G-L ≥ T ER Or communication is interrupted, it switches to local protection logic and performs the opening operation independently;
[0171] Among them, before reclosing, the status of the adjacent feeder terminal is checked. If the switch is not upstream of the fault, the opening command is cancelled and the closing delay time is ≤100ms;
[0172] Step A3: When the non-fault point feeder terminal detects the neighborhood protection start threshold but there is no local fault, it immediately sends a no-fault status to the adjacent feeder terminal to assist in fault section analysis;
[0173] Step A4: Based on the fault section location results, drive the upstream switch to open and isolate the fault, correct the error and close the downstream switch to restore power supply, and upload the event record to the master station.
[0174] In this embodiment, the fault section analysis logic in step A2 includes: if feeder terminal m detects a fault and feeder terminal n has no fault, the fault is located between feeder terminal m and feeder terminal n; if feeder terminal m and feeder terminal n both report a fault, the fault is located in the downstream section of feeder terminal n.
[0175] FA protection methods, specifically:
[0176] When a fault occurs, the feeder automation terminal FXU exchanges information with adjacent terminals based on 4G, and only reflects the fault within the protection area between adjacent terminals.
[0177] The system has high sampling synchronization accuracy, high sampling rate and sampling accuracy, so it can realize neighborhood protection based on 4G horizontal communication. At the same time, it integrates traditional fault analysis algorithms such as local zero-sequence power directional protection to accurately judge whether the fault is within the protection area. It has absolute selectivity and automatically implements upstream and downstream tripping of the fault point on site, isolates the fault section and subsequent self-healing control.
[0178] The implementation of neighborhood protection relies on two key factors: low-latency communication between the ends of a section and synchronized data measurement at each point. The former ensures quick protection action, while the latter is the basis for current differential calculation. Since both fiber-optic and 5G communications, for their own reasons, are not yet ready for large-scale deployment on overhead distribution lines, this device integrates 4G neighborhood protection with local overcurrent protection and automatic reclosing.
[0179] The core of this distributed intelligent protection and control technology, based on 4G communications, is the exchange of electrical quantity data between adjacent switches via 4G lateral communication. The switches' built-in current differential elements calculate the current longitudinal differential to determine whether the fault is within the switch's protection zone. This result is highly selective. Due to the potentially long latency of 4G communications, neighborhood protection cannot reliably serve as primary protection for phase-to-phase short-circuit faults. It must be combined with local overcurrent protection to adaptively apply either neighborhood protection or overcurrent protection to clear the fault.
[0180] For phase-to-phase short-circuit faults, the terminal completes the entire protection set action within 150ms using 4G, implementing distributed fast-acting FA, clearing the fault before the substation outlet circuit breaker protection (I-stage delay ≥ 200ms) operates. If the substation outlet circuit breaker overcurrent protection delay is short (I-stage delay ≤ 200ms) or the 4G communication delay is extended, distributed slow-acting FA logic is used to clear the fault after the outlet circuit breaker operates. For permanent faults, after clearing the upstream fault point, the terminal neighborhood protection isolates the faulty section downstream and controls the tie switch to transfer power, achieving self-healing within seconds.
[0181] During a short-circuit fault, the line voltage drops, so switches downstream of the fault point use the line voltage fluctuation to initiate operations such as recording, fault diagnosis, and neighborhood communication. Fault current exists upstream of the fault point, so switches upstream of the fault point are directly activated by the current fluctuation.
[0182] When a ground fault occurs, the zero-sequence voltage of the entire line will increase, so all switches use zero-sequence voltage fluctuations to start recording, fault diagnosis, neighborhood communication and other operations.
[0183] The FXU reaches the neighborhood protection start threshold and determines whether there is a short circuit or ground fault. It collects the fault status of adjacent FXUs through neighborhood communication, analyzes the fault area, quickly opens the neighboring switch upstream of the fault point, and corrects and closes other switches.
