Low-voltage electric shock residual current signal feature extraction method and related device

Abstract

The invention provides a low-voltage electric shock residual current signal feature extraction method and a related device, and belongs to the technical field of low-voltage electric shock residual current feature identification, and the method comprises the steps: collecting electric shock fault information; performing adaptive decomposition based on the fault information, and determining K intrinsic mode function (IMF) components; based on the intrinsic mode function IMF component, using an optimal mode component screening method to screen out an optimal component; and extracting electric shock residual current signal features based on the optimal component. According to the invention, after the IPOA-VMD decomposition algorithm is utilized to carry out preliminary noise reduction on the fault information signal acquired in real time, MCKD is utilized to carry out noise reduction on the optimal IMF component, and TEO demodulation analysis is adopted to realize fault feature extraction in a strong background noise environment.

Description

Technical Field

[0001] The present invention relates to the technical field of low-voltage electric shock residual current feature recognition, and in particular to a low-voltage electric shock residual current signal feature extraction method and related devices. Background Art

[0002] With the gradual improvement of domestic power facilities, the power supply and consumption of the whole society has increased rapidly, and the power safety of low-voltage power grids has become a focus. Residual current protection device is a key protection device in the low-voltage power grid system. It can improve the safety and reliability of the low-voltage distribution system. Therefore, it has been widely used and promoted in recent years. However, the existing residual current protection device is based on the effective value of the residual current in the low-voltage power supply circuit as the action criterion, which often leads to false operation or refusal to operate. Therefore, there is an urgent need to study high-reliability residual current protection devices with more reasonable action criteria.

[0003] The transient signals generated when a power system fails are nonlinear and non-stationary. When an electric shock fault occurs, the residual current contains rich information about the operating status and fault. When a low-voltage electric shock fault occurs, there are a large number of harmonics, interharmonics and attenuated DC components in the transient signal of the residual current. How to quickly and accurately extract the digital characteristics of the residual current during an electric shock fault is the key to improving the speed and accuracy of the residual current protection action.

[0004] At present, there are many feature quantities used to characterize transient signals, which can be summarized as follows: time domain features, frequency domain features, time-frequency domain features, nonlinear parameter features, etc. Due to the nonlinear and non-stationary nature of electric shock fault signals, traditional time domain and frequency domain feature extraction methods often cannot produce good results. Among them, the method of fault signal envelope demodulation is widely used. This method has the advantages of enhancing transient characteristics and highlighting the impact component of fault signals. However, the modulated signal is often mixed with carrier signals and noise, which makes the feature extraction effect less than ideal.

[0005] Therefore, there is a technical problem in the prior art that the effect of extracting the characteristics of the low-voltage electric shock residual current signal is not ideal. Summary of the invention

[0006] The main purpose of the present invention is to provide a method and related device for extracting characteristics of low-voltage electric shock residual current signals, so as to solve the technical problem that the extraction effect of characteristics of low-voltage electric shock residual current signals in the related art is not ideal.

[0007] In order to achieve the above-mentioned purpose, according to one aspect of the present invention, a method for extracting features of a low-voltage electric shock residual current signal is provided, comprising the following steps: collecting electric shock fault information; performing adaptive decomposition based on the fault information to determine K intrinsic mode function (IMF) components; using an optimal mode component screening method to screen out optimal components based on the intrinsic mode function (IMF) components; and extracting features of the electric shock residual current signal based on the optimal components.

[0008] Furthermore, the adaptive decomposition is performed based on the fault information to determine K intrinsic modulus function IMF components, including: improving the Pelican optimization algorithm POA using circular mapping and lens imaging reverse learning strategy to obtain an improved Pelican optimization algorithm IPOA; optimizing the penalty factor and the number of modal layers in the VMD algorithm using IPOA to obtain an improved VMD algorithm; and adaptively decomposing the electric shock signal using the improved VMD algorithm to obtain K IMF components.

[0009] Furthermore, the improved Pelican optimization algorithm POA is improved by using the circular mapping and lens imaging reverse learning strategy to obtain the improved Pelican optimization algorithm IPOA, including:

[0010] A circular mapping is used to initialize the position of the pelican population so that each pelican is evenly distributed in the entire search space. The circular mapping definition is: ,

[0011] Among them, numi is the i-th chaotic sequence number, and mod is the remainder operation;

[0012] The expression for initializing the IPOA population is as follows:

[0013] ,

[0014] Among them, num ij is the value of the chaotic sequence generated by the circular mapping on the dimension of the corresponding pelican individual, i=1,2,3...,N, j=1,2,3...,m, N is the number of population members, m is the number of problem variables, x ij is the value of the jth variable specified by the i-th candidate solution, l j is the jth lower limit, u j is the jth upper limit of the problem variable;

[0015] Use population moment X to represent the pelican population:

[0016] ,

[0017] Among them, X iis the position of the i-th pelican, each row of the matrix represents a candidate solution, and the columns of the matrix represent the proposed values ​​of the problem variables;

[0018] The fitness function for a given problem is evaluated based on each candidate solution:

[0019] ,

[0020] Where F is the fitness function vector, and F i is the fitness function value of the i-th candidate solution;

[0021] The Pelican algorithm enters the first stage of the search and begins to explore prey. The Pelican's position strategy formula in the exploration stage is:

[0022] ,

[0023] in, is the position of the jth dimension of the ith pelican after the first phase update, rand is a random number in the interval [0,1], I is a random number of 1 or 2, I is randomly selected by each iteration and each member, p j is the j-th dimension position of the target, F p is the fitness function value;

