Arc prediction method and related device for low-voltage circuit

Abstract

The present application proposes an arc prediction method and related device for a low-voltage circuit. The method includes: the processor receives circuit data within a target period from a current acquisition device; determines an effective current component and a noise current component based on the current data, and determines a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation; determines a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal; determines a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and obtains a feature contribution rate by solving the feature matrix; determines an arc prediction result based on the circuit data, the feature contribution rate, and a pre-constructed arc prediction model. In this way, the arc prediction model has a strong anti-interference ability, can achieve good results in a harsh environment, can extract the characteristics of arc faults, thereby obtaining higher accuracy, and feature optimization reduces the input layer of the algorithm, improving the detection speed and efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of circuits, and in particular, to an arc prediction method and related device for a low-voltage circuit. Background Art

[0002] With the improvement of the technological level, the types of household appliances have increased, and the demand for electric energy by ordinary household users has been continuously rising. Most household products need to be driven by electric energy, and while electric energy brings many conveniences, it also brings many potential safety hazards to people.

[0003] Some of the current related technologies are applied to fixed-point detection scenarios such as distribution cabinets, substations, and switch cabinets in medium and low-voltage distribution systems, and some establish time-varying relationship equations between two or more of current, power, voltage, and energy. However, there are problems such as difficulty in being widely used in various complex low-voltage power consumption scenarios, a gap between the data generated by the model and the actual situation, and insufficiently rapid response. Moreover, they all study whether a fault arc occurs at the current moment. However, once a fault arc occurs, it may lead to an electrical fire. Summary of the Invention

[0004] The embodiments of the present application provide an arc prediction method and related device for a low-voltage circuit, which can be widely used in various complex low-voltage power consumption scenarios, realize rapid prediction and response of fault arcs in advance, and improve the power consumption safety factor.

[0005] In a first aspect, the embodiments of the present application provide an arc prediction method for a low-voltage circuit, which is applied to a processor of a low-voltage circuit system and includes:

[0006] Receiving circuit data within a target period from the current acquisition device, where the circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period;

[0007] Determining an effective current component and a noise current component based on the current data, and determining a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation. The reconstructed signal is a current signal after noise reduction;

[0008] Determining a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal. The frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal;

[0009] Determining a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and solving to obtain a feature contribution rate based on the feature matrix. The feature contribution rate characterizes the importance of each time-domain feature or frequency-domain feature to constrain the feature dimension of a pre-constructed arc prediction model;

[0010] Determine the arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model. The arc prediction result includes a first prediction result. If the arc prediction result is the first prediction result, a warning message is issued; the arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection sequence period.

[0011] In a possible embodiment, the determining the effective current component and the noise current component based on the current data includes:

[0012] Determine the normalized current data based on the current data;

[0013] Perform modal decomposition based on the normalized current data to determine the j-th order eigenmode component;

[0014] Determine a plurality of signal parameters based on the j-th order eigenmode component and / or the normalized current data. The signal parameters include the j-th order variance of the j-th order eigenmode component and / or the variance of the current data;

[0015] Determine the variance contribution rate based on the signal parameters, and determine the effective current component and the noise current component based on the variance contribution rate.

[0016] In a possible embodiment, the determining the reconstructed signal based on the effective current component, the noise current component, and the wavelet threshold denoising operation includes:

[0017] Determine the first type of frequency component based on the noise current component. The first type of frequency component includes high-frequency components;

[0018] Determine the j-th order modal component threshold based on the j-th order eigenmode component; and determine the j-th order denoised modal component based on the first type of frequency component and the j-th order modal component threshold;

[0019] Determine the reconstructed signal based on the j-th order denoised modal component and the effective current component.

[0020] In a possible embodiment, the determining the j-th order modal component threshold based on the j-th order eigenmode component includes:

[0021] Determine the first-order eigenmode component based on the j-th order eigenmode component. The first-order eigenmode component is the eigenmode component when j = 1;

[0022] Determine the first-order energy signal based on the first-order eigenmode component and the following formula:

[0023] ;

[0024] Where, represents finding the median, represents finding the absolute value, is the first-order eigenmode component, and 0.6745 is the 75% quantile of the standard normal distribution;

[0025] Determine the j-th order energy signal based on the first-order energy signal and a preset first empirical parameter;

[0026] Determine the j-th order mode component threshold based on the j-th order energy signal and multiple preset second empirical parameters.

[0027] In a possible embodiment, the multiple time-domain features include at least one of a root mean square parameter, a variance parameter, a kurtosis factor parameter, an impulse factor parameter, and a current change rate parameter, and the multiple frequency-domain features include at least one of a harmonic current content rate and a total harmonic distortion rate of the current; the determining the feature matrix based on the multiple time-domain features and the multiple frequency-domain features, and the solving for the feature contribution rate based on the feature matrix includes:

[0028] Determine a feature matrix with n rows and m columns based on the multiple time-domain features and the multiple frequency-domain features, where m corresponds to the number of the time-domain features and / or the frequency-domain features, and n corresponds to the number of the current data;

[0029] Determine a correlation coefficient matrix based on the n-row m-column feature matrix;

[0030] Determine multiple eigenvalues based on the correlation coefficient matrix, and determine the feature contribution rate corresponding to each of the time-domain features and / or the frequency-domain features based on the multiple eigenvalues.

[0031] In a possible embodiment, the determining the arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model includes:

[0032] Determine the principal component features and the number of principal component features based on the feature contribution rate, where the number of principal component features is the number of the principal component features;

[0033] Update the pre-constructed arc prediction model based on the number of principal component features so that the pre-constructed arc prediction model conforms to the current prediction working condition;

[0034] Determine the arc prediction result based on the circuit data and the pre-constructed arc prediction model.

[0035] In a possible embodiment, the circuit data further includes circuit environment data, and the circuit environment data includes at least one of circuit temperature data, circuit pressure data, and circuit illumination data within a target period; the determining of the arc prediction result based on the circuit data and the pre-constructed arc prediction model includes:

[0036] Performing data standardization processing based on the circuit temperature data and / or circuit pressure data and / or circuit illumination data to determine at least one standard environment data;

[0037] Determining a curve and a plurality of slope parameters corresponding to at least one of the circuit environment data by performing linear fitting on the environment data determined based on the standard environment data;

[0038] Determining a deviation parameter based on the plurality of slope parameters and a preset slope parameter;

[0039] Determining an environmental impact factor based on the deviation parameter;

[0040] Determining the arc prediction result based on the environmental impact factor, the current data, and the pre-constructed arc prediction model.

[0041] In a second aspect, an embodiment of the present application provides an arc prediction device for a low-voltage circuit, which is applied to a processor of a low-voltage circuit system. The low-voltage circuit system includes a first circuit module and a second circuit module. The first circuit module includes an actual scenario circuit, and the second circuit module includes a simulated fault circuit. The low-voltage circuit system further includes a current acquisition device. The device includes:

[0042] A data acquisition module, configured to receive circuit data within a target period from the current acquisition device. The circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period;

[0043] A first determination module, configured to determine an effective current component and a noise current component based on the current data, and determine a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation. The reconstructed signal is a current signal after noise reduction;

[0044] A second determination module, configured to determine a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal. The frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal;

[0045] A third determination module, configured to determine a feature matrix based on the multiple time-domain features and the multiple frequency-domain features, and solve a feature contribution rate based on the feature matrix, where the feature contribution rate characterizes the importance of each of the time-domain features or the frequency-domain features, so as to constrain the feature dimension of a pre-constructed arc prediction model;

[0046] An arc prediction module, configured to determine an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model, where the arc prediction result includes a first prediction result. If the arc prediction result is the first prediction result, a warning message is sent; the arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection sequence period.