[0184] The fault point is between FXU1 and FXU2;
[0185] FXU1 detects a fault and receives a warning from FXU2 through neighboring communication that it is in a normal state. It determines that FXU1 is within the fault assessment boundary for neighboring protection. Based on the neighboring communication delay, the following three situations can be considered:
[0186] Case 1: T NPL <T 4G-L <T LPL ,
[0187] The neighborhood protection action time is to reserve time for branch switch action;
[0188] The neighborhood communication delay is faster than the local protection, so the neighborhood protection starts the tripping operation.
[0189] Case 2: T LPL <T 4G-L < T ER ,
[0190] The neighborhood communication delay is slower than the local protection, so the local protection takes action first to start the tripping operation;
[0191] Before the switch is reclosed, it is confirmed through neighborhood communication that this switch is the upstream neighbor switch of the fault point and the opening is correct.
[0192] Case 3: If T 4G-L ≥ T ER or communication is interrupted,
[0193] Neighborhood protection is exited and local protection logic is adopted.
[0194] FXU2 reaches the neighborhood protection activation threshold but does not detect a fault. FXU2 is determined to be outside the neighborhood protection fault analysis boundary and immediately sends the fault status to FXU1 through neighborhood communication.
[0195] FXU3 reaches the neighborhood protection activation threshold but does not detect a fault. FXU3 is determined to be outside the neighborhood protection fault analysis boundary and immediately sends the fault status to FXU2 through neighborhood communication.
[0196] The fault point is between FXU2 and FXU3;
[0197] FXU1 detects a fault and receives the fault status of FXU2 through neighbor communication. It determines that FXU1 is within the neighborhood protection fault assessment boundary. Based on the neighborhood communication delay, it can be divided into the following three situations:
[0198] Case 1: T NPL <T 4G-L <T LPL ,
[0199] The neighborhood protection action time is to reserve time for branch switch action;
[0200] The neighborhood communication delay is faster than the local protection, and it is determined that the fault area is not between FXU1 and FXU2. The switch is in the closed state, the state is correct, and the closing state is maintained.
[0201] Case 2: TLPL < T 4G-L <TER,
[0202] The neighborhood communication delay is slower than the local protection, so the local protection takes action first to start the tripping operation;
[0203] Before the switch is reclosed, it is confirmed through neighboring communication that the switch is not the upstream neighboring switch of the fault point. The opening error is corrected and the switch is closed.
[0204] Case 3: TER < T 4G-L or communication is interrupted,
[0205] Neighborhood protection is exited and local protection logic is adopted.
[0206] FXU2 detects a fault and receives a non-fault status from FXU3 via neighborhood communication. It determines that FXU2 is within the neighborhood protection fault assessment boundary. For details on the protection logic, see the FXU1 protection logic in "(I) The fault point is between FXU1 and FXU2." FXU2 then simultaneously sends the fault status to FXU1 via neighborhood communication.
[0207] FXU3 reaches the neighborhood protection activation threshold but does not detect a fault. FXU3 is determined to be outside the neighborhood protection fault analysis boundary and immediately sends the fault status to FXU2 through neighborhood communication.
[0208] The beneficial effects of the present invention include the following:
[0209] PMU high-precision network-wide monitoring:
[0210] (1) Through the BeiDou / GPS module, synchronous sampling and time synchronization are performed based on the Universal Time (UTC), so that the terminal sampling pulse and the second pulse are synchronized, making the synchronization error ≤ 1µs;
[0211] (2) The terminal’s fundamental current and voltage phase angle measurement error limit is 0.2°;
[0212] (3) The terminal sampling frequency is above 12.8kHz, that is, synchronous sampling is performed at more than 256 sampling points per cycle;
[0213] (4) The clock error of the recorded data time stamp shall not exceed 10μs.
[0214] Single-phase grounding identification algorithm:
[0215] By introducing a multi-point zero-sequence current waveform comparison method based on domain communication on the basis of zero-sequence overcurrent, phase current asymmetry method, and zero-sequence power direction method, the accuracy of identifying ground faults with weak characteristics is effectively improved, and the recognition rate of ground faults with zero-sequence current above 1A is achieved with 100% no false alarms.
