A method for identifying the triggering cause of single-phase grounding signals in distribution networks based on waveform matching.

Abstract

This invention discloses a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching. The method includes: acquiring a target sample library; collecting waveforms corresponding to the original transient waveform data to be identified; processing the waveforms corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; calculating the similarity distance between the standard analysis waveform and waveforms corresponding to multiple historical transient waveform data to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that meets preset conditions from the multiple similarity distances based on the target sample library. This invention solves the technical problem of insufficient accuracy in determining the triggering cause of a single-phase grounding signal.

Description

Technical Field

[0001] This invention relates to the field of power distribution network detection technology, and more specifically, to a method for identifying the triggering cause of a single-phase grounding signal in a power distribution network based on waveform matching. Background Technology

[0002] In modern power systems, the operational safety and stability of distribution networks are of paramount importance. Single-phase grounding incidents are among the most common fault types in distribution networks, with diverse and complex triggering causes, including but not limited to tree branches touching wires, animals touching wires, and lightning strikes to surge arresters. Accurately identifying the triggering causes of these grounding incidents is of great significance for timely maintenance measures, preventing more serious faults, and ensuring the reliability of power supply.

[0003] Currently, the technical solutions for identifying the triggering causes of single-phase grounding signals in distribution networks are mainly divided into three categories:

[0004] 1. Manual Judgment Based on Expert Experience: This method relies on experienced power engineers who manually determine the triggering cause of grounding events by analyzing transient waveform data, combining on-site conditions, and past experience. However, manual judgment has significant limitations, such as high subjectivity, low efficiency, and difficulty in large-scale application, especially in scenarios requiring real-time analysis of large amounts of data.

[0005] 2. Signal processing analysis based on mathematical tools: This method utilizes mathematical tools such as Fourier transform and wavelet analysis to perform time-frequency domain analysis on recorded waveform data, extract waveform features, and then analyze the triggering causes. This type of method has a high recognition rate for ideal waveforms with obvious characteristics, but its accuracy and generalization ability are limited when dealing with complex nonlinear waveforms in real-world conditions. Waveforms in actual working conditions often contain noise, time axis offsets, and nonlinear distortions, all of which pose challenges that signal processing analysis methods struggle to address effectively.

[0006] 3. Deep Learning-Based Intelligent Algorithms: In recent years, deep learning has demonstrated strong potential in pattern recognition, image analysis, and other fields, and has also been attempted to be applied to the identification of triggering causes of single-phase grounding signals in distribution networks. However, the efficient training and application of deep learning models rely on massive amounts of high-quality labeled data, while the distribution network field faces the dilemma of "high-quality, small sample size." Although transient waveform recording terminals are widely deployed, resulting in a huge amount of raw signal data, truly high-quality samples with "accurate causes and reliable labels" are extremely limited because on-site verification is costly and many events are instantaneous and unreproducible. This data situation makes it difficult for neural network models that require large amounts of data to achieve the expected performance, reducing the effectiveness of intelligent identification methods in practical engineering applications.

[0007] Therefore, neither manual analysis based on expert experience, signal processing using mathematical tools, nor intelligent algorithms incorporating deep learning have effectively solved the problem of accurate identification under the dilemma of high-quality, small sample sizes. In particular, when faced with complex nonlinear waveforms in real-world scenarios, the accuracy and generalization ability of existing technologies decline, thus limiting their application in practical engineering scenarios and increasing the uncertainty and risk of misjudgment in fault analysis.

[0008] There is currently no effective solution to the above problems. Summary of the Invention

[0009] This invention provides a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, so as to at least solve the technical problem of insufficient accuracy in judging the triggering cause of a single-phase grounding signal.

[0010] According to one aspect of the present invention, a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching is provided, comprising: acquiring a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signal triggering in the distribution network and triggering causes corresponding to multiple historical transient waveform data; acquiring waveforms corresponding to the original transient waveform data to be identified; processing the waveforms corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; calculating the similarity distances between the standard analysis waveforms and the waveforms corresponding to the multiple historical transient waveform data to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that satisfies a preset condition from the multiple similarity distances based on the target sample library as the triggering cause corresponding to the original transient waveform data.

[0011] Optionally, acquiring the target sample library includes: acquiring multiple historical transient waveform data; determining the waveforms corresponding to the multiple historical transient waveform data based on the multiple historical transient waveform data; aligning the waveforms corresponding to the multiple historical transient waveform data based on a preset time reference to obtain the adjusted waveforms corresponding to each of the multiple historical transient waveform data; determining the sample correspondence relationship based on the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes corresponding to each of the multiple historical transient waveform data, wherein the sample correspondence relationship characterizes the correspondence between the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes; and determining the target sample library based on the sample correspondence relationship.

[0012] Optionally, the original transient waveform data to be identified is collected, including: obtaining the trigger time of the single-phase grounding signal; based on the trigger time of the single-phase grounding signal, according to the preset number of cycles before the fault and the number of cycles after the fault, extracting waveform data subsequences of fixed time lengths to determine the original transient waveform data.

[0013] Optionally, the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data is calculated, wherein the target historical transient waveform data is any one of multiple historical transient waveform data. This includes: initializing a distance matrix, wherein the elements in the distance matrix represent the distance between sampling points in the standard analysis waveform and sampling points in the waveform corresponding to the target historical transient waveform data; calculating the distance between the first sampling point in the standard analysis waveform and multiple sampling points in the waveform corresponding to the target historical transient waveform data, and storing them in the distance matrix to obtain a first matrix; calculating the distance between the first sampling point in the waveform corresponding to the target historical transient waveform and multiple sampling points in the standard analysis waveform, and storing them in the first matrix to obtain a second matrix; using a dynamic programming algorithm, calculating the remaining elements in the second matrix to obtain a third matrix; and determining the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data based on the third matrix.

[0014] Optionally, based on the target sample library, the triggering cause corresponding to the historical transient waveform data that meets the preset conditions is selected from multiple similarity distances as the triggering cause corresponding to the original transient waveform data. This includes: selecting waveforms corresponding to the historical transient waveform data that meet the preset conditions from multiple similarity distances based on the target sample library to form a high-similarity sample set; calculating the weight of the historical transient waveform data in the high-similarity sample set based on the similarity of the historical transient waveform data in the high-similarity sample set; and determining the triggering cause corresponding to the original transient waveform data based on the weight and triggering cause of the historical transient waveform data in the high-similarity sample set.