[0047] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. The computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to execute some or all of the steps described in the first aspect.

[0048] In a fourth aspect, an embodiment of the present application provides an electronic device, including a processor, a memory, a communication interface, and one or more programs. Among them, the one or more programs are stored in the memory and are configured to be executed by the processor. The programs include instructions for executing some or all of the steps described in the first aspect of the embodiments of the present application.

[0049] In a fifth aspect, an embodiment of the present application provides a computer program product. The computer program product includes a non-transitory computer-readable storage medium storing a computer program. The computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application. The computer program product can be a software installation package.

[0050] By implementing the embodiments of the present application, the processor receives circuit data within a target period from the current acquisition device. The circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period. Based on the current data, an effective current component and a noise current component are determined, and a reconstructed signal is determined based on the effective current component, the noise current component, and a wavelet threshold denoising operation. The reconstructed signal is a current signal after denoising. Based on the reconstructed signal, a plurality of time-domain features and a plurality of frequency-domain features are determined. The frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal. Based on the plurality of time-domain features and the plurality of frequency-domain features, a feature matrix is determined, and a feature contribution rate is obtained by solving the feature matrix. The feature contribution rate characterizes the importance of each time-domain feature or frequency-domain feature to constrain the feature dimension of a pre-constructed arc prediction model. Based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model, an arc prediction result is determined. The arc prediction result includes a first prediction result. If the arc prediction result is the first prediction result, a warning message is issued. The arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection period. In this way, by performing empirical mode decomposition and wavelet threshold denoising on the current data of low-voltage power consumption monitoring, the arc prediction model has strong anti-interference ability and can achieve good results in harsh environments. By performing time-domain and frequency-domain features on the denoised current data, the features of arc faults can be mined, thereby obtaining higher accuracy. Feature optimization reduces the input layer of the algorithm and improves the detection speed and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the following will describe the drawings required to be used in the embodiments of the present invention or the background art.

[0052] Figure 1a is a system architecture diagram of a low-voltage circuit system provided by an embodiment of the present application;

[0053] Figure 1b is another system architecture diagram of a low-voltage circuit system provided by an embodiment of the present application;

[0054] Figure 1c is a scenario schematic diagram of a low-voltage circuit system provided by an embodiment of the present application;

[0055] Figure 1d is a structural schematic diagram of an electronic device provided by an embodiment of the present application;

[0056] Figure 2 is a flowchart of an arc prediction method for a low-voltage circuit provided by an embodiment of the present application;

[0057] Figure 3 It is a coordinate schematic diagram of a feature contribution rate provided by an embodiment of the present application;

[0058] Figure 4 It is a schematic diagram of the model architecture of an arc prediction model provided by an embodiment of the present application;

[0059] Figure 5 It is a schematic diagram of the hit rate of an arc prediction model proposed by an embodiment of the present application;

[0060] Figure 6 It is a schematic diagram of the structure of an arc prediction device for a low-voltage circuit provided by an embodiment of the present application. Detailed implementation manners

[0061] In order to enable those skilled in the art to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0062] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or electronic device that includes a series of steps or units is not limited to the listed steps or units, but in an alternative example also includes steps or units not listed, or in an alternative example also includes other steps or units inherent to these processes, methods, products or electronic devices.

[0063] Referring to "embodiment" herein means that a specific feature, structure or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0064] With the improvement of technology, the types of household appliances have increased, and the demand for electric energy by ordinary household users has been rising continuously. Most household products need to be driven by electric energy, which brings many conveniences while also bringing many potential safety hazards. Each year, there are more than 100,000 electrical fires, and 70% of major and extremely large fires are electrical fires. Among them, more than 80% of electrical fires are caused by faulty arcs, which pose a great threat to people's economic property and personal safety. Some of the current related technologies are applied to fixed-point detection scenarios such as distribution cabinets, substations, and switch cabinets in medium and low-voltage distribution systems, and some establish time-varying relationship equations between two or more of current, power, voltage, and energy. However, there are problems such as being difficult to be widely used in various complex low-voltage power consumption scenarios, there is a gap between the data generated by the model and the actual situation, and the response is not fast and timely enough. Moreover, they all study whether a faulty arc occurs at the current moment. However, once a faulty arc occurs, it may lead to an electrical fire.

[0065] In view of the above problems, the embodiments of the present application provide an arc prediction method and related device for a low-voltage circuit. By performing empirical mode decomposition and wavelet threshold denoising on the current data of low-voltage power consumption monitoring, the arc prediction model has strong anti-interference ability and can achieve good results in harsh environments. By performing time-domain and frequency-domain features on the denoised current data, the features of arc faults can be mined, so as to obtain higher accuracy. Feature optimization reduces the input layer of the algorithm, improving the detection speed and efficiency.

[0066] The arc prediction method and related device for a low-voltage circuit provided by the embodiments of the present application can be applied to a low-voltage circuit system such as Figure 1a or Figure 1b as shown.

[0067] Please refer to Figure 1a , Figure 1a which is a system architecture diagram of a low-voltage circuit system provided by the embodiments of the present application. As shown in Figure 1a , the low-voltage circuit system 100 includes: a circuit device 110 and a processor 120, where the circuit device 110 further includes a circuit module 111 and a current acquisition device 112.

[0068] Among them, the circuit device 110 is a low-voltage power consumption scenario. For example, the circuit of a certain power consumption area. The circuit module 111 included in the circuit device 110 is the actual circuit part. The current acquisition device 112 includes a device for current acquisition, which can be set at a fixed position in the circuit or led out from the circuit. The arc prediction model is deployed on the processor 120 for analyzing the occurrence of arcs in the circuit module 111. The processor 120 can also be used to collect data during the use of the model to facilitate subsequent optimization of the arc prediction model.

[0069] Among them, the current acquisition device 112 can be a current sensing device or a device with current acquisition function, which is not limited herein.

[0070] Please refer to Figure 1b , Figure 1b FIG. is a system architecture diagram of another low-voltage circuit system provided by an embodiment of the present application. As Figure 1b shown, the low-voltage circuit system 100 includes: a circuit device 110 and a processor 120. The circuit device 110 further includes a circuit module 111 and a current acquisition device 112. Among them, the circuit module 111 includes a first circuit module 1111 and a second circuit module 1112.

[0071] Among them, the first circuit module is an actual scenario circuit, and the second circuit module is an analog fault circuit. Various low-voltage loads are powered in the first circuit module; a fault arc simulation device is provided in the second circuit module for generating fault data corresponding to the above low-voltage loads and simulating the fault arc situation. The high-frequency current signals generated in both cases are collected by the above current acquisition device 112 and transmitted to the above processor 120.

[0072] Exemplarily, please refer to Figure 1c , Figure 1c FIG. is a scenario schematic diagram of a low-voltage circuit system provided by an embodiment of the present application. As Figure 1c shown, in the low-voltage circuit system, the circuit module may include various low-voltage loads such as a fan, a hair dryer, and a computer. On the left is a low-voltage power supply system. In a part, that is, in the first circuit module, the 220V low-voltage power supply system supplies power to various low-voltage loads such as a fan, a hair dryer, and a computer through switch A1; in another part, that is, in the second circuit module, the 220V low-voltage power supply system is connected to a fault arc simulation device through switch A2 to simulate the fault arc situation. The high-frequency current signals generated in both cases are collected by a high-frequency current sensor, that is, the current acquisition device. Then the current acquisition device can send the acquired circuit data to the processor, and the processor analyzes based on the current data to obtain an arc prediction result.

