A method and system for identifying neutral line faults in a distribution network

Abstract

This invention discloses a method and system for identifying neutral line faults in a distribution network. The method includes: acquiring low-frequency channel data and high-frequency channel data; constructing a zero-sequence loop time-domain differential equation model based on the low-frequency channel data using a recursive least squares method with a forgetting factor, and identifying the pure resistance and inductance parameters of the neutral line in real time; extracting intrinsic mode components whose center frequencies fall within a preset frequency range using a variational mode decomposition algorithm based on the high-frequency channel data, and calculating their multi-scale permutation entropy as a micro-arc characteristic index; performing multi-mode comprehensive judgment based on the identified parameters and characteristic index; when the alarm criteria are met, constructing a first evidence body based on the pure resistance parameters and micro-arc characteristic index, and constructing a second evidence body based on the standard deviation calculated from the voltage data of the user-side smart meter; fusing the evidence using Dempster's synthesis rules, and outputting the final judgment result. This invention achieves a leap from post-disconnection alarm to pre-loosening warning, effectively reducing the false alarm rate.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network fault monitoring technology, and particularly relates to a method and system for identifying neutral line faults in power distribution networks. Background Technology

[0002] As the "last mile" connecting the power grid and users, the reliability of low-voltage distribution networks directly affects the quality of power supply for people's livelihoods. With the accelerated advancement of digital construction of distribution networks, a large number of distribution substation smart integration terminals, smart meters, and low-voltage IoT sensors have been widely deployed, enabling distribution networks to gradually lay the foundation for transformation from "blind adjustment and blind management" to "data-driven perception".

[0003] Currently, neutral line fault monitoring technology in low-voltage distribution networks is at a critical stage of evolution from steady-state scalar monitoring to transient characteristic analysis. Traditional monitoring methods based on simple Ohm's law and power frequency RMS values ​​often struggle to effectively identify the effects of distributed inductance on long lines and the weak arc characteristics caused by early loosening. The existing technologies have the following main drawbacks: First, communication bottlenecks lead to poor timeliness. The existing master station analysis mode relies heavily on low-frequency uploaded data, which easily misses key features at the moment of fault occurrence for transient and gradual faults such as neutral wire loosening and poor contact. Second, the lack of data dimensions leads to a high false alarm rate. The uploaded data is mostly the effective value of voltage / current, which loses phase information and waveform distortion features, making it difficult to distinguish between zero-point drift caused by normal three-phase imbalance and drift caused by fault. Third, the simplification of the physical model makes it impossible to eliminate misjudgments. Existing technologies mostly calculate impedance magnitude based on a simple Ohm's law model, which cannot achieve decoupling of resistance and inductance, making it impossible to distinguish between "natural high impedance" and "fault high resistance". Fourth, the steady-state monitoring mechanism masks early micro-arc characteristics. Existing algorithms usually filter out high-frequency noise, resulting in the system's lack of ability to detect the initial loosening.

[0004] Therefore, how to utilize edge computing power to evolve from simple impedance calculation to decoupled identification of resistance / inductance parameters, from single power frequency monitoring to broad-spectrum monitoring including high-frequency micro-arc characteristics, and from simple logic comparison to multi-source decision fusion based on uncertainty reasoning has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] The present invention aims to solve at least one problem existing in the prior art and provides a method and system for identifying neutral line faults in distribution networks. By decoupling parameter identification and weak feature extraction, combined with multi-source evidence theory fusion, the method achieves collaborative judgment of neutral line breakage and high-resistance contact faults.

[0006] In a first aspect, the present invention provides a method for identifying neutral line faults in a distribution network, comprising: Based on a preset dual-channel processing strategy, low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction are obtained. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. Based on the low-frequency channel data, a zero-sequence loop time-domain differential equation model is constructed using the recursive least squares method with a forgetting factor, and the pure resistance parameters and inductance parameters of the neutral line are decoupled and identified in real time. Based on the high-frequency channel data, the variational mode decomposition algorithm is used to extract the intrinsic mode components whose center frequencies fall within a preset frequency range, and the multi-scale permutation entropy of the intrinsic mode components is calculated as the micro-arc characteristic index. Based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and real-time three-phase current values, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions, a multi-mode comprehensive analysis is performed. If the alarm criteria are met, a first evidence body is constructed based on the pure resistance parameters and the micro-arc characteristic index, and a second evidence body is constructed based on the real-time voltage data of the user-side smart meter to calculate the standard deviation of the three-phase user voltage. The first and second evidence bodies are fused using Dempster's synthesis rules. The overall reliability of the fused evidence is calculated, and the final fault assessment result is output based on the principle of maximum reliability.