[0024] After the first stage, if the value of the objective function improves at the current position, the position is accepted and updated:

[0025] ,

[0026] in, is the new position of the i-th pelican, The fitness function value of the i-th pelican after the position is updated in the first stage;

[0027] The second stage is the pelican hunting development stage, and the mathematical simulation formula is:

[0028]

[0029] in, is the position of the jth dimension of the ith pelican after the second stage based update, rand is a random number in the interval [0,1], R is a random number between 0 and 2, t is the current number of iterations, and T is the maximum number of iterations;

[0030] The position of each pelican is optimized using the inverse learning strategy of dimension-by-dimension lens imaging. The position of each dimension is mapped into space to obtain the reverse position. The expression of lens imaging is:

[0031] ,

[0032] Among them, aj , b j Indicates the search range [a j , b j ], represents the j-th dimension position of the i-th pelican after the second stage update, which is represented as the projection of an object with height h on the x-axis;

[0033] Order on the top , which can be transformed into:

[0034] ;

[0035] When n=1, a fixed reverse solution is obtained :

[0036] ;

[0037] This formula (10) is called the termination condition formula. The fitness value of the previous position is compared with the fitness value after reverse learning. If the fitness value after reverse learning is better than the previous position, the reverse position is selected to replace the previous position; otherwise, the original position is retained for the next generation. The improved second-stage Pelican position update strategy is:

[0038] ,

[0039] in, is the reverse solution of the new position in the jth dimension after the second stage of the i-th pelican, is the objective function value after the second stage of reverse learning;

[0040] The new position update strategy mentioned above is used to promote POA to escape from the local optimal area, forming an improved Pelican optimization algorithm IPOA.

[0041] Furthermore, the penalty factor and the number of modal layers in the VMD algorithm are optimized by using IPOA to obtain an improved VMD algorithm including:

[0042] Using the enhanced Lagrangian function, the restricted variation problem of the VMD technique is transformed into an unrestricted variation problem, which is expressed as follows:

[0043] ,

[0044] Where L represents the constraint function, {u k} represents the kth IMF component generated from VMD decomposition, f represents the original signal, {ω k} represents the center frequency associated with each IMF component, δ(t) represents the unit impulse function, * represents the convolution operation, λ is the Lagrange multiplier, α is the penalty factor, To find the sign of partial derivative, t refers to the number of algorithm iterations, and the initial value of t is 1;

[0045] The alternating direction method of multipliers (ADMMs) is used to solve the problem of solving unconstrained variational models, {ω k}、{u k The update formula of} and λ, as well as the iteration termination condition are as follows:

[0046] ,

[0047] in , and Respectively, , and the Fourier transform corresponding to f, n is the value of the signal at the nth moment during the processing of this stage, ε is the convergence accuracy, Represents the square of the Euclidean coefficient of the IMF component. Use the above steps to update repeatedly until the termination condition formula is met, and finally obtain K IMF components;

[0048] IPOA is used to optimize the penalty factor α and the number of modal layers K in the VMD algorithm, and the optimal parameter combination is adaptively obtained.

[0049] Furthermore, the adaptively improved VMD algorithm decomposition of the electric shock signal to obtain K IMF components includes: S501, setting the maximum number of iterations, spatial dimension, population size, decomposition layer and penalty factor of IPOA, and using the circular mapping strategy to initialize the position of the population; S502, decomposing the fault information signal collected in real time by the improved VMD algorithm to generate multiple IMF components, calculating the envelope spectral entropy ESE value of each IMF component, and determining the fitness function of the global exploration by selecting the component with the lowest ESE value; S503, after each round of iteration, calculating the ESE value corresponding to each group of parameter combinations, and comparing them with the current ESE value. If it is less than the current ESE value, the current ESE value will be updated; S504, determine whether the stopping condition of the iteration is reached. If the maximum number of iterations is not reached, set t=t+1, where t refers to the number of algorithm iterations and the initial value of t is 1. The optimal individual position of the pelican is updated using the inverse learning strategy of dimension-by-dimensional lens imaging, where the pelican population position updated in the previous round of iteration will be used as the initial population position for the next round. Repeat the above steps S503-S504 until the iteration condition is reached. Assign the optimal parameter combination output by IPOA to VMD, and adaptively decompose to obtain K IMF components.

[0050] Furthermore, the optimal component is screened out using the optimal mode component screening method based on the intrinsic mode function IMF component, including: determining the Gini coefficient SEGI and kurtosis based on the intrinsic mode function IMF component; and screening out the optimal component based on the Gini coefficient SEGI and kurtosis.

[0051] According to another aspect of the present invention, a low-voltage electric shock residual current signal feature extraction device is provided, including: an acquisition module, used to collect electric shock fault information; a determination module, which performs adaptive decomposition based on the fault information to determine K intrinsic mode function IMF components; a screening module, which uses the optimal mode component screening method based on the intrinsic mode function IMF components to screen out the optimal component; and an extraction module, which extracts the electric shock residual current signal feature based on the optimal component.

[0052] According to another aspect of the present invention, the present invention also provides a seat belt storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the steps of the low-voltage electric shock residual current signal feature extraction method described in any of the above items are implemented.

[0053] According to another aspect of the present invention, the present invention also provides an electronic device, comprising at least a memory and a processor, wherein a computer program is stored in the memory, and wherein the processor implements the steps of any of the aforementioned methods for extracting characteristics of low-voltage electric shock residual current signals when executing the computer program on the memory.

