Fault analysis method and system for new energy station current collection line

Abstract

The invention relates to the technical field of operation and maintenance of a new energy power system, and discloses a fault analysis method and system for a current collection line of a new energy station, and the method comprises a signal synchronous collection step, an early defect recognition step, a branch network modeling step, a fault interval judgment step and a visual mapping step. The system corresponds to the method. The system can accurately adapt to the complex topology of multiple branches, multiple power generation ends and short spacing of the new energy station current collection line, can effectively identify early gradual defects such as joint damping and insulation reduction and can give an early warning in advance, and through non-intrusive deployment of the distributed traveling wave sensors, an existing power transmission system does not need to be transformed, and the reliability of the system is improved. Manual patrol workload and fault recovery time are greatly reduced, operation and maintenance efficiency is remarkably improved, risks of non-planned shutdown and unit off-network can be reduced, station power generation reliability is practically guaranteed, and the method is integrally suitable for full-period operation and maintenance of a new energy station current collection line.

Description

Technical Field

[0001] This application relates to the field of operation and maintenance technology of new energy power systems, specifically a fault analysis method and system for the collection lines of new energy power plants. Background Technology

[0002] With the large-scale development of new energy power plants, the collection lines, as the core carriers for power collection and transmission, operate in harsh environments and have increasingly complex structures. Large-scale new energy power plants typically contain multiple collection lines, each connecting to multiple generator terminals (such as wind turbines and photovoltaic inverters), forming a topology similar to a small-scale multi-terminal transmission system with extremely short spacing between generator terminals. Furthermore, collection lines are often laid in complex terrains such as mountains and plains, and are frequently affected by factors such as lightning, partial discharge, and changes in ambient temperature and humidity, resulting in frequent faults, with single-phase grounding faults accounting for as much as 80%. If such faults are not addressed promptly, non-effectively grounded new energy power plants (such as wind farms) are prone to exacerbating the fault due to continuous operation with the fault, leading to large-scale unit disconnection from the grid, causing power outages, damage to power equipment, and huge economic and social losses. Therefore, the need for prevention, early warning, and rapid location and repair of collection line faults is extremely urgent.

[0003] Currently, fault analysis and location of power collection lines in new energy power plants mainly rely on two types of technical methods:

[0004] One method is offline location, which uses offline fault ranging instruments (such as the RD / T series from Reddy's Instruments in the UK, the TRY type instrument from SPIRENT, and ranging equipment developed by Xi'an University of Electronic Science and Technology in China) to locate faults after the fact. This method requires manual on-site operation, which is time-consuming, labor-intensive, and has low positioning efficiency. It is difficult to meet the needs of quickly restoring power generation after a fault, and in some scenarios, it cannot even achieve effective positioning.

[0005] Secondly, online monitoring technology is needed. Although online ranging devices based on the traveling wave principle have emerged, conventional traveling wave ranging algorithms are designed for simple line designs and cannot adapt to the complex structure of multi-branch (T-connection) collector lines in new energy power plants. This makes it difficult to accurately determine the branch where the fault is located, leading to an expansion of the fault investigation scope and a longer recovery time. Furthermore, gradual defects such as moisture absorption and reduced insulation are widespread in collector lines. The partial discharge signals of these defects are so weak that existing partial discharge detection technologies cannot capture them. Although dielectric loss detection can detect defects, it cannot locate their positions. At the same time, the insulation layer of these defects has strong tolerance to high-voltage pulses. After multiple pulse impacts, the insulation performance may even temporarily recover, further increasing the difficulty of early warning and making defects prone to developing into sudden faults.

[0006] Therefore, there is an urgent need for a fault analysis technology for power collection lines that can adapt to the complex topology of multiple branches in new energy power plants and has the ability to identify early defects, so as to fill the gap in existing technologies and ensure the safe and stable operation of new energy power plants. Summary of the Invention

[0007] The purpose of this application is to provide a fault analysis method and system for the power collection lines of new energy power plants, so as to solve the technical problems mentioned in the background art.

[0008] To achieve the above objectives, this application discloses the following technical solutions:

[0009] This application discloses a fault analysis method for power collection lines in new energy power plants in its first aspect, the method comprising:

[0010] Signal synchronization acquisition steps: Deploy distributed traveling wave sensors and acquire transient traveling wave signals of the power collection lines of the new energy power station in real time through the distributed traveling wave sensors; perform high-speed temporary storage of the acquired transient traveling wave signals and ensure real-time transmission of the acquired data to eliminate signal loss during the acquisition process of transient traveling wave signals;

[0011] Early defect identification steps: The transient traveling wave signal is processed by wavelet denoising technology, and the high-frequency decomposition coefficient of the signal is processed by a preset high-frequency threshold and then globally reconstructed. Based on the preset signal feature map of common defects in the collector line, low-frequency information with fault characteristics is extracted to determine the type of early defects in the line and the branch where the corresponding defects are located.

[0012] Branch network modeling steps: Based on the transmission equation, construct a multi-port network model adapted to the multi-branch collector lines of new energy power plants. The multi-port network model corresponds to the actual branch topology of the collector lines. Then, process the traveling wave reflection and transmission characteristics of the multi-branch lines through parameter estimation methods and output the traveling wave propagation characteristic parameters of each branch line.

[0013] Fault section determination steps: Preset different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines. Simulate the traveling wave waveform of the hypothetical fault point based on the multi-port network model and compare it with the actual transient traveling wave signal waveform after global reconstruction. Measure the degree of deviation between the actual waveform corresponding to the real fault and the traveling wave waveform corresponding to the hypothetical fault point through the vector deviation objective function, and take the branch line where the hypothetical fault point with the smallest deviation is located and its corresponding physical location as the fault section determination result.

[0014] Visualization mapping steps: Based on the fault range determination results, and using the physical location information of the collector line topology, cable joints, and cable cross-connection boxes, a visual fault map with the fault location marked is generated.

[0015] Preferably, the method of arranging distributed traveling wave sensors and synchronously acquiring transient traveling wave signals of the power collection lines of new energy power plants in real time includes the following steps:

[0016] Based on the characteristics of the multi-branch T-connection structure of the power collection lines of new energy power stations and the short spacing between the generating ends, traveling wave sensors are deployed at the T-joints of the power collection lines, the outlets of the transformer substations, near the cable cross-connection boxes, and at the segment nodes of the lines.

[0017] Time synchronization control of each traveling wave sensor is achieved through a satellite navigation system;

[0018] When the traveling wave sensor acquires transient traveling wave signals, an absolute timestamp is added to the acquired signal; and a second pulse synchronization signal is periodically sent to each traveling wave sensor through the satellite navigation system to trigger the internal timestamp counter of the traveling wave sensor to be cleared and recounted; and the counting deviation of the timestamp counter is periodically corrected through the carrier phase signal of the satellite navigation system to control the timestamp deviation of the signals acquired by different traveling wave sensors to be within a preset range.

