Intelligent oscillograph cooperative relay protection multi-port fault positioning method, system and device and medium

Abstract

The invention relates to the technical field of oscillograph fault identification, and discloses an intelligent oscillograph cooperative relay protection multi-port fault positioning method and system, and the method comprises the steps: setting multiple triggering conditions and multiple sampling windows of an intelligent oscillograph, and setting relay protection as a three-section protection form; the intelligent oscillograph triggers sampling and sends a cooperative trigger signal, collected fault recording data and relay protection action information to relay protection, and time synchronization is carried out through a unified timestamp mark; and performing Fourier decomposition on the fault recording data with the timestamp mark, constructing a fault feature vector, and completing fault type identification through a fault classification model in combination with the relay protection action information. According to the method, the fault type can be identified more accurately, the fault point position can be calculated more accurately, sampling can be triggered quickly, fault information can be collected, the fault positioning time is shortened, the fault processing efficiency is improved, fault features can be analyzed more comprehensively, and complex fault types can be identified more comprehensively.

Description

Technical Field

[0001] The present invention relates to the technical field of oscilloscope fault identification, and in particular to a method and system for locating multi-port faults of an intelligent oscilloscope in coordination with relay protection. Background Art

[0002] As power systems continue to expand and distribution grids become increasingly complex, the accuracy and timeliness of power line fault location are crucial to ensuring safe and stable power system operation. Intelligent oscilloscopes can record waveform data of electrical quantities such as voltage and current during power system faults, while relay protection devices can quickly detect and isolate faults. Combining intelligent oscilloscopes with relay protection devices to achieve collaborative collection and analysis of fault data is an important research direction for improving fault location accuracy.

[0003] The intelligent recorder and relay protection device operate independently and lack an effective data coordination mechanism, which makes it difficult to accurately match fault recording data with protection action information, affecting the reliability of fault location. Existing fault location methods mainly rely on a single fault characteristic parameter and do not fully utilize the multi-dimensional characteristic information of the fault process. They are easily affected by changes in system operating conditions and fault types. The fault location algorithm lacks adaptive capabilities and cannot dynamically adjust the location strategy according to different fault types and protection action characteristics, which reduces the accuracy of fault location. Summary of the Invention

[0004] In view of the above-mentioned existing problems, the present invention is proposed. Therefore, the present invention provides a multi-port fault location method for intelligent oscilloscopes in coordination with relay protection to solve the problems of insufficient coordination between intelligent oscilloscopes and relay protection devices, incomplete fault feature extraction, and poor adaptability of fault location algorithms. It provides a multi-port fault location method for intelligent oscilloscopes in coordination with relay protection based on multi-dimensional feature analysis and adaptive location strategy to improve the accuracy and reliability of fault location.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a multi-port fault location method using an intelligent oscilloscope and coordinated relay protection, comprising: setting multiple trigger conditions and multiple sampling windows for the intelligent oscilloscope, and setting the relay protection to a three-stage protection form;

[0007] Setting a determination threshold of the multiple trigger conditions; if the determination threshold of the multiple trigger conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection, and the collected fault recording data and relay protection action information are synchronized through a unified timestamp mark;

[0008] The fault recording data with timestamp is subjected to Fourier decomposition to extract the fundamental component and harmonic component, and the fault feature vector is constructed. Combined with the relay protection action information, the fault type is identified through the fault classification model, and a fault analysis report is generated.

[0009] As a preferred solution of the multi-port fault location method of the intelligent oscilloscope cooperative relay protection described in the present invention, wherein: multiple trigger conditions and multiple sampling windows of the intelligent oscilloscope are set, and the relay protection is set to a three-stage protection form, including:

[0010] The multiple trigger conditions of the intelligent recorder include voltage amplitude trigger condition, current amplitude trigger condition, phasor mutation trigger condition and frequency deviation trigger condition;

[0011] The sampling window of the intelligent wave recorder includes a pre-trigger sampling section and a post-trigger sampling section;

[0012] The three-stage protection of the relay protection includes an instantaneous action stage, a delayed action stage and a backup protection stage.

[0013] As a preferred solution of the multi-port fault location method of the intelligent oscilloscope coordinated with relay protection described in the present invention, collecting relay protection action information includes:

[0014] The relay protection action information includes protection start time, protection action time, fault type judgment result, fault phase, protection action phase, and protection action stage;

[0015] After receiving the coordinated trigger signal, the relay protection records the protection start time;

[0016] Monitor the relay protection action status and record the protection action time when a trip command is issued to the circuit breaker;

[0017] Generating a fault type judgment result based on the detection results of the zero-sequence current and the negative-sequence current;

[0018] By comparing the amplitude change and phase angle change of each phase voltage and each phase current before and after the fault occurs, the fault phase and protection action phase are determined;

[0019] The protection action stage is determined by comparing the fault current amplitude with the protection setting value.

[0020] As a preferred solution of the multi-port fault location method of the intelligent oscilloscope cooperative relay protection described in the present invention, time synchronization through a unified timestamp tag includes:

[0021] The relay protection receives the clock synchronization signal sent by the master clock server and calibrates the local clock;

[0022] The relay protection adopts a double buffer, the first buffer receives the fault recording data and protection action information sent by the intelligent recorder, and the second buffer performs data synchronization processing;

[0023] A timestamp in a unified format is added to the fault recording data and the protection action information written into the first buffer. When the first buffer is full, the fault recording data and relay protection action information with the timestamp are transferred to the second buffer. The data are time-aligned according to the timestamp to generate a synchronous data set containing the fault recording data and relay protection action information.

[0024] It can improve the accuracy of fault analysis. Through time synchronization, it can accurately associate fault recording data with protection action information, facilitate the analysis of the cause, process and action of the fault protection device, and improve the accuracy of fault analysis; it can improve the efficiency of fault location. Accurate time synchronization can help quickly locate the fault point, shorten the fault investigation time, and improve the reliability of power system operation.