[0216] Intelligent distributed FA and neighborhood communication:
[0217] (1) The terminal has a neighborhood protection function based on 4G wireless communication, which enables accurate determination and rapid isolation of fault sections, with a minimum action delay of ≤200ms. Power supply to non-fault sections is quickly restored, enabling self-healing of distribution network lines.
[0218] (2) In the event of a phase-to-phase short circuit fault, the terminal implements adaptive handling according to the delay configuration of different protections of the outgoing line switches in the station, completes the distributed fast-acting FA and slow-acting FA action logic respectively, correctly outputs the opening / closing signal, completes the isolation of the fault section and the restoration of power supply to the non-fault section, without false operation;
[0219] Fast-acting FA logic: The upstream sectionalizing switch trips to isolate the faulty section, while power is restored to the healthy upstream area. The downstream sectionalizing switch automatically trips to isolate the faulty section, and the tie switch closes to restore power to the healthy downstream area.
[0220] Slow-acting FA logic: The station's outgoing line switch trips to clear the fault. The upstream sectionalizing switch operates to isolate the fault point. The outgoing line switch successfully recloses, restoring power to the healthy upstream area. The downstream sectionalizing switch automatically opens to isolate the faulty section. The tie switch closes to resume power, restoring power to the healthy downstream area.
[0221] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.
Claims
1. A synchronous phasor measurement type distribution automation device, characterized in that: Including circuit boards, which are divided into core board, main board, acquisition board, analog signal sampling board, key display board, power module, residual voltage board, power supply voltage measurement switching board, indicator board and PMU board; The core board circuit includes a main control chip, a FLASH chip, an SRAM, an RTC circuit, an encryption chip, a 24C02 memory, a temperature measurement circuit and a parallel port drive circuit. The interface of the core board includes I2C communication, SPI communication, UART communication, parallel port communication and an I / O interface; The mainboard includes a core board interface, a system power circuit, a switch remote control circuit, a switch remote signaling circuit, a power module remote control remote signaling circuit, an energy storage current, an operating current, a battery voltage, and an operating voltage measurement circuit, a communication network port, a maintenance network port, a 4G module serial port communication, a PMU communication serial port, a reserved 232 communication interface and a reserved serial port communication, an external indicator light drive circuit, and the mainboard is connected to the key display board through a soft connection; The mainboard and the analog signal sampling board are connected by a flexible cable, which provides a 5V power supply to the analog signal sampling board. The analog signal sampling board sends the voltage and current measurement signals to the AD7616 interface of the mainboard through the flexible cable. The mainboard also includes a line loss module drive circuit, a frequency measurement circuit, a GPS, a Bluetooth circuit, and a communication interface circuit of the PMU board. The acquisition board includes a primary voltage measurement circuit, a secondary voltage measurement circuit, a primary current measurement circuit, and a zero-voltage zero-current measurement circuit, wherein the primary voltage measurement circuit is compatible with the electromagnetic FTU voltage measurement input, and the primary current measurement circuit is compatible with the electromagnetic FTU current measurement input; the primary voltage measurement circuit, the secondary voltage measurement circuit, the primary current measurement circuit, and the zero-voltage zero-current measurement circuit are all electromagnetic acquisition transformers, and the voltage and current measurement signals output by the electromagnetic acquisition transformers are sent to the op amp follower circuit, and each measurement signal is divided into two mutually non-interfering measurement signals by the follower, and respectively sent to the main board and the PMU board through flexible cables; The acquisition board also includes a ±15V power supply circuit. The 5V power from the mainboard is sent to the Goldensun A0515S-1WR3 power module, and the ±15V power output by the module provides working power for the on-board operational amplifier. Power module, the structure is compatible with electromagnetic power module and capacitor power module; Residual pressure board, the residual pressure board interface is compatible with electromagnetic FTU, electronic FTU and deep fusion FTU; The PMU board, based on Beidou synchronization technology, uses 12.8 kHz high-frequency sampling and 0.5-level measurement accuracy to continuously sample and record the current and voltage at the measurement point, and time-tags each sampling point with a 1μs synchronization accuracy with the Universal Time Coordinated (UTC).