[0019] By using time-synchronized traveling wave sensors, transient traveling wave signals of the power collection lines of new energy power plants are collected in real time.

[0020] Preferably, the high-speed temporary storage of the acquired transient traveling wave signal and the guarantee of real-time transmission of the acquired data include the following steps:

[0021] The acquired transient traveling wave signals are stored in a high-speed buffer with a preset storage capacity, and the transient traveling wave signals acquired first are given priority to be stored in the buffer.

[0022] When the number of transient traveling wave signals stored in the buffer reaches a preset proportion of the total storage capacity of the buffer, the data is triggered to be transmitted to the background in real time with a preset transmission bandwidth. During the data transmission process, the buffer continuously receives and stores newly acquired transient traveling wave signals.

[0023] Preferably, the process of processing the transient traveling wave signal using wavelet denoising technology and then globally reconstructing the signal after processing the high-frequency decomposition coefficients by setting a preset high-frequency threshold includes the following steps:

[0024] The acquired transient traveling wave signal is decomposed into multiple scales with a preset number of layers using a wavelet basis adapted to the characteristics of the transient traveling wave signal, to obtain low-frequency decomposition coefficients and high-frequency decomposition coefficients.

[0025] The high-frequency decomposition coefficients are subjected to soft thresholding by a preset high-frequency threshold. The processing rule is as follows: when the absolute value of the high-frequency decomposition coefficient is greater than the preset high-frequency threshold, the difference between the high-frequency decomposition coefficient and the preset high-frequency threshold is output as the new high-frequency decomposition coefficient; when the absolute value of the high-frequency decomposition coefficient is less than or equal to the preset high-frequency threshold, 0 is output as the new high-frequency decomposition coefficient.

[0026] The new high-frequency decomposition coefficients are globally reconstructed with the low-frequency decomposition coefficients to obtain the denoised transient traveling wave signal.

[0027] Preferably, the global reconstruction of the new high-frequency decomposition coefficients and the low-frequency decomposition coefficients includes the following steps:

[0028] The new high-frequency decomposition coefficients are classified and matched with the low-frequency decomposition coefficients according to the wavelet basis type at the time of decomposition.

[0029] By using the inverse transform algorithm corresponding to multi-scale decomposition, the newly classified high-frequency decomposition coefficients and low-frequency decomposition coefficients are combined into a signal, and the frequency band information corresponding to each coefficient is kept the same as the frequency band distribution of the original transient traveling wave signal during the synthesis process.

[0030] The complete global reconstructed signal is output through the inverse transform algorithm as the denoised transient traveling wave signal.

[0031] Preferably, the method of extracting low-frequency information with fault characteristics based on a preset signal feature map of common defects in power collection lines, and determining the type of early defects in the line and the corresponding branch where the defects are located, includes the following steps:

[0032] Based on the frequency bands of fault features corresponding to different defect types in the reference signal feature spectrum, the range of low-frequency information extraction frequency bands is set.

[0033] The denoised transient traveling wave signal is subjected to frequency band separation, and a signal segment matching the extracted frequency band range is extracted. This segment is defined as low-frequency information with fault characteristics.

[0034] Based on the feature types corresponding to different defects in the signal feature spectrum, the corresponding feature parameters are extracted from the separated low-frequency information.

[0035] The extracted feature parameters are compared with typical defect features in the signal feature map band by band. When the comparison accuracy exceeds the preset ratio, the corresponding defect type is determined, and the branch where the defect is located is located by the sensor deployment location.

[0036] Preferably, the method for constructing a multi-port network model adapted to multi-branch collection lines of new energy power plants based on transmission equations includes the following steps:

[0037] The number of model ports and the corresponding relationships between each port are determined based on the transmission equation in transmission line theory and the actual branch topology of the collection lines of new energy power plants.

[0038] Each branch of the collector line is equivalent to a transmission line unit containing distributed resistance, distributed inductance, and distributed capacitance. Based on the actual connection method of each branch, the corresponding relationship between each transmission line unit and the model port is established.

[0039] By describing the transmission, reflection, and transmission relationships of transient traveling waves between ports using transmission equations, a multi-port network model with the same model structure as the actual physical structure of the collector line is obtained.

[0040] Preferably, the method of processing the traveling wave reflection and transmission characteristics of multi-branch lines through parameter estimation and outputting the traveling wave propagation characteristic parameters of each branch line includes the following steps:

[0041] The line parameters in the multiport network model are initially estimated using parameter estimation methods. The line parameters include resistance parameters, inductance parameters, and capacitance parameters.

[0042] Based on the degree of influence of different terrains on line parameters and the variation law of line parameters under different climatic conditions, corresponding parameter correction coefficients are set to compensate for the preliminary estimated line parameters.

[0043] Based on the compensated line parameters and the structural characteristics of the mixed section of overhead lines and cables, the propagation speed of the traveling wave in the mixed section is corrected.

[0044] The corrected traveling wave propagation velocity, traveling wave reflection coefficient, and traveling wave transmission coefficient are output as traveling wave propagation characteristic parameters.

[0045] As a preferred embodiment, the method of presetting different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines includes: preferentially presetting hypothetical fault points at a preset interval on the branch where the early defect type is located, and presetting hypothetical fault points at another preset interval on other branch lines.

[0046] The method of measuring the degree of deviation using the vector deviation objective function includes: simulating the traveling wave waveform generated at each hypothetical fault point and the path of propagation to each sensor based on the traveling wave propagation characteristic parameters output from the branch network modeling step; extracting the wavefront arrival time, peak amplitude, and fluctuation period characteristics of the actual transient traveling wave signal waveform after global reconstruction; and calculating the vector angle between the simulated waveform and the actual waveform at each hypothetical fault point using the vector deviation objective function, wherein a smaller vector angle value indicates a smaller degree of deviation.

[0047] Secondly, this application discloses a fault analysis system for power collection lines in new energy power plants, which applies the fault analysis method for power collection lines in new energy power plants as described above. The system includes:

[0048] The signal acquisition module is configured to: synchronously acquire transient traveling wave signals of the power collection lines of the new energy power station in real time based on the distributed traveling wave sensors; perform high-speed temporary storage of the acquired transient traveling wave signals and ensure real-time transmission of the acquired data, thereby eliminating signal loss during the acquisition of transient traveling wave signals;

[0049] The early identification module is configured to: process the transient traveling wave signal using wavelet denoising technology, and perform global reconstruction after processing the high-frequency decomposition coefficients of the signal using a preset high-frequency threshold; extract low-frequency information with fault characteristics based on the preset signal feature map of common defects in the power line, and determine the type of early defects in the line and the branch where the corresponding defects are located.