[0007] Secondly, the present invention provides a neutral line fault identification system for a distribution network, comprising: The acquisition module is configured to acquire low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction based on a preset dual-channel processing strategy. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. The decoupling module is configured to construct a zero-sequence loop time-domain differential equation model based on the low-frequency channel data using a recursive least squares method with a forgetting factor, and to decouple and identify the pure resistance and inductance parameters of the neutral line in real time. The calculation module is configured to extract intrinsic mode components whose center frequencies fall within a preset frequency range based on the high-frequency channel data using a variational mode decomposition algorithm, and calculate the multi-scale permutation entropy of the intrinsic mode components as a micro-arc feature index. The analysis module is configured to perform multi-mode comprehensive analysis based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, the real-time three-phase voltage value, and the real-time three-phase current value, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions. The construction module is configured to construct a first evidence body based on the pure resistance parameters and the micro-arc characteristic index when the alarm criteria are met, and to calculate the standard deviation of the three-phase user voltage based on the real-time voltage data of the user-side smart meter and construct a second evidence body. The output module is configured to fuse the first evidence body and the second evidence body using the Dempster synthesis rule, calculate the overall reliability of the fused evidence body, and output the final fault assessment result according to the principle of maximum reliability.

[0008] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the distribution network neutral line fault identification method according to any embodiment of the present invention.

[0009] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of the distribution network neutral line fault identification method according to any embodiment of the present invention.

[0010] The distribution network neutral line fault identification method and system of this application adopts the recursive least squares method with forgetting factor to construct the zero-sequence loop time domain differential equation model, and decouples and identifies the pure resistance parameters and inductance parameters of the neutral line in real time. It solves the industry problem of "natural high impedance" and "fault high resistance" from the physical mechanism, accurately removes the influence of line distributed inductance, and greatly reduces the false alarm rate. The variational mode decomposition algorithm is used to extract the intrinsic mode components whose center frequency falls within the range of 1kHz to 3kHz, and their multi-scale arrangement entropy is calculated as the micro-arc characteristic index. It can keenly detect anomalies and issue warnings in the early stage when only slight loosening or poor contact occurs in the neutral line, realizing the leap from "post-disconnection alarm" to "pre-loosening warning". By introducing the Dempster evidence theory, the pure resistance parameters identified at the edge and the micro-arc characteristic index are constructed as the first evidence body, and the standard deviation of user meter voltages detected in the cloud is constructed as the second evidence body. The fusion decision is carried out through the Dempster synthesis rule, which effectively handles the conflict of multi-source data and achieves highly reliable fault judgment under weak observability conditions. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart of a method for identifying neutral line faults in a distribution network according to an embodiment of the present invention; Figure 2 This is a structural block diagram of a neutral line fault identification system for a power distribution network according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Please see Figure 1 The diagram shows a flowchart of a method for identifying neutral line faults in a distribution network according to this application.

[0015] like Figure 1 As shown, the method for identifying neutral line faults in a distribution network specifically includes the following steps: Step S101: Based on a preset dual-channel processing strategy, acquire low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction, wherein the low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data.

[0016] In this step, the three-phase voltage and three-phase current on the low-voltage side of the distribution transformer are continuously sampled to obtain the original discrete time-domain signal with a preset sampling rate. The three-phase voltage and three-phase current in the original discrete time domain signal are low-pass filtered and downsampled to obtain low-frequency voltage data and low-frequency current data. High-frequency noise above 3kHz is filtered out from the three-phase voltage in the original discrete time domain signal, and data in the frequency band of 1kHz to 3kHz is retained to obtain high-frequency voltage data.

[0017] Step S102: Based on the low-frequency channel data, a zero-sequence loop time-domain differential equation model is constructed using the recursive least squares method with a forgetting factor, and the pure resistance parameters and inductance parameters of the neutral line are decoupled and identified in real time.