2. The synchronized phasor measurement type distribution automation device according to claim 1, characterized in that: The PMU board includes: Beidou / GPS module, used to provide high-precision time synchronization and geographic location information; AD conversion module, used to convert the analog quantity of the power grid into high-precision digital signals; Vector measurement module, used to measure the synchronized phasors, frequency and frequency change rate of the power grid; Data communication unit, used to transmit measurement data to the master station or other terminals in real time.
3. The synchronized phasor measurement type distribution automation device according to claim 1, characterized in that: The power supply voltage measurement switching board adopts an independent module design, and the circuit is divided into an AC 220V voltage measurement circuit and a two-way power supply switching circuit.
4. The synchronized phasor measurement type distribution automation device according to claim 1, characterized in that: The key-controlled display panel includes an LCD adapter board, a 4G module communication interface, a maintenance serial port interface, a maintenance network port interface, a remote local switch, a fault throw-in / out switch, a centralized local switch, and LCD function buttons.
5. A power distribution implementation method, using the synchronized phasor measurement type power distribution automation device according to any one of claims 1 to 4, characterized in that: It is used to realize single-phase grounding identification protection, including the following steps: Step S1: Collect zero-sequence current data of each monitoring point on the line, and synchronize the timestamp based on Beidou / GPS timing alignment; Step S2: According to the zero sequence current Amplitude classification fault type: Strong characteristic fault: ,Adopt an improved transient power directional protection algorithm and combine the local voltage and current phase difference to determine the fault direction; Weak feature failure: , obtain zero-sequence current waveform data of adjacent terminals through neighborhood protection, calculate waveform similarity and comprehensively judge the fault section; Very weak characteristic fault: , based on the waveform similarity analysis of multi-point synchronous zero-sequence current data, the grounding section is determined; Step S3: For strong characteristic faults, a zero-sequence overcurrent algorithm with adaptive threshold adjustment is used to dynamically optimize the protection action threshold; Step S4: For weak and very weak characteristic faults, generate protection output action instructions, drive the circuit breaker to isolate the fault section, and upload the fault waveform and positioning results to the master station.
6. A power distribution implementation method according to claim 5, characterized in that: The improved transient power direction protection algorithm in step S2 includes: extracting the zero-sequence voltage and current transient components within 1 / 4 cycle after the fault, calculating the instantaneous power direction; if the power direction is continuously reversed and the duration is ≥5ms, it is determined to be an internal fault, specifically: Using db4 wavelet to analyze zero sequence voltage Perform 3-layer decomposition and calculate the maximum modulus of detail coefficients at each scale. When the maximum modulus of a certain scale exceeds 3 Determine the fault starting time ;in, is the instantaneous value of zero-sequence voltage at time t, is the standard deviation of the zero-sequence voltage noise during normal operation; extract Zero-sequence voltage within the time window and zero sequence current , the transient component is separated by a 4th-order Butterworth high-pass filter and ,in: is the instantaneous value of zero-sequence current at time t, is the transient zero-sequence voltage component, is the transient zero-sequence current component; Calculate instantaneous power , define the power direction function: ; represents the power direction determination result at time t; Statistics within a 10ms sliding time window = -1 duration , if both ≥5ms and >0.5A, it is determined to be an internal fault; is the duration of power direction reversal.
