Electric transmission line discharge signal identification method, apparatus and device, and storage medium
By extracting electrical signals from vibration signals and combining database filtering and model optimization, the problem of inaccurate discharge signal identification results in UAV inspections has been solved, enabling accurate identification and efficient operation and maintenance of power transmission lines.
Patent Information
- Application Number
- CN202511322326.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-16
- Publication Date
- 2026-01-06
AI Technical Summary
In existing technologies, the direct combination of acoustic, electrical, and optical signals obtained by drone inspections leads to low reliability of discharge signal identification results, making misjudgments easy and affecting the accurate and efficient operation and maintenance of power transmission lines.
By acquiring vibration signal data from transmission lines, removing environmental interference, and converting it into electrical signals, discharge signal samples are selected from the database based on preset conditions. Then, a discharge signal recognition model is used for feature analysis and model parameter optimization to achieve accurate identification of discharge signals.
It improves the accuracy of discharge signal identification, avoids misjudgment, and supports precise and efficient operation and maintenance of transmission lines.
Smart Images
Figure CN121278464A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power transmission line discharge signal identification technology, and in particular to a method, apparatus, equipment and storage medium for identifying power transmission line discharge signals. Background Technology
[0002] High-voltage transmission lines are prone to discharge phenomena during long-term operation due to insulation aging, dirt accumulation, mechanical damage, or environmental erosion. Discharge phenomena are an early sign of insulation deterioration, which may subsequently lead to flashover, short circuit, or even power grid accidents. When discharge phenomena occur, they are inevitably accompanied by a variety of physical and chemical effects, which are the discharge signals that we can capture and analyze.
[0003] Discharge signals typically exist in the form of electrical signals (high-frequency pulse current, radio frequency electric field), acoustic signals (ultrasonic vibration), and optical signals (discharge arc). However, in existing technologies, the acoustic signals obtained by UAV inspections are directly combined with the aforementioned electrical and optical signals to obtain the final discharge signal identification result. This leads to a decrease in data reliability and may even result in misjudgments, causing the early warning mechanism to fail and hindering the precise and efficient operation and maintenance of power transmission lines.
[0004] Therefore, how to accurately identify the discharge signals of transmission lines has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0005] This invention provides a method, apparatus, device, and storage medium for identifying discharge signals of transmission lines, in order to solve the technical problem that the current identification results of discharge signals have low reliability and cannot accurately identify discharge signals of transmission lines, thereby achieving precise and efficient operation and maintenance of transmission lines.
[0006] To address the aforementioned technical problems, embodiments of the present invention provide a method for identifying discharge signals in transmission lines, the method comprising:
[0007] Acquire vibration signal data of the target transmission line;
[0008] The vibration signal data is processed by removing interference caused by environmental factors to obtain the vibration signal data to be converted.
[0009] The vibration signal data to be converted is processed to obtain electrical signal data;
[0010] Based on preset screening conditions, discharge signal sample data is selected from a typical database, and the discharge signal sample data is integrated with the electrical signal data to obtain the data to be identified.
[0011] The data to be identified is input into the discharge signal identification model for processing to obtain the discharge signal identification result. The processing is designed to extract each discharge feature of the data to be identified and analyze the overlap of each discharge feature. The analysis results determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode. The discharge signal identification model is updated based on the model parameters.
[0012] As one preferred embodiment, the step of removing interference caused by environmental factors from the vibration signal data to obtain the vibration signal data to be converted includes:
[0013] The vibration signal data is processed using short-time Fourier transform to perform time-frequency analysis, resulting in a spectrum.
[0014] Based on a preset energy threshold, the spectrum is processed, and the processed spectrum is subjected to inverse short-time Fourier transform to obtain the vibration signal data to be converted.
[0015] As one preferred embodiment, the process of processing the vibration signal data to be converted to obtain electrical signal data includes:
[0016] The vibration signal data to be converted is processed by a sensor to obtain an analog voltage signal;
[0017] The analog voltage signal is digitized using analog-to-digital conversion technology to obtain the electrical signal data.
