Electromagnetic transient simulation abnormal waveform identification method, system, device and medium

Abstract

The invention discloses an electromagnetic transient simulation abnormal waveform recognition method, system and device and a medium, and the method comprises the steps: carrying out the recording of continuous waveform data of electromagnetic transient simulation, obtaining recording data, and carrying out the preprocessing of the recording data; sampling the preprocessed recording data to obtain sampling data, storing the sampling data as a data set, and dividing the data set into a training set and a to-be-detected waveform data set; training the self-encoding neural network through the training set to obtain an abnormal waveform recognition model; and inputting a to-be-detected waveform data set into the abnormal waveform recognition model, outputting reconstructed data, and performing comparative analysis on the reconstructed data and the to-be-detected waveform data set to determine the state of the waveform. Therefore, the problem that the abnormal waveform in the electromagnetic transient simulation cannot be accurately and efficiently identified due to the lack of self-adaptability and difficulty in comprehensively covering all abnormal conditions in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a method, system, device and medium for identifying abnormal waveforms in electromagnetic transient simulation. Background Art

[0002] As a key power system analysis tool, electromagnetic transient simulation plays an important role in simulating and analyzing the behavior of power systems under various complex fault conditions. Through simulation, we can deeply understand the dynamic characteristics of power systems, optimize design and operation strategies, and ensure the stability and safety of power systems.

[0003] However, in the process of electromagnetic transient simulation, abnormal waveforms may be generated due to the influence of various factors such as system parameters, model accuracy, calculation errors and external interference. These abnormal waveforms not only reflect the abnormal state of the power system, but also pose a potential threat to the stable operation of the system. Therefore, accurate and efficient identification of these abnormal waveforms is of great significance for monitoring, early warning and fault handling of power systems.

[0004] Traditionally, the identification of abnormal waveforms mainly relies on the method of setting thresholds. That is, according to the normal operating status of the power system, a series of thresholds of characteristic information are set. When certain characteristic information of the simulation waveform data exceeds or falls below these thresholds, the waveform is considered to be an abnormal waveform. However, traditional methods have obvious limitations. On the one hand, the setting of thresholds often relies on experience and lacks adaptability; on the other hand, in the face of complex and changeable power system faults, a single threshold is difficult to fully cover all abnormal situations. Therefore, it is urgent to design an accurate and efficient abnormal waveform identification scheme in electromagnetic transient simulation. Summary of the invention

[0005] The present application provides a method, system, device and medium for identifying abnormal waveforms in electromagnetic transient simulation, which are used to solve the problem that the existing technology lacks adaptability and is difficult to fully cover all abnormal situations, resulting in the inability to accurately and efficiently identify abnormal waveforms in electromagnetic transient simulation.

[0006] In view of this, the first aspect of the present application provides a method for identifying abnormal waveforms in electromagnetic transient simulation, the method comprising:

[0007] Recording continuous waveform data of electromagnetic transient simulation to obtain recorded data, and performing preprocessing operations on the recorded data;

[0008] Sampling the preprocessed recorded data to obtain sampled data, storing the sampled data as a data set, and dividing the data set into a training set and a waveform data set to be detected;

[0009] The autoencoder neural network is trained by using the training set to obtain an abnormal waveform recognition model;

[0010] The waveform data set to be detected is input into an abnormal waveform recognition model, reconstructed data is output, and the reconstructed data is compared and analyzed with the waveform data set to be detected to determine the state of the waveform.

[0011] Optionally, the preprocessing operation on the recorded data includes: denoising, filtering, and normalizing the recorded data.

[0012] Optionally, sampling the preprocessed recorded data to obtain sampled data, and storing the sampled data as a data set, comprises:

[0013] The preprocessed recorded data is sampled based on a stepping sliding frame algorithm to obtain sampled data, and the sampled data is stored in a csv format text to obtain a data set.

[0014] Optionally, comparing and analyzing the reconstructed data with the waveform data set to be detected to determine the state of the waveform includes:

[0015] The reconstructed data is compared with the waveform data set to be detected to obtain a difference, the difference is compared with a preset threshold, and whether the waveform is abnormal is determined based on the comparison.

