Electric shock identification and protection method, system and equipment based on wavelet transformation, medium and product

Abstract

The invention relates to the technical field of electric power systems, and discloses an electric shock identification and protection method, system and device based on wavelet transformation, a medium and a product. According to the method, wavelet decomposition is carried out on current and voltage signals during electric shock, low-frequency and high-frequency transient characteristics at the moment of electric shock can be captured; and performing feature extraction on the low-frequency component and the high-frequency component obtained by wavelet decomposition, inputting the extracted signal component features into a pre-trained electric shock diagnosis model for electric shock diagnosis, and performing electric shock diagnosis on the condition that the electric shock diagnosis result is judged to be an electric shock event. According to the electric shock protection method and device, the protection action instruction is generated and issued to the leakage protector electrically connected with the target circuit, and the leakage protector is triggered to execute the protection action and cut off the power supply of the target circuit when the leakage protector receives the protection action instruction and the residual current meets the preset condition, so that the accuracy and reliability of electric shock protection are improved.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method, system, device, medium and product for electric shock identification and protection based on wavelet transform. Background Technology

[0002] In live-line operation skills training in the power industry, personnel safety is always the primary technical challenge. Statistics show that the main causes of electric shock accidents include equipment insulation failure and accidental contact with exposed conductors.

[0003] From a technical perspective, live-line training scenarios have strong interference characteristics: on the one hand, the normal contact current (such as induced current) generated during the operation of trainees overlaps with the frequency band of the actual electric shock signal, which can easily trigger malfunctions; on the other hand, complex working conditions (such as the use of high-frequency tools and cross contact of multiple conductors) lead to nonlinear distortion of current characteristics.

[0004] Currently, electric shock identification technology struggles to identify faults and execute protective actions quickly and promptly, making it difficult to meet the real-time, anti-interference, and accuracy requirements for electric shock identification in live-line training scenarios. Summary of the Invention

[0005] In view of this, the present invention provides a wavelet transform-based method, system, device, medium and product for electric shock identification and protection, which solves the technical problem that electric shock identification technology is difficult to perform timely and rapid fault identification and execution of protection actions, resulting in difficulty in meeting the requirements of real-time performance, anti-interference and accuracy of electric shock identification in live training scenarios.

[0006] The first aspect of this invention provides a method for electric shock identification and protection based on wavelet transform, comprising:

[0007] Acquire the current and voltage signals of the target circuit when an electric shock occurs;

[0008] Wavelet decomposition is performed on the current and voltage signals to obtain low-frequency and high-frequency components at multiple levels.

[0009] Features are extracted from the low-frequency component and the high-frequency component respectively to obtain signal component features;

[0010] The signal component features are input into a pre-trained electric shock diagnosis model, which then outputs the electric shock diagnosis result.

[0011] If the electric shock diagnosis result indicates an electric shock event, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. When the RCD receives the protection action command and the residual current meets the preset conditions, the RCD is triggered to perform a protection action and cut off the power supply to the target circuit.

[0012] Optionally, the step of performing wavelet decomposition on the current and voltage signals to obtain multiple levels of low-frequency and high-frequency components further includes:

[0013] The current and voltage signals are preprocessed; wherein the preprocessing includes at least one of downsampling and wavelet denoising.

[0014] Optionally, the wavelet decomposition of the current and voltage signals to obtain multiple levels of low-frequency and high-frequency components includes:

[0015] Determine the wavelet basis functions and the number of wavelet decomposition layers;

[0016] Using a defined wavelet basis function and a number of decomposition levels, the current and voltage signals are decomposed using wavelet decomposition to obtain low-frequency and high-frequency components at multiple levels.

[0017] Optionally, the signal component features include low-frequency component features and high-frequency component features; the step of extracting features from the low-frequency component and the high-frequency component respectively to obtain the signal component features includes:

[0018] The mean and variance of the low-frequency component and the high-frequency component features are extracted using a statistical feature extraction method.

[0019] The low-frequency component features and the high-frequency component features are converted into low-frequency component frequency domain features and high-frequency component frequency domain features, respectively. Then, using the sliding window method, the spectral distribution features and second harmonic amplitudes of the low-frequency component frequency domain features and the high-frequency component frequency domain features are extracted, respectively.