[0018] In this step, the instantaneous value of zero-sequence voltage is calculated based on the low-frequency voltage data, and the instantaneous value of zero-sequence current is calculated based on the low-frequency current data; The neutral line is equivalent to a resistor and an inductor connected in series. Based on the instantaneous values ​​of the zero-sequence voltage and zero-sequence current, a time-domain differential equation model of the zero-sequence loop is constructed, and its expression is: , In the formula, For zero-sequence voltage at The instantaneous sampled value at time t. In order to be in The pure resistance parameters of the neutral line at any given time include the resistance of the conductor and the resistance of fault contacts. In order to be in The algebraic sum of the instantaneous values ​​of the three-phase currents at time t. The inductance parameters of the neutral line, In order to be in White noise measured at specific times, Indicates time; The differential equation is discretized using the backward difference method, resulting in a discrete equation, expressed as: , In the formula, The sampling period is In order to be in The algebraic sum of the instantaneous values ​​of the three-phase currents at time 1; The discrete equation is rewritten as a linear regression equation, expressed as: , In the formula, For observing scalars, , For data vectors, , The parameter vector to be identified, , These are the pure resistance parameters of the neutral line. It is the transpose symbol; Introducing a forgetting factor, the recursive least squares method is used to update the parameter vector to be identified online, obtaining the pure resistance and inductance parameters of the neutral line. The expression for the online update of the parameter vector to be identified is: , , , In the formula, In order to be in The parameter estimate vector at time step, containing The pure resistance and inductance parameters of the neutral line at any given time. In order to be in The vector of parameter estimates at time t, In order to be in Gain matrix at time step Forgetting factor, In order to be in The covariance matrix at time t, In order to be in The covariance matrix at time t, It is an identity matrix.

[0019] Step S103: Based on the high-frequency channel data, the variational mode decomposition algorithm is used to extract the intrinsic mode components whose center frequencies fall within the preset frequency range, and the multi-scale permutation entropy of the intrinsic mode components is calculated as the micro-arc characteristic index.

[0020] In this step, the zero-sequence voltage high-frequency signal is calculated based on the high-frequency voltage data. ; Using variational mode decomposition algorithm to Decomposed into The constrained variational problem for the intrinsic modal components is constructed as follows: , , In the formula, This represents a minimization operation, and the goal of the solution is to find the optimal set of intrinsic mode components. and its corresponding set of center frequencies , For the Dirac function, For time Find the partial derivative. The imaginary unit, Pi The first decomposition obtained at time t The time-domain signal of the intrinsic mode components, For the first The center frequency corresponding to each modal component This is the convolution operator; The constrained variational problem is solved iteratively by alternating direction multiplier method to obtain each intrinsic modal component and its corresponding center frequency; The intrinsic mode components whose center frequencies fall within the range of 1kHz to 3kHz are selected as arc characteristic components to obtain the arc characteristic component set. Treating each arc feature component in the arc feature component set as a discrete-time series and reconstructing its phase space, the multi-scale permutation entropy of each arc feature component is calculated, expressed as: , , , In the formula, The index of the intrinsic modal component. For the embedding dimension, The sequence number of the arrangement pattern, To reconstruct the element index step size within the vector, No. Under the modal component, the first Transient energy weighting factors for various arrangement patterns To reconstruct the discrete-time series sampling points in phase space, To delay time, For the first The average of a number of local sequence vectors for The absolute value of the instantaneous zero-sequence current at time t. This is the current amplification factor, used to adjust the activation sensitivity of the feature weights to changes in zero-sequence current. For the first The original statistical probability of the occurrence of the permutation pattern For weighted relative probabilities, The transient energy sensing multi-scale arrangement entropy, i.e., the micro-arc characteristic index, This is the regularization penalty coefficient, used to amplify the probability distribution non-uniformity during sudden changes in arc energy. It is a non-linear exponent; The maximum value of the entropy of all arc characteristic components is taken as the micro-arc characteristic index. .

[0021] Step S104: Based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and current values, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions, a multi-mode comprehensive analysis is performed.

[0022] In this step, when ,or And if it continues for more than 60 seconds, the alarm criteria are met, among which, These are the pure resistance parameters of the neutral line. For micro-arc characteristic index; when And the duration exceeds 100ms, or any single-phase voltage is greater than 265V or less than 160V and the phase correlation between the zero-sequence voltage and the phase with the lowest voltage is... The alarm criteria are met when the angle is within 180°±15°.