[0016] A second aspect of the present application provides an electromagnetic transient simulation abnormal waveform recognition system, the system comprising:

[0017] A wave recording unit, used for recording continuous waveform data of electromagnetic transient simulation to obtain recorded wave data, and performing preprocessing operations on the recorded wave data;

[0018] An acquisition unit is used to sample the preprocessed recorded data to obtain sampled data, store the sampled data as a data set, and divide the data set into a training set and a waveform data set to be detected;

[0019] A training unit, used for training the autoencoding neural network through the training set to obtain an abnormal waveform recognition model;

[0020] The recognition unit is used to input the waveform data set to be detected into the abnormal waveform recognition model, output reconstructed data, compare and analyze the reconstructed data with the waveform data set to be detected, and determine the state of the waveform.

[0021] Optionally, the preprocessing operation on the recorded data includes: denoising, filtering, and normalizing the recorded data.

[0022] Optionally, sampling the preprocessed recorded data to obtain sampled data, and storing the sampled data as a data set, comprises:

[0023] The preprocessed recorded data is sampled based on a stepping sliding frame algorithm to obtain sampled data, and the sampled data is stored in a csv format text to obtain a data set.

[0024] Optionally, comparing and analyzing the reconstructed data with the waveform data set to be detected to determine the state of the waveform includes:

[0025] The reconstructed data is compared with the waveform data set to be detected to obtain a difference, the difference is compared with a preset threshold, and whether the waveform is abnormal is determined based on the comparison.

[0026] A third aspect of the present application provides an electromagnetic transient simulation abnormal waveform recognition device, the device comprising a processor and a memory:

[0027] The memory is used to store program code and transmit the program code to the processor;

[0028] The processor is used to execute the steps of the electromagnetic transient simulation abnormal waveform identification method as described in the first aspect according to the instructions in the program code.

[0029] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the electromagnetic transient simulation abnormal waveform identification method described in the first aspect above.

[0030] It can be seen from the above technical solutions that this application has the following advantages:

[0031] 1) This application uses a step-sliding frame algorithm to process the recorded data. By dynamically adjusting the sampling window and step size, the key characteristics of the waveform, such as peak value, frequency, phase, etc., are effectively retained. The specific implementation method and parameter setting of the waveform characteristics are maintained to ensure that the sampled data can truly reflect the dynamic changes of the original waveform, providing a high-quality data foundation for subsequent machine learning model training.

[0032] 2) The electromagnetic transient simulation abnormal waveform recognition method based on the autoencoder neural network in this application can automatically adjust the recognition strategy by learning the characteristic representation of the normal waveform, so as to realize the accurate recognition of the abnormal waveform. And it can handle different types of abnormal waveforms. This means that the model can not only recognize known fault types, but also make certain predictions and recognitions of unknown fault types. Furthermore, it can realize the rapid processing and analysis of waveform data. This enables this method to play an important role in real-time monitoring and early warning, and timely discover and process abnormal waveforms in the power system. In summary, the electromagnetic transient simulation abnormal waveform recognition method based on the autoencoder neural network adopted in this application has the advantages of improving recognition accuracy, enhancing generalization ability, improving recognition efficiency, and promoting intelligent development. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic diagram of a flow chart of an abnormal waveform identification method for electromagnetic transient simulation provided in an embodiment of the present application;

[0034] Figure 2 This is a structural schematic diagram of an electromagnetic transient simulation abnormal waveform identification system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0036] See also Figure 1 , a method for identifying abnormal waveforms in electromagnetic transient simulation provided in an embodiment of the present application includes:

[0037] Step 101: Record the continuous waveform data of the electromagnetic transient simulation to obtain recorded data, and perform preprocessing operations on the recorded data.

[0038] In one embodiment, in step 101, the recorded data is preprocessed, including: denoising, filtering, and normalizing the recorded data.

[0039] It should be noted that a large amount of continuous waveform data is collected from the electromagnetic transient simulation experiment. The collected data is preprocessed, including denoising, filtering, normalization, etc., to improve the data quality.

[0040] Step 102: Sampling the preprocessed recorded data to obtain sampled data, storing the sampled data as a data set, and dividing the data set into a training set and a waveform data set to be detected.

[0041] In one embodiment, step 102 includes:

[0042] The preprocessed recorded data is sampled based on the stepping sliding frame algorithm to obtain sampled data, and the sampled data is stored in csv format text to obtain a data set.