[0015] Optionally, based on the similarity of the historical transient waveform data in the high-similarity sample set, the weights of the historical transient waveform data in the high-similarity sample set are calculated according to a preset formula, wherein the preset formula is as follows:

[0016] ;

[0017] in, For the set of highly similar samples, the first j The weights of historical transient waveform data d j For the first j The similarity distance between historical transient waveform data S p A set of highly similar samples, It is a positive number that is lower than the preset threshold.

[0018] According to another aspect of the present invention, a device for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching is also provided, comprising: an acquisition module for acquiring a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signal triggering in the distribution network and triggering causes corresponding to multiple historical transient waveform data; an acquisition module for acquiring waveforms corresponding to original transient waveform data to be identified; a processing module for processing the waveforms corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; a calculation module for calculating the similarity distances between the standard analysis waveforms and the waveforms corresponding to the multiple historical transient waveform data to obtain multiple similarity distances; and a selection module for selecting, based on the target sample library, the triggering cause corresponding to the historical transient waveform data that satisfies a preset condition from the multiple similarity distances as the triggering cause corresponding to the original transient waveform data.

[0019] According to another aspect of the present invention, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored program, wherein, when the program is running, the device where the non-volatile storage medium is located is controlled to execute any of the above-described waveform matching-based single-phase grounding signal triggering cause identification methods in power distribution networks.

[0020] According to another aspect of the present invention, a computer device is also provided, the computer device including a processor, the processor being configured to run a program, wherein the program, when running, executes any of the above-described methods for identifying the triggering cause of a single-phase grounding signal in a power distribution network based on waveform matching.

[0021] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements any of the above-described methods for identifying the triggering cause of a single-phase grounding signal in a power distribution network based on waveform matching.

[0022] In this embodiment of the invention, a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching is adopted. This involves acquiring a target sample library, which includes waveforms and triggering causes corresponding to multiple historical transient waveforms generated by historical single-phase grounding signals in the distribution network; collecting the waveform corresponding to the original transient waveform data to be identified; processing the waveform corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; calculating the similarity distance between the standard analysis waveform and the waveforms corresponding to the multiple historical transient waveforms to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that meets preset conditions from the multiple similarity distances based on the target sample library. This achieves the goal of determining the triggering cause of a single-phase grounding signal by calculating the similarity between the waveform to be identified and the sample waveforms, thereby improving the accuracy of determining the triggering cause of a single-phase grounding signal and solving the technical problem of insufficient accuracy in judging the triggering cause of a single-phase grounding signal. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0024] Figure 1 A hardware block diagram of a computer terminal for implementing a waveform matching-based method for identifying the triggering cause of a single-phase grounding signal in a power distribution network is shown.

[0025] Figure 2 This is a flowchart illustrating the method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, according to an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, provided by an optional embodiment of the present invention.

[0027] Figure 4 This is a structural block diagram of a distribution network single-phase grounding signal triggering cause identification device based on waveform matching according to an embodiment of the present invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] According to an embodiment of the present invention, a method embodiment for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a waveform matching-based method for identifying the triggering cause of a single-phase grounding signal in a distribution network is shown. Figure 1 As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0032] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0033] The memory 104 can be used to store software programs and modules for application software, such as the program instructions / data storage device corresponding to the waveform matching-based distribution network single-phase grounding signal triggering cause identification method in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby implementing the waveform matching-based distribution network single-phase grounding signal triggering cause identification method described above. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 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.

[0034] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0035] Figure 2 This is a flowchart illustrating the method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, as provided in an embodiment of the present invention. Figure 2 As shown, the method includes the following steps:

[0036] Step S202: Obtain the target sample library, which includes the waveforms corresponding to multiple historical transient waveform data generated by the historical single-phase grounding signal of the distribution network and the triggering reasons corresponding to each of the multiple historical transient waveform data.

[0037] In this step, all relevant data records of historical single-phase grounding events can be collected from the transient waveform recording system or fault recorder of the distribution network. These records typically contain voltage and current waveform data at the time of the fault, as well as fault-related metadata (such as timestamps, location, equipment information, etc.). Simultaneously, on-site investigation reports, maintenance logs, and accident investigation reports after the fault are collected; these documents will contain the triggering cause of the fault and detailed on-site conditions. Ensure that each report matches the corresponding waveform data. Check that the collected waveform data is complete, without data loss or corruption, and that the signal clarity is sufficient for subsequent analysis. Any data or reports that do not meet quality standards must be discarded to ensure the reliability of the sample library.

[0038] A clear classification system for fault triggering causes can be defined, which will help you standardize fault causes and ensure consistency. For example, it can be categorized into "tree branch touching the wire," "animal contact," "lightning strike," and "insulation aging." Then, based on this classification system, the correct trigger can be labeled for each fault event, transforming the filtered transient waveform data into a unified format and specification to ensure data consistency and comparability. Relevant information for each historical fault event, including standardized transient waveform data, trigger cause labels, time information, geographical location, and other necessary metadata, can be integrated into a structured entry and stored in a dedicated database.

[0039] Through the above steps, a rich, diverse, and accurately labeled sample library of fault triggering causes can be established, which will provide a solid data foundation for subsequent intelligent identification systems based on similar waveform matching.

[0040] The sample library can display sample correspondences, which clearly define the link between specific waveform patterns and fault triggering causes, providing a crucial reference template for subsequent fault identification and diagnosis. It can establish the correspondence between historical transient waveform data and tagged triggering causes. A unique identifier (ID) is created for each historical transient waveform data record. Then, each ID is associated with its waveform data and triggering cause tag, forming a structured data record. These records should include, but are not limited to: ID, standardized waveform data, triggering cause type, date and time of fault occurrence, geographical coordinates, and information on affected power distribution segments.

[0041] Step S204: Collect the waveform corresponding to the original transient waveform data to be identified.