[0054] The low-voltage electric shock residual current signal feature extraction method and related device provided by the present invention are applied to collect electric shock fault information; adaptively decompose based on the fault information to determine K intrinsic mode function IMF components; and use the optimal mode component screening method based on the intrinsic mode function IMF components to screen out the optimal component; thereby extracting the electric shock residual current signal feature based on the optimal component. The present invention uses IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm, adaptively obtains the optimal parameter combination, and combines the advantages of SEGI and kurtosis screening indicators to effectively screen out the modal components with the most periodic shocks. After using the IPOA-VMD decomposition algorithm to perform preliminary noise reduction on the fault information signal collected in real time, MCKD is used to denoise the optimal IMF component, and TEO demodulation analysis is used to achieve fault feature extraction in a strong background noise environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0056] Figure 1The algorithm flow chart of the method for extracting the characteristics of the low-voltage electric shock residual current signal provided by the present invention is shown;

[0057] Figure 2 A flow chart showing the steps of the method for extracting the characteristics of a low-voltage electric shock residual current signal provided by the present invention is shown;

[0058] Figure 3 A flowchart showing the steps of adaptively decomposing and determining K intrinsic mode function IMF components based on fault information provided by the present invention is shown;

[0059] Figure 4 The IPOA-VMD fitness function iteration curve in the present invention is shown;

[0060] Figure 5 The time domain waveform of each IMF component after IPOA-VMD decomposition in the present invention is shown;

[0061] Figure 6 The frequency domain waveform of each IMF component after IPOA-VMD decomposition in the present invention is shown;

[0062] Figure 7 The flowchart of the steps of selecting the optimal component by using the optimal modal component screening method based on the intrinsic modal function IMF component provided in the present invention is shown;

[0063] Figure 8 The time domain diagram of IMF3 after MCKD processing in the present invention is shown;

[0064] Fig. 9 The TEO spectrum of IMF3 after MCKD treatment in the present invention is shown;

[0065] Fig.10 The structural block diagram of the low-voltage electric shock residual current signal feature extraction device provided in the present invention is shown;

[0066] Fig.11 The figure shows a schematic diagram of the structure of the electronic device provided in the present invention. DETAILED DESCRIPTION

[0067] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings, but are not intended to limit the present disclosure.

[0068] It should be understood that various modifications may be made to the embodiments disclosed herein. Therefore, the above description should not be considered as limiting, but merely as an example of an embodiment. Other modifications within the scope and spirit of the present disclosure will occur to those skilled in the art.

[0069] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the general description of the present disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the present disclosure.

[0070] These and other characteristics of the present disclosure will become apparent from the following description of a preferred form of embodiment given as a non-limiting example, with reference to the attached drawings.

[0071] It should also be understood that, although the present disclosure has been described with reference to some specific examples, those skilled in the art will be able to realize many other equivalent forms of the present disclosure that have the features described in the claims and are therefore within the scope of protection defined thereby.

[0072] The above and other aspects, features and advantages of the present disclosure will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

[0073] Specific embodiments of the present disclosure are described hereinafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid obscuring the present disclosure with unnecessary or redundant details. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but merely serve as a basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of ways with substantially any suitable detailed structure.

[0074] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0075] This specification may use the phrases "in one embodiment," "in another embodiment," "in a further embodiment," or "in other embodiments," all of which may refer to one or more of the same or different embodiments according to the present disclosure.

[0076] The present disclosure is further described below in conjunction with the accompanying drawings and specific embodiments.

[0077] In order to solve the technical problem in the prior art that the feature extraction effect of a low-voltage electric shock residual current signal is not ideal, the present invention provides a method for extracting the feature of a low-voltage electric shock residual current signal and a related device.

[0078] Figure 1 The algorithm flow chart of the low-voltage electric shock residual current signal feature extraction method provided by the present invention is shown. Figure 2 The flowchart of the method for extracting the characteristics of the low-voltage electric shock residual current signal provided by the present invention is shown as follows: Figure 1 and Figure 2 As shown, the low-voltage electric shock residual current signal feature extraction method includes the following steps:

[0079] S101, collecting electric shock fault information.

[0080] In this step, electric shock fault information is collected, wherein the fault information includes a leakage signal and an electric shock fault signal.

[0081] Specifically, an electric shock physical experiment platform is established, including a symmetrical three-phase circuit, a current transformer, an equivalent load, a three-phase inherent leakage equivalent resistance and capacitance, a fault filter, etc.; the electric shock residual current signal is collected, where the sampling frequency is 10Khz, and both normal leakage and electric shock fault signals are collected, and data is intercepted to build an electric shock signal database.

[0082] S102, performing adaptive decomposition based on the fault information to determine K intrinsic modulus function (IMF) components.

[0083] After completing the above step S101, in this step, adaptive decomposition is performed based on the fault information to determine K intrinsic mode function IMF components.

[0084] Further, Figure 3 FIG. 4 shows a flowchart of the steps of adaptively decomposing and determining K intrinsic mode function IMF components based on fault information in the present invention, as shown in FIG. Figure 3 As shown, the adaptive decomposition based on the fault information to determine K intrinsic mode function IMF components includes the following steps:

[0085] S201, using circle mapping and lens imaging reverse learning strategy to improve the Pelican Optimization Algorithm POA to obtain an improved Pelican Optimization Algorithm IPOA.

[0086] In this step, the pelican optimization algorithm POA is improved by using circular mapping and lens imaging reverse learning strategy to obtain the improved pelican optimization algorithm IPOA; the pelican optimization algorithm POA is used to simulate the hunting process of the pelican population in terrestrial organisms to achieve optimal search, and circular mapping and reverse learning strategy are used to perform dimension-by-dimensional lens imaging to enhance population diversity, thereby forming an improved pelican optimization algorithm IPOA, improving the convergence rate of the algorithm, reducing the risk of local optimal traps, and avoiding the problem of decreased population diversity that may occur in the later stage of iteration of the original method.

