Power quality interference signal type detection method, detection device and storage medium

By decomposing and denoising power quality interference signals using VMD and optimizing support vector machines with IGWO, the method addresses noise interference, improving detection accuracy and ensuring power system stability.

CN119089285APending Publication Date: 2024-12-06GUANGDONG POWER GRID CO LTD +1

Patent Information

Application Number
CN202411131040.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Power quality interference signals suffer from noise interference, which reduces the accuracy of detection and hampers targeted improvement measures, posing a threat to the stability of power systems.

Method used

The method employs a VMD algorithm to decompose power quality interference signals, extract denoised intrinsic modal components, and reconstruct them to determine the signal type, using a type recognition model trained with support vector machines optimized by the IGWO algorithm.

Benefits of technology

The method effectively denoises power quality interference signals, improving detection accuracy and enabling precise identification of signal types, thereby enhancing power system stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119089285A_ABST
    Figure CN119089285A_ABST
Patent Text Reader

Abstract

The invention provides a power quality interference signal type detection method and device and a storage medium. The method comprises the steps that a VMD algorithm is adopted to decompose an electric energy quality interference signal to obtain target intrinsic mode data, and the target intrinsic mode data comprises a plurality of intrinsic mode components and the center frequency of each intrinsic mode component; extracting a plurality of de-noised intrinsic mode components from the plurality of intrinsic mode components, the de-noised intrinsic mode components being intrinsic mode components satisfying a predetermined condition, and the predetermined condition being that the center frequency is less than or equal to a predetermined frequency; reconstructing the de-noised intrinsic mode component to obtain a de-noised power quality interference signal; and determining the type of the electric energy quality interference signal according to the de-noised electric energy quality interference signal. The method solves the problem that the accuracy of the type detection of the electric energy quality interference signal is reduced due to the noise interference of the electric energy quality interference signal in the prior art.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power quality interference signal type detection, and in particular to a power quality interference signal type detection method, a power quality interference signal type detection device, a computer-readable storage medium and a computer program product. Background Art

[0002] In the detection of power quality interference signal types, since the power quality interference signal itself has noise interference, it will reduce the accuracy of power quality interference signal type detection and may also lead to the lack of targeted improvement measures, thus posing a potential threat to the stable operation of the power system.

[0003] Currently, there is no solution to the above problem. Summary of the invention

[0004] The main purpose of the present application is to provide a method for detecting the type of a power quality interference signal, a device for detecting the type of a power quality interference signal, a computer-readable storage medium and a computer program product, so as to at least solve the problem in the prior art that the power quality interference signal itself has noise interference that reduces the accuracy of the type detection of the power quality interference signal.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for detecting the type of power quality interference signal is provided, the method comprising: decomposing the power quality interference signal by using a VMD algorithm to obtain target intrinsic modal data, the target intrinsic modal data comprising: multiple intrinsic modal components and the center frequency of each of the intrinsic modal components; extracting multiple denoised intrinsic modal components from the multiple intrinsic modal components, the denoised intrinsic modal components being the intrinsic modal components that meet a predetermined condition, the predetermined condition being that the center frequency is less than or equal to a predetermined frequency; reconstructing the denoised intrinsic modal components to obtain a denoised power quality interference signal; and determining the type of the power quality interference signal according to the denoised power quality interference signal.

[0006] Optionally, the VMD algorithm is used to decompose the power quality interference signal to obtain target intrinsic mode data, including: a first decomposition step, when the number of decomposed modes is m, the VMD algorithm is used to decompose the power quality interference signal to obtain the mth candidate intrinsic mode data, the mth candidate intrinsic mode data includes: m said intrinsic mode components and the center frequencies of the eigenmode components, m is an integer, and the initial value of m is 2; a second decomposition step, when the number of decomposed modes is m+1, the VMD algorithm is used to decompose the power quality interference signal to obtain the m+1th candidate intrinsic mode data, the m+1th candidate intrinsic mode data includes: m+1 said intrinsic mode components and the center frequencies of each of the eigenmode components; a first determination step, according to the mth candidate intrinsic mode data, the m+1th candidate intrinsic mode data and Determine the energy convergence factor, where Δ mm+1 is the energy convergence factor, s is the power quality interference signal, u m is the mth eigenmode component in the mth candidate eigenmode data, u m+1 is the m+1th eigenmode component in the m+1th candidate eigenmode data, is the square of the 2-norm of s, for u m The square of the 2-norm of for u m+1 The square of the 2-norm of the energy convergence factor; a first repetitive step, sequentially increasing the value of m, repeating the first decomposition step, the second decomposition step and the determination step until the energy convergence factor is less than or equal to a predetermined value; a second determination step, when the energy convergence factor is less than the predetermined value, determining the mth candidate intrinsic mode data as the target intrinsic mode data.

[0007] Optionally, determining the type of the power quality interference signal based on the denoised power quality interference signal includes: inputting the denoised power quality interference signal into a type recognition model to obtain the type of the power quality interference signal, the type recognition model is obtained by training a classification structure using multiple groups of training data, each group of the training data at least includes: the power quality interference signal, the type of the power quality interference signal acquired in a historical period, and the classification structure is composed of multiple support vector machines.

[0008] Optionally, the classification structure is composed of N support vector machines, and the denoised power quality interference signal is input into a type recognition model to obtain the type of the power quality interference signal, including: an input step, inputting the denoised power quality interference signal into the nth support vector machine to obtain a classification result, the classification result is that the type of the power quality interference signal is the type corresponding to the nth support vector machine or the type of the power quality interference signal is not the type corresponding to the nth support vector machine, one support vector machine corresponds to one type, n is a positive integer, the initial value of N is 1, and n≤N; a second repetition step step, increasing the value of N in sequence, repeating the input step at least once until the classification result is the type corresponding to the Nth support vector machine or n is greater than N; a third determination step, when the classification result is the type corresponding to the Nth support vector machine, determining that the type corresponding to the Nth support vector machine is the type of the power quality interference signal, when n is greater than N and the classification result is not the type corresponding to the Nth support vector machine, determining that the non-identification type is the type of the power quality interference signal, and the non-identification type is a type other than the types corresponding to all the support vector machines.