[0023] Step S105: If the alarm criteria are met, construct a first evidence body based on the pure resistance parameters and the micro-arc characteristic index, and calculate the three-phase user voltage standard deviation based on the real-time voltage data of the user-side smart meter and construct a second evidence body.

[0024] In this step, an identification framework for fault assessment is constructed. ,in This indicates a true fault in the neutral line. This indicates normal zero-point drift caused by severe three-phase load imbalance. This indicates an abnormality in the terminal measurement circuit. Indicates an uncertain state; The confidence levels of the pure resistance parameters and micro-arc feature indices for each proposition in the recognition framework are calculated using a preset fuzzy membership function. Then, the two confidence levels are weighted and fused to obtain the first evidence body. Basic probability assignment Among them, the first piece of evidence The basic probability assignment satisfies , For identification framework Any non-empty subset of; Calculate the standard deviation of real-time voltage data from user-side smart meters. According to standard deviation The size is mapped to the confidence level of each proposition using a preset membership function, thus obtaining the second evidence body. Basic probability assignment Among them, the second piece of evidence The basic probability assignment satisfies .

[0025] Step S106: The first evidence body and the second evidence body are fused using the Dempster synthesis rule, the overall reliability of the fused evidence body is calculated, and the final fault assessment result is output according to the maximum reliability principle.

[0026] In this step, Dempster's composition rules are used to analyze the first body of evidence. Second evidence body Fusion is performed for the recognition framework. any non-empty subset of The overall reliability after fusion The expression is: , , , In the formula, The dynamic conflict coefficient, For the first body of evidence against the proposition Basic probability assignment, For the second body of evidence against the proposition Basic probability assignment, These are the timestamps for obtaining the features of the first and second pieces of evidence, respectively. For the dynamic time constant of the distribution radio station area topology, To identify the frame, , , All are recognition frameworks The proposition in, and satisfying , For the overall reliability after fusion, These are the reliability factors for the first source of evidence and the reliability factors for the second source of evidence, respectively. Adaptive allocation of bootstrap functions for conflict resolution. The pure resistance parameters of the neutral line, This is the impedance sensitivity coefficient.

[0027] Decision-making is based on the overall reliability after fusion: like and and If the fault is found, it is determined to be a real fault in the neutral line, and an emergency repair work order is automatically generated. in, This represents the overall confidence level (probability of confidence) for a "real neutral line fault" (i.e., proposition F1). Physical meaning: After fusing multi-source data from the edge side (resistance / arc) and the cloud (smart meter voltage) using the DS rule, the system ultimately determines the probability that a real neutral line break or high-resistance fault has occurred in the current distribution network. If this value exceeds a set threshold (e.g., >0.75), the system will automatically generate an emergency repair work order. This represents the overall confidence level for "normal zero-sequence drift caused by severe three-phase load imbalance" (i.e., proposition F2). Physical meaning: Uneven distribution of single-phase loads in a distribution network can also generate zero-sequence voltage and current. m(F2) represents the probability that the system determines the current anomaly is caused by normal physical imbalance, rather than equipment failure. This represents the overall reliability of "terminal measurement loop anomaly" (i.e., proposition F3). Physical meaning: It indicates the probability that the system judges the anomaly to be caused by a problem with the measurement wiring or hardware of the metering terminal or sensor itself, resulting in abnormal data. If this value is too high, no line repair work order will be issued; instead, a metering terminal troubleshooting work order will be generated. like If so, it is determined that the terminal measurement circuit is abnormal, and a metering terminal troubleshooting work order is generated; Otherwise, remain in a state of uncertainty and wait for the next round of updated evidence.