[0020] The mean, variance, spectral distribution characteristics, and second harmonic amplitude of the low-frequency components are integrated into the low-frequency component characteristics, and the mean, variance, spectral distribution characteristics, and second harmonic amplitude of the high-frequency components are integrated into the high-frequency component characteristics.

[0021] Optionally, the training process of the electric shock diagnostic model includes:

[0022] Collect electric shock sample data and non-electric shock sample data to construct a sample dataset; wherein, the electric shock sample data includes current and voltage signal sample data when electric shock occurs, and the non-electric shock sample data includes current and voltage signal sample data under normal conditions;

[0023] Wavelet decomposition is performed on the current and voltage signal sample data in the sample dataset to obtain low-frequency component samples and high-frequency component samples.

[0024] Feature extraction is performed on the low-frequency component samples and the high-frequency component samples respectively to obtain the sample signal component features;

[0025] The sample signal component features are classified according to the electric shock sample data and the non-electric shock sample data, and the classified sample signal component features are labeled with whether or not electric shock has occurred.

[0026] A training dataset is constructed based on the characteristics of the sample signal components and the label indicating whether an electric shock has occurred.

[0027] The electric shock diagnosis model is obtained by training a preset machine learning model using the training dataset; wherein the preset machine learning model is a GRU recurrent neural network or a PNN probabilistic neural network.

[0028] Optionally, the step of generating a protection action command and sending it to a residual current device (RCD) electrically connected to the target circuit when the electric shock diagnosis result indicates an electric shock event, and triggering the RCD to perform a protection action and cut off the power supply to the target circuit when the RCD receives the protection action command and the residual current meets a preset condition, includes:

[0029] When the electric shock diagnosis result is determined to be an electric shock event, a protective action command is generated and sent to the leakage current protection device electrically connected to the target circuit.

[0030] After receiving the protection action command, the residual current device detects whether the residual current of the target circuit exceeds the preset safe current threshold. If the residual current of the target circuit exceeds the preset safe current threshold, it performs the protection action of cutting off the power supply to the target circuit.

[0031] Secondly, the present invention also provides an electric shock identification and protection system based on wavelet transform, comprising:

[0032] The signal acquisition module is used to acquire the current and voltage signals of the target circuit when an electric shock occurs;

[0033] The wavelet decomposition module is used to perform wavelet decomposition on the current and voltage signals to obtain low-frequency and high-frequency components at multiple levels.

[0034] The feature extraction module is used to extract features from the low-frequency component and the high-frequency component respectively to obtain signal component features;

[0035] The electric shock diagnosis module is used to input the signal component features into a pre-trained electric shock diagnosis model, so that the electric shock diagnosis model outputs the electric shock diagnosis result;

[0036] The electric shock protection module is used to generate a protection action command and send it to the leakage current protector electrically connected to the target circuit when the electric shock diagnosis result determines that an electric shock event has occurred. When the leakage current protector receives the protection action command and the residual current meets the preset conditions, it triggers the leakage current protector to perform a protection action and cut off the power supply to the target circuit.

[0037] Thirdly, the present invention also provides an electronic device, the electronic device including a memory and a processor, the memory storing a computer program, the computer program being executed by the processor causing the processor to perform the steps of the wavelet transform-based electric shock identification and protection method as described in the first aspect.

[0038] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of the wavelet transform-based electric shock identification and protection method as described in the first aspect.

[0039] Fifthly, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein, when the program instructions are executed by a computer, the computer performs the steps of the wavelet transform-based electric shock identification and protection method as described in the first aspect.