[0042] In this step, when signs of a single-phase ground fault are detected, the corresponding waveform recording terminal will automatically start recording voltage and current signals for a certain period before and after the fault. The collected transient waveform data can be uploaded to a data center or server via a communication network. After receiving the data, the original waveform is previewed to confirm that the waveform data does indeed contain fault characteristics, i.e., there are significant voltage changes or current spikes in the waveform, indicating successful data acquisition. At the same time, metadata related to the waveform data is recorded, including the exact time and location of the fault, information on affected lines and equipment, and possible environmental conditions (such as weather conditions).

[0043] It can check waveform data for missing or outliers, and perform interpolation to fill in gaps or remove outliers when necessary, ensuring the accuracy of subsequent analysis. It can also convert waveform data into a standardized format and file type for easy reading and processing by computer systems.

[0044] Step S206: Based on a preset time reference, process the waveform corresponding to the original transient waveform data to be identified to obtain a standard analysis waveform.

[0045] In this step, based on a preset time reference, the raw transient waveform data to be identified is processed. First, taking the precise trigger time of the single-phase grounding event as the center, two power frequency cycles are extracted forward and four power frequency cycles are extracted backward to form a waveform segment of fixed duration. Then, the sampling frequency of this waveform is uniformly adjusted to the system's preset standard sampling rate. Linear interpolation is used to eliminate differences in sampling density caused by different acquisition devices, ensuring that all waveforms have completely consistent point counts and intervals on the time axis. Next, the adjusted waveform undergoes amplitude normalization processing, mapping the minimum value of the waveform to negative one and the maximum value to positive one, making the amplitude range of all waveforms uniform and eliminating amplitude interference caused by differences in fault location, grounding resistance, or measurement device gain, retaining only the shape characteristics of the waveform. Finally, a waveform with a fixed duration, consistent number of sampling points, and standardized amplitude range is obtained, which serves as the input for subsequent similarity matching and is called the standard analysis waveform.

[0046] Step S208: Calculate the similarity distance between the standard analysis waveform and the waveforms corresponding to multiple historical transient waveform data to obtain multiple similarity distances.

[0047] In this step, the similarity distance between the standard analysis waveform and the waveforms corresponding to multiple historical transient waveform records is calculated. This primarily utilizes time series analysis methods, especially the Dynamic Time Warping (DTW) algorithm, to accommodate waveform temporal distortions and length differences. Let the waveform sample library be... DB , which includes M A standard sample waveform { DB 1 DB 2 ,...,DB M}. Standard analysis waveform W norm With each sample in the sample library DB j (in j Similarity calculations are performed on pairs (e.g., 1, 2, ..., M); an elastic time series matching algorithm can be used to calculate the similarity. W norm Each DB j similarity distance value d j ; All calculated distance values d j and its corresponding sample index j Store in a result set .

[0048] The above steps allow for the efficient and accurate calculation of the similarity distance between the standard analysis waveform and multiple historical samples. This provides crucial data support for selecting highly similar samples and making decisions about triggering causes based on these samples. The robustness of the DTW algorithm makes it particularly suitable for comparing waveform data; even with small time delays or distortions between waveforms, it can find the optimal alignment, thereby obtaining meaningful similarity distance values.

[0049] Step S210: Based on the target sample library, select the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from multiple similarity distances as the triggering cause corresponding to the original transient waveform data.

[0050] In this step, all calculated similarity distance values ​​can be sorted numerically, typically in ascending order (i.e., smaller distances rank higher), to identify historical samples most similar to the waveform to be identified. The triggering cause corresponding to the historical transient waveform data with the lowest similarity distance value (representing the highest similarity) can be selected as the triggering cause corresponding to the original transient waveform data. Alternatively, a threshold or ranking strategy can be defined to determine which historical samples are considered "highly similar." This could involve setting a distance threshold, with only samples below this threshold considered highly similar; or selecting a certain number of top-ranked samples (e.g., the top 5% or top 10) as reference objects. Based on preset conditions, a high-similarity sample set is selected from the sorted result set. The waveforms of these samples are considered to have high morphological similarity to the waveform to be identified. The triggering cause categories included in the high-similarity sample set are statistically analyzed, and then a weight is assigned to each sample in the set based on the similarity distance value. The smaller the distance value, the higher the weight, thus quantifying the influence of each reference sample. The weights of each triggering reason are aggregated, and a comprehensive score for each possible triggering reason is calculated using strategies such as weighted average or weighted voting.

[0051] Compare the comprehensive scores of all possible triggering causes and select the cause with the highest score as the most likely triggering cause of the event to be identified. A confidence index can be added to assess the reliability of the determination result based on the distribution of highly similar samples and the dispersion of the comprehensive scores. The finally determined triggering cause, along with its confidence score, is output as the intelligent identification result of the triggering cause for this original transient waveform data.

[0052] Through the above steps, the goal of determining the triggering cause of a single-phase grounding signal can be achieved by calculating the similarity between the waveform to be identified and the sample waveform. This improves the accuracy of determining the triggering cause of a single-phase grounding signal and solves the technical problem of insufficient accuracy in judging the triggering cause of a single-phase grounding signal.

[0053] As an optional embodiment, obtaining the target sample library includes: acquiring multiple historical transient waveform data; determining the waveforms corresponding to the multiple historical transient waveform data based on the multiple historical transient waveform data; aligning the waveforms corresponding to the multiple historical transient waveform data based on a preset time reference to obtain the adjusted waveforms corresponding to each of the multiple historical transient waveform data; determining the sample correspondence relationship based on the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes corresponding to each of the multiple historical transient waveform data, wherein the sample correspondence relationship characterizes the correspondence between the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes; and determining the target sample library based on the sample correspondence relationship.

[0054] Optionally, raw transient waveform data of historical single-phase grounding events and corresponding on-site investigation records can be collected. The collected raw data is screened based on the criteria of complete waveform data, clear on-site images, and reasonable electrical characteristics to ensure the basic quality of the samples to be included in the database. Then, according to a predefined trigger cause labeling system, the investigation records are compared to assign the most matching cause label to the selected samples. The waveform data with the labeled cause, along with their corresponding cause labels, equipment information, time information, etc., are stored in a structured database to form a high-quality waveform sample library.