[0087] Specifically, a circular mapping is used to initialize the position of the pelican population so that each pelican is evenly distributed in the entire search space. The circular mapping definition is defined as formula (1), as follows: ,

[0088] Among them, num i is the i-th chaotic sequence number, and mod is the remainder operation.

[0089] The expression for initialization of IPOA population is defined as formula (2), as follows:

[0090] ,

[0091] Among them, num ij is the value of the chaotic sequence generated by the circular mapping on the dimension of the corresponding pelican individual, i=1,2,3...,N, j=1,2,3...,m, N is the number of population members, m is the number of problem variables, x ij is the value of the jth variable specified by the i-th candidate solution, l j is the jth lower limit, u j is the j-th upper bound of the problem variable.

[0092] The population moment X is used to represent the pelican population, which is defined as formula (3):

[0093] ,

[0094] Among them, X i is the position of the i-th pelican, each row of the matrix represents a candidate solution, and the columns of the matrix represent the proposed values ​​of the problem variables.

[0095] The fitness function of a given problem is evaluated based on each candidate solution, which is defined as formula (4) as follows:

[0096] ,

[0097] Where F is the fitness function vector, and F i is the fitness function value of the i-th candidate solution.

[0098] The Pelican algorithm enters the first stage of the search and begins to explore prey. The Pelican position strategy formula in the exploration stage is defined as formula (5), as follows:

[0099] ,

[0100] in, is the position of the jth dimension of the ith pelican after the first phase update, rand is a random number in the interval [0,1], I is a random number of 1 or 2, I is randomly selected by each iteration and each member, p j is the j-th dimension position of the target, F p is the fitness function value.

[0101] After the first stage, if the value of the objective function increases at the current position, the position is accepted and updated, which is defined as formula (6), as follows:

[0102] ,

[0103] in, is the new position of the i-th pelican, The fitness function value of the i-th pelican after the position is updated in the first phase.

[0104] The second stage is the pelican hunting development stage, and the mathematical simulation formula is defined as formula (7), as follows:

[0105] ,

[0106] in, is the position of the j-th dimension of the ith pelican after the second stage is updated based on the second stage, rand is a random number in the interval [0,1], R is a random number between 0 and 2, t is the current number of iterations, and T is the maximum number of iterations.

[0107] The position of each pelican is optimized by using the inverse learning strategy of dimension-by-dimension lens imaging. The position of each dimension is mapped into space to obtain the reverse position. The expression of lens imaging is defined as formula (8), as follows:

[0108] ,

[0109] Among them, a j , b j Indicates the search range [a j , b j ], represents the j-th dimension position of the i-th pelican after the second stage update, which is represented as the projection of an object with height h on the x-axis.

[0110] Order on the top , the above formula can be transformed into formula (9), as follows:

[0111] ;

[0112] When n=1, a fixed inverse solution is obtained Formula (10) is as follows:

[0113] ;

[0114] This formula (10) is called the termination condition formula. The fitness value of the previous position is compared with the fitness value after reverse learning. If the fitness value after reverse learning is better than the previous position, the reverse position is selected to replace the previous position; otherwise, the original position is retained for the next generation; the improved second-stage Pelican position update strategy is formula (11), as follows:

[0115] ,

[0116] in, is the reverse solution of the new position in the jth dimension after the second stage of the i-th pelican, is the objective function value after the second stage of reverse learning;

[0117] The new position update strategy mentioned above is used to promote POA to escape from the local optimal area, forming an improved Pelican optimization algorithm IPOA.

[0118] S202, using IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm to obtain an improved VMD algorithm.

[0119] After completing the above step S201, in this step, the penalty factor and the number of modal layers in the VMD algorithm are optimized using IPOA to obtain an improved VMD algorithm.

[0120] Using the enhanced Lagrangian function, the constrained variational problem of the VMD technique is transformed into an unconstrained variational problem. The expression is defined as formula (12), as follows:

[0121] ,

[0122] Where L represents the constraint function, {u k} represents the kth IMF component generated from VMD decomposition, f represents the fault information signal collected in real time, {ω k} represents the center frequency associated with each IMF component, δ(t) represents the unit impulse function, * represents the convolution operation, λ is the Lagrange multiplier, α is the penalty factor, To find the sign of the partial derivative.

[0123] The alternating direction method of multipliers ADMMs is used to solve the problem of solving unconstrained variational models, {ω k}、{u k The update formulas of} and λ are defined as formula (13), formula (14), and formula (15) respectively, and the iteration termination condition is defined as formula (16), as follows:

[0124]

[0125] in , and Respectively, , and the Fourier transform corresponding to f, ε is the convergence accuracy, represents the square of the Euclidean coefficient of the IMF component, n is the value of the signal at the nth moment during the processing of this stage, and the above steps are used for cyclic update until the termination condition formula is met, and finally K IMF components are obtained.

[0126] IPOA is used to optimize the penalty factor α and the number of modal layers K in the VMD algorithm, and the optimal parameter combination is adaptively obtained.

[0127] Specifically, the envelope spectral entropy ESE is used as the fitness function of the IPOA-VMD optimization model, that is, the fitness function vector F described in formula (4) above, which is defined as formula (17), as follows:

[0128] ,

[0129] Wherein, a(i) represents the envelope signal of the fault information signal collected immediately after Hilbert demodulation, N represents the length of the signal, and ε(i) is the normalized form of a(i).