[0018] As one preferred embodiment, the step of selecting discharge signal sample data from a typical database based on preset screening conditions includes:
[0019] The electrical signal data is subjected to deep feature extraction processing to obtain sample embedding vectors;
[0020] Using the sample embedding vector as the filtering condition, the discharge signal sample data is selected from a typical database using vectorized retrieval technology.
[0021] As a preferred embodiment, before integrating the discharge signal sample data with the electrical signal data, the discharge signal identification method for the transmission line further includes:
[0022] The discharge signal sample data and the electrical signal data are preprocessed, and the preprocessing steps include data cleaning, data standardization and data normalization.
[0023] As one preferred embodiment, the step of analyzing the overlap of each of the discharge characteristics to determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode based on the analysis results includes:
[0024] The discharge features are processed using the distribution divergence technique to obtain the overlap of each discharge feature;
[0025] The overlap of each of the discharge characteristics is weighted and analyzed to obtain the analysis results;
[0026] Based on the analysis results, the discharge signal identification model is adaptively optimized using the adaptive bimodal whale optimization algorithm to determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode.
[0027] As one preferred embodiment, after obtaining the discharge signal identification result, the discharge signal identification method for the transmission line further includes:
[0028] The discharge signal recognition results are processed using a natural language model to generate a natural language report, which is then sent to the corresponding client terminal for storage.
[0029] Another embodiment of the present invention provides a discharge signal identification device for transmission lines, the device comprising:
[0030] The acquisition module is used to acquire vibration signal data of the target transmission line;
[0031] The elimination module is used to remove interference caused by environmental factors from the vibration signal data to obtain the vibration signal data to be converted;
[0032] The processing module is used to process the vibration signal data to be converted to obtain electrical signal data;
[0033] An integration module is used to select discharge signal sample data from a typical database based on preset screening conditions, and integrate the discharge signal sample data with the electrical signal data to obtain the data to be identified.
[0034] The identification module is used to input the data to be identified into the discharge signal identification model for processing to obtain the discharge signal identification result. The processing is designed to extract each discharge feature of the data to be identified and analyze the overlap of each discharge feature. The analysis results determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode. The discharge signal identification model is updated based on the model parameters.
[0035] Another embodiment of the present invention provides a discharge signal identification device for transmission lines, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the discharge signal identification method for transmission lines as described above.
[0036] In another embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, the discharge signal identification method for transmission lines as described above is implemented.
[0037] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:
[0038] Compared with existing technologies, this invention addresses the problem of low data reliability and susceptibility to misjudgment caused by directly combining acoustic signals from UAV inspections with electrical / optical signals. It achieves accurate identification of transmission line discharge signals through end-to-end optimization: First, it acquires the vibration signal of the target transmission line, eliminating environmental factors such as wind noise and mechanical interference to ensure data purity, resulting in the vibration signal to be converted. Then, it converts this signal into an easily quantifiable and analyzable electrical signal, resolving the clutter caused by directly splicing acoustic signals with other signals. Subsequently, based on preset conditions such as the current line voltage level and operating conditions, it selects discharge signal samples matching the scenario from a typical database and integrates them with the converted electrical signal to form high-quality data to be identified, avoiding untargeted signal combination. Finally, the data to be identified is input into a discharge signal identification model. The model first extracts discharge features and analyzes feature overlap, then determines the parameters of the coarse exploration mode and the fine exploration mode based on the analysis results and updates the model, solving the problem of rigid identification by a fixed model. Ultimately, this improves data reliability, avoids misjudgment, and achieves accurate discharge signal identification, supporting precise operation and maintenance of transmission lines. Attached Figure Description
[0039] Figure 1 This is a flowchart illustrating a method for identifying discharge signals of transmission lines in one embodiment of the present invention.