[0025] As a preferred solution of the multi-port fault location method of the intelligent recorder in cooperation with relay protection of the present invention, wherein: performing Fourier decomposition on the fault recording data with timestamp mark, extracting the fundamental component and harmonic component, and constructing the fault feature vector includes:

[0026] Establishing sliding data windows for the pre-trigger sampling segment and the post-trigger sampling segment respectively according to the timestamp mark, wherein the length of the sliding data window is one power frequency cycle, and extracting the fundamental component and the harmonic component by recursive discrete Fourier transform within the sliding data window, wherein the harmonic component includes the third harmonic to the thirteenth harmonic;

[0027] The phasor change rate, harmonic content change rate and zero-sequence current change rate are calculated based on the fundamental component and the harmonic component; wherein the phasor change rate includes the amplitude change rate and the phase angle change rate, the amplitude change rate is obtained by calculating the amplitude mean of the pre-trigger sampling segment and the amplitude mean of the post-trigger sampling segment, the phase angle change rate is obtained by calculating the phase angle mean of the pre-trigger sampling segment and the phase angle mean of the post-trigger sampling segment, the harmonic content change rate is obtained by calculating the harmonic content mean of the pre-trigger sampling segment and the harmonic content mean of the post-trigger sampling segment, and the zero-sequence current change rate is obtained by calculating the zero-sequence current mean of the pre-trigger sampling segment and the zero-sequence current mean of the post-trigger sampling segment;

[0028] A fault characteristic vector is constructed, and the phasor change rate, harmonic content change rate and zero-sequence current change rate are combined to generate the fault characteristic vector.

[0029] It can improve the accuracy of fault identification. By extracting multiple characteristic parameters and constructing a fault feature vector, it can more comprehensively describe the fault characteristics, thereby improving the accuracy of fault identification; achieving fine classification of fault types, improving the efficiency of fault analysis, and shortening fault handling time.

[0030] As a preferred solution of the multi-port fault location method of the intelligent oscilloscope coordinated with relay protection of the present invention, wherein: combining the relay protection action information, completing the fault type identification through the fault classification model includes:

[0031] According to the protection action stage and fault type judgment result in the relay protection action information, combined with the fault feature vector, the fault type is identified by a pre-trained fault classification model, and the feature data is input into the fault classification model;

[0032] The pre-trained fault classification model is constructed using a support vector machine algorithm to output fault type identification probabilities, including single-phase grounding fault probability, two-phase short circuit fault probability, two-phase grounding fault probability, and three-phase short circuit fault probability;

[0033] The difference between the maximum probability value and the second largest probability value in the fault type identification probability is calculated. When the difference is greater than a preset confidence threshold, the fault type corresponding to the maximum probability value is used as the final identification result; when the difference is less than or equal to the preset confidence threshold, the data is input into the preset expert rule library to determine the fault type and obtain the final identification result.

[0034] As a preferred solution of the multi-port fault location method of the intelligent oscilloscope cooperative relay protection described in the present invention, generating a fault analysis report includes:

[0035] When the matching degree between the recognition result of the fault classification model and the fault type judgment result of the relay protection action information exceeds a preset matching threshold, the final fault type is confirmed;

[0036] According to the fault type, fault characteristic vector and protection action stage, an adaptive fault location algorithm is used to calculate the fault point location, and a fault analysis report including the final fault type, fault point location, fault phase and fault occurrence time is generated.

[0037] In a second aspect, the present invention provides a multi-port fault location system using an intelligent oscilloscope and coordinated relay protection, comprising:

[0038] Configuration module, used to set multiple trigger conditions and sampling windows of the intelligent recorder, and set the relay protection to three-stage protection form;

[0039] An acquisition module is used to set the judgment threshold of the multiple trigger conditions. If the judgment threshold of the multiple trigger conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection. The collected fault recording data and relay protection action information are synchronized through a unified timestamp mark;

[0040] The identification module is used to perform Fourier decomposition on the fault recording data with time stamp marks, extract the fundamental component and harmonic component, construct the fault feature vector, and combine it with the relay protection action information to complete the fault type identification through the fault classification model and generate a fault analysis report.

[0041] In a third aspect, the present invention provides an electronic device, comprising:

[0042] memory and processor;

[0043] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the intelligent recorder collaborative relay protection multi-port fault locating method are implemented.

[0044] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the intelligent recorder collaborative relay protection multi-port fault locating method.

[0045] Compared with the existing technology, the beneficial effects of the present invention are as follows: by combining the high-precision recording data of the intelligent recorder and the protection action information of the relay protection, and using Fourier decomposition to extract the fault feature vector, the present invention can more accurately identify the fault type and calculate the fault point location, avoiding the errors of the traditional method; the intelligent recorder and the relay protection device work together to quickly trigger sampling and collect fault information, and adopt an adaptive fault location algorithm to shorten the fault location time and improve the fault handling efficiency; by constructing a fault feature vector and adopting a pre-trained fault classification model, the fault characteristics can be analyzed more comprehensively, complex fault types can be identified, and a detailed fault analysis report can be generated, providing more reliable data support for fault cause analysis and prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0047] Figure 1The figure is a schematic diagram of the overall process of a multi-port fault location method using an intelligent oscilloscope recorder in collaboration with relay protection according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0049] Reference Figure 1 , as one embodiment of the present invention, provides a multi-port fault location method using an intelligent oscilloscope and coordinated relay protection, comprising:

[0050] S101, set multiple trigger conditions and sampling windows for the intelligent recorder, and set the relay protection to three-stage protection;

[0051] S102, setting the judgment threshold of multiple trigger conditions. If the judgment threshold of multiple trigger conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection. The collected fault recording data and relay protection action information are synchronized through a unified timestamp mark;

[0052] S103, performing Fourier decomposition on the fault recording data with time stamp, extracting the fundamental component and harmonic component, constructing the fault feature vector, and combining it with the relay protection action information, completing the fault type identification through the fault classification model, and generating a fault analysis report.

[0053] In a preferred embodiment, in step S101, multiple trigger conditions and multiple sampling windows of the intelligent recorder are set, and the relay protection is set to a three-stage protection form, including:

[0054] The multiple trigger conditions of the intelligent recorder include voltage amplitude trigger condition, current amplitude trigger condition, phasor mutation trigger condition and frequency deviation trigger condition;

[0055] The sampling window of the intelligent recorder includes the pre-trigger sampling section and the post-trigger sampling section;

[0056] The three-stage protection of relay protection includes instantaneous action stage, delayed action stage and backup protection stage.

[0057] In an optional embodiment, the sampling duration of the pre-trigger sampling segment is 100ms before the fault occurs, and the sampling duration of the post-trigger sampling segment is 400ms after the fault occurs; the relay protection is configured with three-stage protection, including instantaneous action segment, delayed action segment and backup protection segment, and a communication connection is established between the intelligent recorder and the relay protection device.

[0058] In another optional embodiment, the multiple trigger conditions of the intelligent recorder can also be set according to the characteristics of the electrical quantity, which can be a sudden change quantity trigger, a sequence component trigger, etc. The sampling window of the intelligent recorder can also automatically switch the window length according to the trigger condition type. The three-stage protection of the relay protection can be set according to the protection function positioning, and the core parameters can be set to achieve a balance between selectivity and speed.