7. A power distribution implementation method according to claim 5, characterized in that: The waveform similarity calculation in step S2 includes: Normalize the zero-sequence current waveforms of adjacent terminals; extract the zero-sequence current waveform data of adjacent terminals within one cycle after the fault occurs, and define: represents the instantaneous value of zero-sequence current of the ith terminal at the nth sampling point; represents the instantaneous value of zero-sequence current of the jth terminal adjacent to terminal i at the nth sampling point; N is the total number of sampling points collected for each terminal, which is 128 points; Preprocess the waveform and calculate the waveform mean , remove the DC component; After normalization, we get , scale the amplitude to the interval [-1, 1]; Similarity is calculated using the mutual correlation coefficient method : ; in: , is the mean of the normalized waveform; When there is a sampling delay, the dynamic time warping (DTW) algorithm is used: ; in, is the end point value of the distance matrix, which is calculated by dynamic programming; is the normalization factor, and the similarity range is [0, 1]; Calculate the waveform similarity between poles. For adjacent terminals m, n, k distributed along the line, if the similarity between pole m and pole n is ≥0.9, and the similarity between rod n and rod k ≤0.2, the fault is determined to be between rods n and k.
8. A power distribution implementation method according to claim 5, characterized in that: The step S3 adopts a zero-sequence overcurrent algorithm with adaptive threshold adjustment to dynamically optimize the protection action threshold, specifically: Real-time acquisition of zero-sequence current after a strong characteristic fault occurs Data, sampling frequency is not less than 1000Hz; Calculate the maximum value of zero-sequence current in one power frequency cycle ,average value , and the zero-sequence current change rate ,in is the sampling interval; Using the formula Calculate dynamic protection action threshold ,in, 、 、 is the weight coefficient, satisfying + + =1, and dynamically adjusted according to the time t after the fault occurs: 50ms before the fault occurs, = 0.6, = 0.3 , = 0.1; 50ms after the fault occurs, the weight coefficient is adjusted in real time according to the fluctuation of zero-sequence current.
9. A power distribution implementation method according to claim 5, characterized in that: The method further includes implementing feeder automation protection, specifically the following steps: Step A1: Each feeder terminal monitors the line current / voltage in real time. If a fault characteristic is detected and reaches the neighborhood protection activation threshold, the neighborhood protection logic is triggered; Step A2: The feeder terminal where the fault point is located collects the fault status of the adjacent feeder terminals through neighborhood communication, and calculates the fault status of the adjacent feeder terminals according to the 4G communication delay T 4G-L and local protection action time T LPL , Neighborhood protection action time T NPL , reclosing time T ER Dynamically select the protection action mode: Case 1: If T NPL <T 4G-L <T LPL , then the neighborhood protection starts, the upstream neighboring switch of the fault point is quickly opened, and the branch switch reserves time for action; Case 2: If T LPL <T 4G-L < T ER , then the local protection trips and verifies the correctness of the trip through neighborhood communication before reclosing. If there is an error, the error correction closing is triggered; Case 3: If T 4G-L ≥ T ER Or communication is interrupted, it switches to local protection logic and performs the opening operation independently; Among them, before reclosing, the status of the adjacent feeder terminal is checked. If the switch is not upstream of the fault, the opening command is cancelled and the closing delay time is ≤100ms; Step A3: When the non-fault point feeder terminal detects the neighborhood protection start threshold but there is no local fault, it immediately sends a no-fault status to the adjacent feeder terminal to assist in fault section analysis; Step A4: Based on the fault section location results, drive the upstream switch to open and isolate the fault, correct the error and close the downstream switch to restore power supply, and upload the event record to the master station.
10. A power distribution implementation method according to claim 9, characterized in that: The fault section analysis logic in step A2 includes: if feeder terminal m detects a fault and feeder terminal n has no fault, the fault is located between feeder terminal m and feeder terminal n; if feeder terminal m and feeder terminal n both report a fault, the fault is located in the downstream section of feeder terminal n.
Citation Information
Patent Citations
Power distribution network single-phase earth fault locating method based on transient state
CN102944814A
Method for acquiring zero-sequence current transient starting point during single-phase earth fault of power distribution network
CN110907752A
Active power distribution network fault positioning method based on direct current dynamic time bending distance
CN112130027A
Fault studying and judging method and device based on edge computing FTU and fault indicator
CN114384374A
Single-phase earth fault feeder protection method based on wide-area synchronous measurement
CN115173377A
Cited By
Primary and secondary fusion type pole-mounted switch intelligent terminal
CN121395719A