[0073] Please refer to Figure 1d , Figure 1dIt is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 200 includes a processor 120, a memory 130, a communication interface 140, and one or more programs 131. Among them, the one or more programs 131 are stored in the memory 130 and are configured to be executed by the processor 120. The one or more programs 131 include instructions for executing any step in the following method embodiments. The processor 120, the memory 130, and the communication interface 140 are interconnected and complete communication with each other. Among them, the memory 130 can be a volatile memory such as a dynamic random access memory DRAM, or a non-volatile memory such as a mechanical hard disk. In a specific implementation, the processor 120 is used to execute any step executed by the infusion device in the following method embodiments, and when performing data transmission such as sending, the communication interface 140 can be selectively called to complete the corresponding operation.

[0074] Please refer to Figure 2 , Figure 2 It is a schematic flowchart of an arc prediction method for a low-voltage circuit provided by an embodiment of the present application. This method is applied to a processor of a low-voltage circuit system. The low-voltage circuit system includes a first circuit module and a second circuit module. The first circuit module includes an actual scenario circuit, and the second circuit module includes a simulated fault circuit. The low-voltage circuit system further includes a current acquisition device. The method includes:

[0075] S210, Receive circuit data within a target period from the current acquisition device. The circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period.

[0076] Among them, the current data includes the current data of the first circuit module and / or the second circuit module. The current data generated by the first circuit module is the actual current data generated by the actual application scenario, and the current data generated by the second circuit module is the simulated current data generated by the fault simulation device. In actual applications, it can be to obtain the current data of the first circuit module, or to obtain the current data of the first circuit module and the second circuit module. The target period can be the current period, and the period length can be a time cycle length. The specific cycle length can be preset, such as 1 second, 5 seconds, 1 millisecond, etc., which is not limited here.

[0077] S220, Determine an effective current component and a noise current component based on the current data, and determine a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation. The reconstructed signal is a current signal after denoising.

[0078] Among them, the effective current component and the noise current component are distinguished based on the current data. Specifically, data feature analysis can be first performed through one or more of time-domain analysis, frequency-domain analysis, and statistical characteristic analysis. Among them, time-domain analysis can be to analyze the amplitude, pulse width, periodicity, etc. of the current waveform, and initially distinguish steady-state signals, such as normal load current, and transient noise, such as spikes and glitches. Frequency-domain analysis can be to transform the signal to the frequency domain through fast Fourier transform (FFT), plot the frequency spectrum diagram, and identify the frequency band where the noise is mainly distributed. Statistical characteristics can be to calculate statistics such as mean, variance, and kurtosis, and judge whether the noise is Gaussian distribution or impulse-type noise. Then filtering is performed to distinguish the current data, mainly to filter out the noise component to obtain an effective current component with less noise. Specifically, wavelet filtering can be used for filtering, or other methods can be used, which are not limited here.

[0079] Among them, then the current component with some noise removed after filtering is integrated to obtain a reconstructed signal.

[0080] In a possible embodiment, determining the effective current component and the noise current component based on the current data includes: determining normalized current data based on the current data; performing modal decomposition based on the normalized current data to determine the j-th order intrinsic mode component; determining a plurality of signal parameters based on the j-th order intrinsic mode component and / or the normalized current data, where the signal parameters include the j-th order variance of the j-th order intrinsic mode component and / or the variance of the current data; determining the variance contribution rate based on the signal parameters, and determining the effective current component and the noise current component based on the variance contribution rate.

[0081] Among them, since each piece of low-voltage power consumption current data (t = 1, 2,..., T) varies greatly in magnitude under different loads, to eliminate its influence, data normalization processing is first performed. Specifically, the normalization processing can be performed according to the following first formula:

[0082] ;

[0083] Among them, the obtained above is the normalized current data obtained by normalizing the current data ; is the minimum value in is the maximum value in

[0084] Among them, since the collected low-voltage power consumption current data usually includes various noises, the existence of noises is not conducive to the accurate detection of fault arcs. Therefore, the current signal is preprocessed first, specifically, the current signal can be denoised. The specific process of denoising can include: First, perform empirical mode decomposition on the current data to decompose the current data into h intrinsic mode components of order j arranged from high frequency to low frequency and a trend term component , specifically, the empirical mode decomposition can be performed according to the following second formula: .

[0085] Among them, multiple signal parameters include at least one of the j-th order variance of the j-th order intrinsic mode component and the variance of the current data x(t). In some possible implementation manners, the j-th order variance of the above-mentioned j-th order intrinsic mode component is determined based on the following third formula , and the variance of the current data is determined according to the following fourth formula;

[0086] Third formula: ;

[0087] Fourth formula: ;

[0088] Among them, T is the length of the collected signal, that is, the target time period.

[0089] Among them, the present invention uses the variance contribution rate to describe the contribution of each mode component to the fluctuation of the original current data, so as to characterize the influence degree of different periodic components on the original data. Therefore, the present invention takes the variance contribution rate as the screening condition. The variance contribution rate (VCR) of a component refers to the proportion of the variance that a component can explain in the total variance. The larger its value, the stronger the ability of the component to synthesize the information of the original current signal. Specifically, the variance contribution rate is determined according to the following fifth formula:

[0090] ;

[0091] Among them, represents the variance contribution rate of the j-th order mode component, j = 1, 2,..., h.

[0092] Among them, when the variance contribution rate is not greater than the preset contribution rate threshold, it can be considered that the mode component is a noise-dominated component. The above-mentioned preset contribution rate threshold can be 0.02, and the above-mentioned preset contribution rate threshold can also be other values, which are not limited herein. Then, the q-th order mode component is calculated as the boundary line between the noise component and the effective component, that is of When at the q-th order and at the (q - 1)-th order , then the current components from the q-th order to the h-th order are effective current components, and the first (q - 1) orders are noise current components.

[0093] It can be seen that in this embodiment, through normalization, modal decomposition and signal parameter calculation, key features are gradually extracted from the original current data, avoiding the limitations of traditional methods for non-stationary signals. Based on the quantitative analysis of the variance contribution rate, it can objectively distinguish effective current components and noise components, reducing the dependence on human experience or subjective judgment. This method has strong robustness to non-linear and non-stationary current signals and is applicable to various low-voltage power consumption scenarios.

[0094] In a possible embodiment, determining the reconstructed signal based on the effective current component, the noise current component and the wavelet threshold denoising operation includes: determining a first type of frequency component based on the noise current component, where the first type of frequency component includes high-frequency components; determining the j-th order modal component threshold based on the j-th order intrinsic mode component; and determining the j-th order denoised modal component based on the first type of frequency component and the j-th order modal component threshold; determining the reconstructed signal based on the j-th order denoised modal component and the effective current component.

[0095] Wherein, based on a preset base wavelet and the decomposition level, the noise current component is decomposed to obtain high-frequency coefficients and low-frequency coefficients.

[0096] Specifically, the base wavelet is the core function of wavelet transform, such as Daubechies wavelet, Symlets wavelet, etc., which are not limited herein. Its characteristics, such as time-frequency localization ability, orthogonality, etc., will affect the signal decomposition effect. An appropriate wavelet can be selected according to signal characteristics (such as non-linearity, non-stationarity) and analysis objectives (such as denoising, feature extraction). The decomposition level determines the scale range in which the signal is decomposed. Each level of decomposition will divide the signal into a low-frequency approximation component and a high-frequency detail component. Specifically, the decomposition level selection is based on covering the main frequency range of the signal. Then, based on the decomposition level, the noise current component is decomposed to obtain the first type of frequency component, that is, the high-frequency component, and, secondarily, the second type of frequency component, that is, the low-frequency component, can also be obtained.