[0059] In an optional implementation, in step S102, setting the determination thresholds for multiple trigger conditions includes:

[0060] If any phase voltage is set to be within the rated voltage threshold range, or the phase change between two adjacent sampling periods exceeds the preset mutation threshold, or the frequency deviation exceeds the frequency deviation threshold, and any of the above judgment threshold conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection.

[0061] Specifically, the rated voltage threshold interval is expressed as when it is detected that any phase voltage is lower than 0.7 times the rated value, or any phase current exceeds 2 times the rated value. The preset mutation threshold is expressed as the phase quantity change between two adjacent sampling cycles exceeds plus or minus 300 amperes, and the frequency deviation threshold is expressed as the frequency deviation exceeds plus or minus 0.5 Hz.

[0062] In a preferred embodiment, collecting relay protection action information includes:

[0063] Relay protection action information includes protection start time, protection action time, fault type judgment result, fault phase, protection action phase, and protection action stage;

[0064] After the relay protection receives the coordinated trigger signal, it records the protection start time;

[0065] Monitor the relay protection action status and record the protection action time when a trip command is issued to the circuit breaker;

[0066] Generating a fault type judgment result based on the detection results of the zero-sequence current and the negative-sequence current;

[0067] By comparing the amplitude change and phase angle change of each phase voltage and each phase current before and after the fault occurs, the fault phase and protection action phase are determined;

[0068] The protection action stage is determined by comparing the fault current amplitude with the protection setting value.

[0069] In an optional embodiment, when the relay protection receives the coordinated trigger signal, the protection start time is recorded, the operating state of the relay protection device is monitored, and the protection action time is recorded when the relay protection issues a trip command to the circuit breaker;

[0070] The relay protection determines the fault type based on the detection results of zero-sequence current and negative-sequence current. When the zero-sequence current exceeds the first preset threshold and the negative-sequence current is less than the second preset threshold, it is determined to be a single-phase grounding fault, and a fault type judgment result is generated. The fault phase and protection action phase are determined by comparing the amplitude changes and phase angle changes of each phase voltage and each phase current before and after the fault occurs. The protection action stage is determined based on the comparison result of the fault current amplitude and the protection setting value; the protection start time, protection action time, fault type judgment result, fault phase, protection action phase and protection action stage constitute the protection action information.

[0071] In another optional implementation, the relay protection action information can also be determined based on the protection measurement values ​​and fault parameters, protection action logic and coordination relationship, fault recording characteristics, operating status data, protection device abnormal data, fault duration and action timing, etc. Through the multi-dimensional fusion of protection measurement values, action logic, recording characteristics, system status, abnormal information and timing coordination, the fault type (such as single-phase grounding, phase-to-phase short circuit, three-phase short circuit, excitation inrush current, high-resistance fault, etc.) can be identified more accurately, and the fault nature (instantaneous / permanent) and severity can be assisted in locating, providing a comprehensive basis for fault analysis and power grid restoration.

[0072] In a preferred embodiment, time synchronization through a unified timestamp includes:

[0073] The relay protection receives the clock synchronization signal sent by the master clock server and calibrates the local clock;

[0074] The relay protection adopts a double buffer. The first buffer receives the fault recording data and protection action information sent by the intelligent recorder, and the second buffer performs data synchronization processing.

[0075] A timestamp in a unified format is added to the fault recording data and protection action information written into the first buffer. When the first buffer is full, the fault recording data and relay protection action information with the timestamp are transferred to the second buffer. The data are time-aligned according to the timestamp to generate a synchronous data set containing the fault recording data and relay protection action information.

[0076] Specifically, the relay protection device is connected to the master clock server through an optical fiber network, receives the clock synchronization signal sent by the master clock server, and calibrates the local clock according to the transmission delay compensation algorithm, wherein the resolution of the timestamp mark is 1 microsecond.

[0077] In an optional implementation manner, the relay protection receives a clock synchronization signal sent by a master clock server and calibrates a local clock, including:

[0078] The relay protection device is connected to the master clock server via a fiber optic network. The relay protection device receives the clock synchronization signal sent by the master clock server via the fiber optic network and calibrates the local clock according to the transmission delay compensation algorithm. The relay protection device sends a time request message to the master clock server and records the time when the time request message is sent. The master clock server records the time when the time request message is received and the time when the reply message is sent. The relay protection device records the time when the reply message is received.

[0079] A sliding window of 64 sampling points was established. Transmission delay data was calculated based on the sending time, receiving time, reply message sending time, and reply message receiving time. The transmission delay data was sorted and the delay data with the maximum and minimum values, each accounting for 25%, were removed. The arithmetic average of the remaining delay data in the sliding window was calculated to obtain the initial transmission delay. Fiber temperature data was collected using temperature sensors installed at key nodes in the fiber optic network, and a mapping relationship between the fiber temperature data and the initial transmission delay was established.

[0080] The initial transmission delay is compensated according to the mapping relationship to obtain the compensated transmission delay, and the clock deviation between the local clock and the master clock server is calculated based on the compensated transmission delay. When the clock deviation is less than one millisecond, the frequency is fine-tuned by adjusting the crystal oscillator division coefficient of the local clock to achieve clock calibration. When the clock deviation is greater than or equal to one millisecond, the clock deviation is decomposed into multiple clock adjustment steps of less than one millisecond, and the frequency is fine-tuned for each clock adjustment step to achieve clock calibration.

[0081] The relay protection system connects to the master clock server via a fiber optic network to precisely calibrate the local clock. First, the relay protection device receives the clock synchronization signal from the master clock server via the fiber optic network. To accurately measure transmission delay, the relay protection system sends a time request message to the master clock server and records the sending time. Upon receiving the message, the master clock server records the receiving time and the sending time of the reply message. The relay protection system then records the receiving time of the reply message.

[0082] To improve the accuracy of transmission delay measurements, a sliding window mechanism and temperature compensation algorithm are employed. A sliding window consisting of 64 sampling points is established. After each transmission delay measurement, new delay data is added to the window while the oldest data is removed. The transmission delay data for each sampling point is calculated using the recorded send and receive times. To mitigate the influence of outliers, the delay data within the window are sorted, and the maximum and minimum values, each accounting for 25%, are removed. For example, if the delay data are sorted from smallest to largest as 1ms, 2ms, 3ms, and so on to 64ms, the data between 1ms and 16ms and between 49ms and 64ms are removed. The remaining delay data are then arithmetic averaged to obtain the initial transmission delay.

[0083] To compensate for the impact of fiber temperature fluctuations on transmission delay, temperature sensors are deployed at key nodes in the fiber network to collect fiber temperature data. Through long-term data accumulation, a mapping relationship between fiber temperature data and initial transmission delay is established. For example, a lookup table can be created to store initial transmission delay correction values ​​corresponding to different temperatures. Based on the currently measured fiber temperature, the corresponding correction value is found in the lookup table and used to compensate for the initial transmission delay, resulting in the compensated transmission delay.