[0040] Figure 2 This is a flowchart illustrating a method for identifying discharge signals of transmission lines according to another embodiment of the present invention.
[0041] Figure 3 This is a schematic diagram of the structure of a discharge signal identification device for a power transmission line in one embodiment of the present invention;
[0042] Figure 4 This is a schematic diagram of the structure of a discharge signal identification device for a power transmission line in one embodiment of the present invention;
[0043] Figure label:
[0044] Among them, 11 is the acquisition module; 12 is the rejection module; 13 is the processing module; 14 is the integration module; 15 is the identification module; 21 is the processor; and 22 is the memory. Detailed Implementation
[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0046] In the description of this invention, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0047] One embodiment of the present invention provides a method for identifying discharge signals in transmission lines. For details, please refer to [link to relevant documentation]. Figure 1 , Figure 1 The diagram shown is a flowchart illustrating a method for identifying discharge signals in transmission lines according to one embodiment of the present invention. The method includes:
[0048] S1: Acquire vibration signal data of the target transmission line;
[0049] S2: Remove interference caused by environmental factors from the vibration signal data to obtain the vibration signal data to be converted;
[0050] S3: Process the vibration signal data to be converted to obtain electrical signal data;
[0051] S4: Based on preset screening conditions, select discharge signal sample data from a typical database, integrate the discharge signal sample data with the electrical signal data to obtain the data to be identified;
[0052] S5: Input the data to be identified into the discharge signal identification model for processing to obtain the discharge signal identification result. The processing is designed to extract each discharge feature of the data to be identified and analyze the overlap of each discharge feature. The analysis results determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode. The discharge signal identification model is updated based on the model parameters.
[0053] Specifically, obtaining vibration signal data of a target transmission line requires consideration of the line's deployment environment, such as plains, mountains, or river crossings; monitoring needs; and signal characteristics. Generally, dedicated vibration sensors are installed at key locations along the transmission line to directly capture vibration signals from components such as conductors, insulators, and towers. For sections of the line with complex terrain that are difficult to access manually, drones equipped with lightweight vibration detection equipment are used to achieve mobile data acquisition. Fixed vibration monitoring stations are set up at substations, inspection base stations, or dedicated monitoring towers along the transmission line, combined with meteorological monitoring equipment, to achieve synchronous acquisition of vibration signals and environmental factors.
[0054] In step S2, interference caused by environmental factors is removed from the vibration signal data to obtain the vibration signal data to be converted. This includes: performing time-frequency analysis on the vibration signal data using short-time Fourier transform to obtain a spectrum; processing the spectrum based on a preset energy threshold; and performing inverse short-time Fourier transform on the processed spectrum to obtain the vibration signal data to be converted.
[0055] The specific process is as follows: Considering that the original vibration signal data exists in the time domain in the form of "time-amplitude", environmental interference, such as low-frequency continuous vibration caused by a breeze, external random noise, and other vibration signals related to the discharge of the transmission line are superimposed in the time domain. It is impossible to clearly separate the two from the amplitude change alone. For example, the low-frequency vibration of wind noise may mask the medium and high-frequency pulses of the discharge vibration, making it difficult to directly determine which are interferences and which are useful signals.
[0056] By using the Short Time Fourier Transform (STFT) with a sliding time window, the entire time-domain vibration signal is divided into multiple short-segment local signals. Fourier transform is performed on each short-segment signal, and finally a spectrum is generated with time as the horizontal axis, frequency as the vertical axis, and amplitude as the third dimension.
[0057] The preset energy threshold is not set randomly, but is determined based on historical monitoring data of the transmission line, typical energy characteristics of environmental interference, and known energy range of the target discharge signal. Usually, the energy of the time-frequency region of environmental interference will be lower than the threshold, while the energy of the time-frequency region of discharge-related vibration will be higher than the threshold.