[0130] S203, performing adaptive improved VMD algorithm decomposition on the electric shock signal to obtain K IMF components.

[0131] After completing the above step S202, in this step, the electric shock signal is decomposed by an adaptively improved VMD algorithm to obtain K IMF components, which specifically includes the following steps:

[0132] S501, setting the maximum number of iterations, spatial dimension, population size, decomposition layer and penalty factor of IPOA, and initializing the position of the population using a circle mapping strategy;

[0133] S502, decomposing the fault information signal collected in real time by using an improved VMD algorithm to generate multiple IMF components, calculating the envelope spectrum entropy ESE value of each IMF component, and determining the fitness function of the global exploration by selecting the component with the lowest ESE value;

[0134] S503, after each round of iteration, the ESE value corresponding to each parameter combination is calculated and compared with the current ESE value. If it is less than the current ESE value, the current ESE value will be updated;

[0135] S504, determining whether the stopping condition of the iteration is reached, if the maximum number of iterations is not reached, setting t=t+1, where t refers to the number of algorithm iterations, and the initial value of t is 1, and using the inverse learning strategy of dimension-by-dimension lens imaging to update the optimal individual position of the pelican, where the pelican population position updated in the previous round of iterations will be used as the initial population position of the next round;

[0136] Repeat the above steps S503-S504 until the iteration condition is met; the iteration curve of this embodiment is shown in Figure 4 ,Depend on Figure 4 It can be seen that the fitness function value decreases with the increase of the number of iterations. The minimum fitness value appears in the 10th generation, and then the minimum value remains unchanged. After the cycle iteration is completed, the optimal parameter combination (that is, the optimal penalty factor and modal layer number) is output. The optimal parameter combination iteratively output in this embodiment is [2634,4].

[0137] The optimal parameter combination output by IPOA is assigned to VMD, and K IMF components are obtained through adaptive decomposition. The electric shock data is decomposed into K IMF components in the time domain and frequency domain by IPOA-VMD. Figure 5 and Figure 6 ,in, Figure 6 In the figure, the spectrum of IMF4 is concentrated on the y-axis.

[0138] S103, using an optimal modal component screening method based on the intrinsic modal function IMF component to screen out the optimal component.

[0139] After completing the above step S102, in this step, an optimal modal component screening method is used to screen out the optimal component based on the intrinsic modal function IMF component.

[0140] Further, Figure 7 The flowchart of the steps of using the optimal modal component screening method to screen out the optimal component based on the intrinsic modal function IMF component provided by the present disclosure is shown as follows: Figure 7 As shown, the optimal component is selected by the optimal modal component selection method based on the intrinsic modal function IMF component, including the following steps:

[0141] S301, determining the Gini coefficient SEGI and kurtosis based on the intrinsic mode function IMF component.

[0142] In this step, the Gini coefficient SEGI and the kurtosis are determined based on the intrinsic mode function IMF component.

[0143] Specifically, the Gini coefficient SEGI is an indicator used to measure the sparsity of the sequence. It has high stability against noise interference and its value is between 0 and 1. The closer it is to 1, the better the balance of the sequence. The expression of SEGI is defined as formula (18) and formula (19), as follows:

[0144] ,

[0145] Among them, SE represents the square envelope of the fault information signal collected in real time, It is the L1 paradigm of SE.

[0146] Kurtosis is sensitive to shock signals and is a good indicator for detecting periodic shocks. The expression of kurtosis can be described as follows:

[0147] ,

[0148] Where x is the expected value, the value of N represents the total number of signal points, and σ represents the standard deviation.

[0149] S302, selecting the optimal component based on the Gini coefficient SEGI and kurtosis.

[0150] After completing the above step S301, in this step, the optimal component is screened out based on the Gini coefficient SEGI and kurtosis.

[0151] Specifically, the modal components are screened based on the Gini coefficient SEGI and kurtosis advantage. The calculation formula of the K-SEGI screening standard is as follows:

[0152]

[0153] in, and is the amplitude of the two indicators after normalization.

[0154] The modal component with the largest calculated value of the K IMF components after decomposition is taken as the best IMF component. In this embodiment, the K-SEGI evaluation index of each IMF component is shown in Table 1:

[0155] Table 1:

[0156]

[0157] According to Table 1, IMF3 has the largest K-SEGI value, indicating that this component contains more impact components and fewer noise components, so it is selected as the best modal component.

[0158] S104, extracting electric shock residual current signal characteristics based on the optimal component.

[0159] After completing the above step S103, in this step, the electric shock residual current signal characteristics are extracted based on the optimal component.

[0160] Furthermore, the extracting of electric shock residual current signal characteristics based on the optimal component comprises the following steps:

[0161] S401, denoising the optimal IMF component using the maximum correlation kurtosis deconvolution algorithm MCKD.

[0162] In this step, the maximum correlation kurtosis deconvolution algorithm MCKD is used to denoise the optimal IMF component. The maximum correlation kurtosis deconvolution algorithm MCKD is a weak signal enhancement method that can effectively reduce noise and highlight the impact component of the signal. The optimal filter is searched in an iterative manner so that the correlation kurtosis of the output signal after filtering is maximized.

[0163] Specifically, the IMF component is reconstructed to obtain the signal y, which satisfies the following expression:

[0164] ,

[0165] Where x is the periodic impulse signal, h is the transfer function, e is the environmental noise, and * represents the convolution operation.