[0028] In summary, the method of this application, through a dual-channel processing strategy, acquires low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction, respectively, realizing parallel processing of steady-state parameter identification and transient feature extraction. It takes into account the needs of accurate identification of resistance / inductance parameters and early weak arc feature capture, effectively solving the contradiction in existing technologies where a single channel cannot simultaneously meet the requirements of steady-state accuracy and transient sensitivity. Furthermore, it employs a recursive least squares method with a forgetting factor to construct a zero-sequence loop time-domain differential equation model, and decouples and identifies the pure resistance and inductance parameters of the neutral line in real time, thus solving the industry problem of the superposition of "natural high impedance" and "fault high resistance" from a physical mechanism perspective. By accurately isolating the influence of distributed inductance in the line, it is possible to effectively distinguish between normal inductive reactance and poor contact resistance in long lines, significantly reducing the false alarm rate caused by line length or three-phase imbalance, and significantly improving the accuracy of fault identification. A variational mode decomposition algorithm is used to extract intrinsic mode components with center frequencies falling within the 1kHz to 3kHz range, and their multi-scale permutation entropy is calculated as a micro-arc characteristic index. This allows for the sensitive detection of weak non-stationary arc characteristics and the issuance of early warnings even in the early stages of slight loosening or poor contact in the neutral line of the distribution network. This overcomes the limitation of traditional FFT methods that only monitor power frequency signals, achieving a leap from "post-disconnection alarm" to "pre-loosening warning," effectively avoiding equipment damage and property loss to users due to fault escalation. The Dempster evidence theory is introduced, constructing the first evidence body from the pure resistance parameters identified on the edge side and the micro-arc characteristic index, and constructing the second evidence body from the standard deviation of user meter voltages detected in the cloud. A fusion decision is then made using Dempster synthesis rules. This method can effectively handle conflicts between multiple data sources. Even in complex nonlinear load environments or when single sensor data is missing or subject to transient shocks, it can still arrive at the most reliable diagnostic conclusion by evaluating the confidence of different evidence bodies, which greatly improves the robustness and fault tolerance of the judgment model.

[0029] Please see Figure 2 The diagram shows a structural block diagram of a neutral line fault identification system for a power distribution network according to this application.

[0030] like Figure 2As shown, the neutral line fault identification system 200 of the distribution network includes an acquisition module 210, a decoupling module 220, a calculation module 230, an analysis module 240, a construction module 250, and an output module 260.

[0031] The acquisition module 210 is configured to acquire low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction based on a preset dual-channel processing strategy. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. The decoupling module 220 is configured to construct a zero-sequence loop time-domain differential equation model based on the low-frequency channel data using a recursive least squares method with a forgetting factor, and to decouple and identify the pure resistance and inductance parameters of the neutral line in real time. The calculation module 230 is configured to extract intrinsic mode components whose center frequencies fall within a preset frequency range using a variational mode decomposition algorithm based on the high-frequency channel data, and to calculate the multi-scale permutation entropy of the intrinsic mode components as a micro-arc feature index. The analysis module 240 is configured to perform multi-mode comprehensive analysis based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and current values, combined with a preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions; the construction module 250 is configured to construct a first evidence body based on the pure resistance parameters and the micro-arc characteristic index when the alarm criteria are met, and to calculate the standard deviation of the three-phase user voltage based on the real-time voltage data of the user-side smart meter and construct a second evidence body; the output module 260 is configured to fuse the first evidence body and the second evidence body using the Dempster synthesis rule, calculate the comprehensive reliability after fusion, and output the final fault analysis result according to the maximum reliability principle.

[0032] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0033] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the distribution network neutral line fault identification method in any of the above method embodiments. In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows: Based on a preset dual-channel processing strategy, low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction are obtained. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. Based on the low-frequency channel data, a zero-sequence loop time-domain differential equation model is constructed using the recursive least squares method with a forgetting factor, and the pure resistance parameters and inductance parameters of the neutral line are decoupled and identified in real time. Based on the high-frequency channel data, the variational mode decomposition algorithm is used to extract the intrinsic mode components whose center frequencies fall within a preset frequency range, and the multi-scale permutation entropy of the intrinsic mode components is calculated as the micro-arc characteristic index. Based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and real-time three-phase current values, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions, a multi-mode comprehensive analysis is performed. If the alarm criteria are met, a first evidence body is constructed based on the pure resistance parameters and the micro-arc characteristic index, and a second evidence body is constructed based on the real-time voltage data of the user-side smart meter to calculate the standard deviation of the three-phase user voltage. The first and second evidence bodies are fused using Dempster's synthesis rules. The overall reliability of the fused evidence is calculated, and the final fault assessment result is output based on the principle of maximum reliability.

[0034] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and application programs required for at least one function; the stored data area may store data created based on the use of the distribution network neutral fault identification system, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, and this remote memory may be connected to the distribution network neutral fault identification system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0035] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the distribution network neutral line fault identification method described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the distribution network neutral line fault identification system. The output device 340 may include a display screen or other display device.