[0058] Through this processing, all time-frequency points with energy below the threshold in the spectrum will be filtered out, and only the time-frequency regions with energy above the threshold and a high probability of containing useful discharge vibration signals will be retained, thus achieving the goal of removing environmental interference from the time-frequency domain.
[0059] Subsequently, the Inverse Short-Time Fourier Transform (ISTFT) is used to process the spectrum, transforming the filtered time-frequency domain data back into a time-amplitude time-domain vibration signal. This is because subsequent processing of the vibration signal needs to be based on the time domain signal. Therefore, ISTFT, through inverse operation, restores the continuous time-domain vibration signal from the threshold-filtered data that retains only useful time-frequency information. At this point, the time-domain signal has eliminated low-energy interference caused by environmental factors, retaining only the high-energy useful vibration components, i.e., the vibration signal data to be converted.
[0060] The vibration signal data to be converted is processed to obtain electrical signal data, including sensor conversion processing of the vibration signal data to be converted to obtain an analog voltage signal; and analog-to-digital conversion technology is used to digitize the analog voltage signal to obtain the electrical signal data.
[0061] The vibration signal data to be converted is essentially the physical quantity of mechanical vibration of the transmission line. This type of mechanical quantity cannot be directly identified and processed by subsequent digital equipment and signal analysis algorithms. It must be converted into a corresponding electrical quantity by a sensor. Therefore, vibration sensors adapted to the transmission line scenario, such as piezoelectric vibration sensors, are used. These sensors contain a core component (piezoelectric ceramic) that can convert mechanical vibration into an electrical signal. When the vibration signal to be converted is applied to the sensor, the sensor will output a continuous voltage signal proportional to the amplitude, frequency and other characteristics of the vibration.
[0062] Analog-to-digital conversion (ADC) technology is used to digitize continuous voltage signals to obtain electrical signal data. In this process, ADC mainly completes two steps: sampling and quantization. Sampling involves taking snapshots of the continuous analog voltage signal at fixed time intervals to obtain the instantaneous voltage value at each time point. Quantization, on the other hand, converts the continuous voltage value at each sampling point into a finite number of discrete digital quantities.
[0063] Based on preset screening conditions, discharge signal sample data is selected from a typical database. The discharge signal sample data is then integrated with the electrical signal data to obtain the data to be identified. In this process, deep feature extraction is performed on the electrical signal data to obtain a sample embedding vector. Using the sample embedding vector as the screening condition, vectorized retrieval technology is used to select the discharge signal sample data from the typical database.
[0064] Considering that the electrical signal data obtained after analog-to-digital conversion is essentially a discrete sequence of digital voltage values, this type of raw data contains a large amount of potential information related to discharge. However, it also suffers from high data dimensionality and a lot of redundant information. If the raw electrical signal data is directly used as the screening criterion, not only will the curse of dimensionality lead to extremely low subsequent retrieval efficiency, but the interference of redundant information will also prevent the accurate location of discharge samples matching the current scenario. Therefore, it is necessary to use deep feature extraction to screen the electrical signal data and condense these high-dimensional and complex features into a low-dimensional, highly recognizable vector form, namely, the sample embedding vector.
[0065] It should be noted that deep feature extraction usually relies on deep learning models, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs).
[0066] Using sample embedding vectors as the filtering criteria essentially transforms the comparison of raw data into a comparison of feature vector similarity. Vectorized retrieval technology calculates the similarity between the sample embedding vector of the current electrical signal and the embedding vectors of all discharge signal samples in a typical database, and quickly locates samples with similarity higher than a preset threshold, i.e., discharge signal sample data.
[0067] Preferably, machine learning-driven classification retrieval techniques can also be used to select discharge signal sample data from typical databases. However, such techniques are suitable for scenarios where samples in typical databases have already been clearly labeled with category tags.
[0068] It should also be noted that before integrating the discharge signal sample data and the electrical signal data, data preprocessing is required. The preprocessing steps include data cleaning, data standardization, and data normalization.