[0043] It should be noted that after using the continuous waveform data collection and preprocessing method to collect enough normal waveform data in step 101, step 102 uses the data set generation method to collect waveform data based on the stepping sliding frame algorithm for the collected normal waveform data and generates a normal waveform data set file.

[0044] The step-by-step sliding frame algorithm specifically includes the following steps:

[0045] Define the step length S for continuous waveform sampling and the sampling length L of the continuous waveform. The step length S represents the displacement between every two adjacent samples in the continuous waveform. The sampling length L represents the number of data collected continuously in one sampling operation. This can be regarded as a sliding window that moves along the continuous waveform, each time moving the step length S, and collects L continuous waveform data. When sampling from continuous waveform data using a sliding box with a sampling length of L, the sliding box moves along the waveform with a step length of S. After each movement, L samples are collected starting from the current position. This process can be repeated to obtain a discrete representation of the entire continuous waveform. After each data collection is completed, the collected data is written to the dataset file according to the storage format of the dataset.

[0046] It should be noted here that the sampling length L of the continuous waveform should be able to include the length of the typical abnormal waveform. The step length S should be smaller than the length of the typical abnormal wavelength. This ensures that the sampled data can fully represent the abnormal situation of the waveform.

[0047] Furthermore, the training data set can be stored in CSV format, where each row of data is a specified number of continuous recording data, separated by commas, and each data is a single-step recording data of a channel. It should be noted that CSV (Comma-Separated Values) is a simple and practical file format used to store and represent various types of data including text, numerical values, etc.

[0048] Step 103: train the autoencoder neural network using the training set to obtain an abnormal waveform recognition model.

[0049] It should be noted that the autoencoder neural network is trained using the collected training set (normal waveform data set) so that the model can learn the characteristic representation of the normal waveform. During the training process, the performance of the model can be improved by adjusting the model parameters, optimizing the algorithm, etc.

[0050] Step 104: input the waveform data set to be detected into the abnormal waveform recognition model, output the reconstructed data, compare and analyze the reconstructed data with the waveform data set to be detected, and determine the state of the waveform.

[0051] In one embodiment, in step 104, the reconstructed data is compared and analyzed with the waveform data set to be detected to determine the state of the waveform, including:

[0052] The reconstructed data is compared with the waveform data set to be detected to obtain the difference, the difference is compared with the preset threshold, and whether the waveform is abnormal is determined based on the comparison.

[0053] It should be noted that in step 102, the waveform data to be detected has been collected using the continuous waveform data collection and preprocessing method, and the collected normal waveform data has been collected using the data set generation method, and a waveform data set file to be detected has been generated. In this step, the waveform data set file to be detected is input into the trained model, and by comparing the difference between the reconstructed data output by the model and the original input data, it can be determined whether the waveform is normal. If the difference exceeds the set threshold, the waveform is considered to be an abnormal waveform.

[0054] The present application provides a method for identifying abnormal waveforms in electromagnetic transient simulation. The method uses a step-sliding frame algorithm to process the recorded data. By dynamically adjusting the sampling window and step size, the key characteristics of the waveform, such as peak value, frequency, phase, etc., are effectively retained. The specific implementation method and parameter settings of the waveform characteristics are maintained to ensure that the sampled data can truly reflect the dynamic changes of the original waveform, and provide a high-quality data foundation for subsequent machine learning model training. Furthermore, an autoencoder neural network is used to identify abnormal waveforms in electromagnetic transient simulation, which improves the recognition accuracy, enhances the generalization ability, and improves the recognition efficiency.

[0055] The above is a method for identifying abnormal waveforms in electromagnetic transient simulation provided in an embodiment of the present application, and the following is a system for identifying abnormal waveforms in electromagnetic transient simulation provided in an embodiment of the present application.

[0056] See also Figure 2 , an electromagnetic transient simulation abnormal waveform recognition system provided in an embodiment of the present application includes:

[0057] The recording unit 201 is used to record the continuous waveform data of the electromagnetic transient simulation to obtain recorded data, and perform preprocessing operations on the recorded data.

[0058] The acquisition unit 202 is used to sample the preprocessed recorded data to obtain sampled data, store the sampled data as a data set, and divide the data set into a training set and a waveform data set to be detected.

[0059] The training unit 203 is used to train the autoencoder neural network through a training set to obtain an abnormal waveform recognition model.