[0009] Optionally, before inputting the denoised power quality interference signal into a type recognition model to obtain the type of the power quality interference signal, the method further includes: an optimization step, taking the maximum classification accuracy of the target support vector machine as the goal, and using the IGWO algorithm to optimize the parameters of the target support vector machine, the parameters including: a penalty factor and a kernel function parameter, and the target support vector machine is one of the support vector machines in the classification structure; a third repetition step, repeating the optimization step until the optimization of all the support vector machines in the classification structure is completed.

[0010] Optionally, the optimization step, with the maximum classification accuracy of the target support vector machine as the goal, uses the IGWO algorithm to optimize the parameters of the target support vector machine, including: an acquisition step, acquiring a data set, and randomly dividing the data set into a training set and a validation set according to the ratio of a:b, wherein the data set, the training set and the validation set all include multiple groups of the training data; an initialization step, initializing the size of the gray wolf population, the position of the gray wolves in the gray wolf population, the optimization interval of the position of the gray wolf and the maximum number of iterations, wherein the position of the gray wolf contains a value of the parameter; a training determination step, in which the parameters of the target support vector machine are In the case where the values ​​are the positions of the gray wolves respectively, the training set is used to train the target support vector machine, and based on the validation set and the trained target support vector machine, the fitness value of each gray wolf is determined, and the fitness value is the classification accuracy of the target support vector machine when the parameter value of the target support vector machine is the position of the gray wolf; in the fourth determination step, in the order of the fitness values ​​of the gray wolves from large to small, the gray wolves with the top three fitness values ​​in the gray wolf population are respectively determined as the gray wolves of the first level, the gray wolves of the second level and the gray wolves of the third level; in the updating step, according to and Update the position of each low-level gray wolf, where A = 2a·r1-a, C = 2·r2, is the position of the gray wolf at the first level during the t-th iteration, is the position of the gray wolf at the second level during the t-th iteration, is the position of the gray wolf at the third level during the t-th iteration, is the position of the i-th low-level gray wolf in the t-th iteration, is the position of the i-th low-level gray wolf in the t+1-th iteration process, r1, r2 and r3 are all random numbers in [0,1], and the low-level gray wolves are the gray wolves in the gray wolf population except the first-level gray wolves, the second-level gray wolves and the third-level gray wolves; the fourth repetition step, successively increasing the value of t, repeating the training determination step, the fourth determination step and the update step, until the value of t is equal to or greater than the maximum number of iterations, and outputting the best position, the best position is the position of the gray wolf with the largest fitness value in the gray wolf population.

[0011] Optionally, the types include at least: normal, transient rise, transient drop, interruption, pulse, and harmonic.

[0012] According to another aspect of the present application, a device for detecting the type of a power quality interference signal is provided, the device comprising: a decomposition unit, configured to decompose the power quality interference signal by using a VMD algorithm to obtain target intrinsic modal data, wherein the target intrinsic modal data comprises: a plurality of intrinsic modal components and a center frequency of each of the intrinsic modal components; an extraction unit, configured to extract a plurality of denoised intrinsic modal components from the plurality of intrinsic modal components, wherein the denoised intrinsic modal components are the intrinsic modal components that meet a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency; a reconstruction unit, configured to reconstruct the denoised intrinsic modal components to obtain a denoised power quality interference signal; and a determination unit, configured to determine the type of the power quality interference signal according to the denoised power quality interference signal.

[0013] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the power quality interference signal type detection methods.

[0014] According to another aspect of the present application, a computer program product is provided, comprising a computer program / instruction, which implements any one of the power quality interference signal type detection methods when executed by a processor.

[0015] By applying the technical solution of the present application, a VMD algorithm is used to denoise the power quality interference signal to obtain a denoised power quality interference signal, and the denoised power quality interference signal is used to determine the type of the power quality interference signal, thereby solving the problem in the prior art that the power quality interference signal itself has noise interference that reduces the accuracy of the type detection of the power quality interference signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for detecting a power quality interference signal type provided in an embodiment of the present application is shown;

[0017] Figure 2 A schematic diagram of a flow chart of a method for detecting a type of power quality interference signal provided according to an embodiment of the present application is shown;

[0018] Figure 3 A structural block diagram of a type identification model provided according to an embodiment of the present application is shown;

[0019] Figure 4 A VMD decomposition schematic diagram provided according to an embodiment of the present application is shown;

[0020] Figure 5A schematic diagram of VMD decomposition and reconstruction provided according to an embodiment of the present application is shown;

[0021] Figure 6 A schematic diagram showing the recognition accuracy of different models provided according to an embodiment of the present application;

[0022] Figure 7 A structural block diagram of an improved IEEE-33 node system provided according to an embodiment of the present application is shown;

[0023] Figure 8 A schematic diagram showing the recognition accuracy of different nodes provided according to an embodiment of the present application is shown;

[0024] Fig. 9 A structural block diagram of a power quality interference signal type detection device provided according to an embodiment of the present application is shown.

[0025] The above drawings include the following reference numerals:

[0026] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0027] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0031] VMD (Variational Mode Decomposition) algorithm: It is a signal decomposition method that analyzes the structure and characteristics of the signal by decomposing it into a series of modes with different frequency and amplitude characteristics. The VMD algorithm achieves signal decomposition by minimizing the difference between the signal and the mode function.

[0032] The number of decomposition modes of the VMD algorithm is determined according to the specific application requirements and signal characteristics. Usually, it is necessary to select the appropriate number of decomposition modes based on factors such as the frequency range, amplitude characteristics and decomposition accuracy of the signal. In practical applications, the optimal number of decomposition modes can be determined through experiments and adjustments.