[0069] This preprocessing is to ensure that the data quality and numerical scale of the discharge signal sample data are consistent with those of the electrical signal data.
[0070] The data to be identified is input into the discharge signal identification model for processing to obtain the discharge signal identification result. The processing is designed to extract each discharge feature of the data to be identified and analyze the overlap of each discharge feature. The analysis results determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode. The discharge signal identification model is updated based on the model parameters.
[0071] Specifically, key discharge features are extracted from the data to be identified. These key discharge features include at least: the high-frequency pulse amplitude / frequency of the electrical signal, the discharge duration, and the energy percentage of the characteristic frequency band.
[0072] The discharge features are processed using the distribution divergence technique to obtain the overlap of each discharge feature. If the feature overlap is low, a coarse exploration mode is enabled and the model parameters are adjusted to fast screening; if the feature overlap is high, a fine exploration mode is enabled and the model parameters are adjusted to fine differentiation.
[0073] The coarse / fine exploration parameters determined in this study are fed back to the model to update the model's decision logic, allowing the model to gradually adapt to the discharge characteristics of different scenarios, avoiding the rigidity problem of a fixed model dealing with all scenarios, and thus improving the recognition accuracy in the long term.
[0074] Preferably, the overlap of each of the discharge features is analyzed to determine the model parameters corresponding to the coarse exploration mode and the fine exploration mode, including: processing the discharge features using the distribution divergence technique to obtain the overlap of each of the discharge features; performing a weighted analysis on the overlap of each of the discharge features to obtain the analysis result; and based on the analysis result, using the adaptive bimodal whale optimization algorithm to adaptively optimize the discharge signal recognition model to determine the model parameters corresponding to the coarse exploration mode and the fine exploration mode.
[0075] Specifically, the core function of distribution divergence techniques, such as KL divergence, JS divergence, and Wasserstein distance, is to quantitatively measure the degree of distribution difference between different discharge characteristics, and then inversely deduce the degree of overlap of discharge characteristics.
[0076] Considering that different discharge characteristics contribute differently to discharge signal identification, the overlap of all features cannot be treated equally. Each feature needs to be assigned a corresponding weight, which can be determined through domain knowledge, feature importance assessment models, etc.
[0077] The adaptive bimodal whale optimization algorithm adjusts its search strategy based on the analysis results from the second step. If the analysis results show high feature overlap and difficulty in differentiation, it indicates that the model needs a wider range of parameter searches to find parameters that fit the complex features. In this case, the search step size in the coarse exploration mode will be increased to expand the search space, while the step size in the fine exploration mode will be reduced to avoid missing the optimal solution. If the analysis results show low feature overlap and difficulty in differentiation, it indicates that the model does not need to search a large range. In this case, the search space in the coarse exploration mode will be reduced to improve efficiency, and the step size in the fine exploration mode can also be appropriately increased to accelerate convergence.
[0078] Finally, through iterative optimization of the algorithm, the optimal model parameters corresponding to the coarse exploration mode and the fine exploration mode can be determined respectively, and the discharge signal identification model can be updated based on the model parameters.
[0079] After obtaining the discharge signal identification result, the discharge signal identification method of the transmission line further includes: processing the discharge signal identification result using a natural language model to generate a natural language report, and sending the natural language report to the corresponding client terminal for storage.
[0080] Another embodiment of the present invention provides a method for identifying discharge signals of transmission lines. For details, please refer to [link to relevant documentation]. Figure 2 , Figure 2 The diagram shown is a flowchart illustrating a method for identifying discharge signals in transmission lines according to one embodiment of the present invention. The method includes:
[0081] The acquired high-voltage discharge signal dataset is input into the constructed first classification model for processing to obtain the first recognition and classification accuracy. Based on the first recognition and classification accuracy, the high-voltage discharge signal dataset is processed using adaptive sample synthesis technology to obtain a synthetic signal dataset. The first classification model is then optimized using an adaptive bimodal whale optimization algorithm to obtain adaptive optimization parameters. The optimization process is configured to determine the optimization mode based on spatial distribution entropy and convergence stagnation index, with the optimization modes being coarse exploration mode and fine exploration mode. The first classification model is updated based on the adaptive optimization parameters to obtain a second classification model. The synthetic signal dataset is then input into the second classification model for processing to obtain the classification and recognition results.