[0036] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0037] In one implementation, the above-described electronic device is applied to a neutral line fault identification system in a distribution network for a client application, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Based on a preset dual-channel processing strategy, low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction are obtained. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. Based on the low-frequency channel data, a zero-sequence loop time-domain differential equation model is constructed using the recursive least squares method with a forgetting factor, and the pure resistance parameters and inductance parameters of the neutral line are decoupled and identified in real time. Based on the high-frequency channel data, the variational mode decomposition algorithm is used to extract the intrinsic mode components whose center frequencies fall within a preset frequency range, and the multi-scale permutation entropy of the intrinsic mode components is calculated as the micro-arc characteristic index. Based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and real-time three-phase current values, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions, a multi-mode comprehensive analysis is performed. If the alarm criteria are met, a first evidence body is constructed based on the pure resistance parameters and the micro-arc characteristic index, and a second evidence body is constructed based on the real-time voltage data of the user-side smart meter to calculate the standard deviation of the three-phase user voltage. The first and second evidence bodies are fused using Dempster's synthesis rules. The overall reliability of the fused evidence is calculated, and the final fault assessment result is output based on the principle of maximum reliability.

[0038] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying neutral line faults in a distribution network, characterized in that, include: Based on a preset dual-channel processing strategy, low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction are obtained. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. Based on the low-frequency channel data, a zero-sequence loop time-domain differential equation model is constructed using the recursive least squares method with a forgetting factor, and the pure resistance parameters and inductance parameters of the neutral line are decoupled and identified in real time. Based on the high-frequency channel data, the variational mode decomposition algorithm is used to extract the intrinsic mode components whose center frequencies fall within a preset frequency range, and the multi-scale permutation entropy of the intrinsic mode components is calculated as the micro-arc characteristic index. Based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and real-time three-phase current values, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions, a multi-mode comprehensive analysis is performed. If the alarm criteria are met, a first evidence body is constructed based on the pure resistance parameters and the micro-arc characteristic index, and a second evidence body is constructed based on the real-time voltage data of the user-side smart meter, calculating the standard deviation of the three-phase user voltage. Specifically, this includes: Build an identification framework for fault assessment ,in This indicates a true fault in the neutral line. This indicates normal zero-point drift caused by severe three-phase load imbalance. This indicates an abnormality in the terminal measurement circuit. Indicates an uncertain state; The confidence levels of the pure resistance parameters and micro-arc feature indices for each proposition in the recognition framework are calculated using a preset fuzzy membership function. Then, the two confidence levels are weighted and fused to obtain the first evidence body. Basic probability assignment Among them, the first piece of evidence The basic probability assignment satisfies , For identification framework Any non-empty subset of; Calculate the standard deviation of real-time voltage data from user-side smart meters. According to standard deviation The size is mapped to the confidence level of each proposition using a preset membership function, thus obtaining the second evidence body. Basic probability assignment Among them, the second piece of evidence The basic probability assignment satisfies ; The first and second pieces of evidence are fused using Dempster's rules of composition. The overall reliability of the fused evidence is calculated, and the final fault assessment result is output according to the principle of maximum reliability. Specifically, the calculation of the overall reliability includes: Dempster's composition rules were applied to the first body of evidence. Second evidence Fusion is performed for the recognition framework. any non-empty subset of The overall reliability after fusion The expression is: ， ， ， In the formula, The dynamic conflict coefficient, For the first body of evidence against the proposition Basic probability assignment, For the second body of evidence against the proposition Basic probability assignment, These are the timestamps for obtaining the features of the first and second pieces of evidence, respectively. The dynamic time constant of the distribution radio station area topology. To identify the frame, , , All are recognition frameworks The proposition in, and satisfying , For the overall reliability after fusion, These are the reliability factors for the first source of evidence and the reliability factors for the second source of evidence, respectively. Adaptive allocation of bootstrap functions for conflict resolution. The pure resistance parameters of the neutral line, This is the impedance sensitivity coefficient.