[0040] As can be seen from the above technical solution, this invention, by performing wavelet decomposition on the current and voltage signals during electric shock, can capture the low-frequency and high-frequency transient features at the moment of electric shock. Features are extracted from the low-frequency and high-frequency components obtained from the wavelet decomposition, and the extracted signal features are input into a pre-trained electric shock diagnosis model for electric shock diagnosis. This enables rapid response to electric shock events and meets the requirements of real-time protection. If the electric shock diagnosis result determines that an electric shock event has occurred, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. Upon receiving the protection action command and if the residual current meets preset conditions, the RCD is triggered to execute a protection action, cutting off the power supply to the target circuit. This ensures rapid execution of the protection action. Furthermore, the combined effect of the protection action command and the residual current effectively avoids false or non-operational situations that may result from a single condition judgment, improving the accuracy and reliability of electric shock protection. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 An application environment diagram of an electric shock identification and protection method based on wavelet transform provided in an embodiment of the present invention;

[0043] Figure 2 A flowchart illustrating an electric shock identification and protection method based on wavelet transform, provided as an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of the structure of an electric shock identification and protection system based on wavelet transform provided in an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0046] 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 are within the scope of protection of the present invention.

[0047] The wavelet transform-based electric shock identification and protection method provided in this application can be applied to, for example... Figure 1In the application environment shown, terminal 101 communicates with server 102 via a network. A data storage system can store the data that server 102 needs to process. The data storage system can be integrated onto server 102 or placed in the cloud or on another network server. Terminal 101 or server 102 acquires the current and voltage signals of the target circuit when an electric shock occurs; performs wavelet decomposition on the current and voltage signals to obtain multiple levels of low-frequency and high-frequency components; extracts features from the low-frequency and high-frequency components respectively to obtain signal component features; inputs the signal component features into a pre-trained electric shock diagnosis model, causing the electric shock diagnosis model to output an electric shock diagnosis result; if the electric shock diagnosis result indicates an electric shock event, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit; and when the RCD receives the protection action command and the residual current meets preset conditions, the RCD is triggered to execute a protection action, cutting off the power supply to the target circuit.

[0048] Terminal 101 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets.

[0049] Server 102 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0050] like Figure 2 As shown, this application provides a wavelet transform-based method for electric shock identification and protection, which is applied to... Figure 1 Taking terminal 101 or server 102 as an example, the explanation includes the following steps S1 to S5. Wherein:

[0051] Step S1: Obtain the current and voltage signals of the target circuit when an electric shock occurs.

[0052] The target circuit is a specific circuit loop involved in live-line operation skills training in the power industry. This circuit may include various electrical devices and connecting lines. In actual operation scenarios, current and voltage signal data are collected in real time through devices such as current transformers and voltage sensors installed at key nodes of the target circuit. These sensors must have high precision and fast response characteristics to ensure that the collected signals accurately reflect the real-time status of the circuit and provide a reliable data foundation for subsequent analysis.

[0053] To make the current and voltage signals more accurate, preprocessing is performed on them; the preprocessing includes at least one of downsampling and wavelet denoising.

[0054] Downsampling reduces the amount of data by lowering the signal sampling rate while retaining key information, preventing increased computational complexity and reduced efficiency in subsequent processing due to excessive data volume. Wavelet denoising utilizes the multi-scale analysis characteristics of wavelet transform to decompose the signal into different frequency sub-bands. Thresholding or coefficient shrinking is applied to the noise-dominated high-frequency sub-bands to effectively suppress noise interference, improve the signal-to-noise ratio, and make the acquired current and voltage signals purer and more accurate.

[0055] Step S2: Perform wavelet decomposition on the current and voltage signals to obtain low-frequency and high-frequency components at multiple levels.

[0056] By decomposing the signal into wavelet coefficients at different scales and locations, multi-resolution analysis of the signal is achieved. Unlike the traditional Fourier transform, the wavelet transform can not only analyze the frequency components of the signal, but also capture the local features of the signal, making it particularly suitable for processing non-stationary signals (such as electric shock signals).

[0057] The core of wavelet transform is the wavelet function (or mother wavelet), which generates a series of sub-wavelets through scaling and translation operations. These sub-wavelets can cover different frequency ranges and time positions, thus enabling localized signal analysis. The mathematical expression for wavelet transform is:

[0058]

[0059] In the formula, f(t) is the signal to be analyzed. It is the wavelet function after scaling and translation, where a is the scaling factor and b is the translation factor.