[0084] Based on the compensated transmission delay, the clock deviation between the local clock and the master clock server is calculated. If the clock deviation is less than 1 millisecond, the frequency is fine-tuned by adjusting the crystal oscillator division coefficient of the local clock to achieve clock calibration. For example, if the clock deviation is 0.5 milliseconds, the crystal oscillator division coefficient can be fine-tuned according to the pre-set adjustment coefficient so that the local clock gradually catches up with the master clock. If the clock deviation is greater than or equal to 1 millisecond, the clock deviation is decomposed into multiple clock adjustment steps of less than 1 millisecond, and the frequency is fine-tuned for each step to gradually achieve clock calibration. For example, if the clock deviation is 2 milliseconds, it can be decomposed into two 1 millisecond steps and calibrated in two steps.

[0085] It should be noted that the beneficial effects of this technical solution are reflected in three aspects:

[0086] 1. Improved clock synchronization accuracy. Through the sliding window mechanism and temperature compensation algorithm, the transmission delay measurement error is effectively reduced, thereby improving the clock synchronization accuracy and ensuring the reliable operation of the relay protection device.

[0087] 2. Enhanced system stability. Through the step-by-step calibration mechanism, the impact caused by large step-size adjustment is avoided, the system stability is enhanced, and the risk of clock desynchronization is reduced.

[0088] 3. Simplified calibration process. This solution realizes the automated clock calibration process without manual intervention, simplifies the operation process and reduces maintenance costs.

[0089] In another optional embodiment, a synchronous method for collaboratively collecting fault information by relay protection and intelligent recorders is used to synchronize the fault recording data with the protection action information through a unified timestamp mark, thereby improving the accuracy of fault analysis. The specific steps are as follows:

[0090] First, a connection is configured and established. The relay protection device connects to the master clock server via a fiber optic network, receives the clock synchronization signal, and uses a transmission delay compensation algorithm to calibrate the local clock to ensure time accuracy. Specifically, the relay protection device can receive the clock synchronization signal from the master clock server every second and calculate the deviation of the local clock based on the signal transmission time for real-time calibration. The intelligent oscilloscope recorder is also synchronized with the master clock server to ensure time consistency. Two buffers are set up within the relay protection device: the first buffer is used to receive fault recording data and protection action information sent by the intelligent oscilloscope recorder, and the second buffer is used for data synchronization processing.

[0091] Then, fault recording data and protection action information are collected. The intelligent recorder continuously collects analog quantities such as voltage and current on the line and converts the collected data into digital quantities to form fault recording data. When a fault occurs, the relay protection device receives a preset collaborative trigger signal. Specifically, in this embodiment, it can be assumed that the collaborative trigger signal is a specific digital code. After receiving the collaborative trigger signal, the relay protection device records the protection start time, for example, "2024-07-27 10:00:00.123456"; at the same time, the relay protection device monitors its own status and records the protection action time when it issues a trip command to the circuit breaker, for example, "2024-07-27 10:00:00.123789".

[0092] Next, the fault type, phase, and stage are determined. The relay protection device determines the fault type based on the zero-sequence current and negative-sequence current detection results. Optionally, the preset first threshold value can be 100 amperes, and the second threshold value can be 50 amperes. If the zero-sequence current is 150 amperes and the negative-sequence current is 30 amperes, it is determined to be a single-phase ground fault. Furthermore, the fault phase and protection action phase are determined by comparing the amplitude and phase angle changes of each phase voltage and current before and after the fault. For example, if the voltage of phase A suddenly drops and the current of phase A suddenly increases, the fault phase and protection action phase are both determined to be phase A. Furthermore, the protection action stage is determined based on the comparison of the fault current amplitude with the protection setting. For example, in this embodiment, if the fault current amplitude is 2000 amperes, which exceeds the first protection stage setting of 1500 amperes, the protection action stage is determined to be the first stage. The above information constitutes the protection action information.

[0093] Subsequently, data synchronization is performed. The relay protection device adds a unified timestamp format to the fault recording data and protection action information written into the first buffer, with a resolution of 1 microsecond. Specifically, for example, the fault recording data collected at a certain moment is timestamped "2024-07-27 10:00:00.123456." When the first buffer is full, the timestamped data is transferred to the second buffer. Within the second buffer, the fault recording data and protection action information are time-aligned based on the timestamp, generating a synchronized data set containing the fault recording data and protection action information. For example, the fault recording data with the timestamp "2024-07-27 10:00:00.123456" is associated with the protection action information with the same timestamp.

[0094] It should be noted that this step can improve the accuracy of fault analysis. Through time synchronization, the fault recording data can be accurately associated with the protection action information, which is convenient for analyzing the cause, process and action of the fault device, and improving the accuracy of fault analysis; it can improve the efficiency of fault location. Accurate time synchronization can help quickly locate the fault point, shorten the fault investigation time, and improve the reliability of power system operation; it can simplify the data processing process. The unified timestamp tag simplifies the data processing process and facilitates subsequent fault analysis and data management.

[0095] In a preferred embodiment, in step S103, Fourier decomposition is performed on the fault recording data with the timestamp to extract the fundamental component and the harmonic component, and the fault feature vector is constructed, which includes:

[0096] A sliding data window is established for the pre-trigger sampling segment and the post-trigger sampling segment according to the timestamp mark. The length of the sliding data window is one power frequency cycle. Recursive discrete Fourier transform is used within the sliding data window to extract the fundamental component and harmonic components, wherein the harmonic components include the third harmonic to the thirteenth harmonic;

[0097] The phasor change rate, harmonic content change rate and zero-sequence current change rate are calculated based on the fundamental component and the harmonic component; wherein the phasor change rate includes the amplitude change rate and the phase angle change rate, the amplitude change rate is obtained by calculating the amplitude mean of the pre-trigger sampling segment and the amplitude mean of the post-trigger sampling segment, the phase angle change rate is obtained by calculating the phase angle mean of the pre-trigger sampling segment and the phase angle mean of the post-trigger sampling segment, the harmonic content change rate is obtained by calculating the harmonic content mean of the pre-trigger sampling segment and the harmonic content mean of the post-trigger sampling segment, and the zero-sequence current change rate is obtained by calculating the zero-sequence current mean of the pre-trigger sampling segment and the zero-sequence current mean of the post-trigger sampling segment;

[0098] Construct a fault feature vector by combining the phasor change rate, harmonic content change rate and zero-sequence current change rate to generate the fault feature vector.