[0097] Wherein, soft threshold denoising is performed based on the following sixth formula to filter out the noise component part to obtain the j-th order denoised modal component. Sixth formula:

[0098] ;

[0099] Wherein, is the j-th order modal component after denoising, is the sign function, is the first type of frequency component of the j-th order after wavelet decomposition, is the threshold corresponding to the j-th order modal component, that is, the j-th order modal component threshold. For the coefficients with amplitudes greater than the j-th order modal component threshold, subtract the threshold (retaining the useful signal but reducing its amplitude). For the coefficients with amplitudes less than the j-th order modal component threshold, directly set them to zero (removing the noise).

[0100] Among them, the determination of the above-mentioned j-th order modal component threshold can be determined by parameters such as the signal length and the standard deviation of the noise, or can be determined by parameters such as the signal length and the energy corresponding to the modal component.

[0101] Among them, after obtaining the j-th order denoised modal component then integrate the j-th order denoised modal component and the effective current component to obtain a reconstructed signal containing the effective signal. Specifically, signal reconstruction is performed through the following seventh formula:

[0102] ;

[0103] Among them, is the current signal after noise reduction, is the j-th order modal component after wavelet denoising, is the retained effective component, is the residual component.

[0104] It can be seen that in this embodiment, by accurately identifying the noise frequency range through wavelet decomposition, it can be determined that the noise is mainly distributed in the high-frequency region. Thus, the high-frequency components are classified as "the first type of frequency components", and then the j-th order modal component threshold is dynamically calculated according to the parameters related to the j-th order intrinsic modal component, making the threshold match the signal characteristics, so that the noise can be more accurately decomposed by the threshold to meet the usage scenario, dealing with the noise separately at different scales (frequencies), improving the denoising accuracy, and avoiding the loss of high-frequency features caused by traditional low-pass filtering.

[0105] In a possible embodiment, determining the reconstructed signal based on the effective current component, the noise current component, and the wavelet threshold denoising operation includes: determining the j-th order modal component threshold based on the j-th order intrinsic modal component; and determining the j-th order denoised modal component based on the j-th order intrinsic modal component and the j-th order modal component threshold; determining the reconstructed signal based on the j-th order denoised modal component and the effective current component.

[0106] Among them, soft thresholding is used for denoising based on the j-th order intrinsic modal component, and the soft threshold function is the following eighth formula:

[0107] ;

[0108] Among them, is the j-th order modal component after denoising, is the sign function, is the j-th order modal noise component before denoising. is the threshold corresponding to the j-th order modal component. Among them, the determination of the above j-th order modal component threshold can be determined by parameters such as signal length and noise standard deviation, or can be determined by parameters such as signal length and energy corresponding to the modal component.

[0109] Among them, after obtaining the j-th order denoised modal component the j-th order denoised modal component and the effective current component are integrated to obtain a reconstructed signal containing the effective signal. Specifically, signal reconstruction is performed through the following seventh formula:

[0110] ;

[0111] Among them, is the current signal after noise reduction, is the j-th order modal component after wavelet denoising, is the retained effective component, is the residual component.

[0112] It can be seen that in this embodiment, the j-th order modal component threshold is dynamically calculated through parameters related to the j-th order intrinsic modal component, so that the threshold matches the signal characteristics, making the noise decomposition by the threshold more accurately conform to the usage scenario, processing the noise separately at different scales (frequencies), improving the denoising accuracy, and avoiding the loss of high-frequency features caused by traditional low-pass filtering.

[0113] In a possible embodiment, the determination of the j-th order modal component threshold is performed through the following steps: determining the first-order intrinsic modal component based on the j-th order intrinsic modal component, where the first-order intrinsic modal component is the intrinsic modal component when j = 1; determining the first-order energy signal based on the first-order intrinsic modal component and the following ninth formula:

[0114] ;

[0115] Among them, represents finding the median, represents finding the absolute value, is the first-order intrinsic modal component, and 0.6745 is the 75% quantile of the standard normal distribution; determining the j-th order energy signal based on the first-order energy signal and a preset first empirical parameter; determining the j-th order modal component threshold based on the j-th order energy signal and multiple preset second empirical parameters.

[0116] Among them, determining the j-th order modal component threshold based on the j-th order energy signal and multiple preset second empirical parameters specifically includes the following steps: determining the j-th order energy signal based on the first-order energy signal and a preset first empirical parameter, specifically determined according to the following tenth formula:

[0117] ;

[0118] Among them, and are the first empirical parameters, which are constants and are preset to 0.719, 2.01 respectively according to engineering experience.

[0119] Among them, determining the j-th order modal component threshold based on the j-th order energy signal and multiple preset second empirical parameters specifically includes the following steps: The j-th order modal component threshold is determined based on the j-th order energy signal, the signal length, and the preset second empirical parameters according to the following eleventh formula:

[0120] ;

[0121] Among them, D is the second empirical parameter, which is a constant and generally takes 0.3, and T is the signal length.

[0122] It should be noted that the above first empirical parameter and second empirical parameter are variable and can be set based on the actual application scenario, which is not limited here.

[0123] It can be seen that in this embodiment, by combining the signal statistical characteristics and engineering experience, the dual goals of efficient denoising and feature retention are achieved. When calculating the threshold, the first empirical parameter and the second empirical parameter set based on experience are introduced, making the threshold more suitable for the actual scenario.

[0124] S230, determining multiple time-domain features and multiple frequency-domain features based on the reconstructed signal, where the frequency-domain features include waveform distortion features used to characterize the frequency domain of the reconstructed signal.

[0125] Among them, the current waveforms of normal operation of low-voltage power consumption and the occurrence of fault arcs change greatly, and the time-domain and frequency-domain characteristics of the current change significantly before and after the occurrence of the arc fault. Therefore, the present invention extracts the time-domain and frequency-domain characteristics of the current signal respectively.

[0126] In a possible embodiment, the multiple time-domain features include at least one of root mean square parameter, variance parameter, kurtosis factor parameter, impulse factor parameter, current change rate parameter, and the multiple frequency-domain features include at least one of harmonic current content rate and total harmonic distortion rate of the current.

[0127] Among them, the time-domain features can be calculated and determined according to the signal obtained after reconstruction according to the seventh formula; specifically, the root mean square parameter can be calculated according to the following twelfth formula: ; The variance parameter can be calculated according to the following thirteenth formula: ; The kurtosis factor parameter can be calculated according to the following fourteenth formula: ; The impulse factor parameter can be calculated according to the following fifteenth formula: ; The rate of change of current can be calculated according to the following sixteenth formula: . Wherein, n represents the nth sample.

[0128] Among them, the frequency-domain features include at least one of the harmonic current content rate and the total harmonic distortion rate of current.

[0129] Specifically, since during a low-voltage arc fault, the amplitude of odd harmonics of the current increases significantly compared to normal, and the increase in the amplitude of odd harmonics is not affected by the load type, some harmonic current content rates can be selected as one of the frequency-domain feature indicators for detecting arc faults. Specifically, the content rate of the first harmonic current and the content rate of the third harmonic current can be selected: Based on the signal obtained after reconstruction and the following seventeenth formula to calculate the content rate of the first harmonic current:

[0130] ;

[0131] Based on the signal obtained after reconstruction and the following eighteenth formula to calculate the content rate of the third harmonic current:

[0132] ;

[0133] Among them, and are the effective values of the first and third harmonic currents respectively, is the effective value of the fundamental current.