[0166] By constructing a filter, the noise is suppressed and the fault impulse component is highlighted. The expression is as follows:

[0167] ,

[0168] in, is the filter coefficient, y n is the output signal, L is the filter length, and n is the value of the corresponding signal at the nth moment.

[0169] S402, performing Teager energy operator demodulation on the reconstructed signal to obtain a Teager energy operator energy spectrum.

[0170] After completing the above step S401, in this step, Teager energy operator demodulation is performed on the reconstructed signal to obtain a Teager energy operator energy spectrum.

[0171] Specifically, the definition of the kurtosis of any signal is:

[0172] ,

[0173] Among them, T s is the deconvolution period, M is the number of filter shifts, and n is the value of the corresponding signal at the nth moment.

[0174] The maximum value of the maximum correlation kurtosis is the objective function, and the formula is as follows:

[0175]

[0176] From this, the filter coefficients can be obtained as:

[0177] .

[0178] in, ,

[0179] , , represents a norm operation, and X0 represents a normalization transformation, which is usually used to adjust the measurement scale in the data space.

[0180] The filter coefficient combination f is obtained by the above formula (26), and is input into formulas (23) and (24) to obtain the deconvolution enhanced signal after noise reduction. In this embodiment, the IMF3 obtained in step S302 is subjected to MCKD signal enhancement processing. Figure 8 .

[0181] The Teager Energy Operator (TEO) is used to demodulate and analyze the deconvoluted signal to obtain the TEO energy spectrum, and finally accurately extract the fault signal features. The TEO spectrum of IMF3 after MCKD processing can be found in Fig. 9 As shown in the figure, the fault frequency characteristics of the electric shock signal, the peak and octave frequency are very obvious. This shows that the algorithm proposed in the present disclosure can accurately extract the low-voltage electric shock residual current signal characteristics.

[0182] The method for extracting features of residual current signals of low-voltage electric shock provided by the present invention is applied to collect electric shock fault information; adaptively decompose based on the fault information to determine K intrinsic mode function IMF components; and use the optimal mode component screening method based on the intrinsic mode function IMF components to screen out the optimal components; thereby extracting features of residual current signals of electric shock based on the optimal components. The present invention utilizes IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm, adaptively obtains the optimal parameter combination, and combines the advantages of SEGI and kurtosis screening indicators to effectively screen out the modal components with the most periodic shocks. After the IPOA-VMD decomposition algorithm is used to perform preliminary noise reduction on the fault information signal collected in real time, the optimal IMF components are denoised using MCKD, and TEO demodulation analysis is used to achieve fault feature extraction in a strong background noise environment.

[0183] In order to better implement the above method, the second aspect of the present disclosure further provides a low-voltage electric shock residual current signal feature extraction device, which can be integrated on an electronic device.

[0184] For example, Fig.10 As shown, the extraction device may include: a collection module 210, a determination module 220, a screening module 230 and a judgment module 240, which are specifically as follows:

[0185] (1) A collection module 210 , which is used to collect electric shock fault information.

[0186] (2) A determination module 220 , wherein the determination module 220 performs adaptive decomposition based on the fault information to determine K intrinsic mode function (IMF) components.

[0187] (3) A screening module 230, wherein the screening module 230 uses an optimal modal component screening method to screen out the optimal component based on the intrinsic modal function (IMF) component.

[0188] (4) An extraction module 240, wherein the extraction module 240 extracts electric shock residual current signal characteristics based on the optimal component.

[0189] Furthermore, the determination module 220 includes an improvement unit, an optimization unit and a decomposition unit, wherein the improvement unit is used to improve the Pelican optimization algorithm POA by using circular mapping and lens imaging reverse learning strategies to obtain an improved Pelican optimization algorithm IPOA, the optimization unit uses IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm to obtain an improved VMD algorithm, and the decomposition unit performs an adaptively improved VMD algorithm decomposition on the electric shock signal to obtain K IMF components.

[0190] Furthermore, the screening module 230 includes a determination unit and a screening unit, wherein the determination unit determines the Gini coefficient SEGI and the kurtosis based on the intrinsic mode function IMF component; and the screening unit screens out the optimal component based on the Gini coefficient SEGI and the kurtosis.

[0191] Furthermore, the extraction module 240 includes a denoising unit and a demodulation unit, wherein the denoising unit uses a maximum correlation kurtosis deconvolution algorithm MCKD to denoise the optimal IMF component, and the demodulation unit is used to perform Teager energy operator demodulation on the reconstructed signal to obtain a Teager energy operator energy spectrum.

[0192] The low-voltage electric shock residual current signal feature extraction device provided by the present invention is applied to collect electric shock fault information; adaptively decompose based on the fault information to determine K intrinsic mode function IMF components; and use the optimal mode component screening method based on the intrinsic mode function IMF components to screen out the optimal component; thereby extracting the electric shock residual current signal feature based on the optimal component. The present invention uses IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm, adaptively obtains the optimal parameter combination, and combines the advantages of SEGI and kurtosis screening indicators to effectively screen out the modal components with the most periodic shocks. After using the IPOA-VMD decomposition algorithm to perform preliminary noise reduction on the fault information signal collected in real time, MCKD is used to denoise the optimal IMF component, and TEO demodulation analysis is used to achieve fault feature extraction in a strong background noise environment.

[0193] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0194] To this end, a third embodiment of the present disclosure provides a storage medium, which is a computer-readable medium and stores a computer program. When the computer program is executed by a processor, the method provided by the embodiment of the present disclosure is implemented, including the following steps S11 to S14:

[0195] S11, collecting electric shock fault information;

[0196] S12, performing adaptive decomposition based on the fault information to determine K intrinsic mode function IMF components;

[0197] S13, using an optimal modal component screening method to screen out an optimal component based on the intrinsic modal function IMF component;

[0198] S14, extracting electric shock residual current signal characteristics based on the optimal component.