[0060] By selecting different wavelet functions, optimized analysis can be performed on various signal characteristics. The level of wavelet decomposition determines the precision of the signal analysis. Generally, more decomposition levels result in more detailed analysis of the high-frequency components of the signal, but also increase computational load and complexity. In practical applications, the decomposition level should be selected appropriately based on signal characteristics and analysis requirements. For example, for non-stationary signals like electric shock signals that contain transient characteristics, appropriately increasing the decomposition level helps capture the details of abrupt changes in the signal, but excessive decomposition should be avoided to prevent information redundancy.

[0061] After determining the wavelet basis functions and the number of wavelet decomposition levels, the acquired current and voltage signals are decomposed using these parameters. During the decomposition process, the signals are progressively decomposed into low-frequency and high-frequency components at different levels. The low-frequency components reflect the overall trend of the signal, while the high-frequency components contain the detailed features of the signal.

[0062] Step S3: Extract features from the low-frequency component and the high-frequency component respectively to obtain signal component features.

[0063] Step S4: Input the signal component features into the pre-trained electric shock diagnosis model, so that the electric shock diagnosis model outputs the electric shock diagnosis result.

[0064] The training process of the electric shock diagnosis model includes:

[0065] Step S41: Collect electric shock sample data and non-electric shock sample data to construct a sample dataset; wherein, electric shock sample data includes current and voltage signal sample data when electric shock occurs, and non-electric shock sample data includes current and voltage signal sample data under normal conditions;

[0066] Step S42: Perform wavelet decomposition on the current and voltage signal sample data in the sample dataset to obtain low-frequency component samples and high-frequency component samples;

[0067] Step S43: Extract features from the low-frequency component samples and the high-frequency component samples respectively to obtain the sample signal component features;

[0068] Step S44: Classify the sample signal component features according to the electric shock sample data and the non-electric shock sample data, and assign a label to the classified sample signal component features indicating whether or not electric shock has occurred.

[0069] Among them, the sample signal component features of electric shock sample data are labeled as electric shock, and the sample signal component features of non-electric shock sample data are labeled as non-electric shock.

[0070] Step S45: Construct a training dataset based on the characteristics of the sample signal components and the label indicating whether an electric shock occurred;

[0071] Step S46: Train the preset machine learning model using the training dataset to obtain the electric shock diagnosis model; wherein the preset machine learning model is a GRU recurrent neural network or a PNN probabilistic neural network.

[0072] Among them, the GRU (Gated Recurrent Unit) recurrent neural network controls the flow of information by introducing reset and update gates, effectively solving the gradient vanishing and gradient exploding problems existing in traditional recurrent neural networks, and can better capture long-term dependencies in sequence data. Applying the GRU recurrent neural network in electric shock diagnosis models can improve the model's ability to recognize the temporal characteristics of electric shock signals, enabling it to learn the temporal characteristics of electric shock signals and thus achieve high-precision electric shock identification.

[0073] PNN (Probabilistic Neural Network) is a type of neural network based on a Bayesian classifier. It can quickly learn the distribution characteristics of sample data and classify new input data. The PNN model classifies and identifies the feature vectors of electric shock signals by constructing a probability density function, exhibiting high recognition rate and noise resistance.

[0074] Step S5: If the electric shock diagnosis result indicates an electric shock event, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. When the RCD receives the protection action command and the residual current meets the preset conditions, the RCD is triggered to perform the protection action and cut off the power supply to the target circuit.

[0075] It should be noted that, in this embodiment, wavelet decomposition of the current and voltage signals during electric shock is performed to capture the low-frequency and high-frequency transient features at the moment of electric shock. Features are extracted from the low-frequency and high-frequency components obtained by wavelet decomposition, and the extracted signal component features are input into a pre-trained electric shock diagnosis model for electric shock diagnosis. This enables rapid response to electric shock events and meets the requirements of real-time protection. When the electric shock diagnosis result determines that an electric shock event has occurred, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. When the RCD receives the protection action command and the residual current meets the preset conditions, it is triggered to execute the protection action and cut off the power supply to the target circuit. This ensures rapid execution of the protection action. Through the combined effect of the protection action command and the residual current, false or non-operational situations that may occur due to a single condition judgment are effectively avoided, improving the accuracy and reliability of electric shock protection.