[0099] In another optional implementation, constructing a fault feature vector requires combining the fault signal characteristics (steady state / transient, linear / nonlinear) and the identification target (fault type, location, nature). It can also be achieved by integrating multiple methods such as time-frequency domain analysis, statistical parameters, sequence components, entropy, and machine learning to form a feature set containing multi-dimensional information such as amplitude, phase, frequency, energy, and complexity, providing a more comprehensive input basis for subsequent fault classification models. Among them, time-frequency domain analysis uses wavelet transform to perform multi-resolution analysis on non-stationary fault signals, extracting energy, amplitude, and phase coefficients in different frequency bands (such as high-frequency transients and low-frequency steady states), capturing detailed features at the moment of waveform mutation, and statistically analyzing waveform characteristic parameters. Entropy can be used to calculate the probability distribution entropy of the signal in the time domain or frequency domain through information entropy.

[0100] In an optional embodiment, the harmonic components include the third to thirteenth harmonics and are calculated as follows:

[0101] The fundamental component is calculated as follows:

[0102] X1(k)=X1(k-1)+[x(k)-x(kN)]·e (-j2π / N)

[0103] Where X1(k) is the fundamental component at time k, N is the length of the sliding data window, x(k) is the sampled data, and j represents the imaginary unit;

[0104] The harmonic components include the third to the thirteenth harmonics, which are expressed as:

[0105] X n (k) = X n (k-1)+[x(k)-x(kN)]·e (-jn2π / N)

[0106] Where n is the harmonic order, X n (k) is the nth harmonic component at time k.

[0107] In an optional embodiment, calculating the phasor change rate, the harmonic content change rate, and the zero-sequence current change rate includes:

[0108] The phasor change rate includes the amplitude change rate and the phase angle change rate. The amplitude change rate is calculated by multiplying (M2-M1) / M1 by 100%, where M1 is the amplitude mean of the pre-trigger sampling segment and M2 is the amplitude mean of the post-trigger sampling segment.

[0109] Phase angle change rate is Calculated, where is the phase angle mean value of the pre-trigger sampling segment, is the phase angle mean value of the post-trigger sampling segment;

[0110] The rate of change of harmonic content is expressed by (H n2 -H n1 ) / H n1 Multiply by 100% to get the result, where H n1 is the mean value of the nth harmonic content in the pre-trigger sampling period, H n2 is the mean value of the nth harmonic content in the post-trigger sampling section;

[0111] The calculation of zero-sequence current change rate includes: calculating the zero-sequence current based on the fundamental component of the three-phase current, and the zero-sequence current is calculated by (IA+IB+IC) / 3, where IA, IB, and IC are the fundamental components of the current of phase A, phase B, and phase C respectively, and the zero-sequence current change rate is calculated by (I 02 -I 01 ) / I 01 Multiply by 100% to get the value, where I 01 is the mean value of zero-sequence current in the pre-trigger sampling section, I 02 is the mean value of zero-sequence current in the post-trigger sampling section.

[0112] It should be noted that it can accurately identify and classify power system faults, use fault recording data with timestamps, perform Fourier decomposition on the data, extract fundamental and harmonic components, and calculate the phasor change rate, harmonic content change rate and zero-sequence current change rate to construct the fault feature vector.

[0113] First, preprocess the timestamped fault recording data, including data cleaning and alignment. Based on the timestamp, the recording data is divided into pre-trigger and post-trigger sampling segments. The pre-trigger sampling segment refers to the sampling period before the fault occurs, while the post-trigger sampling segment refers to the sampling period after the fault occurs. For example, the pre-trigger sampling segment can be 0.1 seconds before the fault occurs, and the post-trigger sampling segment can be 0.1 seconds after the fault occurs.

[0114] Next, Fourier decomposition is performed on the pre-trigger sampling segment and the post-trigger sampling segment respectively. A sliding data window with a length of one power frequency cycle is established. For example, for a 50Hz power frequency, the window length is 20 milliseconds, which contains data samples of a complete power frequency cycle. Within the sliding data window, a recursive discrete Fourier transform is used to extract the fundamental component and harmonic components. The fundamental component is calculated as follows: the fundamental component at the current moment is equal to the fundamental component at the previous moment plus the difference between the sampling data at the current moment and the sampling data at the corresponding moment of the previous power frequency cycle, multiplied by a complex coefficient. The calculation method of the harmonic component is similar to that of the fundamental component, except that the complex coefficient is different. Harmonic orders can include third harmonics, fifth harmonics, seventh harmonics, ninth harmonics, eleventh harmonics, and thirteenth harmonics.

[0115] Then, the phasor change rate, harmonic content change rate and zero-sequence current change rate are calculated based on the extracted fundamental component and harmonic component. The phasor change rate includes the amplitude change rate and the phase angle change rate. The amplitude change rate is calculated by dividing the difference between the amplitude mean of the post-trigger sampling segment and the amplitude mean of the pre-trigger sampling segment by the amplitude mean of the pre-trigger sampling segment, and then multiplying by 100%. The phase angle change rate is calculated by subtracting the phase angle mean of the pre-trigger sampling segment from the phase angle mean of the post-trigger sampling segment. The harmonic content change rate is calculated by dividing the difference between the nth harmonic content mean of the post-trigger sampling segment and the nth harmonic content mean of the pre-trigger sampling segment by the nth harmonic content mean of the pre-trigger sampling segment, and then multiplying by 100%. The zero-sequence current is calculated by dividing the sum of the fundamental components of the three-phase current by 3. The calculation method of the zero-sequence current change rate is: the difference between the zero-sequence current mean value of the post-trigger sampling section and the zero-sequence current mean value of the pre-trigger sampling section is divided by the zero-sequence current mean value of the pre-trigger sampling section, and then multiplied by 100%.

[0116] Finally, the calculated phasor change rates, harmonic content change rates, and zero-sequence current change rates are combined to form a fault feature vector. For example, a fault feature vector can include features such as the fundamental amplitude change rate, fundamental phase angle change rate, third harmonic content change rate, fifth harmonic content change rate, and zero-sequence current change rate. Assuming the mean fundamental amplitude in the pre-trigger sampling period is 10 and the mean fundamental amplitude in the post-trigger sampling period is 12, the fundamental amplitude change rate is (12-10) / 10*100% = 20%.

[0117] The beneficial effects of this program are reflected in the following three aspects:

[0118] Improve the accuracy of fault identification: By extracting multiple characteristic parameters and constructing a fault feature vector, the fault characteristics can be described more comprehensively, thereby improving the accuracy of fault identification;

[0119] Achieve fine classification of fault types: Different fault types correspond to different feature vectors. By analyzing the fault feature vectors, fine classification of fault types can be achieved, such as distinguishing single-phase grounding faults, two-phase short circuit faults, and three-phase short circuit faults.