[0199] Furthermore, when the computer program is executed by a processor, other methods provided by any of the above embodiments of the present disclosure are implemented.

[0200] The method for extracting features of residual current signals of low-voltage electric shock provided by the present invention is applied to collect electric shock fault information; adaptively decompose based on the fault information to determine K intrinsic mode function IMF components; and use the optimal mode component screening method based on the intrinsic mode function IMF components to screen out the optimal components; thereby extracting features of residual current signals of electric shock based on the optimal components. The present invention utilizes IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm, adaptively obtains the optimal parameter combination, and combines the advantages of SEGI and kurtosis screening indicators to effectively screen out the modal components with the most periodic shocks. After the IPOA-VMD decomposition algorithm is used to perform preliminary noise reduction on the fault information signal collected in real time, the optimal IMF components are denoised using MCKD, and TEO demodulation analysis is used to achieve fault feature extraction in a strong background noise environment.

[0201] A fourth embodiment of the present disclosure provides an electronic device, such as Fig.11 As shown, the electronic device at least includes a memory 310 and a processor 320. The memory 310 stores a computer program. The processor 320 implements the method provided by any embodiment of the present disclosure when executing the computer program on the memory 310. Exemplarily, the method executed by the electronic device computer program is as follows:

[0202] S21, collecting electric shock fault information;

[0203] S22, performing adaptive decomposition based on the fault information to determine K intrinsic mode function (IMF) components;

[0204] S23, using an optimal modal component screening method to screen out an optimal component based on the intrinsic modal function IMF component;

[0205] S24, extracting electric shock residual current signal characteristics based on the optimal component.

[0206] In specific implementation, the acquisition module 210, the determination module 220, the screening module 230 and the judgment module 240 are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions.

[0207] The method for extracting features of residual current signals of low-voltage electric shock provided by the present invention is applied to collect electric shock fault information; adaptively decompose based on the fault information to determine K intrinsic mode function IMF components; and use the optimal mode component screening method based on the intrinsic mode function IMF components to screen out the optimal components; thereby extracting features of residual current signals of electric shock based on the optimal components. The present invention utilizes IPOA to optimize the penalty factor and the number of modal layers in the VMD algorithm, adaptively obtains the optimal parameter combination, and combines the advantages of SEGI and kurtosis screening indicators to effectively screen out the modal components with the most periodic shocks. After the IPOA-VMD decomposition algorithm is used to perform preliminary noise reduction on the fault information signal collected in real time, the optimal IMF components are denoised using MCKD, and TEO demodulation analysis is used to achieve fault feature extraction in a strong background noise environment.

[0208] The storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0209] The storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0210] Alternatively, the storage medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0211] Computer program code for performing operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the passenger computer, partially on the passenger computer, as a separate software package, partially on the passenger computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the passenger computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0212] It should be noted that the storage medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any storage medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0213] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0214] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0215] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0216] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0217] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0218] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0219] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

[0220] Multiple embodiments of the present disclosure are described in detail above, but the present disclosure is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications to the embodiments based on the concepts of the present disclosure, and these variations and modifications should all fall within the scope of protection required by the present disclosure.

Claims

1. A method for extracting characteristics of low-voltage electric shock residual current signal, characterized in that: The steps include: Collect electric shock fault information; Based on the fault information, adaptive decomposition is performed to determine K intrinsic mode function (IMF) components; Based on the intrinsic mode function IMF component, an optimal component is selected by using an optimal mode component selection method; The electric shock residual current signal characteristics are extracted based on the optimal component.

2. The method for extracting the characteristics of low-voltage electric shock residual current signal according to claim 1 is characterized in that: The adaptive decomposition based on the fault information to determine K intrinsic mode function IMF components includes: The improved Pelican Optimization Algorithm IPOA is obtained by improving the Pelican Optimization Algorithm POA using the circular mapping and lens imaging reverse learning strategy. IPOA is used to optimize the penalty factor and modal layers in the VMD algorithm to obtain an improved VMD algorithm. The fault information is adaptively decomposed by the improved VMD algorithm to obtain K IMF components.