[0033] Support Vector Machine (SVM): is a commonly used machine learning algorithm used for classification and regression analysis. Its basic idea is to separate samples of different categories by finding an optimal hyperplane (or multiple hyperplanes) to maximize the interval between the sample points and the hyperplane. In the classification problem, SVM can divide the data points into two categories and map the unknown sample points to one side of the hyperplane for classification.

[0034] IGWO algorithm: is an optimization algorithm based on the Grey Wolf Optimization Algorithm (GWO). It improves the performance and convergence speed of the algorithm by introducing different improvement measures. Some common improvements include: selection of optimization parameters, improvement of group update strategy, optimization of convergence criteria, etc. These improvement measures can help the algorithm find the global optimal solution faster and improve the stability and robustness of the algorithm. The IGWO algorithm has been widely used in many fields and has achieved good optimization results.

[0035] SVD algorithm (Singular Value Decomposition): It is a commonly used matrix decomposition algorithm, which is used to decompose a matrix into the product of three matrices: A = UΣV^T, where U and V are orthogonal matrices, Σ is a diagonal matrix. The SVD algorithm can be used in many fields such as dimensionality reduction, feature extraction, and recommendation systems. In the recommendation system, the SVD algorithm is usually used to decompose the user-item rating matrix to obtain the implicit feature vectors of users and items, and then make recommendations.

[0036] CNN model (Convolutional Neural Network): is a deep learning model, usually used to process image and video data. The CNN model consists of multiple convolutional layers and pooling layers. It achieves image classification, object detection, image segmentation and other tasks by learning the hierarchical representation of features.

[0037] As introduced in the background technology, in the prior art, the power quality interference signal itself has noise interference, which reduces the accuracy of the type detection of the power quality interference signal. In order to solve the problem in the prior art that the power quality interference signal itself has noise interference, which reduces the accuracy of the type detection of the power quality interference signal, the embodiments of the present application provide a power quality interference signal type detection method, a power quality interference signal type detection device, a computer-readable storage medium and a computer program product.

[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0039] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for detecting a type of power quality interference signal according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0040] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0041] In this embodiment, a method for detecting the type of power quality interference signal running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0042] Figure 2 1 is a flow chart of a method for detecting a power quality interference signal type according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0043] Step S201, using a VMD algorithm to decompose the power quality interference signal to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components;

[0044] In an optional solution, the above step S201 can be implemented as follows:

[0045] In a first decomposition step, when the number of decomposed modes is m, the VMD algorithm is used to decompose the power quality interference signal to obtain m-th candidate intrinsic mode data, wherein the m-th candidate intrinsic mode data includes: m-th intrinsic mode components and the center frequencies of the m-th intrinsic mode components, where m is an integer and the initial value of m is 2;

[0046] A second decomposition step, when the number of decomposition modes is m+1, the power quality interference signal is decomposed by using the VMD algorithm to obtain the m+1th candidate eigenmode data, wherein the m+1th candidate eigenmode data includes: the m+1th eigenmode components and the center frequency of each of the eigenmode components;

[0047] The first determination step is to determine the mth candidate eigenmode data, the m+1th candidate eigenmode data and Determine the energy convergence factor, where Δ mm+1 is the energy convergence factor, s is the power quality interference signal, u m is the mth eigenmode component in the mth candidate eigenmode data, u m+1 is the m+1th eigenmode component in the m+1th candidate eigenmode data, is the square of the 2-norm of s, for u m The square of the 2-norm of for u m+1 The square of the 2-norm of ;

[0048] A first repetition step, sequentially increasing the value of m, repeating the first decomposition step, the second decomposition step and the determination step until the energy convergence factor is less than or equal to a predetermined value;

[0049] The second determination step is to determine the mth candidate eigenmode data as the target eigenmode data when the energy convergence factor is less than the predetermined value.

[0050] Specifically, the selection of the number of decomposition modes in VMD has a significant impact. Due to the complexity of the environment of the power quality interference detection point, it is difficult to determine the optimal value of the number of decomposition modes when using the VMD algorithm to decompose the power quality interference signal. At present, the determination of the number of decomposition modes mainly relies on empirical observation of the center frequency. However, this method lacks quantitative basis. When the center rate frequencies of adjacent intrinsic mode components are too close, it will lead to over-decomposition or under-decomposition, which will have a significant impact on the adaptability and effectiveness of the VMD decomposition algorithm. This application calculates the energy of the residual signal after decomposition based on the adjacent decomposition mode numbers. or Energy of power quality interference signal When the calculated difference is less than or equal to the predetermined value (0.02), the energy is considered to have converged. At this time, the signal decomposition effect is relatively good and the decomposition is considered to be effective.

[0051] Specifically, the VMD algorithm is used to decompose the power quality interference signal. The essence of the VMD algorithm is to transform the signal decomposition process into the process of solving the variational problem, and decompose the input power quality interference signal f into k single-component amplitude-frequency modulation intrinsic mode components u k (t), the constraint condition of the objective function of the corresponding variational problem is to minimize the sum of the estimated bandwidths of the k intrinsic mode components and ensure that the sum of the intrinsic mode components is equal to f. The specific steps of decomposing the power quality interference signal using the VMD algorithm are as follows:

[0052] Detailed steps of VMD algorithm:

[0053] Step 1: pre-specify the number of decomposed intrinsic mode components and initialize the intrinsic mode components The center frequency of the eigenmode component and Lagrange multipliers (n=0);

[0054] Step 2, n←n+1, and according to the formula renew According to the formula renew

[0055] Step 3, according to the formula Update the lambda value;

[0056] Step 4, repeat steps 2 and 3 until When , the iteration is stopped and k eigenmode components and their center frequencies are obtained.

[0057] Step S202, extracting a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency;

[0058] Step S203, reconstructing the denoised intrinsic mode components to obtain a denoised power quality interference signal;

[0059] Step S204: determining the type of the power quality interference signal according to the denoised power quality interference signal.