[0055] Specifically, the corresponding waveform can be determined using a standardization process based on historical transient waveform data. This means that each historical event is converted into a waveform with a uniform format, the same sampling rate, and normalized values. The standardization process involves performing a time window truncation operation on the historical transient waveform data. This operation extracts a waveform data subsequence of fixed length based on the known fault trigger time and preset pre-fault and post-fault cycle numbers, resulting in the truncated waveform. This step aims to focus the analysis window on the critical time period before and after the fault. Then, waveform data with different sampling rates are unified to a fixed standard sampling rate. A system-level standard sampling rate is set. If the original sampling rate is inconsistent with the standard sampling rate, linear interpolation is used for resampling, generating a new waveform sequence with the standard sampling rate, resulting in the resampled waveform. A numerical normalization operation is then performed on the resampled waveform. This operation maps the overall amplitude of the waveform sequence to a standard numerical range (e.g., [-1, 1]) using a preset numerical transformation function. This step aims to eliminate absolute amplitude differences in the waveform, thereby highlighting its inherent, relative shape characteristics, and obtaining the final standard analysis waveform.

[0056] A time baseline can also be set, using the fault trigger point as a reference. For all historical waveform data, their starting points are adjusted to ensure time alignment of the data windows before and after the fault. During the alignment process, for waveforms with differences in length or time scale, the DTW algorithm is used for alignment, allowing non-linear mapping of time points to find the optimal alignment path between two sequences. The adjusted historical waveform data, along with the corresponding trigger cause tags, is stored in a structured database. Each data record includes a standardized waveform, trigger cause, event timestamp, and possibly other descriptive information. In the database, a one-to-one correspondence is formed between each historical waveform and its trigger cause; that is, each waveform data is explicitly associated with a trigger cause tag. This correspondence is reflected in the form of the data record, such as: waveform data, trigger cause type, fault date and time, geographical location information, and the IDs of affected lines and equipment.

[0057] Regular data reviews can be conducted to verify the accuracy of waveform and trigger cause annotations, and to remove outliers or suspected errors to maintain the high quality of the sample correspondence database. Over time, new historical data is continuously added to update the sample correspondences, adapting to changes in the power grid environment and equipment status.

[0058] By following the steps above, a database containing a large number of standardized historical waveforms and their trigger cause labels can be established. This database not only provides a foundation for the learning and validation of the identification algorithm, but also allows it to be continuously optimized as new data is added, thereby improving the accuracy and reliability of fault trigger cause identification.

[0059] As an optional embodiment, the acquisition of the original transient waveform data to be identified includes: obtaining the trigger time of the single-phase grounding signal; based on the trigger time of the single-phase grounding signal, extracting a waveform data subsequence of fixed time length according to the preset number of cycles before and after the fault, and determining the original transient waveform data.

[0060] Optionally, the system can monitor the distribution network's operating status in real time. When an abnormal increase in zero-sequence current or the appearance of negative-sequence voltage, or other electrical characteristic changes that may indicate a single-phase grounding event, is detected, the system will automatically trigger an alarm and record the precise moment of the event. This moment is the trigger moment of the single-phase grounding signal. Based on historical data analysis, the number of pre-fault cycles and the number of post-fault cycles are preset. These two factors together determine the length of the extracted data subsequence to cover the critical time periods before and after the fault trigger. Then, with the trigger moment of the single-phase grounding signal as the center, waveform data for the preset number of cycles is extracted forward and backward. This ensures that the extracted time window covers a sufficient pre-fault background and post-fault reaction period, thus capturing the complete fault development process and related electrical characteristic changes. The extracted waveform data subsequences are integrated into the original transient waveform data. If the data sampling frequency differs from the preset standard, the sampling rate needs to be further adjusted to ensure that all waveform data participating in the comparison have the same sampling density and time specification. Through the above steps, the exact moment of the single-phase grounding event can be accurately located, and a standardized fault waveform data subsequence can be extracted based on this moment. This standardized data not only includes the immediate response at the moment the fault occurs, but also the trend of electrical characteristic changes over a period of time before and after the fault. In practical applications, this subsequence will be further processed to meet the needs of waveform comparison, such as unifying the sampling rate and normalizing the values.

[0061] As an optional embodiment, the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data is calculated, wherein the target historical transient waveform data is any one of multiple historical transient waveform data. This includes: initializing a distance matrix, where each element in the distance matrix represents the distance between a sampling point in the standard analysis waveform and a sampling point in the waveform corresponding to the target historical transient waveform data; calculating the distance between the first sampling point in the standard analysis waveform and multiple sampling points in the waveform corresponding to the target historical transient waveform data, and storing this distance in the distance matrix to obtain a first matrix; calculating the distance between the first sampling point in the waveform corresponding to the target historical transient waveform data and multiple sampling points in the standard analysis waveform, and storing this distance in the first matrix to obtain a second matrix; using a dynamic programming algorithm to calculate the remaining elements in the second matrix to obtain a third matrix; and determining the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data based on the third matrix.

[0062] Optionally, transient waveforms of instantaneous grounding events exhibit significant non-stationary characteristics; even with the same triggering cause, their waveforms often show non-linear scaling or shifting on the time axis. Traditional measurement methods, such as Euclidean distance, employ rigorous point-by-point comparisons and are highly sensitive to time axis drift, thus unsuitable for measuring the similarity of transient waveforms in instantaneous grounding events. Dynamic Time Warping (DTW) algorithms, by allowing non-linear mapping of time points, find the optimal alignment path between two sequences, effectively overcoming the influence of time axis drift. Therefore, the DTW algorithm can be used to measure the true morphological similarity of waveforms. The waveforms corresponding to the original transient recording data can be converted into time series. W norm Its length is n , W norm ={ w 1 ,w 2 ,...,w n At the same time, a sample library can be set up based on the sample correspondence. DB , which includes M A set of standard sample waveforms, wherein the standard sample waveforms can be any sample waveform selected from the waveforms corresponding to historical transient waveform data. DB j (in j =1,2,..., M ( ) is a time series with a length of n , The standard analysis waveform is calculated using the DTW algorithm. W norm With each sample waveform DB jsimilarity distance value between d j The detailed calculation and execution process is as follows:

[0063] (1) Initialize the distance matrix D Initialize the distance matrix used to store the cumulative distance. D Given that the lengths of the two waveforms being compared are both... n Distance matrix D The dimension is n×n .