[0134] Specifically, since the current waveform is distorted during an arc fault, the total harmonic distortion rate of current can be selected as one of the frequency-domain feature indicators of the arc current. Specifically, based on the signal obtained after reconstruction and the following nineteenth formula to calculate the total harmonic distortion rate of current:

[0135] ;

[0136] Among them, is the effective value of the pth harmonic current, is the effective value of the fundamental current, L is the specified maximum harmonic order. In actual use, L can be taken as 30, and in some cases, other values can also be taken, which is not limited here.

[0137] It can be seen that in this embodiment, the time-domain and frequency-domain features are combined to cover multi-dimensional information of the signal. The time-domain features are sensitive to faults, and the frequency-domain features are suitable for harmonic analysis. The overall method has high practicability and reliability and is suitable for the diverse needs of power system monitoring.

[0138] S240, determine a feature matrix based on the multiple time-domain features and the multiple frequency-domain features, and solve for a feature contribution rate based on the feature matrix, where the feature contribution rate characterizes the importance of each of the time-domain features or the frequency-domain features, so as to constrain the feature dimension of a pre-constructed arc prediction model.

[0139] Among them, a matrix is formed by combining the extracted multiple time-domain features and multiple frequency-domain features, and this matrix is the feature matrix. Suppose a time-domain features and b frequency-domain features are extracted, then the size of the feature matrix is (a + b)×n, where n is the number of samples, that is, the number of current data obtained by the current acquisition device and sent to the processor for analysis.

[0140] Among them, the feature contribution rate can be determined based on the feature matrix by the principal component analysis method. Through linear transformation, the original feature space is mapped to a new coordinate system, so that each coordinate axis (i.e., the principal component) in the new coordinate system is arranged in order of the magnitude of the data change amplitude. In the new coordinate system, the first principal component can explain the largest variance in the data, the second principal component explains the largest part of the remaining variance, and so on. In some cases, the feature contribution rate can also be determined by various methods such as mutual information, feature importance analysis, regularization analysis, variance analysis, neural network analysis, etc., which are not limited herein.

[0141] In a possible embodiment, the multiple time-domain features include at least one of a root mean square parameter, a variance parameter, a kurtosis factor parameter, a pulse factor parameter, and a current change rate parameter, and the multiple frequency-domain features include at least one of a harmonic current content rate and a total harmonic distortion rate of the current; the determining the feature matrix based on the multiple time-domain features and the multiple frequency-domain features, and solving for the feature contribution rate based on the feature matrix includes: determining an n-row and m-column feature matrix based on the multiple time-domain features and the multiple frequency-domain features, where m corresponds to the number of the time-domain features and / or the frequency-domain features, and n corresponds to the number of the current data; determining a correlation coefficient matrix based on the n-row and m-column feature matrix; determining a plurality of eigenvalues based on the correlation coefficient matrix, and determining the feature contribution rate corresponding to each of the time-domain features and / or the frequency-domain features based on the plurality of eigenvalues.

[0142] Among them, there may be a time-domain features and b frequency-domain features, and the above m = a + b. Based on the a time-domain features and b frequency-domain features corresponding to each sample, an n×m feature matrix is formed. Each row represents the m features corresponding to one sample, specifically including the values of a time-domain features and the values of b frequency-domain features, and there are n rows in total. In this embodiment, m = a + b = 8. Specifically, the n×m feature matrix can be determined according to the twentieth formula:

[0143] 。

[0144] Among them, the twentieth formula calculates the correlation coefficient corresponding to the feature matrix. Specifically, first, the correlation coefficient can be calculated according to the following twenty-first formula :

[0145] ;

[0146] Then, based on the correlation coefficient calculated by the above twenty-first formula a correlation coefficient matrix R is established. See the following twenty-second formula:

[0147] ;

[0148] Among them, the above 、 represent the mean values of time-domain features or frequency-domain features 、 .

[0149] Among them, based on the above correlation coefficient matrix R, the characteristic equation is solved through , where I is the identity matrix with the same dimension as the correlation coefficient matrix R, and the dimension here is 8×8, is the eigenvalue, and finally at least one value is obtained, that is , and here it can be 、 、 、 、 、 、 、 ; Then the eigenvalues are sorted so that is satisfied; Based on the obtained , the characteristic contribution rate is calculated. Specifically, it can be calculated through the following twenty-third formula:

[0150] ;

[0151] Among them, is the characteristic contribution rate corresponding to 、corresponding to .

[0152] It can be seen that in this embodiment, through eigenvalue decomposition and contribution rate calculation, the importance of each feature can be quantified, which is applicable to multi-feature analysis and feature screening scenarios, so as to obtain higher accuracy. Feature optimization reduces the input layer of the algorithm and improves the detection speed and efficiency.

[0153] S250. Determine an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model. The arc prediction result includes a first prediction result. If the arc prediction result is the first prediction result, a warning message is sent. The arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection period.

[0154] Among them, based on the feature contribution rate, at least one feature with a larger contribution rate can be determined, and then the pre-constructed arc prediction model is optimized and constructed based on the number of features with a larger contribution rate, so that the dimension of the arc prediction model is more concise under the current usage conditions. After optimizing the pre-constructed arc prediction model, the analysis of the arc generation situation is carried out based on the current data of the first circuit module, and a first prediction result is obtained. The first prediction result indicates that an arc may occur in the current circuit. In some cases, a second prediction result can also be obtained, and the second prediction result indicates that it is less likely that an arc will occur in the current circuit. When the first prediction result is obtained, a warning message is sent. If the second prediction result is obtained, no message may be sent or a message indicating that no arc occurs may be sent.

[0155] In a possible embodiment, the determining the arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model includes: determining the principal component features and the number of principal component features based on the feature contribution rate, where the number of principal component features is the number of the principal component features; updating the pre-constructed arc prediction model based on the number of principal component features so that the pre-constructed arc prediction model conforms to the current prediction working condition; and determining the arc prediction result based on the circuit data and the pre-constructed arc prediction model.

[0156] Among them, a cumulative contribution rate threshold (such as 90% or 95%) is set, and the first k features with a cumulative contribution rate reaching this threshold are selected as the principal component features. If the cumulative contribution rate of the first 3 features is 0.3 + 0.25 + 0.2 = 0.75 (75%) and does not reach 90%, then features are continuously added. If the cumulative contribution rate of the first 4 features is 0.75 + 0.15 = 0.9 (90%), then the first 4 features are selected as the principal component features. The number of principal component features (k) is the number of the selected features. The principal components corresponding to the contribution rates exceeding this cumulative contribution rate threshold can effectively cover the main information of a time-domain features and b frequency-domain features. In this embodiment, the cumulative contribution rate threshold can be set to 85%.

[0157] For example, please refer to Figure 3 , Figure 3 which is a coordinate schematic diagram of a feature contribution rate provided by an embodiment of the present application. In this embodiment, if , , , , , , , . Since the preset cumulative contribution rate threshold is 85%, and the cumulative contribution rate of the first three principal components reaches more than 90%, using the first three principal components can cover the main information of the selected 8 time-domain and frequency-domain characteristic parameters.

[0158] Among them, please refer to Figure 4 , Figure 4 , which is a schematic diagram of the model architecture of an arc prediction model provided by an embodiment of the present application. As shown in Figure 4 , the above-preconstructed arc prediction model can be a multi-layer neural network model. Specifically, it can be a three-layer BP neural network model, which can include an input layer, a hidden layer, and an output layer. Among them, the number of neurons in the input layer is multiple. In this embodiment, the number of neurons in the input layer can be determined by the number of principal component features determined above. Specifically, the number of neurons in the input layer can be equal to the number of principal component features; the number of neurons in the output layer can also be multiple. Specifically, it is associated with the situation of the output result. For example, if the output arc prediction result includes two situations, the first output result and the second output result, then the number of neurons in the output layer is 2. If there are multiple situations in the output arc prediction result, the number of neurons in the output layer can also be correspondingly updated to the number of types of the above arc prediction results. The randomly initialized weight from the input layer to the hidden layer of this arc prediction model can be and the weight from the hidden layer to the output layer can be . Initialize the bias and . Then from the input layer to the hidden layer: , where f is the ReLU activation function. From the hidden layer to the output layer: .