[0060] In an optional solution, the above step S204 can be implemented as follows:

[0061] The denoised power quality interference signal is input into a type recognition model to obtain the type of the power quality interference signal. The type recognition model is obtained by training a classification structure using multiple groups of training data. Each group of training data at least includes: the power quality interference signal and the type of the power quality interference signal acquired in a historical period. The classification structure is composed of multiple support vector machines.

[0062] Specifically, the classification object of the support vector machine is mainly binary classification, and there are many types of power quality interference signals. Therefore, multiple support vector machines are used to construct a classification structure, and the classification structure is trained to obtain a type recognition model. The type recognition model is used to realize the classification of the types of power quality interference signals.

[0063] In an optional solution, the classification structure is composed of N support vector machines, and the denoised power quality interference signal is input into a type recognition model to obtain the type of the power quality interference signal, including:

[0064] An input step, inputting the denoised power quality interference signal into the nth support vector machine to obtain a classification result, wherein the classification result is that the type of the power quality interference signal is the type corresponding to the nth support vector machine or the type of the power quality interference signal is not the type corresponding to the nth support vector machine, one support vector machine corresponds to one type, n is a positive integer, the initial value of N is 1, and n≤N;

[0065] The second repetition step is to increase the value of N in sequence, and repeat the input step at least once until the classification result is the type corresponding to the Nth support vector machine or n is greater than N;

[0066] The third determination step, when the above-mentioned classification result is the above-mentioned type corresponding to the above-mentioned Nth support vector machine, determines that the above-mentioned type corresponding to the above-mentioned Nth support vector machine is the above-mentioned type of the above-mentioned power quality interference signal, and when n is greater than N and the above-mentioned classification result is not the above-mentioned type corresponding to the above-mentioned Nth support vector machine, determines that the non-identification type is the above-mentioned type of the above-mentioned power quality interference signal, and the above-mentioned non-identification type is a type other than the above-mentioned types corresponding to all the above-mentioned support vector machines.

[0067] Specifically, in the case where the above types include: normal, transient rise, transient drop, interruption, pulse, harmonic, such as Figure 3As shown, the classification structure is composed of 5 support vector machines, which are denoted as SVM-1, SVM-2, SVM-3, SVM-4 and SVM-5, respectively. Among them, SVM-1 is used to identify whether the type of the power quality interference signal is normal, normal is 1, and abnormal is -1. SVM-2 is used to identify whether the type of the power quality interference signal is a transient rise, transient rise is 1, and non-transient rise is -1. SVM-3 is used to identify whether the type of the power quality interference signal is a transient decrease, transient decrease is 1, and non-transient decrease is -1. SVM-4 is used to identify whether the type of the power quality interference signal is an interruption, interruption is 1, and non-interruption is -1. SVM-5 is used to identify whether the type of the power quality interference signal is a pulse, pulse is 1, and non-pulse is -1. If it is a non-pulse, it can be determined that the type of the power quality interference signal is harmonic.

[0068] In an optional solution, before inputting the denoised power quality interference signal into a type recognition model to obtain the type of the power quality interference signal, the method further includes:

[0069] The optimization step is to maximize the classification accuracy of the target support vector machine, and use the IGWO algorithm to optimize the parameters of the target support vector machine, wherein the parameters include a penalty factor and a kernel function parameter, and the target support vector machine is one of the support vector machines in the classification structure;

[0070] The third repetition step is to repeat the above optimization steps until the optimization of all the above support vector machines in the above classification structure is completed.

[0071] Specifically, the selection of the penalty factor of the support vector machine and the kernel function parameters of the support vector machine directly affects the learning and generalization ability of the support vector machine. By using the IGWO algorithm to optimize the penalty factor of the support vector machine and the kernel function parameters of the support vector machine, the performance of the support vector machine can be effectively adjusted to make the support vector machine more in line with the characteristics of the data, thereby improving the accuracy and stability of the support vector machine prediction.

[0072] In an optional solution, the above optimization steps can be implemented as follows:

[0073] An acquisition step, acquiring a data set, and randomly dividing the data set into a training set and a validation set according to a ratio of a:b, wherein the data set, the training set, and the validation set all include multiple sets of the training data;

[0074] Initialization step, initializing the size of the gray wolf population, the positions of the gray wolves in the gray wolf population, the optimization interval and the maximum number of iterations of the positions of the gray wolves, wherein the position of the gray wolf includes a value of the above parameters;

[0075] A training determination step, in which, when the values ​​of the parameters of the target support vector machine are the positions of the gray wolves, the target support vector machine is trained using the training set, and the fitness value of each of the gray wolves is determined based on the validation set and the trained target support vector machine, wherein the fitness value is the classification accuracy of the target support vector machine when the values ​​of the parameters of the target support vector machine are the positions of the gray wolves;

[0076] A fourth determination step, in descending order of the fitness values ​​of the gray wolves, the gray wolves with the top three fitness values ​​in the gray wolf population are respectively determined as the gray wolves of the first level, the gray wolves of the second level, and the gray wolves of the third level;

[0077] Update steps, according to and Update the position of each low-level gray wolf, where A = 2a·r1-a, C = 2·r2, is the position of the gray wolf at the first level in the t-th iteration, is the position of the gray wolf at the second level in the t-th iteration, is the position of the gray wolf at the third level in the t-th iteration, is the position of the i-th gray wolf of the above low level in the t-th iteration, is the position of the i-th gray wolf of the low level in the t+1-th iteration process, r1, r2 and r3 are all random numbers in the range [0,1], and the gray wolf of the low level is the gray wolf in the gray wolf population except the gray wolf of the first level, the gray wolf of the second level and the gray wolf of the third level;

[0078] Specifically, the gray wolf optimization algorithm is a new type of swarm intelligence algorithm proposed by simulating the hierarchical structure and hunting behavior of gray wolf packs in nature. Gray wolf groups implement a strict hierarchy when hunting prey: the first level is α wolf (α), which is the leader of the wolf pack and is responsible for commanding the activities of the wolf pack. The second level is β wolf (β), which is responsible for assisting α wolf and commanding low-level gray wolves. The third level is δ wolf (δ), which is responsible for executing the orders of α wolf and β and commanding the lowest level ω wolf (ω). In the initial stage of searching for prey, the gray wolf population cannot determine the exact location of the prey. The α wolf with the best physical fitness in the wolf pack will guide β wolf and δ wolf to find prey. By constantly updating their positions, the gray wolf population gradually approaches the location of the prey and finally captures the prey. Assume that the gray wolf population consists of N gray wolves X = [X1, X2, ..., X N ], if the search space is D-dimensional, then the position of the i-th gray wolf in the D-dimensional space is expressed as i=1,2,...,N.