[0082] Specifically, a 20kV high-voltage discharge signal was collected on a discharge test bench using a sound sensor and a signal acquisition card at a sampling frequency of 48kHz. By playing pre-recorded noises such as wind and rain, a dataset of seven types of discharge acoustic signals mixed with noise was obtained.
[0083] The obtained dataset is imported and labeled. The sample data of the noisy discharge signal dataset is truncated according to a certain ratio to meet the data imbalance problem of few fault signal samples encountered in the actual acquisition of transmission line discharge signals. The training set and test set are divided under the standard of ensuring that the number of test sets of the seven datasets are the same.
[0084] A CNN (Convolutional Neural Network) is constructed, with its core structure consisting of an input layer, convolutional layers, pooling layers, flattening layers, fully connected layers, and an output layer. Max pooling is used, and a normalization layer is added between each convolutional and pooling layer to prevent gradient explosion or vanishing. The input layer size is 20×20, and the four convolutional layers have sizes of 32×32, 16×16, 8×8, and 4×4 respectively. The number of convolutional kernels is determined by the variables n1, n2, n3, and n4, with initial values of 20, 64, 64, and 64 respectively. Each convolutional layer is followed by a normalization layer and a max pooling layer with a size of 2×2 and a stride of 2. The flattening layer flattens the multi-dimensional feature map into a one-dimensional vector, preparing for the fully connected layer. The fully connected layer has 64 neurons. The final output layer contains 7 neurons, corresponding to 7 classification tasks.
[0085] An imbalanced dataset of seven signal classes was imported into the model for training and classification, and the recognition and classification accuracy of the seven signal classes was obtained.
[0086] The adaptive sample synthesis method (AF-ADASYN) optimized based on accuracy feedback is used. Guided by the gap between the current number of samples and the target number of samples and the recognition difficulty, minority class samples are synthesized and sampled. The adaptive synthesis sampling method is improved by combining the model training results, prioritizing the supplementation of key difficult samples, and specifically improving the model's recognition ability under complex working conditions.
[0087] The Adaptive Bimodal Whale Optimization (AD-WOA) algorithm is used to optimize the four hyperparameters n1, n2, n3, and n4 of the CNN convolutional neural network. The AD-WOA algorithm achieves adaptive optimization through three techniques: dynamic mode switching, hybrid position update, and parallel elite optimization.
[0088] The dataset synthesized by AF-ADASYN was imported into the AD-WOA optimized CNN convolutional neural network for final pattern recognition. The classification results of the dataset before and after AF-ADASYN synthesis were compared, and the optimized model was compared with other models to obtain the improvement rate of the model's pattern recognition accuracy.
[0089] Principal component analysis (PCA) is a commonly used data dimensionality reduction technique. In this method, PCA is used to visualize the pattern recognition results and draw scatter plots of sample distribution before and after pattern recognition. This allows for a direct observation of the distribution of samples of different categories and the classification effect of the model, effectively verifying the effectiveness and robustness of the model and ensuring that the model can accurately distinguish different types of discharge signals.
[0090] Another embodiment of the present invention provides a discharge signal identification device for transmission lines. For details, please refer to [link to relevant documentation]. Figure 3 , Figure 3 The diagram shown illustrates the structure of a discharge signal identification device for a transmission line according to one embodiment of the present invention. The device includes:
[0091] The acquisition module 11 is used to acquire vibration signal data of the target transmission line;
[0092] The elimination module 12 is used to remove interference caused by environmental factors from the vibration signal data to obtain vibration signal data to be converted.