[0050] The network modeling module is configured to: construct a multi-port network model adapted to the multi-branch collector lines of new energy power plants based on the transmission equation, wherein the multi-port network model corresponds to the actual branch topology of the collector lines; and process the traveling wave reflection and transmission characteristics of the multi-branch lines through parameter estimation methods, and output the traveling wave propagation characteristic parameters of each branch line.

[0051] The interval determination module is configured to: preset different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines; simulate the traveling wave waveform of the hypothetical fault point based on the multi-port network model; and compare it with the actual transient traveling wave signal waveform after global reconstruction; measure the degree of deviation between the actual waveform corresponding to the real fault and the traveling wave waveform corresponding to the hypothetical fault point through the vector deviation objective function; and take the branch line where the hypothetical fault point with the smallest deviation is located and its corresponding physical location as the fault interval determination result;

[0052] The results display module is configured to generate a visual fault map with the location of the fault point marked, based on the fault range determination results and the physical location information of the collector line topology, cable joints, and cable cross-connection boxes.

[0053] Beneficial Effects: The fault analysis method for collection lines in new energy power plants proposed in this application improves fault location accuracy to within the interval of two generator units by constructing a multi-port network model adapted to complex multi-branch topologies, achieving precise location of faulty branches; through wavelet denoising and defect signal feature maps, it effectively identifies early-stage gradual defects that are difficult to capture with existing technologies, such as joint damping and insulation degradation, preventing them from developing into sudden faults and reducing the risk of large-scale unit disconnection; based on the non-intrusive deployment of distributed traveling wave sensors, fault determination is completed by combining the vector deviation objective function without modifying the existing transmission system; through real-time signal acquisition and storage, priority fault investigation based on defect information, and visualization of fault location, the cycle of fault discovery, location, and repair is significantly shortened, reducing economic losses and manual inspection workload caused by unplanned downtime; at the same time, it realizes online and visualized management of the entire process of signal acquisition, defect identification, fault location, and result display, supporting maintenance personnel to remotely monitor line status and track fault trends, improving the controllability of collection line operation and maintenance and the reliability of power generation in new energy power plants. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart illustrating the fault analysis method for power collection lines in new energy power plants provided in this application embodiment;

[0056] Figure 2 This is a structural block diagram of a fault analysis system for power collection lines in new energy power plants, provided in an embodiment of this application. Detailed Implementation

[0057] The technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0058] In this document, the term "comprising" is intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0059] This embodiment aims to address the challenges of accurately locating faults using conventional methods in complex multi-branch topologies of power collection lines at new energy power plants, as well as the difficulty in identifying early defects such as joint dampness / insulation degradation, which can easily lead to sudden faults. It provides a solution... Figure 1 The fault analysis method shown is used for the collection lines of new energy power plants. The method includes a signal synchronization acquisition step, an early defect identification step, a branch network modeling step, a fault interval determination step, and a visualization mapping step.

[0060] In detail

[0061] Signal synchronization acquisition steps: Deploy distributed traveling wave sensors and acquire transient traveling wave signals from the power collection lines of new energy power plants in real time through the distributed traveling wave sensors; perform high-speed temporary storage on the acquired transient traveling wave signals and ensure real-time transmission of the acquired data to eliminate signal loss during the acquisition process of transient traveling wave signals.

[0062] In this step, distributed traveling wave sensors are deployed, and transient traveling wave signals from the power collection lines of the new energy power station are collected in real time using these sensors, including:

[0063] Based on the characteristics of the multi-branch T-connection structure of the power collection lines of new energy power stations and the short spacing between the generating ends, traveling wave sensors are deployed at the T-joints of the power collection lines, the outlets of the transformer substations, near the cable cross-connection boxes, and at the segment nodes of the lines.

[0064] The time synchronization control of each traveling wave sensor is achieved through a satellite navigation system. The time synchronization control includes: integrating a phase-locked module and a low-noise filtering unit into the timing circuit of each traveling wave sensor, and configuring a constant temperature control module and an aging compensation unit for the crystal oscillator of each sensor, so that the sensor outputs a timing signal with nanosecond-level accuracy.

[0065] When the traveling wave sensor acquires transient traveling wave signals, an absolute timestamp is added to the acquired signal; and a second pulse synchronization signal is periodically sent to each traveling wave sensor through the satellite navigation system to trigger the internal timestamp counter of the traveling wave sensor to be cleared and recounted; and the counting deviation of the timestamp counter is periodically corrected through the carrier phase signal of the satellite navigation system to control the timestamp deviation of the signals acquired by different traveling wave sensors to be within a preset range.

[0066] By using time-synchronized traveling wave sensors, transient traveling wave signals of the power collection lines of new energy power plants are collected in real time.

[0067] Based on this, by clearly defining the deployment locations of traveling wave sensors at key nodes such as T-junctions and transformer substation outlets, we can ensure that there are no blind spots in the acquisition of transient traveling wave signals and cover areas with high incidence of power line faults. At the same time, by relying on the satellite navigation system to achieve nanosecond-level time synchronization and microsecond-level timestamp deviation control, we can solve the problem of traveling wave arrival time difference calculation deviation caused by the low time synchronization accuracy of existing acquisition devices. This lays the data acquisition foundation for the accuracy of subsequent fault location and avoids positioning errors caused by unreasonable signal acquisition locations or time asynchrony.

[0068] Furthermore, the acquired transient traveling wave signals are temporarily stored at high speed, and real-time transmission of the acquired data is ensured, including:

[0069] The acquired transient traveling wave signals are stored in a high-speed buffer with a preset storage capacity, and the transient traveling wave signals acquired first are given priority to be stored in the buffer.

[0070] When the number of transient traveling wave signals stored in the buffer reaches a preset proportion of the total storage capacity of the buffer, the data is triggered to be transmitted to the backend in real time with a preset transmission bandwidth. During the data transmission process, the buffer continuously receives and stores newly acquired transient traveling wave signals.

[0071] Based on this, a high-speed buffer design with first-in-first-out logic is used, combined with a mechanism that triggers transmission when storage reaches a threshold and ensures uninterrupted storage during transmission, to solve the problem of dead zones and signal loss risks in existing equipment. This ensures that transient traveling wave signals (especially weak echo signals at the moment of a fault) are completely preserved, providing continuous and complete raw data for subsequent wavelet denoising and defect identification, and avoiding misjudgment of defects or missed faults due to missing data.

[0072] Early defect identification steps: The transient traveling wave signal is processed by wavelet denoising technology, and the high-frequency decomposition coefficients of the signal are processed by a preset high-frequency threshold and then globally reconstructed. Based on the preset signal feature spectrum of common defects in the collector line, low-frequency information with fault characteristics is extracted to determine the type of early defects in the line and the branch where the corresponding defects are located.