[0079] Specifically, the traditional GWO algorithm also has certain limitations. For example, the convergence speed will slow down in the later stage, and it is easy to fall into the local optimal state. By running the domain weighted position update strategy, faster convergence speed and higher convergence accuracy can be achieved. In the GWO algorithm, when A≤1, the gray wolf attacks the prey and conducts a local search, and when A>1, the gray wolf disperses and conducts a global search. It can be seen from A=2a·r1-a that the parameter A changes continuously with the change of the control factor a. Therefore, the entire prey encirclement process is mainly completed by the linear change of the control factor a. The parameter a of the standard GWO algorithm is linearly decreasing, but the search process of the algorithm is not linear. Therefore, the linear change of the parameter a cannot adapt to the actual search situation. Especially when solving multimodal function problems, GWO is prone to fall into local optimization. Therefore, the present application proposes a nonlinear control factor In the early stages of the iteration, the decay rate of a is relatively small, resulting in the equation The fluctuation of A value is large, which makes it easier for the algorithm to get rid of the local optimal state, thereby enhancing its global search ability. In the later stage of iteration, the decay rate of a will become larger, so that the algorithm can find the local optimal solution faster. Therefore, the improved control strategy can effectively balance the global search and local utilization capabilities of the algorithm.

[0080] Specifically, the traditional GWO algorithm uses the average position of the three best gray wolves as the gray wolf position for the next iteration. However, this method does not take into account the level differences within the actual gray wolf population. In the actual optimization process, the fitness of the gray wolves α, β, and δ is guided to decrease in sequence, and the position of the α gray wolf is more likely to be close to the actual optimal solution. In order to highlight the global leadership of the optimal wolf to the population, we increase the position weight of the α wolf and reduce the weight of the δ wolf during the position update process. At the same time, in order to further prevent the algorithm from falling into a local optimum, we use a random weight strategy to update the next generation population, while ensuring that the weight coefficient of the α wolf is greater than the weight coefficient of the δ wolf. This application proposes a random weighted position update strategy:

[0081] The fourth repetitive step increases the value of t in sequence, repeats the training determination step, the fourth determination step and the update step, until the value of t is equal to or greater than the maximum number of iterations, and outputs the best position, which is the position of the gray wolf with the largest fitness value in the gray wolf population.

[0082] In an optional solution, the above types include at least: normal, transient rise, transient drop, interruption, pulse, and harmonic.

[0083] Specifically, the selection of the penalty factor of the support vector machine and the kernel function parameters of the support vector machine directly affects the learning and generalization ability of the support vector machine. By using the IGWO algorithm to optimize the penalty factor of the support vector machine and the kernel function parameters of the support vector machine, the performance of the support vector machine can be effectively adjusted to make the support vector machine more in line with the characteristics of the data, thereby improving the accuracy and stability of the support vector machine prediction.

[0084] Through the above embodiment, the VMD algorithm is used to denoise the power quality interference signal to obtain a denoised power quality interference signal, and the denoised power quality interference signal is used to determine the type of the above power quality interference signal, thereby solving the problem in the prior art that the power quality interference signal itself has noise interference that reduces the accuracy of type detection of the power quality interference signal.

[0085] Example 1

[0086] This application adopts an improved VMD algorithm to achieve denoising of power quality interference signals containing noise. Its ideal mathematical model is as follows: x(t) = 310.2×(coswt+0.65e -18(t-0.8) )(ε(t-0.125))×cos8wt), first, we use the VMD decomposition method to determine the k value, which represents the number of intrinsic modal components of the noisy power quality interference signal. Through the energy convergence factor method mentioned above, we calculate the energy convergence factor Δ value of different modal components. Given that Δ12=0.9325 and Δ23=0.0174<η, the decomposed k value is set to 2. The simulation diagram of the noisy power quality interference signal after VMD decomposition is shown in the figure below. Figure 4 As shown in FIG. 1 , in addition, the denoised intrinsic mode component components are superimposed and reconstructed to obtain the final denoised power quality interference signal. The comparison between the original power quality interference signal, the power quality interference signal with added noise and the denoised power quality interference signal is shown in FIG. Figure 5 .

[0087] The VMD algorithm uses a variable mode decomposition model to decompose complex power quality interference signals into a series of intrinsic mode components with different frequency characteristics. These intrinsic mode components accurately reflect the characteristics of the interference signal, enabling us to more accurately identify and analyze the properties of the power quality interference signal. In addition, the VMD algorithm can effectively suppress noise interference in the decomposition process and improve the signal-to-noise ratio of the power quality interference signal, thereby more accurately extracting useful information from the power quality interference signal. The transient oscillation interference signal is denoised using the VMD algorithm and SVD algorithm of the present application. As shown in Table 1, the calculation results of the signal-to-noise ratio (SNR, Signal-to-Noise Ratio) and root mean square error (RMSE, Root Mean Square Error) of the denoised power quality interference signal are compared. The results show that the VMD algorithm of the present application achieves a higher signal-to-noise ratio and a lower root mean square error for the denoised power quality interference signal. Therefore, the VMD algorithm of the present application has better denoising performance for complex component power quality interference signals containing multiple frequencies.