[0159] Among them, the number of neurons in the hidden layer is also associated with the above number of principal component features. Specifically, it can be determined by the twenty-fourth formula:

[0160] ;

[0161] Among them, is the number of neurons in the hidden layer, p is is a number between 1 and 10; among them, the above is an empirical formula for estimating the benchmark value of the number of neurons in the hidden layer. is used for fine-tuning to avoid insufficient model complexity or overfitting. The experiment selects the optimal value, for example, performs a grid search within the range of 1 to 10. If the model is underfitting, can be increased.; If overfitting occurs, it can be reduced .

[0162] Among them, the output result of the model, that is, the arc prediction result, includes a first prediction result and a second prediction result. The first prediction result indicates that an arc may be generated, and the second prediction result indicates that an arc may not be generated. Specifically, the first prediction result can be represented by 1, and the second prediction result can be represented by 0.

[0163] It should be noted that the hit rate of this model is calculated every time it runs for a period of time. The hit rate is calculated according to the following formula:

[0164] ;

[0165] Where is the number of times of accurately identified low-voltage arc occurrences, and B is the total number of system model identifications. Please refer to Figure 5 , Figure 5 is a schematic diagram of the hit rate of an arc prediction model provided by an embodiment of the present application. As Figure 5 shown, in the results output by the model, "0" represents a fault, that is, the first prediction result, and "1" represents normal, that is, the second prediction result. The value 103 in the upper left cell represents the number of samples that are actually normal and correctly predicted as normal. The value 2 in the upper right cell represents the number of samples that are actually normal but wrongly predicted as faults. The value 0 in the lower left cell, that is, the number of samples that are actually faulty but wrongly predicted as normal. The value 95 in the lower right cell refers to the number of samples that are actually faulty and correctly predicted as faults. The color depth represents the numerical size, and the color scale on the right indicates that the darker the color, the larger the value, which can intuitively reflect the numerical situation of each cell. It can be seen from this matrix that the overall prediction effect of the arc prediction model is good and the misjudgment situation is less.

[0166] It can be seen that in this embodiment, through reasonable structural design, feature optimization, and dynamic adjustment of the number of hidden layer neurons, the number of layers of the model is streamlined to eliminate redundant parts, the computational complexity is low, the amount of calculation is reduced, it is suitable for real-time monitoring tasks, and the arc is more efficient, accurate, and flexible during model discrimination, improving the analysis efficiency and robustness.

[0167] In a possible embodiment, the circuit data further includes circuit environment data, and the circuit environment data includes at least one of circuit temperature data, circuit pressure data, and circuit light data within a target period; determining an arc prediction result based on the circuit data and the pre-constructed arc prediction model includes: performing data standardization processing based on the circuit temperature data and / or circuit pressure data and / or circuit light data to determine at least one standard environment data; performing linear fitting on the environment data determined based on the standard environment data to determine a curve and multiple slope parameters corresponding to at least one of the circuit environment data; determining a deviation parameter based on the multiple slope parameters and a preset slope parameter; determining an environment impact factor based on the deviation parameter; and determining an arc prediction result based on the environment impact factor, the current data, and the pre-constructed arc prediction model.

[0168] Among them, the circuit environment data includes circuit temperature data, circuit pressure data, and circuit light data within a target period. Assume that the environment data is T (temperature), P (pressure), and L (light). Perform normalization processing on each type of environment data, and the formula is:

[0169] ;

[0170] where X is the original environment data, and are the minimum and maximum values of the data respectively. The standardized data is denoted as 、 、 . Use the standardized environment data as new input features and form an input vector together with the current data: , where x is the current data.

[0171] Among them, perform linear fitting on each type of environment data (temperature, pressure, light), and the fitting formula can be: y = kt + b, or other curve formulas. Among them, y is the environment data (such as temperature T); t is time or sampling point; k is the slope parameter, indicating the change trend of the environment data; b is the intercept. Calculate the slope parameter k of each type of environment data through the least squares method or other fitting methods T 、k P 、k L . Then determine a deviation parameter based on a preset normal slope parameter. Among them, the preset normal slope parameter can be a parameter range. For example, the normal temperature change slope 、the normal pressure change slope 、the normal light change slope . Then calculate the deviation δ T 、δ P 、δ L, which can be determined based on the following formula:

[0172] , , ;

[0173] Then, the total deviation parameter is determined based on the above deviation parameters. Specifically, the environmental impact factor can be comprehensively determined based on the deviation magnitudes of each deviation to characterize the degree of environmental impact. Specifically, different weight coefficients can be assigned to each deviation to obtain the final environmental impact factor G(t). The environmental impact factor G(t) is used as a new input feature and together with the current data to form an extended input vector: . The number of neurons in the original input layer is the number of principal component features, and the number of neurons in the extended input layer is the number of principal component features + 1 (the environmental impact factor is added).

[0174] Among them, the arc prediction result includes a first output result and a second output result. In addition to outputting 1, the first output result also outputs the current environmental impact factor, which is used to characterize the possibility or severity of the arc transmission.

[0175] It can be seen that in this embodiment, by integrating circuit environment data (temperature, pressure, light) into the neural network model and combining steps such as linear fitting, deviation calculation, and environmental impact factor, an arc prediction model with multi-dimensional feature fusion is realized. This method not only improves the accuracy and robustness of the model but also enhances the adaptability of the model to complex environments, and has high engineering practical value.

[0176] It can be seen that through the embodiments of the present application, the processor receives circuit data within a target period from the current acquisition device. The circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period. Based on the current data, an effective current component and a noise current component are determined, and a reconstructed signal is determined based on the effective current component, the noise current component, and a wavelet threshold denoising operation. The reconstructed signal is a current signal after denoising. Based on the reconstructed signal, a plurality of time-domain features and a plurality of frequency-domain features are determined. The frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal. Based on the plurality of time-domain features and the plurality of frequency-domain features, a feature matrix is determined, and a feature contribution rate is obtained by solving the feature matrix. The feature contribution rate characterizes the importance of each time-domain feature or frequency-domain feature to constrain the feature dimension of a pre-constructed arc prediction model. Based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model, an arc prediction result is determined. The arc prediction result includes a first prediction result. If the arc prediction result is the first prediction result, a warning message is sent. The arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection period. In this way, through empirical mode decomposition and wavelet threshold denoising of the current data of low-voltage power consumption monitoring, the arc prediction model has strong anti-interference ability and can achieve good results in harsh environments. By performing time-domain and frequency-domain features on the denoised current data, the features of arc faults can be excavated, thereby obtaining higher accuracy. Feature optimization reduces the input layer of the algorithm and improves the detection speed and efficiency.