[0064] (2) Calculate the distance matrix D The starting element D 1,1 :

[0065] ,

[0066] in, express W norm The i sampling points

[0067] (3) Calculate the first row elements of the distance matrix D D 1,k Using a loop from k =2 to n, calculate the remaining elements in the first row. D 1,k :

[0068] ;

[0069] (4) Calculate the first column elements of the distance matrix D D i,1 Using a loop from i =2 to n, calculate the remaining elements in the first row. D i,1 :

[0070] ;

[0071] (5) Calculate the matrix D All other elements in the matrix. Using two nested loops (outer loop i from 2 to n, inner loop k from 2 to n), calculate the matrix according to the following recursive formula. D All other elements in:

[0072] ;

[0073] (6) Based on the final elements of the matrix ,Sure W norm and DBj similarity distance between d j :

[0074] ;

[0075] For all in the sample library M After repeating the above steps for each sample waveform, a waveform containing... M A set of similarity distance values ​​{ d 1 ,d 2 ,…,d M To facilitate subsequent sorting and filtering, this step binds each calculated distance value dj with its corresponding unique index j in the database, forming a tuple ( d j , j Ultimately, all of them M Store the tuples into a result set. R middle: .

[0076] This series of steps ensures that the similarity distance calculation between the original transient waveform data and the target historical transient waveform data is both accurate and comprehensive. Taking into account the nonlinear differences on the time axis, it provides a solid foundation for subsequent screening of high-similarity samples and intelligent decision-making on triggering causes. The use of the DTW algorithm enables the finding of the optimal alignment path between two waveforms even if there are slight temporal misalignments or distortions, thereby accurately measuring the similarity of the waveforms.

[0077] As an optional embodiment, based on a target sample library, the triggering cause corresponding to the historical transient waveform data corresponding to a similarity distance that meets preset conditions is selected from multiple similarity distances as the triggering cause corresponding to the original transient waveform data. This includes: selecting waveforms corresponding to the historical transient waveform data corresponding to a similarity distance that meets preset conditions from multiple similarity distances based on the target sample library, forming a high-similarity sample set; calculating the weights corresponding to the historical transient waveform data in the high-similarity sample set based on the similarity of the historical transient waveform data in the high-similarity sample set; and determining the triggering cause corresponding to the original transient waveform data based on the weights and triggering causes of the historical transient waveform data in the high-similarity sample set.

[0078] Optionally, a set of highly similar samples can be selected from multiple similarity distances in the sample library. Weights can then be calculated based on these samples to infer the triggering cause of the original transient waveform data. A threshold or ranking mechanism for selecting highly similar samples can be defined first. For example, the top few samples with the smallest distances can be selected, or a distance threshold can be set to retain only samples with distances less than that threshold. From the calculated set of similarity distances, historical waveform data is selected according to preset conditions to form a set of highly similar samples. Each sample includes not only the corresponding waveform data but also its triggering cause label and other relevant information. For each sample in the set of highly similar samples, the weight of its associated triggering cause is calculated. This can be done by simply summing the weights of all samples with the same triggering cause label. For each historical waveform data in the set of highly similar samples, its weight can be calculated by the reciprocal of the similarity distance value to reflect the principle that smaller distances result in greater weights. To avoid division by zero, a small positive number (close to zero but not zero) is usually added to the formula for calculating the weight. The triggering reasons are sorted in descending order based on their cumulative weights, and the triggering reason with the highest weight is the most likely triggering reason for the original transient waveform data. If multiple triggering reasons have similar weights, a threshold can be set according to the specific situation, and triggering reasons exceeding this threshold are considered as possible reasons.

[0079] Finally, the determined triggering cause and its corresponding cumulative weight are output as the final judgment result. This result not only provides a qualitative judgment of the triggering cause, but also quantitative support for its weight, which helps operations and maintenance personnel make more accurate risk assessments and operational decisions.

[0080] This process leverages the collective wisdom of highly similar samples, effectively mitigating the uncertainty risks inherent in small sample sizes and improving the accuracy and practicality of the intelligent identification system for single-phase grounding signal triggering causes.

[0081] For example, the result set generated above Based on distance value d j Sort the results in ascending order. From the sorted results, select the top... P The samples with the smallest distances form a set of highly similar samples. S p . P Let be a pre-defined natural number representing the number of reference samples used for decision-making. Based on a set of highly similar samples. S p Calculate the weighted score for each possible triggering cause. For S p For each sample in the dataset, its weight is... It is defined as the reciprocal of its similarity distance dj. To avoid calculation errors caused by a distance value of zero, a very small positive number is added to the denominator. Weight The calculation formula is: Identify high-similarity sample sets. S p All unique trigger cause categories contained herein constitute a cause set. Then, for each candidate reason... , and put it in S p The total weight of the cause is obtained by summing the weights of all corresponding samples. :

[0082] ;

[0083] in, Indicates that the index is j The reason label for the sample.

[0084] The aggregated weights can be converted into normalized probability scores, and the final identification conclusion can be determined.

[0085] Calculate the sum of the total weights of all candidate causes. Then, the aggregate weights for each cause are normalized to obtain the final probability score for that cause. :

[0086] ;

[0087] All candidate causes and their corresponding probability scores The results are sorted in descending order, and the causes and their probabilities are output together as the final judgment result of the single-phase grounding signal triggering cause.

[0088] As an optional embodiment, based on the similarity of historical transient waveform data in the high-similarity sample set, the weights of the historical transient waveform data in the high-similarity sample set are calculated according to a preset formula, wherein the preset formula is as follows:

[0089] ;

[0090] in, For the set of highly similar samples, the first j The weights of historical transient waveform data d j For the first j The similarity distance between historical transient waveform data S p A set of highly similar samples, It is a positive number that is lower than the preset threshold.

[0091] The following is a specific example. Figure 3 This is a schematic diagram of a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, provided by an optional embodiment of the present invention. Figure 3 As shown, it includes:

[0092] Step 1: Construct a high-quality sample library. Collect historical single-phase grounding events of the distribution network, and construct a sample library containing standardized transient waveform data and corresponding cause labels through manual verification and fine annotation.