[0088] Table 1

[0089] Denoising algorithm Signal-to-noise ratio before denoising SNR after denoising Root mean square error / % VMD algorithm of this application 14 19.27 12.3 SVD method 14 16.56 15.8

[0090] In summary, compared with other algorithms, the advantages of VMD algorithm in noise reduction are mainly reflected in the following aspects:

[0091] Adaptive decomposition capability: The VMD algorithm can adaptively decompose complex signals into a series of intrinsic mode components. This adaptive decomposition feature enables the VMD algorithm to effectively handle different types of noise without the need to preset a decomposition mode.

[0092] Noise robustness: The VMD algorithm can effectively suppress noise interference and improve the signal-to-noise ratio during the decomposition process. This means that even in a high-noise environment, the VMD algorithm can accurately extract useful signals and achieve high-quality noise reduction effects.

[0093] Avoiding endpoint effects: Compared with some traditional signal decomposition algorithms, the VMD algorithm avoids the endpoint effect problem. Traditional signal decomposition methods may lead to inaccurate decomposition results due to insufficient signal endpoint information. However, the VMD algorithm can reversely solve this problem, thereby improving the accuracy and stability of noise reduction.

[0094] Low computational complexity: The computational complexity of the VMD algorithm is relatively low, and it can process large-scale data quickly. This makes the VMD algorithm more efficient and practical in practical applications, especially in scenarios that require real-time processing.

[0095] Example 2

[0096] MATLAB software was used to randomly generate 6000 data samples, including 6 types of power quality interference signals, namely normal, temporary rise, temporary drop, interruption, pulse and harmonic. The number of data samples for each interference signal was 1000. 700 interference signals from different sources were selected as training sets, and the rest were used as test sets. The classification accuracy of the type recognition model of the present application for various interference signals under different noise intensities is shown in Table 3. It can be seen from the table that the accuracy of various signals is relatively high, with an average of 99.20%. At 0dB and 40dB, the highest accuracy can reach 100%. The overall analysis shows that the type recognition model of the present application can achieve good classification and recognition effects, indicating that the type recognition model of the present application has high robustness. The numerical example successfully verified the effectiveness of the type recognition model of the present application, and can realize accurate recognition of power quality interference signals.

[0097] Table 2

[0098]

[0099] In order to further prove the effectiveness of the type recognition model of this application, we tested the perturbation recognition accuracy of the deep learning model, CNN model and the proposed model (type recognition model of this application) under different data samples. The relevant test results are as follows: Figure 6 As shown in the figure above, it can be seen that the power quality disturbance recognition accuracy of the type recognition model of the present application is better than that of the traditional deep learning model and CNN model, and its advantages are more prominent in the case of small sample training sets. In fact, the support vector machine is good at processing small sample data, and its classification effect is not only related to the number of training samples, but also to the distribution of training samples. When the number of training samples is small, the support vector machine can better handle uneven data distribution, thereby maintaining a high accuracy rate. In the case of fewer samples, deep learning models and CNN models tend to overfit, resulting in a decrease in test data performance. However, by using the principle of structural risk minimization and the selection of the optimal hyperplane, the support vector machine can reduce the generalization error while ensuring the training error, thereby showing a higher accuracy rate in the case of small samples.

[0100] In order to further verify the feasibility of the type identification model of this application in the distribution network, a corresponding case verification was carried out in the IEEE-33 distribution network. The structural diagram of the IEEE-33 node system is as follows: Figure 7 As shown in the figure, the random disturbance signal (noise source) is superimposed on the voltage curves of nodes 14, 18 and 26 under normal conditions to obtain the fault disturbance data set. The final test results are as follows Figure 8As shown, rectangles of different grayscales represent recognition accuracy under different numbers of test samples. It is not difficult to find that the recognition accuracy of power quality interference of node 18 is higher than that of nodes 14 and 26. The noise signal source near the root node will affect the subsequent child nodes, making the randomness of the self-added noise more prominent. The noise signal source of node 18 will affect nodes 14 and 26, thereby increasing the difficulty of recognition. However, in general, under complex conditions, the proposed method can still maintain a recognition accuracy of more than 91%.

[0101] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0102] The embodiment of the present application also provides a power quality interference signal type detection device. It should be noted that the power quality interference signal type detection device of the embodiment of the present application can be used to execute the power quality interference signal type detection method provided by the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0103] The following is an introduction to the power quality interference signal type detection device provided in an embodiment of the present application.

[0104] Fig. 9 is a structural block diagram of a device for detecting the type of power quality interference signal according to an embodiment of the present application. Fig. 9 As shown, the device comprises:

[0105] The decomposition unit 10 is used to decompose the power quality interference signal by using the VMD algorithm to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components;

[0106] In an optional solution, the above decomposition unit includes:

[0107] A first decomposition module is used to perform a first decomposition step, and when the number of decomposition modes is m, the VMD algorithm is used to decompose the power quality interference signal to obtain an mth candidate intrinsic mode data, wherein the mth candidate intrinsic mode data includes: mth intrinsic mode components and the center frequencies of the mth intrinsic mode components, where m is an integer and an initial value of m is 2;

[0108] A second decomposition module is used to perform a second decomposition step, and when the number of decomposition modes is m+1, the power quality interference signal is decomposed by using the VMD algorithm to obtain the m+1th candidate eigenmode data, wherein the m+1th candidate eigenmode data includes: the m+1th eigenmode components and the center frequency of each of the eigenmode components;

[0109] a first determining module, configured to execute a first determining step, based on the mth candidate intrinsic mode data, the m+1th candidate intrinsic mode data and Determine the energy convergence factor, where Δ mm+1 is the energy convergence factor, s is the power quality interference signal, u m is the mth eigenmode component in the mth candidate eigenmode data, u m+1 is the m+1th eigenmode component in the m+1th candidate eigenmode data, is the square of the 2-norm of s, for u m The square of the 2-norm of for u m+1 The square of the 2-norm of ;

[0110] A first repetition module is used to execute the first repetition step, sequentially increase the value of m, and repeat the first decomposition step, the second decomposition step, and the determination step until the energy convergence factor is less than or equal to a predetermined value;

[0111] The second determination module is used to perform a second determination step, and when the energy convergence factor is less than the predetermined value, determine the mth candidate eigenmode data as the target eigenmode data.