[0073] In this step, the transient traveling wave signal is processed using wavelet denoising technology, and the high-frequency decomposition coefficients of the signal are processed by a preset high-frequency threshold before global reconstruction, including:

[0074] The acquired transient traveling wave signal is decomposed into multiple scales with a preset number of layers using a wavelet basis adapted to the characteristics of the transient traveling wave signal (such as the db4 wavelet basis) to obtain low-frequency decomposition coefficients and high-frequency decomposition coefficients.

[0075] The high-frequency decomposition coefficients are subjected to soft thresholding by a preset high-frequency threshold. The processing rule is as follows: when the absolute value of the high-frequency decomposition coefficient is greater than the preset high-frequency threshold, the difference between the high-frequency decomposition coefficient and the preset high-frequency threshold is output as the new high-frequency decomposition coefficient; when the absolute value of the high-frequency decomposition coefficient is less than or equal to the preset high-frequency threshold, 0 is output as the new high-frequency decomposition coefficient.

[0076] The new high-frequency decomposition coefficients and low-frequency decomposition coefficients are globally reconstructed to obtain the denoised transient traveling wave signal.

[0077] It is feasible to preset the high-frequency threshold based on the background noise intensity of the collector line under fault-free conditions, calculate it using risk estimation criteria, and use the acquisition period of the transient traveling wave signal as the adjustment period. The threshold is dynamically corrected within a preset range by an adaptive algorithm based on the noise intensity of the current acquired signal and the average noise intensity of the historical acquired signals.

[0078] Based on this, by adapting wavelet basis selection to transient traveling wave characteristics and using soft thresholding of high-frequency decomposition coefficients, high-frequency noise (such as interference caused by high acquisition frequency) in the acquired signal is accurately filtered out, while retaining low-frequency components containing fault characteristics. This solves the problem that existing denoising techniques easily filter out useful signals or residual noise, outputting high-quality denoised signals. This provides a clean signal basis for subsequent extraction of low-frequency information of fault characteristics and determination of early defects, thereby improving the accuracy of defect identification.

[0079] Furthermore, the new high-frequency decomposition coefficients are globally reconstructed with the low-frequency decomposition coefficients, including:

[0080] The new high-frequency decomposition coefficients and low-frequency decomposition coefficients are classified according to the wavelet basis type at the time of decomposition.

[0081] By using the inverse transform algorithm corresponding to multi-scale decomposition, the newly classified high-frequency decomposition coefficients and low-frequency decomposition coefficients are combined into a signal, and the frequency band information corresponding to each coefficient is kept the same as the frequency band distribution of the original transient traveling wave signal during the synthesis process.

[0082] The complete global reconstructed signal is output through the inverse transform algorithm as a denoised transient traveling wave signal, which is then used to extract low-frequency information with fault characteristics from the signal feature map based on common defects in collector lines.

[0083] The signal feature spectrum of common defects in power collection lines is a pre-established transient traveling wave feature library of typical defects (such as joint dampness and insulation degradation). Different defect types correspond to characteristic waveforms of specific frequency bands in the reconstructed signal (such as characteristic peaks of multiple consecutive sampling periods, fluctuations of a specific period, etc.). When determining the early defect type, the extracted low-frequency information is compared with the feature library band by band. When the comparison accuracy exceeds the preset ratio, the corresponding defect type is determined, and the branch where the defect is located is located by the sensor deployment position.

[0084] After global reconstruction, low-frequency information refers to the signal separated from the reconstructed transient traveling wave signal by the frequency band corresponding to the low-frequency coefficients. This low-frequency information carries the core fault characteristics in the transient traveling wave signal and is different from the low-frequency decomposition coefficients in the multi-scale decomposition stage (the low-frequency decomposition coefficients are intermediate data in the decomposition process, while the low-frequency information is a signal segment that can be directly used for feature recognition after reconstruction).

[0085] Based on this, by standardizing the steps of coefficient classification, inverse transformation synthesis, and output reconstructed signal, we can ensure that the frequency band distribution of the reconstructed signal is consistent with that of the original transient traveling wave signal, and avoid the loss of fault characteristics (such as specific frequency band peaks of joint dampness and fluctuation patterns of insulation reduction) during the reconstruction process. This solves the problem of signal distortion caused by non-standard reconstruction methods, and ensures that the denoised signal can truly reflect the line fault or defect status, providing a reliable signal carrier for early defect identification.

[0086] Secondly, based on the pre-defined signal feature spectrum of common defects in collector lines, low-frequency information with fault characteristics is extracted to determine the type of early defects in the line and the corresponding branch where the defects are located, including:

[0087] Based on the frequency bands of fault features corresponding to different defect types in the reference signal feature spectrum, the range of low-frequency information extraction frequency bands is set.

[0088] The denoised transient traveling wave signal is subjected to frequency band separation, and a signal segment matching the extracted frequency band range is extracted. This segment is defined as low-frequency information with fault characteristics.

[0089] Based on the feature types corresponding to different defects in the signal feature spectrum (such as feature peaks corresponding to multiple consecutive sampling periods for one type of defect, and fluctuations corresponding to a specific pattern for another type of defect), corresponding feature parameters (such as peak shape, fluctuation pattern, etc.) are extracted from the separated low-frequency information.

[0090] The extracted feature parameters are compared with typical defect features in the signal feature map band by band. When the comparison accuracy exceeds the preset ratio, the corresponding defect type is determined, and the branch where the defect is located is located by the sensor deployment location.

[0091] Based on this, by referring to the characteristic spectrum to determine the frequency band, separate the signal, extract parameters, and compare and judge, early defects that are difficult to detect by existing technologies (partial discharge and dielectric loss detection), such as joint dampness and reduced insulation, can be accurately identified, and the branch where the defect is located can be located. This avoids the pain points of difficult defect detection and location, prevents defects from developing into sudden faults such as single-phase grounding, and reduces the risk of unit disconnection from the grid.

[0092] Branch network modeling steps: Based on the transmission equation, construct a multi-port network model adapted to the multi-branch collector lines of new energy power plants. The multi-port network model corresponds to the actual branch topology of the collector lines. Then, process the traveling wave reflection and transmission characteristics of the multi-branch lines through parameter estimation methods and output the traveling wave propagation characteristic parameters of each branch line.

[0093] In this step, a multi-port network model adapted to the multi-branch collector lines of new energy power plants is constructed based on the transmission equation, including:

[0094] Based on the transmission equations in transmission line theory and the actual branch topology of the power collection lines of new energy power plants (including T-joints, generator set access nodes, and cable / overhead line connection sections), the number of model ports and the corresponding relationships between each port are determined.

[0095] Each branch of the collector line is equivalent to a transmission line unit containing distributed resistance, distributed inductance, and distributed capacitance. Based on the actual connection method of each branch, the corresponding relationship between each transmission line unit and the model port is established.

[0096] By describing the transmission, reflection, and transmission relationships of transient traveling waves between ports using transmission equations, a multi-port network model with the same model structure as the actual physical structure of the collector line is obtained.