[0093] Step 2: Standardize transient waveform data. Process the raw transient waveform data to be identified into a standard analysis waveform that has the same specification as the case waveforms in the sample library and is suitable for comparative analysis.

[0094] Step 3: Waveform similarity matching, calculate the similarity distance between the standard analysis waveform and each sample waveform in the high-quality waveform sample library;

[0095] Step 4: Determine the triggering cause of the single-phase grounding signal. Based on the similarity distance, select high similarity samples and determine the triggering cause of the signal to be identified through a preset classification strategy.

[0096] Step 1 includes:

[0097] Step 11, Data Acquisition: Collect raw transient waveform data of historical single-phase grounding events, as well as corresponding on-site investigation records;

[0098] Step 12, Manual screening: The collected raw data is screened based on the criteria of complete waveform data, clear on-site images, and reasonable electrical characteristics to ensure the basic quality of the samples to be included in the database.

[0099] Step 13, Cause Labeling: Based on a predefined trigger cause labeling system, compare the survey records and assign the most matching cause label to the selected samples;

[0100] Step 14, Sample Storage: Store the waveform data with the identified causes, along with their corresponding cause labels, device information, time information, etc., into a structured database to form a high-quality waveform sample library.

[0101] Step 2 specifically includes the following steps:

[0102] Step 21, Data Alignment: The purpose is to solve the heterogeneity problem caused by different recording durations and sampling rates by unifying the time base and sampling density of the waveform, so that all waveform data have a unified time specification and become comparable time series.

[0103] Step 22, numerical normalization: The purpose is to eliminate the influence of non-essential factors such as fault location and grounding resistance by unifying the amplitude scale of the waveform, and to highlight the inherent shape characteristics of the waveform that can reflect the physical process of the fault.

[0104] Step 3 specifically includes the following steps:

[0105] Step 31, Similarity Calculation: Let the waveform sample library be... DB , which includes M A standard sample waveform { DB 1 DB 2 , ..., DB M}. Standard analysis waveform W norm With each sample in the sample library DB j (in j Similarity calculations are performed on pairs (e.g., 1, 2, ..., M); an elastic time series matching algorithm can be used to calculate the similarity. W norm Each DB j similarity distance value d j ; All calculated distance values d j and its corresponding sample index j Store in a result set .

[0106] Step 32: Calculate the similarity distance: The similarity calculation preferably uses an elastic time series matching algorithm to calculate the similarity distance. W norm Each DB j similarity distance value d j ;

[0107] Step 33, Result Set Generation: Generate all calculated distance values d j The corresponding sample index j is stored in a result set. .

[0108] Step 4 specifically includes the following steps:

[0109] Step 41, High Similarity Sample Screening: Sort the result set R in ascending order according to the distance value dj to form a high similarity sample set;

[0110] Step 42, Similarity Screening: Candidate Cause Weight Calculation and Aggregation. Based on the high-similarity sample set Sp, calculate the weight score for each possible triggering cause;

[0111] Step 43, Cause Determination: Cause Probability Normalization and Final Determination. The aggregated weights are converted into normalized probability scores, and the final identification conclusion is determined.

[0112] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0113] Through the above description of the embodiments, those skilled in the art can clearly understand that the method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, 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 is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0114] According to embodiments of the present invention, a waveform matching-based device for identifying the triggering cause of a single-phase grounding signal in a distribution network is also provided for implementing the above-described waveform matching-based method for identifying the triggering cause of a single-phase grounding signal in a distribution network. Figure 4 This is a structural block diagram of a distribution network single-phase grounding signal triggering cause identification device based on waveform matching according to an embodiment of the present invention, as shown below. Figure 4 As shown, the waveform matching-based distribution network single-phase grounding signal triggering cause identification device includes: acquisition module 402, acquisition module 404, processing module 406, calculation module 408 and selection module 410. The following is a description of the waveform matching-based distribution network single-phase grounding signal triggering cause identification device.

[0115] The acquisition module 402 is used to acquire a target sample library, which includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signals of the distribution network and triggering reasons corresponding to multiple historical transient waveform data.

[0116] The acquisition module 404, connected to the acquisition module 402, is used to acquire the waveform corresponding to the original transient waveform data to be identified.

[0117] The processing module 406, connected to the acquisition module 404, is used to process the waveform corresponding to the original transient waveform data to be identified based on a preset time reference, so as to obtain a standard analysis waveform.

[0118] The calculation module 408, connected to the processing module 406, is used to calculate the similarity distance between the standard analysis waveform and the waveforms corresponding to multiple historical transient waveform data, thereby obtaining multiple similarity distances.

[0119] The selection module 410, connected to the calculation module 408, is used to select the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from multiple similarity distances based on the target sample library, as the triggering cause corresponding to the original transient waveform data.

[0120] It should be noted that the acquisition module 402, collection module 404, processing module 406, calculation module 408, and selection module 410 mentioned above correspond to steps S202 to S210 in the embodiments. Multiple modules and their corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments. It should also be noted that the above modules, as part of the device, can run on the computer terminal 10 provided in the embodiments.

[0121] Embodiments of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one of a plurality of network devices in a computer network. The computer device includes a memory and a processor.

[0122] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the waveform matching-based distribution network single-phase grounding signal triggering cause identification method and device in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned waveform matching-based distribution network single-phase grounding signal triggering cause identification method. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to a computer terminal 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.

[0123] The processor can invoke the information and application program stored in the memory through the transmission device to perform the following steps: acquiring a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signals of the distribution network and triggering causes corresponding to multiple historical transient waveform data; acquiring waveforms corresponding to the original transient waveform data to be identified; processing the waveforms corresponding to the original transient waveform data to be identified based on a preset time reference to obtain standard analysis waveforms; calculating the similarity distances between the standard analysis waveforms and the waveforms corresponding to multiple historical transient waveform data to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from the multiple similarity distances based on the target sample library as the triggering cause corresponding to the original transient waveform data.