[0112] An extraction unit 20, configured to extract a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency;

[0113] A reconstruction unit 30, configured to reconstruct the denoised intrinsic mode component to obtain a denoised power quality interference signal;

[0114] The determination unit 40 is used to determine the type of the power quality interference signal according to the de-noised power quality interference signal.

[0115] In an optional solution, the determining unit includes:

[0116] An input module is used to input the above-mentioned denoised power quality interference signal into a type recognition model to obtain the above-mentioned type of the above-mentioned power quality interference signal. The above-mentioned type recognition model is obtained by training a classification structure with multiple groups of training data, and each group of the above-mentioned training data at least includes: the above-mentioned power quality interference signal and the type of the above-mentioned power quality interference signal obtained in a historical period. The above-mentioned classification structure is composed of multiple support vector machines.

[0117] In an optional solution, the classification structure is composed of N support vector machines, and the input module includes:

[0118] An input submodule, used for executing the input step, inputting the denoised power quality interference signal into the nth support vector machine, and obtaining a classification result, wherein the classification result is that the type of the power quality interference signal is the type corresponding to the nth support vector machine or the type of the power quality interference signal is not the type corresponding to the nth support vector machine, one support vector machine corresponds to one type, n is a positive integer, the initial value of N is 1, and n≤N;

[0119] A repeating submodule, used for executing the second repeating step, increasing the value of N in sequence, and repeating the inputting step at least once, until the classification result is the type corresponding to the Nth support vector machine or n is greater than N;

[0120] A determination submodule is used to execute the third determination step. When the above-mentioned classification result is the above-mentioned type corresponding to the above-mentioned Nth support vector machine, the above-mentioned type corresponding to the above-mentioned Nth support vector machine is determined to be the above-mentioned type of the above-mentioned power quality interference signal. When n is greater than N and the above-mentioned classification result is not the above-mentioned type corresponding to the above-mentioned Nth support vector machine, the non-identification type is determined to be the above-mentioned type of the above-mentioned power quality interference signal, and the above-mentioned non-identification type is a type other than the above-mentioned types corresponding to all the above-mentioned support vector machines.

[0121] In an optional solution, the above device also includes:

[0122] An optimization unit, used for the optimization step, taking the maximum classification accuracy of the target support vector machine as the goal, and using the IGWO algorithm to optimize the parameters of the target support vector machine, the parameters including: a penalty factor and a kernel function parameter, the target support vector machine being one of the support vector machines in the classification structure;

[0123] The repeating unit is used for the third repeating step, repeating the above optimization steps until the optimization work of all the above support vector machines in the above classification structure is completed.

[0124] In an optional solution, the optimization unit includes:

[0125] An acquisition module is used to execute the acquisition step, acquire a data set, and randomly divide the data set into a training set and a validation set according to a ratio of a:b, wherein the data set, the training set, and the validation set all include multiple sets of the training data;

[0126] An initialization module is used to perform an initialization step, to initialize the size of the gray wolf population, the positions of the gray wolves in the gray wolf population, the optimization interval of the positions of the gray wolves, and the maximum number of iterations, wherein the positions of the gray wolves include a value of the parameters;

[0127] A training determination module, used to execute the training determination step, in the case where the values ​​of the parameters of the target support vector machine are the positions of the gray wolves, the target support vector machine is trained using the training set, and the fitness value of each of the gray wolves is determined based on the validation set and the trained target support vector machine, the fitness value being the classification accuracy of the target support vector machine in the case where the values ​​of the parameters of the target support vector machine are the positions of the gray wolves;

[0128] A third determination module is used to execute the fourth determination step, and determine the gray wolves with the top three fitness values ​​in the gray wolf population as the gray wolves of the first level, the gray wolves of the second level, and the gray wolves of the third level, respectively, in the order of the fitness values ​​of the gray wolves from large to small;

[0129] Update module, used to perform the update step, according to and Update the position of each low-level gray wolf, where A = 2a·r1-a, C = 2·r2, is the position of the gray wolf at the first level in the t-th iteration, is the position of the gray wolf at the second level in the t-th iteration, is the position of the gray wolf at the third level in the t-th iteration, is the position of the i-th gray wolf of the above low level in the t-th iteration, is the position of the i-th gray wolf of the low level in the t+1-th iteration process, r1, r2 and r3 are all random numbers in the range [0,1], and the gray wolf of the low level is the gray wolf in the gray wolf population except the gray wolf of the first level, the gray wolf of the second level and the gray wolf of the third level;

[0130] The second repetition module is used for the fourth repetition step, sequentially increasing the value of t, repeating the above training determination step, the above fourth determination step and the above updating step until the value of t is equal to or greater than the maximum number of iterations, and outputting the best position, which is the position of the gray wolf with the largest fitness value in the above gray wolf population.

[0131] In an optional solution, the above types include at least: normal, transient rise, transient drop, interruption, pulse, and harmonic.

[0132] Through the above embodiment, the VMD algorithm is used to denoise the power quality interference signal to obtain a denoised power quality interference signal, and the denoised power quality interference signal is used to determine the type of the above power quality interference signal, thereby solving the problem in the prior art that the power quality interference signal itself has noise interference that reduces the accuracy of type detection of the power quality interference signal.

[0133] The above-mentioned power quality interference signal type detection device includes a processor and a memory, and the above-mentioned decomposition unit, extraction unit, reconstruction unit and determination unit are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in the form of any combination.