[0120] Improve the efficiency of fault analysis: The high degree of automation can quickly extract fault characteristics, thereby improving the efficiency of fault analysis and shortening fault handling time.

[0121] In a preferred embodiment, in step S103, combining the relay protection action information and completing the fault type identification through the fault classification model includes:

[0122] Based on the protection action stage and fault type judgment results in the relay protection action information, combined with the fault feature vector, the fault type is identified through a pre-trained fault classification model, and the feature data is input into the fault classification model;

[0123] The pre-trained fault classification model is constructed using the support vector machine algorithm and outputs the fault type identification probability, which includes the single-phase grounding fault probability, two-phase short circuit fault probability, two-phase grounding fault probability, and three-phase short circuit fault probability;

[0124] The difference between the maximum probability value and the second largest probability value in the fault type identification probability is calculated. When the difference is greater than the preset confidence threshold, the fault type corresponding to the maximum probability value is used as the final identification result; when the difference is less than or equal to the preset confidence threshold, the data is input into the preset expert rule library to judge the fault type and obtain the final identification result.

[0125] In an optional embodiment, according to the protection action stage and the fault type judgment result in the protection action information, combined with the fault feature vector, the fault type identification is performed using a pre-trained fault classification model, including:

[0126] According to the protection action stage and fault type judgment result in the protection action information, combined with the fault feature vector, a pre-trained fault classification model is used to identify the fault type. The protection action stage is used as the first input feature, the fault type judgment result is used as the second input feature, and the fault feature vector is used as the third input feature. The first input feature, the second input feature, and the third input feature are input into the pre-trained fault classification model;

[0127] The pre-trained fault classification model is constructed using a support vector machine algorithm. The kernel function of the support vector machine algorithm adopts a radial basis kernel function. The kernel function parameters and penalty factor of the radial basis kernel function are optimized and determined through a grid search method. The pre-trained fault classification model outputs the fault type recognition probability, which includes the probability of single-phase grounding fault, two-phase short circuit fault, two-phase grounding fault, and three-phase short circuit fault.

[0128] The difference between the maximum probability value and the second largest probability value in the fault type identification probability is calculated. When the difference is greater than the preset confidence threshold, the fault type corresponding to the maximum probability value is taken as the final identification result; when the difference is less than or equal to the preset confidence threshold, the first input feature, the second input feature and the third input feature are input into the preset expert rule library to judge the fault type and obtain the final identification result.

[0129] It should be noted that this step utilizes protection action information, fault feature vectors, and a pre-trained fault classification model to achieve accurate identification of power system faults.

[0130] Specifically, first, collect protection action information at the time of the fault, including the protection action stage and fault type determination result. This information can be obtained, for example, from the substation automation system. Suppose a fault occurs at a certain moment, and protection device A is activated. The corresponding protection stage is "Line 1," and the fault type is determined to be "Ground Fault."

[0131] At the same time, the fault feature vector is extracted when the fault occurs. The fault feature vector can contain a variety of information, such as the voltage and current amplitude, phase, harmonic content, etc. at the time of the fault. Assume that the extracted fault feature vector is [0.85, 0.23, 1.12, 0.95, ...].

[0132] Next, the protection action stage, fault type judgment result and fault feature vector are taken as input and sent to the pre-trained fault classification model. The model is constructed using the support vector machine algorithm, and the kernel function adopts the radial basis kernel function. The kernel function parameters and penalty factor of the radial basis kernel function are optimized and determined by the grid search method. Optionally, by performing a grid search on the training set, the kernel function parameters are finally determined to be 0.5 and the penalty factor is 10.

[0133] The pre-trained fault classification model outputs fault type identification probabilities, including single-phase grounding fault probability, two-phase short circuit fault probability, two-phase grounding fault probability, and three-phase short circuit fault probability. Optionally, it is assumed that the probabilities output by the model are: single-phase grounding fault probability is 0.8, two-phase short circuit fault probability is 0.05, two-phase grounding fault probability is 0.1, and three-phase short circuit fault probability is 0.05.

[0134] Then, the difference between the maximum probability value and the second maximum probability value in the fault type identification probability is calculated. In this embodiment, the maximum probability value is 0.8 (single-phase grounding fault), the second maximum probability value is 0.1 (two-phase grounding fault), and the difference is 0.7.

[0135] A confidence threshold is pre-set, which in this embodiment can be 0.5. If the difference is greater than the confidence threshold, the fault type corresponding to the maximum probability value is used as the final identification result. In this embodiment, 0.7>0.5, so the final identification result is a single-phase grounding fault.

[0136] If the difference is less than or equal to the confidence threshold, the protection action stage, fault type determination, and fault feature vector are input into a pre-set expert rule library to determine the fault type and obtain the final identification result. The expert rule library can contain multiple rules, for example: "If the protection action stage is Line 1, the fault type determination result is a ground fault, and the current sudden change in the fault feature vector is greater than a certain threshold, then determine it to be a single-phase ground fault."

[0137] It should be noted that the present invention can improve the accuracy of fault identification, and comprehensively utilizes protection action information, fault feature vectors and pre-trained fault classification models, which can more accurately identify fault types and reduce the misjudgment rate; enhance the reliability of fault identification, and through the combination of confidence threshold and expert rule base, can effectively handle the situation of low confidence in model identification results, thereby improving the reliability of fault identification; improve the efficiency of fault handling, and fast and accurate fault identification can help operation and maintenance personnel locate the fault point more quickly, shorten the fault handling time, and improve the operating efficiency of the power system.

[0138] In a preferred embodiment, generating a fault analysis report includes:

[0139] When the matching degree between the recognition result of the fault classification model and the fault type judgment result of the relay protection action information exceeds the preset matching threshold, the final fault type is confirmed;

[0140] Based on the fault type, fault characteristic vector and protection action stage, an adaptive fault location algorithm is used to calculate the fault point location and generate a fault analysis report including the final fault type, fault point location, fault phase and fault occurrence time.

[0141] In an optional embodiment, generating a fault analysis report including the final fault type, fault location, fault phase, and fault occurrence time includes:

[0142] The matching degree between the fault classification model's identification result and the fault type determination result of the protection action information is calculated. The matching degree is obtained by calculating the cosine similarity of the two results. When the matching degree exceeds a preset matching threshold, the fault classification model's identification result is confirmed as the final fault type. Based on the final fault type, the fault feature vector, and the protection action stage, an adaptive fault location algorithm is used to calculate the fault point location. The adaptive fault location algorithm includes: selecting a basic location model based on the final fault type, using the zero-sequence current distribution method when the final fault type is a single-phase grounding fault, and using the voltage and current distribution method when the final fault type is a phase-to-phase fault.