[0124] Optionally, the processor may also execute program code for the following steps: acquiring a target sample library, including: acquiring multiple historical transient waveform data; determining the waveforms corresponding to the multiple historical transient waveform data based on the multiple historical transient waveform data; aligning the waveforms corresponding to the multiple historical transient waveform data based on a preset time reference to obtain the adjusted waveforms corresponding to each of the multiple historical transient waveform data; determining the sample correspondence based on the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes corresponding to each of the multiple historical transient waveform data, wherein the sample correspondence represents the correspondence between the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes; and determining the target sample library based on the sample correspondence.

[0125] Optionally, the processor may also execute program code for the following steps: acquiring the original transient waveform data to be identified, including: obtaining the trigger time of the single-phase grounding signal; based on the trigger time of the single-phase grounding signal, extracting a waveform data subsequence of fixed time length according to the preset number of cycles before and after the fault, and determining the original transient waveform data.

[0126] Optionally, the processor may also execute program code for the following steps: calculating the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data, wherein the target historical transient waveform data is any one of multiple historical transient waveform data, including: initializing a distance matrix, wherein the elements in the distance matrix represent the distance between the sampling points in the standard analysis waveform and the sampling points in the waveform corresponding to the target historical transient waveform data; calculating the distance between the first sampling point in the standard analysis waveform and multiple sampling points in the waveform corresponding to the target historical transient waveform data, storing them in the distance matrix to obtain a first matrix; calculating the distance between the first sampling point in the waveform corresponding to the target historical transient waveform data and multiple sampling points in the standard analysis waveform, storing them in the first matrix to obtain a second matrix; using a dynamic programming algorithm to calculate the remaining elements in the second matrix to obtain a third matrix; and determining the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data based on the third matrix.

[0127] Optionally, the processor may also execute program code for the following steps: Based on the target sample library, selecting the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from multiple similarity distances as the triggering cause corresponding to the original transient waveform data, including: Based on the target sample library, selecting the waveforms corresponding to the historical transient waveform data that meet the preset conditions from multiple similarity distances to form a high similarity sample set; Calculating the weight of the historical transient waveform data in the high similarity sample set based on the similarity of the historical transient waveform data in the high similarity sample set; Determining the triggering cause corresponding to the original transient waveform data based on the weight and triggering cause of the historical transient waveform data in the high similarity sample set.

[0128] Optionally, the processor may also execute program code that performs the following steps: based on the similarity of the historical transient waveform data in the high-similarity sample set, calculate the weight of the historical transient waveform data in the high-similarity sample set according to a preset formula, wherein the preset formula is as follows:

[0129] ;

[0130] in, For the set of highly similar samples, the first j The weights of historical transient waveform data d j For the first j The similarity distance between historical transient waveform data S p A set of highly similar samples, It is a positive number that is lower than the preset threshold.

[0131] This invention provides a method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching. The method involves acquiring a target sample library, which includes waveforms and triggering causes corresponding to multiple historical transient waveforms generated by historical single-phase grounding signals in the distribution network; collecting the waveform corresponding to the original transient waveform data to be identified; processing the waveform corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; calculating the similarity distance between the standard analysis waveform and the waveforms corresponding to the multiple historical transient waveforms to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that meets preset conditions from the multiple similarity distances based on the target sample library. This achieves the goal of determining the triggering cause of a single-phase grounding signal by calculating the similarity between the waveform to be identified and the sample waveforms, thereby improving the accuracy of determining the triggering cause of a single-phase grounding signal and solving the technical problem of insufficient accuracy in judging the triggering cause of a single-phase grounding signal.

[0132] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a non-volatile storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0133] Embodiments of the present invention also provide a non-volatile storage medium. Optionally, in this embodiment, the aforementioned non-volatile storage medium can be used to store the program code executed by the waveform matching-based distribution network single-phase grounding signal triggering cause identification method provided in the above embodiments.

[0134] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0135] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: acquiring a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signals of the distribution network and triggering causes corresponding to multiple historical transient waveform data; acquiring waveforms corresponding to the original transient waveform data to be identified; processing the waveforms corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; calculating the similarity distances between the standard analysis waveforms and the waveforms corresponding to the multiple historical transient waveform data to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from the multiple similarity distances based on the target sample library as the triggering cause corresponding to the original transient waveform data.

[0136] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: obtaining a target sample library, including: obtaining multiple historical transient waveform data; determining the waveforms corresponding to the multiple historical transient waveform data based on the multiple historical transient waveform data; aligning the waveforms corresponding to the multiple historical transient waveform data based on a preset time reference to obtain the adjusted waveforms corresponding to each of the multiple historical transient waveform data; determining the sample correspondence relationship based on the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes corresponding to each of the multiple historical transient waveform data, wherein the sample correspondence relationship characterizes the correspondence relationship between the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes; and determining the target sample library based on the sample correspondence relationship.

[0137] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: acquiring the original transient waveform data to be identified, including: obtaining the trigger time of the single-phase grounding signal; based on the trigger time of the single-phase grounding signal, extracting a waveform data subsequence of fixed time length according to the preset number of cycles before and after the fault, and determining the original transient waveform data.

[0138] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: calculating the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data, wherein the target historical transient waveform data is any one of multiple historical transient waveform data, including: initializing a distance matrix, wherein the elements in the distance matrix represent the distance between the sampling points in the standard analysis waveform and the sampling points in the waveform corresponding to the target historical transient waveform data; calculating the distance between the first sampling point in the standard analysis waveform and multiple sampling points in the waveform corresponding to the target historical transient waveform data, storing it in the distance matrix to obtain a first matrix; calculating the distance between the first sampling point in the waveform corresponding to the target historical transient waveform and multiple sampling points in the standard analysis waveform, storing it in the first matrix to obtain a second matrix; using a dynamic programming algorithm to calculate the remaining elements in the second matrix to obtain a third matrix; and determining the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data based on the third matrix.

[0139] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: selecting, based on the target sample library, the triggering cause corresponding to the historical transient waveform data corresponding to the similarity distance that meets the preset conditions from multiple similarity distances as the triggering cause corresponding to the original transient waveform data, including: selecting the waveforms corresponding to the historical transient waveform data corresponding to the similarity distance that meets the preset conditions from multiple similarity distances based on the target sample library, forming a high similarity sample set; calculating the weight corresponding to the historical transient waveform data in the high similarity sample set based on the similarity of the historical transient waveform data in the high similarity sample set; and determining the triggering cause corresponding to the original transient waveform data based on the weight and triggering cause corresponding to the historical transient waveform data in the high similarity sample set.