[0060] The recognition unit 204 is used to input the waveform data set to be detected into the abnormal waveform recognition model, output the reconstructed data, compare and analyze the reconstructed data with the waveform data set to be detected, and determine the state of the waveform.

[0061] Furthermore, an electromagnetic transient simulation abnormal waveform recognition device is provided in an embodiment of the present application, and the device includes a processor and a memory:

[0062] The memory is used to store program code and transmit the program code to the processor;

[0063] The processor is used to execute the steps of the electromagnetic transient simulation abnormal waveform identification method as described in the above method embodiment according to the instructions in the program code.

[0064] Furthermore, a computer-readable storage medium is provided in an embodiment of the present application, and the computer-readable storage medium is used to store program code, and the program code is used to execute the electromagnetic transient simulation abnormal waveform identification method described in the above method embodiment.

[0065] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0066] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0067] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0068] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0069] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0070] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0071] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially 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. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0072] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying abnormal waveforms in electromagnetic transient simulation, characterized in that: include: Recording continuous waveform data of electromagnetic transient simulation to obtain recorded data, and performing preprocessing operations on the recorded data; Sampling the preprocessed recorded data to obtain sampled data, storing the sampled data as a data set, and dividing the data set into a training set and a waveform data set to be detected; The autoencoder neural network is trained by using the training set to obtain an abnormal waveform recognition model; The waveform data set to be detected is input into an abnormal waveform recognition model, reconstructed data is output, and the reconstructed data is compared and analyzed with the waveform data set to be detected to determine the state of the waveform.

2. The electromagnetic transient simulation abnormal waveform identification method according to claim 1 is characterized in that: The preprocessing operation on the recorded data includes: denoising, filtering, and normalizing the recorded data.

3. The electromagnetic transient simulation abnormal waveform identification method according to claim 1 is characterized in that: The sampling of the preprocessed recorded data to obtain sampled data, and storing the sampled data as a data set, comprises: The preprocessed recorded data is sampled based on a stepping sliding frame algorithm to obtain sampled data, and the sampled data is stored in a csv format text to obtain a data set.

4. The electromagnetic transient simulation abnormal waveform identification method according to claim 1 is characterized in that: The comparing and analyzing the reconstructed data with the waveform data set to be detected to determine the state of the waveform includes: The reconstructed data is compared with the waveform data set to be detected to obtain a difference, the difference is compared with a preset threshold, and whether the waveform is abnormal is determined based on the comparison.

5. An electromagnetic transient simulation abnormal waveform recognition system, characterized in that: include: A wave recording unit, used for recording continuous waveform data of electromagnetic transient simulation to obtain recorded wave data, and performing preprocessing operations on the recorded wave data; An acquisition unit is used to sample the preprocessed recorded data to obtain sampled data, store the sampled data as a data set, and divide the data set into a training set and a waveform data set to be detected; A training unit, used for training the autoencoding neural network through the training set to obtain an abnormal waveform recognition model; The recognition unit is used to input the waveform data set to be detected into the abnormal waveform recognition model, output reconstructed data, compare and analyze the reconstructed data with the waveform data set to be detected, and determine the state of the waveform.

6. The electromagnetic transient simulation abnormal waveform identification system according to claim 5, characterized in that: The preprocessing operation on the recorded data includes: denoising, filtering, and normalizing the recorded data.

7. The electromagnetic transient simulation abnormal waveform identification system according to claim 5, characterized in that: The sampling of the preprocessed recorded data to obtain sampled data, and storing the sampled data as a data set, comprises: The preprocessed recorded data is sampled based on a stepping sliding frame algorithm to obtain sampled data, and the sampled data is stored in a csv format text to obtain a training data set.

8. The electromagnetic transient simulation abnormal waveform identification system according to claim 5, characterized in that: The comparing and analyzing the reconstructed data with the waveform data set to be detected to determine the state of the waveform includes: The reconstructed data is compared with the waveform data set to be detected to obtain a difference, the difference is compared with a preset threshold, and whether the waveform is abnormal is determined based on the comparison.

9. An electromagnetic transient simulation abnormal waveform identification device, characterized in that: The device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the electromagnetic transient simulation abnormal waveform identification method described in any one of claims 1-4 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program codes, and the program codes are used to execute the electromagnetic transient simulation abnormal waveform identification method according to any one of claims 1 to 4.

Images