[0093] Processing module 13 is used to process the vibration signal data to be converted to obtain electrical signal data;
[0094] The integration module 14 is used to select discharge signal sample data from a typical database based on preset screening conditions, and integrate the discharge signal sample data with the electrical signal data to obtain the data to be identified.
[0095] The identification module 15 is used to input the data to be identified into the discharge signal identification model for processing to obtain the discharge signal identification result. The processing is designed to extract each discharge feature of the data to be identified and analyze the overlap of each discharge feature. The analysis results are used to determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode. The discharge signal identification model is updated based on the model parameters.
[0096] See Figure 4 This is a schematic diagram of the structure of a discharge signal identification device for transmission lines provided in an embodiment of the present invention. The discharge signal identification device for transmission lines provided in this embodiment includes a processor 21, a memory 22, and a computer program stored in the memory 22 and configured to be executed by the processor 21. When the processor 21 executes the computer program, it implements the steps described in the above embodiment of the discharge signal identification method for transmission lines. Figure 1 The steps S1 to S5 described above; or, when the processor 21 executes the computer program, it implements the functions of each module in the above-described device embodiments, such as the acquisition module 11.
[0097] For example, the computer program can be divided into one or more modules, which are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the discharge signal identification device of the transmission line. For example, the computer program can be divided into an acquisition module 11, a rejection module 12, a processing module 13, etc., with the specific functions of each module as follows:
[0098] The acquisition module 11 is used to acquire vibration signal data of the target transmission line;
[0099] The elimination module 12 is used to remove interference caused by environmental factors from the vibration signal data to obtain vibration signal data to be converted.
[0100] Processing module 13 is used to process the vibration signal data to be converted to obtain electrical signal data;
[0101] The integration module 14 is used to select discharge signal sample data from a typical database based on preset screening conditions, and integrate the discharge signal sample data with the electrical signal data to obtain the data to be identified.
[0102] The identification module 15 is used to input the data to be identified into the discharge signal identification model for processing to obtain the discharge signal identification result. The processing is designed to extract each discharge feature of the data to be identified and analyze the overlap of each discharge feature. The analysis results are used to determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode. The discharge signal identification model is updated based on the model parameters.
[0103] The discharge signal identification device for the transmission line may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art will understand that the schematic diagram is merely an example of a discharge signal identification device for a transmission line and does not constitute a limitation on the device. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, the discharge signal identification device for a transmission line may also include input / output devices, network access devices, buses, etc.
[0104] The processor 21 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor 21 is the control center of the discharge signal identification equipment for the transmission line, connecting various parts of the discharge signal identification equipment throughout the transmission line via various interfaces and lines.
[0105] The memory 22 can be used to store the computer program and / or modules. The processor 21 implements various functions of the discharge signal identification device of the transmission line by running or executing the computer program and / or modules stored in the memory 22 and calling the data stored in the memory 22. The memory 22 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 22 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
[0106] If the module integrated into the discharge signal identification device of the transmission line is implemented as 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, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.
[0107] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.
[0108] Accordingly, embodiments of the present invention provide a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform steps in the discharge signal identification method for transmission lines as described in the above embodiments, for example... Figure 1 Steps S1 to S5 as described above.
[0109] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.
Claims
1. A method of identifying a discharge signal of a power transmission line, characterized by, The method comprises the following steps: acquiring vibration signal data of a target power transmission line; removing interference caused by environmental factors from the vibration signal data to obtain to-be-converted vibration signal data; processing the to-be-converted vibration signal data to obtain electrical signal data; selecting discharge signal sample data from a typical database based on a preset screening condition, integrating the discharge signal sample data with the electrical signal data, and obtaining to-be-identified data; inputting the to-be-identified data into a discharge signal identification model for processing to obtain a discharge signal identification result, wherein the processing process is designed to extract each discharge feature of the to-be-identified data and analyze the overlap degree of each discharge feature, and the analysis result is used to determine model parameters corresponding to a coarse exploration mode and model parameters corresponding to a fine exploration mode of the model, and the discharge signal identification model is updated based on the model parameters.