[0076] In some embodiments, wavelet decomposition is performed on the current and voltage signals to obtain multiple levels of low-frequency and high-frequency components, including:

[0077] Step S201: Determine the wavelet basis functions and the number of wavelet decomposition layers.

[0078] The wavelet basis functions can be selected from Daubechies wavelets, Morlet wavelets, etc., and the number of wavelet decomposition layers can be 3 to 5, depending on the signal characteristics and actual needs. For example, for more complex electric shock signals, it may be necessary to decompose to five layers to fully capture its high-frequency details.

[0079] Step S202: Using the determined wavelet basis function and decomposition level, perform wavelet decomposition on the current and voltage signals to obtain low-frequency and high-frequency components at multiple levels.

[0080] After determining the wavelet basis functions and the number of decomposition levels, wavelet decomposition is performed on the acquired current and voltage signals. During the decomposition process, based on the characteristics of the wavelet basis functions and the set number of decomposition levels, the signal is gradually analyzed into low-frequency and high-frequency components at different levels. The low-frequency components mainly present the overall trend of the signal, reflecting the stable characteristics of the signal over a longer time scale; while the high-frequency components focus on the detailed features of the signal, containing information about abrupt changes and fluctuations in the signal over a short period of time.

[0081] In some embodiments, the signal component features include low-frequency component features and high-frequency component features; features are extracted from the low-frequency component and the high-frequency component respectively to obtain the signal component features, including:

[0082] Step S301: Use statistical feature extraction methods to extract the mean and variance of the low-frequency component and high-frequency component features respectively.

[0083] The statistical feature extraction method is used to calculate the mean and variance of the low-frequency and high-frequency components, which can help to grasp the overall level and fluctuation of the signal. The mean reflects the average intensity of the signal over a certain period of time and is of great reference value for judging the overall energy level of the electric shock signal; the variance reflects the dispersion of the signal data. A larger variance in the high-frequency components may indicate that there are more transient changes in the signal, which is related to the sudden changes that may occur during electric shock.

[0084] Step S302: Convert the low-frequency component features and high-frequency component features into low-frequency component frequency domain features and high-frequency component frequency domain features respectively. Combine the sliding window method to extract the spectral distribution features and second harmonic amplitude of the low-frequency component frequency domain features and high-frequency component frequency domain features respectively.

[0085] Transforming the low-frequency and high-frequency component characteristics from the time domain to the frequency domain allows for a more in-depth analysis of the signal's frequency characteristics. Frequency domain analysis methods, such as Fourier transform, can convert time-domain signals into frequency-domain signals, thereby obtaining the signal's spectral distribution characteristics.

[0086] The spectral distribution characteristics reflect the energy distribution of a signal at different frequencies. For electric shock signals, certain specific frequency components may be closely related to the electric shock event.

[0087] Furthermore, by combining the sliding window method, the second harmonic amplitudes of both the low-frequency and high-frequency components can be extracted. As an important indicator of signal frequency characteristics, the second harmonic amplitude provides information about the signal's nonlinear characteristics, helping to more accurately identify electric shock signals.

[0088] Step S303: Integrate the mean, variance, spectral distribution characteristics and second harmonic amplitude of the low-frequency components into low-frequency component characteristics, and integrate the mean, variance, spectral distribution characteristics and second harmonic amplitude of the high-frequency components into high-frequency component characteristics.

[0089] In some embodiments, if the electric shock diagnosis result indicates an electric shock event, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. Upon receiving the protection action command and if the residual current meets preset conditions, the RCD is triggered to perform a protection action, cutting off the power supply to the target circuit, including:

[0090] Step S501: When the electric shock diagnosis result is determined to be an electric shock event, a protection action command is generated and sent to the leakage current protection device electrically connected to the target circuit.

[0091] Step S502: After receiving the protection action command, the leakage current protector detects whether the residual current of the target circuit exceeds the preset safe current threshold. If the residual current of the target circuit exceeds the preset safe current threshold, the protection action of cutting off the power supply of the target circuit is executed.