[0097] Based on this, and using the transmission equation and the actual topology of the collector line, the branches are equivalent to transmission line units with distributed parameters, and a multi-port network model with port-branch correspondence is established. This solves the problem that conventional models cannot adapt to complex structures such as multi-T wiring and mixed laying (cable / overhead line), ensuring that the model can truly reflect the transmission, reflection and transmission laws of transient traveling waves in multi-branch lines, and providing accurate model support for subsequent calculation of traveling wave propagation characteristics and fault waveform simulation.

[0098] Furthermore, the traveling wave reflection and transmission characteristics of multi-branch lines are processed using parameter estimation methods, and the traveling wave propagation characteristic parameters of each branch line are output, including:

[0099] The line parameters in the multiport network model are initially estimated using parameter estimation methods. The line parameters include resistance, inductance, and capacitance parameters.

[0100] Based on the degree of influence of different terrains on line parameters and the variation law of line parameters under different climatic conditions, corresponding parameter correction coefficients are set to compensate for the preliminary estimated line parameters.

[0101] Based on the compensated line parameters and the structural characteristics of the mixed section of overhead lines and cables, the propagation speed of the traveling wave in the mixed section is corrected.

[0102] The corrected traveling wave propagation velocity, traveling wave reflection coefficient, and traveling wave transmission coefficient are output as traveling wave propagation characteristic parameters. For the branch where the defect is located, as determined in the early defect identification step, the resistance and capacitance parameters are further corrected to accommodate the defect's influence before being included in the calculation of traveling wave propagation characteristic parameters. Specifically, this requires combining the defect type, defect severity, and line laying environment, and is achieved through a technical path of defect-parameter influence correlation modeling → dynamic matching of correction coefficients → integration of parameter correction and characteristic calculation, as detailed below:

[0103] 1. Establish the correlation between defect types and changes in line parameters.

[0104] Based on the physical characteristics of common early defects in power line collectors at new energy power plants (dampness at joints, reduced insulation), this study, through experimental testing and data accumulation, clarifies the impact of different defects on line resistance parameters (mainly distributed resistance) and capacitance parameters (mainly distributed capacitance).

[0105] Regarding the defects caused by moisture in the joint: Moisture will disrupt the conductive continuity of the joint interface, increase the contact impedance, and lead to an increase in the contact resistance at the joint; at the same time, the dielectric loss of the joint insulation layer will increase, and the equivalent distributed capacitance will increase slightly (for example, when slightly damp, the distributed resistance will increase by 10%-15% and the distributed capacitance will increase by 5%-8% compared with the normal state; when heavily damp, the distributed resistance will increase by 15%-25% and the distributed capacitance will increase by 8%-12%).

[0106] Regarding insulation degradation defects: aging and damage to the insulation layer will increase the leakage current of the insulation layer and increase the equivalent conductive path, resulting in a decrease in the equivalent distributed resistance of the line; at the same time, the change in the dielectric constant of the insulation layer will cause the distributed capacitance to increase significantly (for example, in the case of mild insulation degradation, the distributed resistance decreases by 5%-10% and the distributed capacitance increases by 12%-18%; in the case of severe insulation degradation, the distributed resistance decreases by 10%-18% and the distributed capacitance increases by 18%-25%).

[0107] 2. Construct a defect-adaptive correction coefficient library

[0108] Based on the aforementioned defect-parameter influence patterns, a pre-defined correction coefficient library is constructed. This library contains the correspondence between defect type, defect severity, environmental correction factor, resistance correction coefficient, and capacitance correction coefficient, as detailed below:

[0109] First, for the two core defects of joint dampness and reduced insulation, basic correction coefficients are set respectively (minor joint dampness: resistance basic correction coefficient 1.10-1.15, capacitance basic correction coefficient 1.05-1.08; severe joint dampness: resistance basic correction coefficient 1.15-1.25, capacitance basic correction coefficient 1.08-1.12; minor insulation reduction: resistance basic correction coefficient 0.90-0.95, capacitance basic correction coefficient 1.12-1.18; severe insulation reduction: resistance basic correction coefficient 0.82-0.90, capacitance basic correction coefficient 1.18-1.25).

[0110] Secondly, an environmental correction factor is introduced (to adapt to the terrain and climate conditions of the power line laying): In mountainous terrain, due to the high tension of the line laying and uneven stress on the joints, the correction factor is added by an additional 1.03-1.05 times on the base value; High temperature and high humidity environments will aggravate the impact of defects on parameters, so the correction factor is added by an additional 1.02-1.04 times; In plains and normal temperature and humidity environments, no additional factor is added, and the correction factor is taken as the base value (Note: High temperature and high humidity environment refers to relative humidity > 85% and temperature > 35℃; normal temperature and humidity environment refers to relative humidity ≤ 60% and temperature 15-25℃).

[0111] The correction coefficient library is pre-installed in the branch network modeling module and can be regularly updated and optimized based on the actual terrain and climate conditions and historical defect data of the new energy power station.

[0112] 3. Dynamic matching and parameter correction of correction coefficients

[0113] In the branch network modeling step, after the early defect identification step outputs the determination result of the branch where the defect is located + defect type + defect severity, the correction coefficients are matched and the parameters are corrected according to the following process:

[0114] The first step is to extract key defect information: from the early defect identification results, obtain the identifier of the branch where the defect is located (such as the T-connection branch number), the defect type (dampness in the joint / reduced insulation), and the severity of the defect (mild / severe). At the same time, read the terrain (mountainous / plain) and real-time climate (temperature and humidity) data of the branch's laying location.

[0115] The second step is to match the correction coefficients: based on the above information, call the preset correction coefficient library to determine the corresponding resistance correction coefficient and capacitance correction coefficient; for example, if the T3 branch of a certain mountain station is determined to be severely damp at the joint, and the real-time environment is high temperature and high humidity, then the resistance correction coefficient = the basic resistance coefficient of the severely damp joint (1.15-1.25) × the mountain environment factor (1.03-1.05) × the high temperature and high humidity factor (1.02-1.04), and the capacitance correction coefficient = the basic capacitance coefficient of the severely damp joint (1.08-1.12) × the mountain environment factor (1.03-1.05) × the high temperature and high humidity factor (1.02-1.04).

[0116] The third step is to correct the line parameters: the distributed resistance and distributed capacitance parameters of the branch where the defect is located, which are initially estimated by parameter estimation methods (such as the least squares method), are multiplied by the resistance correction coefficient and capacitance correction coefficient obtained by matching, respectively, to obtain the corrected resistance and capacitance parameters adapted to the effect of the defect; the inductance parameters are not further corrected because they are less affected by the defect, and the initial estimated value is used.