[0143] Adaptively adjust the parameters of the basic positioning model according to the protection action stage. When the protection action stage is the distance protection stage 1 action, the measured voltage and current weight coefficient is set to the first preset value. When the protection action stage is the distance protection stage 2 action, the measured voltage and current weight coefficient is set to the second preset value. When the protection action stage is the distance protection stage 3 action, the measured voltage and current weight coefficient is set to the third preset value.

[0144] The fault phase voltage amplitude change rate, voltage phase angle change rate, current amplitude change rate, and current phase angle change rate are extracted from the fault characteristic vector, and the weight coefficient is dynamically adjusted according to the significance of each change rate. The adjusted weight coefficient is applied to the basic positioning model to calculate the fault point location; a fault analysis report is generated based on the final fault type, fault point location, fault phase information in the fault characteristic vector, and the time of fault occurrence.

[0145] Specifically, first, collect power system operating data at the time of the fault. This data includes voltage and current values ​​at each line point, as well as protection device operation information, such as operation time and operation stage. Assume the following data is collected: the voltage at measuring point 1 on line A suddenly drops to 80kV, the current suddenly increases to 2kA, and the distance protection device's second stage operates 0.1 seconds after the fault occurs; the voltage at measuring point 2 on line B drops to 100kV, and the current increases to 1.5kA.

[0146] The collected data is then preprocessed. This includes operations such as data cleaning, format conversion, and feature extraction. For example, the collected raw voltage and current data is converted into amplitude and phase angle, and features such as the voltage amplitude change rate, voltage phase angle change rate, current amplitude change rate, and current phase angle change rate are calculated to form a fault feature vector. Suppose that the voltage amplitude change rate calculated at measuring point 1 on line A is -0.2, the voltage phase angle change rate is 0.1, the current amplitude change rate is 1, and the current phase angle change rate is 0.05; and the voltage amplitude change rate calculated at measuring point 2 on line B is -0.1, the voltage phase angle change rate is 0.05, the current amplitude change rate is 0.5, and the current phase angle change rate is 0.025.

[0147] Next, the preprocessed fault feature vector is input into a pre-trained fault classification model to identify the fault type. This model can be trained using a machine learning algorithm such as a support vector machine or neural network. Assume that the fault classification model identifies a single-phase ground fault.

[0148] At the same time, the fault type is determined based on the action information of the protection device. For example, based on the tripping information, action time and other information of the protection device, the fault type can be determined to be a single-phase grounding fault.

[0149] Afterwards, the matching degree between the recognition result of the fault classification model and the fault type judgment result of the protection action information is calculated. The matching degree can be obtained by calculating the cosine similarity of the two results. In this embodiment, the calculated cosine similarity is 0.95, and the preset matching threshold is set to 0.9. Since 0.95 is greater than 0.9, the recognition result of the fault classification model is confirmed as the final fault type, that is, a single-phase grounding fault.

[0150] An adaptive fault location algorithm is used to calculate the fault location based on the final fault type, fault characteristic vector, and protection operation stage. Since the final fault type is a single-phase ground fault, the zero-sequence current distribution method is selected as the basic location model. Since the protection operation stage is distance protection stage 2, the measured voltage and current weight coefficients are set to the second preset value, assumed to be 0.8. The voltage amplitude change rate, voltage phase angle change rate, current amplitude change rate, and current phase angle change rate are extracted from the fault characteristic vector for measuring points 1 on line A and 2 on line B. The weight coefficients are dynamically adjusted based on the significance of each change rate. Assume that after adjustment based on significance, the weight coefficient for measuring point 1 on line A is 0.9, and the weight coefficient for measuring point 2 on line B is 0.7. The adjusted weight coefficients are applied to the zero-sequence current distribution method to calculate the fault location. Assume that the calculated fault location is 10 kilometers from the starting point of line A.

[0151] Finally, a fault analysis report is generated based on the final fault type (single-phase grounding fault), the fault point location (10 kilometers from the starting point of Line A), the fault phase information (Phase A) in the fault feature vector, and the fault occurrence time information (8:00 on January 1, 2024).

[0152] It should be noted that the present invention can improve the accuracy of fault location. By combining the recognition results of the fault classification model and the protection action information, and adopting an adaptive fault location algorithm, it can locate the fault point more accurately; shorten the fault handling time, and fast and accurate fault location can help operation and maintenance personnel quickly isolate the fault and restore power supply, thereby shortening the fault handling time and reducing power outage losses; it can improve the operational reliability of the power system. By timely and accurately analyzing the cause of the fault, corresponding preventive measures can be taken to avoid the recurrence of similar faults, thereby improving the operational reliability of the power system.

[0153] The above is a schematic scheme of the intelligent oscilloscope cooperative relay protection multi-port fault locating method of this embodiment. It should be noted that the technical scheme of the intelligent oscilloscope cooperative relay protection multi-port fault locating system and the technical scheme of the intelligent oscilloscope cooperative relay protection multi-port fault locating method described above are based on the same concept. For details not described in detail in the technical scheme of the intelligent oscilloscope cooperative relay protection multi-port fault locating system in this embodiment, please refer to the description of the technical scheme of the intelligent oscilloscope cooperative relay protection multi-port fault locating method described above.

[0154] Example 2

[0155] This embodiment provides a multi-port fault location system for an intelligent oscilloscope coordinated with relay protection, including:

[0156] Configuration module, used to set multiple trigger conditions and sampling windows of the intelligent recorder, and set the relay protection to three-stage protection form;

[0157] The acquisition module is used to set the judgment threshold of multiple trigger conditions. If the judgment threshold of multiple trigger conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection. The collected fault recording data and relay protection action information are synchronized through a unified timestamp mark;

[0158] The identification module is used to perform Fourier decomposition on the fault recording data with timestamp marks, extract the fundamental component and harmonic components, construct the fault feature vector, and combine it with the relay protection action information to complete the fault type identification through the fault classification model and generate a fault analysis report.

[0159] Example 3

[0160] This embodiment provides an electronic device suitable for locating multi-port faults in conjunction with an intelligent oscilloscope and relay protection, including:

[0161] Memory and processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the multi-port fault location method of intelligent recorder coordinated relay protection as proposed in the above embodiment.

[0162] This embodiment further provides a storage medium storing a computer program, which, when executed by a processor, implements the multi-port fault location method of intelligent oscilloscope coordinated relay protection as proposed in the above embodiment.