[0177] Please refer to Figure 6 , Figure 6 FIG. 6 is a schematic structural diagram of an arc prediction device for a low-voltage circuit proposed by an embodiment of the present application, which is applied to a processor of a low-voltage circuit system. The low-voltage circuit system includes a first circuit module and a second circuit module. The first circuit module includes an actual scenario circuit, and the second circuit module includes an analog fault circuit. The low-voltage circuit system further includes a current acquisition device. The arc prediction device 600 for the low-voltage circuit includes:

[0178] A data acquisition module 610, configured to receive circuit data within a target period from the current acquisition device. The circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period;

[0179] The first determination module 620 is configured to determine an effective current component and a noise current component based on the current data, and determine a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation, where the reconstructed signal is a current signal after noise reduction;

[0180] The second determination module 630 is configured to determine a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal, where the frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal;

[0181] The third determination module 640 is configured to determine a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and solve a feature contribution rate based on the feature matrix, where the feature contribution rate characterizes the importance of each of the time-domain features or the frequency-domain features, so as to constrain the feature dimension of a pre-constructed arc prediction model;

[0182] The arc prediction module 650 is configured to determine an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model, where the arc prediction result includes a first prediction result, and if the arc prediction result is the first prediction result, a warning message is sent; the arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection period.

[0183] In a possible embodiment, in terms of determining the effective current component and the noise current component based on the current data, the first determination module 620 is specifically configured to:

[0184] Determine normalized current data based on the current data;

[0185] Perform modal decomposition on the normalized current data to determine the j-th order intrinsic mode component;

[0186] Determine a plurality of signal parameters based on the j-th order intrinsic mode component and / or the normalized current data, where the signal parameters include the j-th order variance of the j-th order intrinsic mode component and / or the variance of the current data;

[0187] Determine a variance contribution rate based on the signal parameters, and determine the effective current component and the noise current component based on the variance contribution rate.

[0188] In a possible embodiment, in terms of determining the reconstructed signal based on the effective current component, the noise current component, and the wavelet threshold denoising operation, the second determination module 630 is specifically configured to:

[0189] Determine a first type of frequency component based on the noise current component, where the first type of frequency component includes high-frequency components;

[0190] Determine the j-th order modal component threshold based on the j-th order eigenmode component; and determine the j-th order denoised modal component based on the first type of frequency component and the j-th order modal component threshold;

[0191] Determine the reconstructed signal based on the j-th order denoised modal component and the effective current component.

[0192] In a possible embodiment, the second determination module 630, in terms of determining the j-th order modal component threshold based on the j-th order eigenmode component, is specifically configured to:

[0193] Determine the first-order eigenmode component based on the j-th order eigenmode component, where the first-order eigenmode component is the eigenmode component when j = 1;

[0194] Determine the first-order energy signal based on the first-order eigenmode component and the following formula:

[0195] ;

[0196] where, represents taking the median, represents taking the absolute value, is the first-order eigenmode component, and 0.6745 is the 75% quantile of the standard normal distribution;

[0197] Determine the j-th order energy signal based on the first-order energy signal and a preset first empirical parameter;

[0198] Determine the j-th order modal component threshold based on the j-th order energy signal and a plurality of preset second empirical parameters.

[0199] In a possible embodiment, the plurality of time-domain features include at least one of a root mean square parameter, a variance parameter, a kurtosis factor parameter, an impulse factor parameter, and a current change rate parameter, and the plurality of frequency-domain features include at least one of a harmonic current content rate and a total harmonic distortion rate of the current; the third determination module 640, in terms of determining a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features and solving for the feature contribution rate based on the feature matrix, is specifically configured to:

[0200] Determine an n-row and m-column feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, where m corresponds to the number of the time-domain features and / or the frequency-domain features, and n corresponds to the number of the current data;

[0201] Determine a correlation coefficient matrix based on the n-row and m-column feature matrix;

[0202] Determine a plurality of eigenvalues based on the correlation coefficient matrix, and determine the feature contribution rate corresponding to each of the time-domain features and / or the frequency-domain features based on the plurality of eigenvalues.

[0203] In a possible embodiment, the arc prediction module 650 is specifically configured to determine an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model as follows:

[0204] Determine the principal component features and the number of principal component features based on the feature contribution rate, where the number of principal component features is the number of the principal component features;

[0205] Update the pre-constructed arc prediction model based on the number of principal component features, so that the pre-constructed arc prediction model conforms to the current prediction working condition;

[0206] Determine the arc prediction result based on the circuit data and the pre-constructed arc prediction model.

[0207] In a possible embodiment, the circuit data further includes circuit environment data, and the circuit environment data includes at least one of circuit temperature data, circuit pressure data, and circuit illumination data within a target period; the arc prediction module 650 is specifically configured to determine an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model as follows:

[0208] Perform data standardization processing based on the circuit temperature data and / or circuit pressure data and / or circuit illumination data to determine at least one standard environment data;

[0209] Perform linear fitting on the environment data determined based on the standard environment data to determine a curve and multiple slope parameters corresponding to at least one of the circuit environment data;

[0210] Determine a deviation parameter based on the multiple slope parameters and a preset slope parameter;

[0211] Determine an environment impact factor based on the deviation parameter;

[0212] Determine the arc prediction result based on the environment impact factor, the current data, and the pre-constructed arc prediction model.

[0213] It should be noted that, for the specific functional implementation manner of the arc prediction device 600 for the low-voltage circuit, refer to the above Figure 2Description of the arc prediction method for the low-voltage circuit shown. For example, the data acquisition module 610 is used to implement the relevant content of executing S210. Each unit or module in the arc prediction device 600 for the low-voltage circuit can be separately or all combined into one or several other units or modules to form, or some of the units or modules can be further split into multiple smaller units or modules in terms of function to form, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present invention. The above units or modules are divided based on logical functions. In practical applications, the function of one unit (or module) is realized by multiple units (or modules), or the functions of multiple units (or modules) are realized by one unit (or module).

[0214] It can be seen that for the arc prediction device for the low-voltage circuit described in the embodiments of the present application, by receiving the circuit data within the target period from the current acquisition device, the circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module. The target period includes the current period; determining the effective current component and the noise current component based on the current data, and determining the reconstructed signal based on the effective current component, the noise current component, and the wavelet threshold denoising operation. The reconstructed signal is the current signal after noise reduction; determining a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal. The frequency-domain features include the waveform distortion features used to characterize the frequency domain of the reconstructed signal; determining the feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and solving the feature contribution rate based on the feature matrix. The feature contribution rate characterizes the importance of each time-domain feature or frequency-domain feature to constrain the feature dimension of the pre-constructed arc prediction model; determining the arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model. The arc prediction result includes the first prediction result. If the arc prediction result is the first prediction result, a warning message is sent; the arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection period. In this way, by performing empirical mode decomposition and wavelet threshold denoising on the current data of low-voltage power consumption monitoring, the arc prediction model has a strong anti-interference ability and can achieve good results in harsh environments. By performing time-domain and frequency-domain features on the denoised current data, the characteristics of arc faults can be mined, thereby obtaining higher accuracy. Feature optimization reduces the input layer of the algorithm and improves the detection speed and efficiency.

[0215] The embodiments of the present application also provide a computer storage medium. Among them, this computer storage medium stores a computer program for electronic data exchange, and this computer program enables the computer to execute part or all of the steps of any method recorded in the above method embodiments. The above computer includes an electronic device.

[0216] The embodiments of the present application also provide a computer program product. The computer program product includes a non-transitory computer-readable storage medium storing a computer program. The computer program is operable to cause a computer to execute some or all of the steps of any of the methods described in the foregoing method embodiments. The computer program product may be a software installation package, and the computer includes an electronic device.

[0217] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps may be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0218] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0219] In several embodiments provided by the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the above division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0220] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0221] In addition, the functional units in the various embodiments of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0222] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer electronic device (which can be a personal computer, an electronic device, or a network electronic device, etc.) to execute all or part of the steps of the above methods in various embodiments of this application. The aforementioned memory includes: various media such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), external hard drives, magnetic disks, or optical discs that can store program codes.