[0092] A residual current device (RCD) is a device that detects and judges leakage current but does not have the function of cutting off or connecting the main circuit. An RCD consists of a zero-sequence current transformer, a trip unit, and auxiliary contacts for output signals. It can be used in conjunction with high-current automatic switches as the main protection for low-voltage power grids or for monitoring and protecting against leakage, grounding, or insulation issues in main circuits.

[0093] When there is leakage current in the main circuit, since the auxiliary contacts and the main circuit switch's trip unit are connected in series to form a circuit, the auxiliary contacts activate the trip unit, disconnecting the air switch, AC contactor, etc., causing them to trip and disconnecting the main circuit. The auxiliary contacts can also activate audible and visual signal devices to issue leakage alarm signals, reflecting the insulation status of the line.

[0094] Residual current devices (RCDs) are highly sensitive and fast-acting in responding to electric shocks and protecting against leakage. During normal operation, the residual current of the system is almost zero, so its operating setting value can be set very small (generally in the mA range). When a person is electrocuted or the equipment casing becomes energized, a large residual current appears.

[0095] In this application, the residual current of the target circuit is monitored in real time by a zero-sequence current transformer. If no leakage occurs, the vector sum of the currents flowing through the phase line and the neutral line is equal to zero. When the residual current device receives the protection action command, if leakage occurs, the vector sum of the currents in the phase line and the neutral line is not equal to zero, and the residual current is detected to exceed the preset safe current threshold. This means that the electric shock situation actually exists and has reached the level that requires the power supply to be cut off to ensure safety. At this time, the residual current device will immediately execute the protection action of cutting off the power supply of the target circuit, thereby effectively preventing the further expansion of the electric shock accident and ensuring the safety of personnel and equipment.

[0096] Therefore, this dual-judgment mechanism fully utilizes the accurate identification capability of the electric shock diagnostic model for electric shock signals, and combines it with the real-time monitoring function of the residual current device (RCD) for residual current, greatly improving the reliability and timeliness of electric shock protection. In practical applications, this dual-judgment mechanism can effectively avoid misjudgments or omissions that may occur due to a single judgment condition.

[0097] Based on the same inventive concept, this application also provides a wavelet transform-based electric shock identification and protection system for implementing the wavelet transform-based electric shock identification and protection method described above.

[0098] The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more wavelet transform-based electric shock identification and protection system embodiments provided below can be found in the limitations of the wavelet transform-based electric shock identification and protection method described above, and will not be repeated here.

[0099] like Figure 3 As shown, this application also provides a wavelet transform-based electric shock identification and protection system, including:

[0100] The signal acquisition module 100 is used to acquire the current and voltage signals of the target circuit when an electric shock occurs;

[0101] The wavelet decomposition module 200 is used to perform wavelet decomposition on current and voltage signals to obtain low-frequency and high-frequency components at multiple levels.

[0102] The feature extraction module 300 is used to extract features from the low-frequency components and the high-frequency components respectively to obtain signal component features;

[0103] The electric shock diagnosis module 400 is used to input signal component features into a pre-trained electric shock diagnosis model, so that the electric shock diagnosis model outputs electric shock diagnosis results.

[0104] The electric shock protection module 500 is used to generate a protection action command and send it to the residual current device (RCD) electrically connected to the target circuit when the electric shock diagnosis result determines that an electric shock event has occurred. When the RCD receives the protection action command and the residual current meets the preset conditions, it triggers the RCD to perform the protection action and cut off the power supply to the target circuit.

[0105] In some embodiments, the system further includes:

[0106] The preprocessing module is used to preprocess current and voltage signals; the preprocessing includes at least one of downsampling and wavelet denoising.

[0107] In some embodiments, the wavelet decomposition module 200 is used for:

[0108] Determine the wavelet basis functions and the number of wavelet decomposition layers;

[0109] Using a defined wavelet basis function and a certain number of decomposition levels, wavelet decomposition is performed on current and voltage signals to obtain low-frequency and high-frequency components at multiple levels.

[0110] In some embodiments, the signal component features include low-frequency component features and high-frequency component features; the feature extraction module 300 is used for:

[0111] Statistical feature extraction methods were used to extract the mean and variance of the low-frequency and high-frequency components, respectively.