[0117] 4. Integration of corrected parameters with traveling wave propagation characteristic calculations

[0118] The corrected resistance and capacitance parameters (and the uncorrected inductance parameters) are substituted into the transmission equations of the multiport network model to participate in the calculation of the traveling wave propagation characteristic parameters, as follows:

[0119] Based on the corrected distribution parameters, the wave velocity of the traveling wave in the branch where the defect is located is recalculated (the wave velocity satisfies the formula for the line parameters). Where L is the corrected inductance and C is the corrected capacitance), the wave velocity deviation caused by the defect is corrected.

[0120] By combining the corrected resistance parameters, the attenuation coefficient of the traveling wave in the branch where the defect is located is adjusted (the attenuation of the traveling wave intensifies when the resistance increases), and the calculation accuracy of the traveling wave reflection coefficient and transmission coefficient is optimized.

[0121] The final output traveling wave propagation characteristic parameters (corrected wave velocity, reflection coefficient, and transmission coefficient) can accurately reflect the impact of the defect on the propagation of the traveling wave in this branch, providing a more realistic model basis for the simulation of the waveform of the hypothetical fault point in the subsequent fault interval determination step.

[0122] Based on this, by combining parameter estimation with terrain and climate compensation, the parameters of line resistance, inductance, and capacitance are corrected. At the same time, the traveling wave velocity in the mixed section is optimized, and accurate traveling wave propagation characteristic parameters (wave velocity, reflection / transmission coefficient) are output. This provides accurate parameter basis for the simulation of the waveform of the hypothetical fault point in the fault section determination, avoids the deviation between the simulated waveform and the actual waveform caused by inaccurate parameters, and improves the fault location accuracy.

[0123] Fault section determination steps: Preset different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines. Simulate the traveling wave waveform of the hypothetical fault point based on the multi-port network model and compare it with the actual transient traveling wave signal waveform after global reconstruction. Measure the degree of deviation between the actual waveform corresponding to the real fault and the traveling wave waveform corresponding to the hypothetical fault point through the vector deviation objective function. The branch line where the hypothetical fault point with the smallest deviation is located and its corresponding physical location are taken as the fault section determination result.

[0124] In this step, different hypothetical fault points are preset on the branch where the early defect type of the line is located and other branch lines, including: firstly, hypothetical fault points are preset at preset intervals on the branch where the early defect type is located, and hypothetical fault points are preset at another preset interval on other branch lines.

[0125] Based on this, by placing points at close intervals on priority defect branches and at sparse intervals on other branches, the scope of fault investigation is narrowed and the efficiency of judgment is improved.

[0126] In this step, the degree of deviation is measured by the vector deviation objective function, including: simulating the traveling wave waveform generated at each hypothetical fault point and the path of propagation to each sensor based on the traveling wave propagation characteristic parameters output from the branch network modeling step; extracting the wavefront arrival time, peak amplitude, and fluctuation period characteristics of the actual transient traveling wave signal waveform after global reconstruction; and calculating the vector angle between the simulated waveform and the actual waveform at each hypothetical fault point using the vector deviation objective function, where a smaller vector angle value indicates a smaller degree of deviation. The formula for calculating the vector deviation objective function is: objective function value = 1 - (dot product of two waveforms) / (product of the magnitudes of two waveforms).

[0127] Based on this, the waveform angle is calculated using the vector deviation objective function to quantify the degree of deviation between the real fault and the hypothetical fault point. This solves the problems of existing fault judgment relying on experience, large investigation scope, and lack of quantitative standards for deviation. It can quickly locate the fault point with the smallest deviation, ensure the efficiency and accuracy of fault range judgment, and avoid the waste of operation and maintenance resources caused by misjudgment.

[0128] Visualization mapping steps: Based on the fault range determination results, and using the physical location information of the collector line topology, cable joints, and cable cross-connection boxes, a visual fault map with the fault location marked is generated.

[0129] Specifically, this step involves constructing a visual fault map of the power collection line using AutoCAD Electrical software. First, the CAD drawing of the power collection line topology (including coordinate parameters of T-joints and cable cross-connection boxes) is imported into the software. Its electrical component library is used to digitally model the line route and equipment locations, ensuring that each branch line matches the actual laying path. Next, fault location information, such as the T2 branch being 250 meters from the transformer outlet, is extracted from the fault section determination results. Using the software's dynamic annotation function, a red triangle icon is overlaid on the corresponding line node, and fault type labels (such as joint moisture defects) are automatically associated. Different colors are used to distinguish line status: faulty branches are displayed with bold red lines, normal branches maintain blue lines, and critical equipment such as cable cross-connection boxes are highlighted with yellow square icons. Finally, an interactive map containing the fault point's latitude and longitude and adjacent equipment models is generated. This map can be exported as a DWG file or synchronized to the site's SCADA system via a software interface. Maintenance personnel can directly click on the fault icon on the map to view detailed traveling wave waveform data and location information.

[0130] In summary, the fault analysis method for power collection lines in new energy power plants presented in this embodiment constructs an analysis framework adapted to the complex topology of multi-branch (T-connection) power plants through a full-process design of signal synchronous acquisition, early defect identification, branch network modeling, fault zone determination, and visualization mapping. This addresses the pain point of conventional methods being unable to adapt to multiple generator terminals and short-distance lines. It integrates early defect identification and fault location to achieve early warning, preventing gradual defects from developing into sudden faults. Furthermore, through the vector deviation objective function and visualization mapping, it achieves accurate fault zone determination and intuitive presentation, significantly shortening fault investigation time and providing core technical support for the integrated operation and maintenance of prevention, location, and repair of power collection lines in new energy power plants.

[0131] In this embodiment, in a second aspect, a fault analysis system for power collection lines in new energy power plants is provided, applying the fault analysis method for power collection lines in new energy power plants as described above, such as... Figure 2 As shown, the system includes:

[0132] The signal acquisition module is configured to: acquire transient traveling wave signals of the power collection lines of the new energy power station in real time based on the deployed distributed traveling wave sensors; perform high-speed temporary storage of the acquired transient traveling wave signals and ensure real-time transmission of the acquired data, eliminating signal loss during the acquisition process of transient traveling wave signals;

[0133] The early identification module is configured to: process the transient traveling wave signal (the complete transient traveling wave signal data composed of the acquired transient traveling wave signal and the temporarily stored transient traveling wave signal) using wavelet denoising technology, and perform global reconstruction after processing the high-frequency decomposition coefficients of the signal through a preset high-frequency threshold; and extract low-frequency information with fault characteristics based on the preset signal feature map of common defects in the collector line to determine the type of early defects in the line and the branch where the corresponding defects are located.

[0134] The network modeling module is configured to: receive complete transient traveling wave signal data and the actual branch topology of the collector lines corresponding to the traveling wave sensor layout transmitted by the signal acquisition module; construct a multi-port network model adapted to the multi-branch collector lines of new energy power plants based on the transmission equation; and process the traveling wave reflection and transmission characteristics of the multi-branch lines through parameter estimation methods, and output the traveling wave propagation characteristic parameters of each branch line.