[0163] The storage medium proposed in this embodiment and the method for realizing multi-port fault location of intelligent recorder and collaborative relay protection proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0164] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a computer's floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0165] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-port fault location method using an intelligent oscilloscope and relay protection, characterized in that: include: Set multiple trigger conditions and sampling windows for the intelligent recorder, and set the relay protection to three-stage protection; Setting a determination threshold of the multiple trigger conditions; if the determination threshold of the multiple trigger conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection, and the collected fault recording data and relay protection action information are synchronized through a unified timestamp mark; The fault recording data with timestamp is subjected to Fourier decomposition to extract the fundamental component and harmonic component, and the fault feature vector is constructed. Combined with the relay protection action information, the fault type is identified through the fault classification model, and a fault analysis report is generated.

2. The multi-port fault location method using an intelligent oscilloscope and relay protection as claimed in claim 1, wherein: include: Set multiple trigger conditions and sampling windows for the intelligent recorder, and set the relay protection to three-stage protection, including: The multiple trigger conditions of the intelligent recorder include voltage amplitude trigger condition, current amplitude trigger condition, phasor mutation trigger condition and frequency deviation trigger condition; The sampling window of the intelligent wave recorder includes a pre-trigger sampling section and a post-trigger sampling section; The three-stage protection of the relay protection includes an instantaneous action stage, a delayed action stage and a backup protection stage.

3. The multi-port fault location method using an intelligent oscilloscope and relay protection as claimed in claim 1, wherein: The collected relay protection action information includes: The relay protection action information includes protection start time, protection action time, fault type judgment result, fault phase, protection action phase, and protection action stage; After receiving the coordinated trigger signal, the relay protection records the protection start time; Monitor the relay protection action status and record the protection action time when a trip command is issued to the circuit breaker; Generating a fault type judgment result based on the detection results of the zero-sequence current and the negative-sequence current; By comparing the amplitude change and phase angle change of each phase voltage and each phase current before and after the fault occurs, the fault phase and protection action phase are determined; The protection action stage is determined by comparing the fault current amplitude with the protection setting value.

4. The method for locating multi-port faults using an intelligent oscilloscope and relay protection according to claim 3, wherein: Time synchronization through unified timestamps includes: The relay protection receives the clock synchronization signal sent by the master clock server and calibrates the local clock; The relay protection adopts a double buffer, the first buffer receives the fault recording data and protection action information sent by the intelligent recorder, and the second buffer performs data synchronization processing; A timestamp in a unified format is added to the fault recording data and the protection action information written into the first buffer. When the first buffer is full, the fault recording data and relay protection action information with the timestamp are transferred to the second buffer. The data are time-aligned according to the timestamp to generate a synchronous data set containing the fault recording data and relay protection action information.

5. The multi-port fault location method using an intelligent oscilloscope and relay protection as claimed in claim 4, characterized in that: Perform Fourier decomposition on the fault recording data with time stamp, extract the fundamental component and harmonic component, and construct the fault feature vector including: Establishing sliding data windows for the pre-trigger sampling segment and the post-trigger sampling segment respectively according to the timestamp mark, wherein the length of the sliding data window is one power frequency cycle, and extracting the fundamental component and the harmonic component by recursive discrete Fourier transform within the sliding data window, wherein the harmonic component includes the third harmonic to the thirteenth harmonic; The phasor change rate, harmonic content change rate and zero-sequence current change rate are calculated based on the fundamental component and the harmonic component; wherein the phasor change rate includes the amplitude change rate and the phase angle change rate, the amplitude change rate is obtained by calculating the amplitude mean of the pre-trigger sampling segment and the amplitude mean of the post-trigger sampling segment, the phase angle change rate is obtained by calculating the phase angle mean of the pre-trigger sampling segment and the phase angle mean of the post-trigger sampling segment, the harmonic content change rate is obtained by calculating the harmonic content mean of the pre-trigger sampling segment and the harmonic content mean of the post-trigger sampling segment, and the zero-sequence current change rate is obtained by calculating the zero-sequence current mean of the pre-trigger sampling segment and the zero-sequence current mean of the post-trigger sampling segment; A fault characteristic vector is constructed, and the phasor change rate, harmonic content change rate and zero-sequence current change rate are combined to generate the fault characteristic vector.

6. The method for locating multi-port faults using an intelligent oscilloscope and relay protection according to claim 5, wherein: In combination with the relay protection action information, the fault type identification is completed through the fault classification model, including: According to the protection action stage and fault type judgment result in the relay protection action information, combined with the fault feature vector, the fault type is identified by a pre-trained fault classification model, and the feature data is input into the fault classification model; The pre-trained fault classification model is constructed using a support vector machine algorithm to output fault type identification probabilities, including single-phase grounding fault probability, two-phase short circuit fault probability, two-phase grounding fault probability, and three-phase short circuit fault probability; The difference between the maximum probability value and the second largest probability value in the fault type identification probability is calculated. When the difference is greater than a preset confidence threshold, the fault type corresponding to the maximum probability value is used as the final identification result; when the difference is less than or equal to the preset confidence threshold, the data is input into the preset expert rule library to determine the fault type and obtain the final identification result.

7. The method for locating multi-port faults using an intelligent oscilloscope and relay protection according to claim 6, wherein: Generate a fault analysis report including: When the matching degree between the recognition result of the fault classification model and the fault type judgment result of the relay protection action information exceeds a preset matching threshold, the final fault type is confirmed; According to the fault type, fault characteristic vector and protection action stage, an adaptive fault location algorithm is used to calculate the fault point location, and a fault analysis report including the final fault type, fault point location, fault phase and fault occurrence time is generated.

8. An intelligent oscilloscope coordinated relay protection multi-port fault location system, using the intelligent oscilloscope coordinated relay protection multi-port fault location method according to any one of claims 1 to 7, characterized in that: include, Configuration module, used to set multiple trigger conditions and sampling windows of the intelligent recorder, and set the relay protection to three-stage protection form; An acquisition module is used to set the judgment threshold of the multiple trigger conditions. If the judgment threshold of the multiple trigger conditions is met, the intelligent recorder triggers sampling and sends a coordinated trigger signal to the relay protection. The collected fault recording data and relay protection action information are synchronized through a unified timestamp mark; The identification module is used to perform Fourier decomposition on the fault recording data with time stamp marks, extract the fundamental component and harmonic component, construct the fault feature vector, and combine it with the relay protection action information to complete the fault type identification through the fault classification model and generate a fault analysis report.

9. An electronic device, characterized in that: include: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the intelligent recorder collaborative relay protection multi-port fault locating method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that It stores computer-executable instructions, which, when executed by a processor, can implement the steps of the intelligent oscilloscope recorder coordinated relay protection multi-port fault locating method described in any one of claims 1 to 7.

Images