[0112] The low-frequency component features and high-frequency component features are converted into low-frequency component frequency domain features and high-frequency component frequency domain features, respectively. Then, using the sliding window method, the spectral distribution features and second harmonic amplitude of the low-frequency component frequency domain features and high-frequency component frequency domain features are extracted, respectively.

[0113] The mean, variance, spectral distribution characteristics, and second harmonic amplitude of the low-frequency components are integrated into low-frequency component characteristics, and the mean, variance, spectral distribution characteristics, and second harmonic amplitude of the high-frequency components are integrated into high-frequency component characteristics.

[0114] In some embodiments, the training process of the electric shock diagnostic model includes:

[0115] Collect electric shock sample data and non-electric shock sample data to construct a sample dataset; wherein, electric shock sample data includes current and voltage signal sample data when electric shock occurs, and non-electric shock sample data includes current and voltage signal sample data under normal conditions;

[0116] Wavelet decomposition is performed on the current and voltage signal sample data in the sample dataset to obtain low-frequency component samples and high-frequency component samples.

[0117] Feature extraction is performed on low-frequency component samples and high-frequency component samples respectively to obtain sample signal component features;

[0118] The signal component features of the samples are classified according to the electric shock sample data and the non-electric shock sample data, and the classified signal component features are labeled with whether or not electric shock has occurred.

[0119] A training dataset is constructed based on the characteristics of the sample signal components and the label indicating whether an electric shock occurred.

[0120] The electric shock diagnosis model is obtained by training a pre-defined machine learning model using a training dataset; the pre-defined machine learning model is either a GRU recurrent neural network or a PNN probabilistic neural network.

[0121] In some embodiments, the electric shock protection module 500 is used for:

[0122] When the electric shock diagnosis result is determined to be an electric shock event, a protective action command is generated and sent to the residual current device (RCD) that is electrically connected to the target circuit.

[0123] After receiving a protection action command, the residual current device (RCD) checks whether the residual current of the target circuit exceeds the preset safe current threshold. If the residual current of the target circuit exceeds the preset safe current threshold, it executes a protection action to cut off the power supply to the target circuit.

[0124] like Figure 4 As shown, this application provides an electronic device. The electronic device 10 includes a memory 20 and a processor 30. The memory 20 stores a computer program. When the computer program is executed by the processor 30, the processor 30 performs the steps of the wavelet transform-based electric shock identification and protection method as described in the above embodiment.

[0125] This application provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed, it implements the steps of the wavelet transform-based electric shock identification and protection method as described in the above embodiments.

[0126] This application provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer performs the steps of the wavelet transform-based electric shock identification and protection method as described in the above embodiments.

[0127] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, electronic devices, computer storage media, and computer program products described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. 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 explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products or devices.

[0129] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0130] In the several embodiments provided by this invention, it should be understood that the disclosed systems, electronic devices, computer storage media, computer program products, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0131] 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 network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0132] 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.

[0133] 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 computer-readable 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 for executing all or part of the steps of the methods described in the various embodiments of the present invention through a computer device (which may be a personal computer, a server, or a network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0134] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for electric shock identification and protection based on wavelet transform, characterized in that, include: Acquire the current and voltage signals of the target circuit when an electric shock occurs; Wavelet decomposition is performed on the current and voltage signals to obtain low-frequency and high-frequency components at multiple levels. Features are extracted from the low-frequency component and the high-frequency component respectively to obtain signal component features; The signal component features are input into a pre-trained electric shock diagnosis model, which then outputs the electric shock diagnosis result. If the electric shock diagnosis result indicates an electric shock event, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. When the RCD receives the protection action command and the residual current meets the preset conditions, the RCD is triggered to perform a protection action and cut off the power supply to the target circuit.

2. The wavelet transform-based method for electric shock identification and protection according to claim 1, characterized in that, The step of performing wavelet decomposition on the current and voltage signals to obtain multiple levels of low-frequency and high-frequency components also includes: The current and voltage signals are preprocessed; wherein the preprocessing includes at least one of downsampling and wavelet denoising.