[0135] The interval determination module is configured to: preset different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines; simulate the traveling wave waveform of the hypothetical fault point based on the multi-port network model; and compare it with the actual transient traveling wave signal waveform after global reconstruction; measure the degree of deviation between the actual waveform corresponding to the real fault and the traveling wave waveform corresponding to the hypothetical fault point through the vector deviation objective function; and take the branch line where the hypothetical fault point with the smallest deviation is located and its corresponding physical location as the fault interval determination result.

[0136] The results display module is configured to generate a visual fault map with the location of the fault point marked, based on the fault range determination results and the physical location information of the collector line topology, cable joints, and cable cross-connection boxes.

[0137] The data flow between modules is as follows: Signal Acquisition Module → Early Identification Module: Transmits real-time synchronously acquired transient traveling wave signals (including timestamps) and complete signal data after high-speed temporary storage; Signal Acquisition Module → Network Modeling Module: Transmits basic information of the collector line topology associated with sensor deployment locations (such as T-joints and transformer outlet node locations); Early Identification Module → Network Modeling Module: Transmits the global reconstruction signal after wavelet denoising, the determined early defect type, and the branch identifier where the defect is located; Early Identification Module → Interval Determination Module: Transmits the global reconstruction signal after wavelet denoising, the determined early defect type, and the branch identifier where the defect is located; Network Modeling Module → Interval Determination Module: Transmits multi-port network model parameters (port correspondence, transmission line unit equivalent parameters), and calculated traveling wave propagation characteristic parameters (corrected wave velocity, reflection coefficient, and transmission coefficient); Interval Determination Module → Result Display Module: Transmits the fault interval determination results (the branch where the hypothetical fault point with the smallest deviation is located and its physical location); Signal Acquisition Module → Result Display Module: Transmits complete topology information of the collector line; Network Modeling Module → Result Display Module: Transmits the physical location information of cable joints and cable cross-connection boxes.

[0138] It should be noted that the fault analysis system for power collection lines of new energy power plants in this embodiment corresponds to the aforementioned fault analysis method for power collection lines of new energy power plants. Therefore, the parts of the fault analysis system for power collection lines of new energy power plants in this embodiment that are not described in detail (including but not limited to specific technical means and technical effects) can be referred to the description in the aforementioned fault analysis method for power collection lines of new energy power plants. This text will not repeat them here.

[0139] In the embodiments provided in this application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor may be implemented in one or more of the following: application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to implement the functions described herein, or combinations thereof. For software implementation, some or all of the processes of the embodiments may be performed by a computer program instructing the associated hardware. During implementation, the program may be stored in a computer-readable storage medium or transmitted as one or more instructions or code on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media may be any available medium accessible to a computer. Computer-readable storage media may include, but are not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer.

[0140] Finally, it should be noted that the above description is only a preferred embodiment of this application and is not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A fault analysis method for power collection lines in new energy power plants, characterized in that, The method includes: Signal synchronization acquisition steps: Deploy distributed traveling wave sensors and acquire transient traveling wave signals of the power collection lines of the new energy power station in real time through the distributed traveling wave sensors; perform high-speed temporary storage of the acquired transient traveling wave signals and ensure real-time transmission of the acquired data to eliminate signal loss during the acquisition process of transient traveling wave signals; Early defect identification steps: The transient traveling wave signal is processed by wavelet denoising technology, and the high-frequency decomposition coefficient of the signal is processed by a preset high-frequency threshold and then globally reconstructed. Based on the preset signal feature map of common defects in the collector line, low-frequency information with fault characteristics is extracted to determine the type of early defects in the line and the branch where the corresponding defects are located. Branch network modeling steps: Based on the transmission equation, construct a multi-port network model adapted to the multi-branch collector lines of new energy power plants. The multi-port network model corresponds to the actual branch topology of the collector lines. Then, process the traveling wave reflection and transmission characteristics of the multi-branch lines through parameter estimation methods and output the traveling wave propagation characteristic parameters of each branch line. Fault section determination steps: Preset different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines. Simulate the traveling wave waveform of the hypothetical fault point based on the multi-port network model and compare it with the actual transient traveling wave signal waveform after global reconstruction. Measure the degree of deviation between the actual waveform corresponding to the real fault and the traveling wave waveform corresponding to the hypothetical fault point through the vector deviation objective function, and take the branch line where the hypothetical fault point with the smallest deviation is located and its corresponding physical location as the fault section determination result. Visualization mapping steps: Based on the fault range determination results, and using the physical location information of the collector line topology, cable joints, and cable cross-connection boxes, a visual fault map with the fault location marked is generated.

2. The fault analysis method for power collection lines in new energy power plants according to claim 1, characterized in that, The aforementioned deployment of distributed traveling wave sensors, and the real-time synchronous acquisition of transient traveling wave signals from the power collection lines of new energy power plants through these distributed traveling wave sensors, includes the following steps: Based on the characteristics of the multi-branch T-connection structure of the power collection lines of new energy power stations and the short spacing between the generating ends, traveling wave sensors are deployed at the T-joints of the power collection lines, the outlets of the transformer substations, near the cable cross-connection boxes, and at the segment nodes of the lines. Time synchronization control of each traveling wave sensor is achieved through a satellite navigation system; When the traveling wave sensor acquires transient traveling wave signals, an absolute timestamp is added to the acquired signal; and a second pulse synchronization signal is periodically sent to each traveling wave sensor through the satellite navigation system to trigger the internal timestamp counter of the traveling wave sensor to be cleared and recounted; and the counting deviation of the timestamp counter is periodically corrected through the carrier phase signal of the satellite navigation system to control the timestamp deviation of the signals acquired by different traveling wave sensors to be within a preset range. By using time-synchronized traveling wave sensors, transient traveling wave signals of the power collection lines of new energy power plants are collected in real time.

3. The fault analysis method for power collection lines in new energy power plants according to claim 2, characterized in that, The high-speed temporary storage of the acquired transient traveling wave signal and the guarantee of real-time transmission of the acquired data include the following steps: The acquired transient traveling wave signals are stored in a high-speed buffer with a preset storage capacity, and the transient traveling wave signals acquired first are given priority to be stored in the buffer. When the number of transient traveling wave signals stored in the buffer reaches a preset proportion of the total storage capacity of the buffer, the data is triggered to be transmitted to the background in real time with a preset transmission bandwidth. During the data transmission process, the buffer continuously receives and stores newly acquired transient traveling wave signals.