[0223] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable memory, and the memory can include: flash drives, read-only memories (English: Read-Only Memory, abbreviated: ROM), random access memories (English: Random Access Memory, abbreviated: RAM), magnetic disks, or optical discs, etc.

[0224] The above has introduced the embodiments of this application in detail. Specific examples are used in this article to elaborate on the principles and implementation manners of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. An arc prediction method for a low-voltage circuit, characterized in that, A processor applied to a low-voltage circuit system, the low-voltage circuit system includes a first circuit module and a second circuit module, the first circuit module includes an actual scenario circuit, the second circuit module includes an analog fault circuit, the low-voltage circuit system further includes a current acquisition device, and the method includes: Receiving circuit data within a target period from the current acquisition device, the circuit data includes current data, the current data includes the current data of the first circuit module and / or the second circuit module, and the target period includes the current period; Determining an effective current component and a noise current component based on the current data, and determining a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation, the reconstructed signal is a current signal after denoising; Determining a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal, and the frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal; Determining a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and solving a feature contribution rate based on the feature matrix, the feature contribution rate characterizes the importance of each of the time-domain features or the frequency-domain features, so as to constrain the feature dimension of a pre-constructed arc prediction model; Determining an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model, the arc prediction result includes a first prediction result, if the arc prediction result is the first prediction result, then a warning message is sent; the arc prediction result is used to characterize whether an arc may be generated in the low-voltage circuit during the connection period; wherein, the circuit data further includes circuit environment data, the circuit environment data includes at least one of circuit temperature data, circuit pressure data, and circuit illumination data within the target period, and determining the arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model includes: determining a principal component feature and a principal component feature number based on the feature contribution rate, the principal component feature number is the number of the principal component features; updating the pre-constructed arc prediction model based on the principal component feature number, so that the pre-constructed arc prediction model conforms to the current prediction working condition; performing data standardization processing on the circuit temperature data and / or the circuit pressure data and / or the circuit illumination data to determine at least one standard environment data; determining a curve and a plurality of slope parameters corresponding to at least one of the circuit environment data by performing linear fitting on the standard environment data to determine the environment data; determining a deviation parameter based on the plurality of slope parameters and a preset slope parameter; determining an environment influence factor based on the deviation parameter; determining the arc prediction result based on the environment influence factor, the current data, and the pre-constructed arc prediction model.

2. The method according to claim 1, wherein The determining the effective current component and the noise current component based on the current data includes: Determining normalized current data based on the current data; Performing modal decomposition on the normalized current data to determine the j-th order intrinsic mode component; Determine a plurality of signal parameters based on the j-th eigenmode component and / or the normalized current data, where the signal parameters include the j-th variance of the j-th eigenmode component and / or the variance of the current data; Determine a variance contribution rate based on the signal parameters, and determine the effective current component and the noise current component based on the variance contribution rate.

3. The method according to claim 2, wherein The determining the reconstructed signal based on the effective current component, the noise current component, and the wavelet threshold denoising operation includes: Determine a first type of frequency component based on the noise current component, where the first type of frequency component includes high-frequency components; Determine a j-th modal component threshold based on the j-th eigenmode component; and determine a j-th denoised modal component based on the first type of frequency component and the j-th modal component threshold; Determine the reconstructed signal based on the j-th denoised modal component and the effective current component.

4. The method according to claim 3, characterized in that The determining the j-th modal component threshold based on the j-th eigenmode component includes: Determine a first-order eigenmode component based on the j-th eigenmode component, where the first-order eigenmode component is the eigenmode component when j = 1; Determine a first-order energy signal based on the first-order eigenmode component and the following formula: ； Among them, represents finding the median value, represents finding the absolute value, is the first-order eigenmode component, and 0.6745 is the 75% quantile of the standard normal distribution; Determine a j-th energy signal based on the first-order energy signal and a preset first empirical parameter; Determine the j-th modal component threshold based on the j-th energy signal and a plurality of preset second empirical parameters.

5. The method according to claim 1 or 4, characterized in that, The plurality of time-domain features include at least one of a root mean square parameter, a variance parameter, a kurtosis factor parameter, an impulse factor parameter, and a current change rate parameter, and the plurality of frequency-domain features include at least one of a harmonic current content rate and a total harmonic distortion rate of the current; The determining a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and solving to obtain a feature contribution rate based on the feature matrix includes: Determine a feature matrix with n rows and m columns based on the plurality of time-domain features and the plurality of frequency-domain features, where m corresponds to the number of the time-domain features and / or the frequency-domain features, and n corresponds to the number of the current data; Determine a correlation coefficient matrix based on the n-row and m-column feature matrix; Determine a plurality of eigenvalues based on the correlation coefficient matrix, and determine the feature contribution rate corresponding to each of the time-domain features and / or the frequency-domain features based on the plurality of eigenvalues.

6. An arc prediction device for a low-voltage circuit, characterized in that, A processor applied to a low-voltage circuit system, where the low-voltage circuit system includes a first circuit module and a second circuit module, the first circuit module includes an actual scenario circuit, the second circuit module includes an analog fault circuit, and the low-voltage circuit system further includes a current acquisition device, and the device includes: A data acquisition module, configured to receive circuit data within a target period from the current acquisition device, where the circuit data includes current data, and the current data includes the current data of the first circuit module and / or the second circuit module, and the target period includes the current period; A first determination module, configured to determine an effective current component and a noise current component based on the current data, and determine a reconstructed signal based on the effective current component, the noise current component, and a wavelet threshold denoising operation, where the reconstructed signal is a current signal after noise reduction; A second determination module, configured to determine a plurality of time-domain features and a plurality of frequency-domain features based on the reconstructed signal, where the frequency-domain features include waveform distortion features for characterizing the frequency domain of the reconstructed signal; A third determination module, configured to determine a feature matrix based on the plurality of time-domain features and the plurality of frequency-domain features, and solve a feature contribution rate based on the feature matrix, where the feature contribution rate characterizes the importance of each of the time-domain features or the frequency-domain features, so as to constrain the feature dimension of a pre-constructed arc prediction model; An arc prediction module, configured to determine an arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model, where the arc prediction result includes a first prediction result, and if the arc prediction result is the first prediction result, a warning message is sent; the arc prediction result is used to characterize whether an arc is likely to occur in the low-voltage circuit during the connection sequence period; where the circuit data further includes circuit environment data, and the circuit environment data includes at least one of circuit temperature data, circuit pressure data, and circuit illumination data within a target period, and determining the arc prediction result based on the circuit data, the feature contribution rate, and the pre-constructed arc prediction model includes: determining a principal component feature and a principal component feature quantity based on the feature contribution rate, where the principal component feature quantity is the number of the principal component features; updating the pre-constructed arc prediction model based on the principal component feature quantity, so that the pre-constructed arc prediction model conforms to the current prediction working condition; performing data standardization processing on the circuit temperature data and / or the circuit pressure data and / or the circuit illumination data to determine at least one standard environment data; determining curve and a plurality of slope parameters corresponding to at least one of the circuit environment data by performing linear fitting on the environment data determined based on the standard environment data; determining a deviation parameter based on the plurality of slope parameters and a preset slope parameter; determining an environment influence factor based on the deviation parameter; and determining the arc prediction result based on the environment influence factor, the current data, and the pre-constructed arc prediction model.

7. A computer-readable storage medium, characterized in that, Stores an arc prediction program for a low-voltage circuit, including execution instructions, and when a processor of an electronic device executes the execution instructions, the processor executes the method according to any one of claims 1 to 5.

8. An electronic device, characterized in that, Includes a processor and a memory storing execution instructions, where the memory stores one or more programs; when the processor executes the execution instructions stored in the memory, the processor executes the method according to any one of claims 1 to 5.

Images