3. The wavelet transform-based method for electric shock identification and protection according to claim 1 or 2, characterized in that, The wavelet decomposition of the current and voltage signals yields multiple levels of low-frequency and high-frequency components, including: Determine the wavelet basis functions and the number of wavelet decomposition layers; Using a defined wavelet basis function and a number of decomposition levels, the current and voltage signals are decomposed using wavelet decomposition to obtain low-frequency and high-frequency components at multiple levels.

4. The wavelet transform-based method for electric shock identification and protection according to claim 1, characterized in that, The signal component characteristics include low-frequency component characteristics and high-frequency component characteristics; The step of extracting features from the low-frequency component and the high-frequency component respectively to obtain signal component features includes: The mean and variance of the low-frequency component and the high-frequency component features are extracted using a statistical feature extraction method. The low-frequency component features and the high-frequency component features are converted into low-frequency component frequency domain features and high-frequency component frequency domain features, respectively. Then, using the sliding window method, the spectral distribution features and second harmonic amplitudes of the low-frequency component frequency domain features and the high-frequency component frequency domain features are extracted, respectively. The mean, variance, spectral distribution characteristics, and second harmonic amplitude of the low-frequency components are integrated into the low-frequency component characteristics, and the mean, variance, spectral distribution characteristics, and second harmonic amplitude of the high-frequency components are integrated into the high-frequency component characteristics.

5. The wavelet transform-based method for electric shock identification and protection according to claim 1, characterized in that, The training process of the electric shock diagnostic model includes: Collect electric shock sample data and non-electric shock sample data to construct a sample dataset; wherein, the electric shock sample data includes current and voltage signal sample data when electric shock occurs, and the non-electric shock sample data includes current and voltage signal sample data under normal conditions; Wavelet decomposition is performed on the current and voltage signal sample data in the sample dataset to obtain low-frequency component samples and high-frequency component samples. Feature extraction is performed on the low-frequency component samples and the high-frequency component samples respectively to obtain the sample signal component features; The sample signal component features are classified according to the electric shock sample data and the non-electric shock sample data, and the classified sample signal component features are labeled with whether or not electric shock has occurred. A training dataset is constructed based on the characteristics of the sample signal components and the label indicating whether an electric shock has occurred. The electric shock diagnosis model is obtained by training a preset machine learning model using the training dataset; wherein the preset machine learning model is a GRU recurrent neural network or a PNN probabilistic neural network.

6. The wavelet transform-based method for electric shock identification and protection according to claim 1, characterized in that, When the electric shock diagnosis result indicates an electric shock event, a protection action command is generated and sent to the residual current device (RCD) electrically connected to the target circuit. Upon receiving the protection action command and if the residual current meets preset conditions, the RCD is triggered to perform a protection action, cutting off the power supply to the target circuit, including: When the electric shock diagnosis result is determined to be an electric shock event, a protective action command is generated and sent to the leakage current protection device electrically connected to the target circuit. After receiving the protection action command, the residual current device detects whether the residual current of the target circuit exceeds the preset safe current threshold. If the residual current of the target circuit exceeds the preset safe current threshold, it performs the protection action of cutting off the power supply to the target circuit.

7. A wavelet transform-based electric shock identification and protection system, characterized in that, include: The signal acquisition module is used to acquire the current and voltage signals of the target circuit when an electric shock occurs; The wavelet decomposition module is used to perform wavelet decomposition on the current and voltage signals to obtain low-frequency and high-frequency components at multiple levels. The feature extraction module is used to extract features from the low-frequency component and the high-frequency component respectively to obtain signal component features; The electric shock diagnosis module is used to input the signal component features into a pre-trained electric shock diagnosis model, so that the electric shock diagnosis model outputs the electric shock diagnosis result; The electric shock protection module is used to generate a protection action command and send it to the leakage current protector electrically connected to the target circuit when the electric shock diagnosis result determines that an electric shock event has occurred. When the leakage current protector receives the protection action command and the residual current meets the preset conditions, it triggers the leakage current protector to perform a protection action and cut off the power supply to the target circuit.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the wavelet transform-based electric shock identification and protection method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the steps of the wavelet transform-based electric shock identification and protection method as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, wherein when the program instructions are executed by a computer, the computer performs the steps of the wavelet transform-based electric shock identification and protection method as described in any one of claims 1-6.

Images