2. The method for identifying neutral line faults in a distribution network according to claim 1, characterized in that, The acquisition of low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction based on the preset dual-channel processing strategy includes: The three-phase voltage and three-phase current on the low-voltage side of the distribution transformer are continuously sampled to obtain the original discrete-time domain signal with a preset sampling rate; The three-phase voltage and three-phase current in the original discrete time domain signal are low-pass filtered and downsampled to obtain low-frequency voltage data and low-frequency current data. High-frequency noise above 3kHz is filtered out from the three-phase voltage in the original discrete-time domain signal, and data in the frequency band of 1kHz to 3kHz is retained to obtain high-frequency voltage data.

3. The method for identifying neutral line faults in a distribution network according to claim 1, characterized in that, Based on the low-frequency channel data, a recursive least squares method with a forgetting factor is used to construct a zero-sequence loop time-domain differential equation model, and the pure resistance and inductance parameters of the neutral line are identified in real time through decoupling. The instantaneous value of zero-sequence voltage is calculated based on the low-frequency voltage data, and the instantaneous value of zero-sequence current is calculated based on the low-frequency current data; The neutral line is equivalent to a resistor and an inductor connected in series. Based on the instantaneous values ​​of the zero-sequence voltage and zero-sequence current, a time-domain differential equation model of the zero-sequence loop is constructed, and its expression is: ， In the formula, For zero-sequence voltage at The instantaneous sampled value at time t. In order to be in The pure resistance parameters of the neutral line at any given time include the resistance of the conductor and the resistance of fault contacts. In order to be in The algebraic sum of the instantaneous values ​​of the three-phase currents at time t. The inductance parameters of the neutral line, In order to be in White noise measured at specific times, Indicates time; The differential equation is discretized using the backward difference method, resulting in a discrete equation, expressed as: ， In the formula, The sampling period is In order to be in The algebraic sum of the instantaneous values ​​of the three-phase currents at time 1; The discrete equation is rewritten as a linear regression equation, expressed as: ， In the formula, For observing scalars, , For data vectors, , The parameter vector to be identified, , These are the pure resistance parameters of the neutral line. It is the transpose symbol; Introducing a forgetting factor, the recursive least squares method is used to update the parameter vector to be identified online, obtaining the pure resistance and inductance parameters of the neutral line. The expression for the online update of the parameter vector to be identified is: ， ， ， In the formula, In order to be in The parameter estimate vector at time step, containing The pure resistance and inductance parameters of the neutral line at any given time. In order to be in The vector of parameter estimates at time t, In order to be in Gain matrix at time step Forgetting factor, In order to be in The covariance matrix at time t, In order to be in The covariance matrix at time t, It is an identity matrix.

4. The method for identifying neutral line faults in a distribution network according to claim 1, characterized in that, The step of extracting intrinsic mode components whose center frequencies fall within a preset frequency range using variational mode decomposition algorithm based on the high-frequency channel data, and calculating the multi-scale permutation entropy of the intrinsic mode components as a micro-arc characteristic index includes: Calculate the zero-sequence voltage high-frequency signal based on the high-frequency voltage data. ; Using variational mode decomposition algorithm to Decomposed into The constrained variational problem for the intrinsic modal components is constructed as follows: ， ， In the formula, This represents a minimization operation, and the goal of the solution is to find the optimal set of intrinsic mode components. and its corresponding set of center frequencies , For the Dirac function, For time Find the partial derivative. The imaginary unit, Pi The first decomposition obtained at time t The time-domain signal of the intrinsic mode components, For the first The center frequency corresponding to each modal component This is the convolution operator; The constrained variational problem is solved iteratively by alternating direction multiplier method to obtain each intrinsic modal component and its corresponding center frequency; The intrinsic mode components whose center frequencies fall within the range of 1kHz to 3kHz are selected as arc characteristic components to obtain the arc characteristic component set. Treating each arc feature component in the arc feature component set as a discrete-time series and reconstructing its phase space, the multi-scale permutation entropy of each arc feature component is calculated, expressed as: ， ， ， In the formula, The index of the intrinsic modal component. For the embedding dimension, The sequence number of the arrangement pattern, To reconstruct the element index step size within the vector, No. Under the modal component, the first Transient energy weighting factors for various arrangement patterns To reconstruct the discrete-time series sampling points in phase space, To delay time, For the first The average of a number of local sequence vectors for The absolute value of the instantaneous zero-sequence current at time t. This is the current amplification factor, used to adjust the activation sensitivity of the feature weights to changes in zero-sequence current. For the first The original statistical probability of the occurrence of the permutation pattern For weighted relative probabilities, The transient energy sensing multi-scale arrangement entropy, i.e., the micro-arc characteristic index, This is the regularization penalty coefficient, used to amplify the probability distribution non-uniformity during sudden changes in arc energy. It is a non-linear exponent; The maximum value of the entropy of all arc characteristic components is taken as the micro-arc characteristic index. .