4. The fault analysis method for power collection lines in new energy power plants according to claim 1, characterized in that, The process of processing transient traveling wave signals using wavelet denoising technology and then globally reconstructing the signal after processing the high-frequency decomposition coefficients by setting a preset high-frequency threshold includes the following steps: The acquired transient traveling wave signal is decomposed into multiple scales with a preset number of layers using a wavelet basis adapted to the characteristics of the transient traveling wave signal, to obtain low-frequency decomposition coefficients and high-frequency decomposition coefficients. The high-frequency decomposition coefficients are subjected to soft thresholding by a preset high-frequency threshold. The processing rule is as follows: when the absolute value of the high-frequency decomposition coefficient is greater than the preset high-frequency threshold, the difference between the high-frequency decomposition coefficient and the preset high-frequency threshold is output as the new high-frequency decomposition coefficient; when the absolute value of the high-frequency decomposition coefficient is less than or equal to the preset high-frequency threshold, 0 is output as the new high-frequency decomposition coefficient. The new high-frequency decomposition coefficients are globally reconstructed with the low-frequency decomposition coefficients to obtain the denoised transient traveling wave signal.

5. The fault analysis method for power collection lines in new energy power plants according to claim 4, characterized in that, The method of globally reconstructing the new high-frequency decomposition coefficients and the low-frequency decomposition coefficients includes the following steps: The new high-frequency decomposition coefficients are classified and matched with the low-frequency decomposition coefficients according to the wavelet basis type at the time of decomposition. By using the inverse transform algorithm corresponding to multi-scale decomposition, the newly classified high-frequency decomposition coefficients and low-frequency decomposition coefficients are combined into a signal, and the frequency band information corresponding to each coefficient is kept the same as the frequency band distribution of the original transient traveling wave signal during the synthesis process. The complete global reconstructed signal is output through the inverse transform algorithm as the denoised transient traveling wave signal.

6. The fault analysis method for power collection lines in new energy power plants according to claim 4, characterized in that, The method for extracting low-frequency information with fault characteristics based on a preset signal feature map of common defects in power collection lines, and determining the type of early defects in the line and the corresponding branch where the defects are located, includes the following steps: Based on the frequency bands of fault features corresponding to different defect types in the reference signal feature spectrum, the range of low-frequency information extraction frequency bands is set. The denoised transient traveling wave signal is subjected to frequency band separation, and a signal segment matching the extracted frequency band range is extracted. This segment is defined as low-frequency information with fault characteristics. Based on the feature types corresponding to different defects in the signal feature spectrum, the corresponding feature parameters are extracted from the separated low-frequency information. The extracted feature parameters are compared with typical defect features in the signal feature map band by band. When the comparison accuracy exceeds the preset ratio, the corresponding defect type is determined, and the branch where the defect is located is located by the sensor deployment location.

7. The fault analysis method for power collection lines in new energy power plants according to claim 1, characterized in that, The aforementioned construction of a multi-port network model adapted to multi-branch collection lines of new energy power plants based on transmission equations includes the following steps: The number of model ports and the corresponding relationships between each port are determined based on the transmission equation in transmission line theory and the actual branch topology of the collection lines of new energy power plants. Each branch of the collector line is equivalent to a transmission line unit containing distributed resistance, distributed inductance, and distributed capacitance. Based on the actual connection method of each branch, the corresponding relationship between each transmission line unit and the model port is established. By describing the transmission, reflection, and transmission relationships of transient traveling waves between ports using transmission equations, a multi-port network model with the same model structure as the actual physical structure of the collector line is obtained.

8. The fault analysis method for power collection lines in new energy power plants according to claim 1, characterized in that, The method described above for processing the traveling wave reflection and transmission characteristics of multi-branch lines using parameter estimation to output the traveling wave propagation characteristic parameters of each branch line includes the following steps: The line parameters in the multiport network model are initially estimated using parameter estimation methods. The line parameters include resistance parameters, inductance parameters, and capacitance parameters. Based on the degree of influence of different terrains on line parameters and the variation law of line parameters under different climatic conditions, corresponding parameter correction coefficients are set to compensate for the preliminary estimated line parameters. Based on the compensated line parameters and the structural characteristics of the mixed section of overhead line and cable, the propagation speed of the traveling wave in the mixed section is corrected. The corrected traveling wave propagation velocity, traveling wave reflection coefficient, and traveling wave transmission coefficient are output as traveling wave propagation characteristic parameters.

9. The fault analysis method for power collection lines in new energy power plants according to claim 1, characterized in that, The aforementioned setting different hypothetical fault points on branches and other branches of the line where early-stage defect types are located includes: preferentially setting hypothetical fault points at preset intervals on branches where early-stage defect types are located, and setting hypothetical fault points at another preset interval on other branches. The method of measuring the degree of deviation using the vector deviation objective function includes: simulating the traveling wave waveform generated at each hypothetical fault point and the path of propagation to each sensor based on the traveling wave propagation characteristic parameters output from the branch network modeling step; extracting the wavefront arrival time, peak amplitude, and fluctuation period characteristics of the actual transient traveling wave signal waveform after global reconstruction; and calculating the vector angle between the simulated waveform and the actual waveform at each hypothetical fault point using the vector deviation objective function, wherein a smaller vector angle value indicates a smaller degree of deviation.

10. A fault analysis system for power collection lines in new energy power plants, employing the fault analysis method for power collection lines in new energy power plants as described in any one of claims 1-9, characterized in that, The system includes: The signal acquisition module is configured to: synchronously acquire transient traveling wave signals of the power collection lines of the new energy power station in real time based on the distributed traveling wave sensors; perform high-speed temporary storage of the acquired transient traveling wave signals and ensure real-time transmission of the acquired data, thereby eliminating signal loss during the acquisition of transient traveling wave signals; The early identification module is configured to: process the transient traveling wave signal using wavelet denoising technology, and perform global reconstruction after processing the high-frequency decomposition coefficients of the signal using a preset high-frequency threshold; extract low-frequency information with fault characteristics based on the preset signal feature map of common defects in the power line, and determine the type of early defects in the line and the branch where the corresponding defects are located. The network modeling module is configured to: construct a multi-port network model adapted to the multi-branch collector lines of new energy power plants based on the transmission equation, wherein the multi-port network model corresponds to the actual branch topology of the collector lines; and process the traveling wave reflection and transmission characteristics of the multi-branch lines through parameter estimation methods, and output the traveling wave propagation characteristic parameters of each branch line. The interval determination module is configured to: preset different hypothetical fault points on the branch where the early defect type of the line is located and other branch lines; simulate the traveling wave waveform of the hypothetical fault point based on the multi-port network model; and compare it with the actual transient traveling wave signal waveform after global reconstruction; measure the degree of deviation between the actual waveform corresponding to the real fault and the traveling wave waveform corresponding to the hypothetical fault point through the vector deviation objective function; and take the branch line where the hypothetical fault point with the smallest deviation is located and its corresponding physical location as the fault interval determination result; The results display module is configured to generate a visual fault map with the location of the fault point marked, based on the fault range determination results and the physical location information of the collector line topology, cable joints, and cable cross-connection boxes.

Images