[0140] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: based on the similarity of the historical transient waveform data in the high-similarity sample set, calculate the weight of the historical transient waveform data in the high-similarity sample set according to a preset formula, wherein the preset formula is as follows:

[0141] ;

[0142] in, For the set of highly similar samples, the first j The weights of historical transient waveform data d j For the first j The similarity distance between historical transient waveform data S p A set of highly similar samples, It is a positive number that is lower than the preset threshold.

[0143] Embodiments of the present invention also provide a computer program product, including a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, it can perform the following: acquiring a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signals of the distribution network and triggering causes corresponding to multiple historical transient waveform data; acquiring waveforms corresponding to the original transient waveform data to be identified; processing the waveforms corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform; calculating the similarity distances between the standard analysis waveforms and the waveforms corresponding to the multiple historical transient waveform data to obtain multiple similarity distances; and selecting the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from the multiple similarity distances based on the target sample library as the triggering cause corresponding to the original transient waveform data.

[0144] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0145] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0146] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0147] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0148] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0149] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0150] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, characterized in that, include: Obtain a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signals of the distribution network and triggering reasons corresponding to each of the multiple historical transient waveform data; Collect the waveform corresponding to the original transient waveform data to be identified; Based on a preset time reference, the waveform corresponding to the original transient waveform data to be identified is processed to obtain a standard analysis waveform. Calculate the similarity distance between the standard analysis waveform and the waveforms corresponding to the multiple historical transient waveform data to obtain multiple similarity distances; Based on the target sample library, the triggering reason corresponding to the historical transient waveform data that meets the preset conditions is selected from the multiple similarity distances as the triggering reason corresponding to the original transient waveform data.

2. The method according to claim 1, characterized in that, The acquisition of the target sample library includes: Acquire the aforementioned multiple historical transient waveform data; Based on the multiple historical transient waveform data, determine the waveforms corresponding to the multiple historical transient waveform data; Based on the preset time reference, the waveforms corresponding to the multiple historical transient waveform data are aligned to obtain the adjusted waveforms corresponding to each of the multiple historical transient waveform data. Based on the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes corresponding to each of the multiple historical transient waveform data, a sample correspondence is determined, wherein the sample correspondence represents the correspondence between the adjusted waveforms corresponding to each of the multiple historical transient waveform data and the triggering causes; Based on the sample correspondence, the target sample library is determined.

3. The method according to claim 1, characterized in that, The acquired raw transient waveform data to be identified includes: Obtain the trigger time of the single-phase grounding signal; Based on the trigger time of the single-phase grounding signal, and according to the preset number of cycles before and after the fault, a waveform data subsequence of fixed time length is extracted to determine the original transient waveform data.

4. The method according to claim 1, characterized in that, Calculate the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data, wherein the target historical transient waveform data is any one of the plurality of historical transient waveform data, including: Initialize a distance matrix, wherein the elements of the distance matrix represent the distance between the sampling points in the standard analysis waveform and the sampling points in the waveform corresponding to the target historical transient waveform data; Calculate the distance between the first sampling point in the standard analysis waveform and multiple sampling points in the waveform corresponding to the target historical transient waveform data, and store them in the distance matrix to obtain the first matrix; Calculate the distance between the first sampling point in the waveform corresponding to the target historical transient waveform data and multiple sampling points in the standard analysis waveform, and store the distance in the first matrix to obtain the second matrix; Using a dynamic programming algorithm, the remaining elements in the second matrix are calculated to obtain the third matrix; Based on the third matrix, the similarity distance between the standard analysis waveform and the waveform corresponding to the target historical transient waveform data is determined.

5. The method according to claim 1, characterized in that, The step of selecting the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from the multiple similarity distances based on the target sample library as the triggering cause corresponding to the original transient waveform data includes: Based on the target sample library, waveforms corresponding to historical transient waveform data that meet the preset conditions are selected from the multiple similarity distances to form a high similarity sample set; Based on the similarity of the historical transient waveform data in the high similarity sample set, the weights of the historical transient waveform data in the high similarity sample set are calculated. The triggering cause corresponding to the original transient waveform data is determined based on the weights and triggering causes of the historical transient waveform data in the high similarity sample set.

6. The method according to claim 5, characterized in that, Based on the similarity of the historical transient waveform data in the high-similarity sample set, the weights of the historical transient waveform data in the high-similarity sample set are calculated according to a preset formula, wherein the preset formula is as follows: ； in, The first high-similarity sample in the set of high-similarity samples j The weights of historical transient waveform data d j For the first j The similarity distance between historical transient waveform data S p The set of highly similar samples, It is a positive number that is lower than the preset threshold.

7. A device for identifying the triggering cause of a single-phase grounding signal in a distribution network based on waveform matching, characterized in that, include: The acquisition module is used to acquire a target sample library, wherein the target sample library includes waveforms corresponding to multiple historical transient waveform data generated by historical single-phase grounding signals of the distribution network and triggering reasons corresponding to each of the multiple historical transient waveform data. The acquisition module is used to acquire the waveforms corresponding to the raw transient waveform data to be identified; The processing module is used to process the waveform corresponding to the original transient waveform data to be identified based on a preset time reference to obtain a standard analysis waveform. The calculation module is used to calculate the similarity distance between the standard analysis waveform and the waveforms corresponding to the multiple historical transient waveform data, and obtain multiple similarity distances; The selection module is used to select, based on the target sample library, the triggering cause corresponding to the historical transient waveform data that meets the preset conditions from the multiple similarity distances as the triggering cause corresponding to the original transient waveform data.

8. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the non-volatile storage medium to execute the waveform matching-based method for identifying the triggering cause of a single-phase grounding signal in a power distribution network as described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: Memory and processor The memory stores computer programs; The processor is configured to execute a computer program stored in the memory, wherein when the computer program is executed, the processor performs the waveform matching-based method for identifying the triggering cause of a single-phase grounding signal in a power distribution network as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for identifying the triggering cause of a single-phase grounding signal in a power distribution network based on waveform matching, as described in any one of claims 1 to 6.

Images