5. The method for identifying neutral line faults in a distribution network according to claim 1, characterized in that, The multi-mode comprehensive analysis based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, and the real-time three-phase voltage and current values, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions, includes: when ,or And if it continues for more than 60 seconds, the alarm criteria are met, among which, These are the pure resistance parameters of the neutral line. For micro-arc characteristic index; when And the duration exceeds 100ms, or any single-phase voltage is greater than 265V or less than 160V and the phase correlation between the zero-sequence voltage and the phase with the lowest voltage is... The alarm criteria are met when the angle is within 180°±15°.

6. A neutral line fault identification system for a distribution network, characterized in that, include: The acquisition module is configured to acquire low-frequency channel data for parameter identification and high-frequency channel data for micro-arc feature extraction based on a preset dual-channel processing strategy. The low-frequency channel data includes low-frequency voltage data and low-frequency current data, and the high-frequency channel data includes high-frequency voltage data. The decoupling module is configured to construct a zero-sequence loop time-domain differential equation model based on the low-frequency channel data using a recursive least squares method with a forgetting factor, and to decouple and identify the pure resistance and inductance parameters of the neutral line in real time. The calculation module is configured to extract intrinsic mode components whose center frequencies fall within a preset frequency range based on the high-frequency channel data using a variational mode decomposition algorithm, and calculate the multi-scale permutation entropy of the intrinsic mode components as a micro-arc feature index. The analysis module is configured to perform multi-mode comprehensive analysis based on the pure resistance parameters, the inductance parameters, the micro-arc characteristic index, the real-time three-phase voltage value, and the real-time three-phase current value, combined with the preset high-load anti-interference blocking mechanism, state transition criteria, and corresponding execution actions. The construction module is configured to, if the alarm criteria are met, construct a first evidence body based on the pure resistance parameters and the micro-arc characteristic index, and calculate the three-phase user voltage standard deviation based on the real-time voltage data of the user-side smart meter to construct a second evidence body, specifically including: Build an identification framework for fault assessment ,in This indicates a true fault in the neutral line. This indicates normal zero-point drift caused by severe three-phase load imbalance. This indicates an abnormality in the terminal measurement circuit. Indicates an uncertain state; The confidence levels of the pure resistance parameters and micro-arc feature indices for each proposition in the recognition framework are calculated using a preset fuzzy membership function. Then, the two confidence levels are weighted and fused to obtain the first evidence body. Basic probability assignment Among them, the first piece of evidence The basic probability assignment satisfies , For identification framework Any non-empty subset of; Calculate the standard deviation of real-time voltage data from user-side smart meters. According to standard deviation The size is mapped to the confidence level of each proposition using a preset membership function, thus obtaining the second evidence body. Basic probability assignment Among them, the second piece of evidence The basic probability assignment satisfies ; The output module is configured to fuse the first and second pieces of evidence using Dempster's synthesis rules, calculate the overall reliability of the fused evidence, and output the final fault assessment result based on the maximum reliability principle. Specifically, calculating the overall reliability of the fused evidence includes: Dempster's composition rules were applied to the first body of evidence. Second evidence Fusion is performed for the recognition framework. any non-empty subset of The overall reliability after fusion The expression is: ， ， ， In the formula, The dynamic conflict coefficient, For the first body of evidence against the proposition Basic probability assignment, For the second body of evidence against the proposition Basic probability assignment, These are the timestamps for obtaining the features of the first and second pieces of evidence, respectively. The dynamic time constant of the distribution radio station area topology. To identify the frame, , , All are recognition frameworks The proposition in, and satisfying , For the overall reliability after fusion, These are the reliability factors for the first source of evidence and the reliability factors for the second source of evidence, respectively. Adaptive allocation of bootstrap functions for conflict resolution. The pure resistance parameters of the neutral line, This is the impedance sensitivity coefficient.

7. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method described in any one of claims 1 to 5.

Images