3. The method for extracting the characteristics of low-voltage electric shock residual current signal according to claim 2 is characterized in that: The improved Pelican optimization algorithm POA is improved by using the circular mapping and lens imaging reverse learning strategy to obtain the improved Pelican optimization algorithm IPOA, including: A circular mapping is used to initialize the position of the pelican population so that each pelican is evenly distributed in the entire search space. The circular mapping definition is: ， Among them, numi is the i-th chaotic sequence number, and mod is the remainder operation; The expression for initializing the IPOA population is as follows: ， Among them, num ij is the value of the chaotic sequence generated by the circular mapping on the dimension of the corresponding pelican individual, i=1,2,3...,N, j=1,2,3...,m, N is the number of population members, m is the number of problem variables, x ij is the value of the jth variable specified by the i-th candidate solution, l j is the jth lower limit, u j is the jth upper limit of the problem variable; Use population moment X to represent the pelican population: ， Among them, X i is the position of the i-th pelican, each row of the matrix represents a candidate solution, and the columns of the matrix represent the proposed values ​​of the problem variables; The fitness function for a given problem is evaluated based on each candidate solution: ， Where F is the fitness function vector, and F i is the fitness function value of the i-th candidate solution; The Pelican algorithm enters the first stage of the search and begins to explore prey. The Pelican's position strategy formula in the exploration stage is: ， in, is the position of the jth dimension of the ith pelican after the first phase update, rand is a random number in the interval [0,1], I is a random number of 1 or 2, I is randomly selected by each iteration and each member, p j is the j-th dimension position of the target, F p is the fitness function value; After the first stage, if the value of the objective function improves at the current position, the position is accepted and updated: ， in, is the new position of the i-th pelican, The fitness function value of the i-th pelican after the position is updated in the first stage; The second stage is the pelican hunting development stage, and the mathematical simulation formula is: ， in, is the position of the jth dimension of the ith pelican after the second stage based update, rand is a random number in the interval [0,1], R is a random number between 0 and 2, t is the current number of iterations, and T is the maximum number of iterations; The position of each pelican is optimized using the inverse learning strategy of dimension-by-dimension lens imaging. The position of each dimension is mapped into space to obtain the reverse position. The expression of lens imaging is: ， Among them, a j , b j Indicates the search range [a j , b j ], represents the j-th dimension position of the i-th pelican after the second stage update, which is represented as the projection of an object with height h on the x-axis; Order on the top , which can be transformed into: ; When n=1, a fixed reverse solution is obtained : ； This formula (10) is called the termination condition formula. The fitness value of the previous position is compared with the fitness value after reverse learning. If the fitness value after reverse learning is better than the previous position, the reverse position is selected to replace the previous position; otherwise, the original position is retained for the next generation. The improved second-stage Pelican position update strategy is: ， in, is the reverse solution of the new position in the jth dimension after the second stage of the i-th pelican, is the objective function value after the second stage of reverse learning; The new position update strategy mentioned above is used to promote POA to escape from the local optimal area, forming an improved Pelican optimization algorithm IPOA.

4. The method for extracting the characteristics of low-voltage electric shock residual current signal according to claim 2 is characterized in that: The improved VMD algorithm obtained by optimizing the penalty factor and the number of modal layers in the VMD algorithm using IPOA includes: Using the enhanced Lagrangian function, the restricted variation problem of the VMD technique is transformed into an unrestricted variation problem, which is expressed as follows: ， Where L represents the constraint function, {u k } represents the kth IMF component generated from VMD decomposition, f represents the original signal, {ω k } represents the center frequency associated with each IMF component, δ(t) represents the unit impulse function, * represents the convolution operation, λ is the Lagrange multiplier, α is the penalty factor, To find the sign of partial derivative, t refers to the number of algorithm iterations, and the initial value of t is 1; The alternating direction method of multipliers ADMMs is used to solve the problem of solving unconstrained variational models, {ω k }、{u k The update formula of} and λ, as well as the iteration termination condition are as follows: ， in , and Respectively, , and the Fourier transform corresponding to f, ε is the convergence accuracy, represents the square of the Euclidean coefficient of the IMF component, n is the value of the signal at the nth moment during the processing of this stage, and the above formulas (11) to (16) are used to perform cyclic updates until the termination condition formula is met, and finally K IMF components are obtained; IPOA is used to optimize the penalty factor α and the number of modal layers K in the VMD algorithm, and the optimal parameter combination is adaptively obtained.

5. The method for extracting characteristics of low-voltage electric shock residual current signal according to claim 2, characterized in that: The adaptively improved VMD algorithm decomposition of the fault information to obtain K IMF components includes: S501, setting the maximum number of iterations, spatial dimension, population size, decomposition layer and penalty factor of IPOA, and initializing the position of the population using a circle mapping strategy; S502, decomposing the fault information signal collected in real time by using an improved VMD algorithm to generate multiple IMF components, calculating the envelope spectrum entropy ESE value of each IMF component, and determining the fitness function of the global exploration by selecting the component with the lowest ESE value; S503, after each round of iteration, the ESE value corresponding to each parameter combination is calculated and compared with the current ESE value. If it is less than the current ESE value, the current ESE value is updated; S504, determining whether the stopping condition of the iteration is reached, if the maximum number of iterations is not reached, setting t=t+1, where t refers to the number of algorithm iterations, and the initial value of t is 1, and using the inverse learning strategy of dimension-by-dimension lens imaging to update the optimal individual position of the pelican, where the pelican population position updated in the previous round of iterations will be used as the initial population position of the next round; Repeat the above steps S503-S504 until the iteration condition is met; The optimal parameter combination output by IPOA is assigned to VMD, and K IMF components are obtained through adaptive decomposition.

6. The method for extracting characteristics of low-voltage electric shock residual current signal according to claim 1, characterized in that: The method of selecting the optimal component based on the intrinsic mode function IMF component by using the optimal mode component screening method comprises: Determine the Gini coefficient SEGI and kurtosis based on the intrinsic mode function IMF component; The optimal components are selected based on the Gini coefficient SEGI and kurtosis.

7. A low-voltage electric shock residual current signal feature extraction device, characterized in that: include: A collection module is used to collect electric shock fault information; A determination module, which performs adaptive decomposition based on the fault information to determine K intrinsic mode function IMF components; A screening module, which uses an optimal modal component screening method to screen out the optimal component based on the intrinsic modal function IMF component; An extraction module extracts the electric shock residual current signal characteristics based on the optimal component.

8. A seat belt storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for extracting characteristics of a low-voltage electric shock residual current signal described in any one of claims 1 to 6 are implemented.

9. An electronic device, comprising at least a memory and a processor, wherein a computer program is stored in the memory, wherein: The processor implements the steps of the low-voltage electric shock residual current signal feature extraction method described in any one of claims 1 to 6 when executing the computer program on the memory.

Images