2. The method of claim 1, wherein the discharge signal of the power transmission line is identified by the steps of: The step of removing interference caused by environmental factors from the vibration signal data to obtain to-be-converted vibration signal data comprises the following steps: performing time-frequency analysis processing on the vibration signal data by using short-time Fourier transform to obtain a frequency spectrum; processing the frequency spectrum based on a preset energy threshold, performing inverse short-time Fourier transform processing on the processed frequency spectrum, and obtaining the to-be-converted vibration signal data.
3. The method for identifying discharge signals of transmission lines as described in claim 1, characterized in that, The step of processing the to-be-converted vibration signal data to obtain electrical signal data comprises the following steps: performing sensor conversion processing on the to-be-converted vibration signal data to obtain an analog voltage signal; digitizing the analog voltage signal by using an analog-to-digital conversion technology to obtain the electrical signal data.
4. The method of claim 1, wherein the discharge signal of the power transmission line is identified by the steps of: The step of selecting discharge signal sample data from a typical database based on a preset screening condition comprises the following steps: performing deep feature extraction processing on the electrical signal data to obtain a sample embedding vector; using the sample embedding vector as the screening condition and using vectorization retrieval technology to select the discharge signal sample data from the typical database.
5. The method of claim 1, wherein the step of identifying the discharge signal of the power transmission line is characterized by, Before the step of integrating the discharge signal sample data with the electrical signal data, the discharge signal identification method for the power transmission line further comprises the following steps: performing data preprocessing on the discharge signal sample data and the electrical signal data, and the preprocessing steps comprise data cleaning, data standardization, and data normalization.
6. The method of claim 1, wherein the step of identifying the discharge signal of the power transmission line is characterized by, The step of analyzing the overlap degree of each discharge feature to determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode based on the analysis result comprises the following steps: processing the discharge features by using distribution divergence technology to obtain the overlap degree of each discharge feature; performing weighted analysis on the overlap degree of each discharge feature to obtain the analysis result; based on the analysis result, performing adaptive optimization processing on the discharge signal identification model by using an adaptive bimodal whale optimization algorithm to determine the model parameters corresponding to the coarse exploration mode and the model parameters corresponding to the fine exploration mode.
7. The method of claim 1, wherein the step of identifying the discharge signal of the power transmission line is characterized by, After obtaining the discharge signal identification result, the discharge signal identification method for the power transmission line further comprises the following steps: The natural language model is used to process the discharge signal recognition result, generate a natural language report, and send the natural language report to a corresponding client terminal for storage.
8. A discharge signal recognition device for a power transmission line, characterized by comprising: The method comprises the steps of: An acquisition module is configured to acquire vibration signal data of a target power transmission line. A rejection module is configured to reject interference caused by environmental factors from the vibration signal data to obtain to-be-converted vibration signal data. A processing module is configured to process the to-be-converted vibration signal data to obtain electric signal data. An integration module is configured to select discharge signal sample data from a typical database based on a preset screening condition, integrate the discharge signal sample data with the electric signal data, and obtain to-be-identified data. An identification module is configured to input the to-be-identified data into a discharge signal identification model for processing to obtain a discharge signal recognition result, wherein the processing process is designed to extract each discharge feature of the to-be-identified data and analyze the overlap degree of each discharge feature to determine model parameters corresponding to a coarse exploration mode and model parameters corresponding to a fine exploration mode of the model, and update the discharge signal identification model based on the model parameters.
9. A discharge signal recognition device for a power transmission line, characterized by comprising: The computer readable storage medium stores a computer program, wherein when the computer program is executed by the device where the computer readable storage medium is located, the power transmission line discharge signal identification method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, wherein when the computer program is executed by the device where the computer readable storage medium is located, the power transmission line discharge signal identification method according to any one of claims 1 to 7 is implemented.