[0134] The processor includes a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of reducing the accuracy of the type detection of the power quality interference signal due to the interference of noise in the power quality interference signal itself in the prior art is solved by adjusting the kernel parameters.

[0135] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0136] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the power quality interference signal type detection method.

[0137] Specifically, the power quality interference signal type detection method includes:

[0138] Step S201, using a VMD algorithm to decompose the power quality interference signal to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components;

[0139] Step S202, extracting a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency;

[0140] Step S203, reconstructing the denoised intrinsic mode components to obtain a denoised power quality interference signal;

[0141] Step S204: determining the type of the power quality interference signal according to the denoised power quality interference signal.

[0142] An embodiment of the present invention provides a processor, and the processor is used to run a program, wherein the power quality interference signal type detection method is executed when the program is running.

[0143] Specifically, the power quality interference signal type detection method includes:

[0144] Step S201, using a VMD algorithm to decompose the power quality interference signal to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components;

[0145] Step S202, extracting a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency;

[0146] Step S203, reconstructing the denoised intrinsic mode components to obtain a denoised power quality interference signal;

[0147] Step S204: determining the type of the power quality interference signal according to the denoised power quality interference signal.

[0148] An embodiment of the present invention provides a device, the device including a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, at least the following steps are implemented:

[0149] Step S201, using a VMD algorithm to decompose the power quality interference signal to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components;

[0150] Step S202, extracting a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency;

[0151] Step S203, reconstructing the denoised intrinsic mode components to obtain a denoised power quality interference signal;

[0152] Step S204: determining the type of the power quality interference signal according to the denoised power quality interference signal.

[0153] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing at least the following method steps:

[0154] Step S201, using a VMD algorithm to decompose the power quality interference signal to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components;

[0155] Step S202, extracting a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency;

[0156] Step S203, reconstructing the denoised intrinsic mode components to obtain a denoised power quality interference signal;

[0157] Step S204: determining the type of the power quality interference signal according to the denoised power quality interference signal.

[0158] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0159] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0160] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0161] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0162] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0163] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0164] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0165] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0166] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0167] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0168] 1) In the power quality interference signal type detection method of the present application, the VMD algorithm is used to denoise the power quality interference signal to obtain a denoised power quality interference signal, and the denoised power quality interference signal is used to determine the type of the above-mentioned power quality interference signal, thereby solving the problem in the prior art that the power quality interference signal itself has noise interference that reduces the accuracy of the type detection of the power quality interference signal.

[0169] 2) In the power quality interference signal type detection device of the present application, the VMD algorithm is used to denoise the power quality interference signal to obtain a denoised power quality interference signal, and the denoised power quality interference signal is used to determine the type of the above-mentioned power quality interference signal, thereby solving the problem in the prior art that the power quality interference signal itself has noise interference that reduces the accuracy of the type detection of the power quality interference signal.

[0170] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting the type of power quality interference signal, characterized in that: The method comprises: The VMD algorithm is used to decompose the power quality interference signal to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components; Extracting a plurality of denoised eigenmode components from the plurality of eigenmode components, the denoised eigenmode components being the eigenmode components satisfying a predetermined condition, the predetermined condition being that the center frequency is less than or equal to a predetermined frequency; Reconstructing the denoised intrinsic mode component to obtain a denoised power quality interference signal; According to the denoised power quality interference signal, a type of the power quality interference signal is determined.

2. The method according to claim 1, characterized in that: The VMD algorithm is used to decompose the power quality interference signal to obtain the target intrinsic mode data, including: In the first decomposition step, when the number of decomposed modes is m, the VMD algorithm is used to decompose the power quality interference signal to obtain the mth candidate eigenmode data, wherein the mth candidate eigenmode data includes: m eigenmode components and the center frequencies of the mth eigenmode components. m is an integer, and the initial value of m is 2; A second decomposition step, when the decomposition mode number is m+1, using the VMD algorithm to decompose the power quality interference signal to obtain the m+1th candidate eigenmode data, wherein the m+1th candidate eigenmode data includes: m+1 eigenmode components and the center frequency of each eigenmode component; The first determination step is to determine the mth candidate eigenmode data, the m+1th candidate eigenmode data and Determine the energy convergence factor, where Δ mm+1 is the energy convergence factor, s is the power quality interference signal, u m is the mth eigenmode component in the mth candidate eigenmode data, u m+1 is the m+1th eigenmode component in the m+1th candidate eigenmode data, is the square of the 2-norm of s, for u m The square of the 2-norm of for u m+1 The square of the 2-norm of ; A first repetition step of sequentially increasing the value of m, repeating the first decomposition step, the second decomposition step and the determination step until the energy convergence factor is less than or equal to a predetermined value; The second determining step is to determine the mth candidate eigenmode data as the target eigenmode data when the energy convergence factor is less than the predetermined value.

3. The method according to claim 1, characterized in that Determining the type of the power quality interference signal according to the denoised power quality interference signal includes: The denoised power quality interference signal is input into a type recognition model to obtain the type of the power quality interference signal. The type recognition model is obtained by training a classification structure using multiple groups of training data. Each group of training data at least includes: the power quality interference signal and the type of the power quality interference signal acquired in a historical period. The classification structure is composed of multiple support vector machines.

4. The method according to claim 3, characterized in that The classification structure is composed of N support vector machines, and the denoised power quality interference signal is input into a type recognition model to obtain the type of the power quality interference signal, including: The input step includes inputting the denoised power quality interference signal into the nth support vector machine to obtain a classification result, wherein the classification result is that the type of the power quality interference signal is the type corresponding to the nth support vector machine or the type of the power quality interference signal is not the type corresponding to the nth support vector machine, one support vector machine corresponds to one type, and n is a positive integer. The initial value of N is 1, and n≤N; A second repetition step of increasing the value of N in sequence, and repeating the input step at least once until the classification result is the type corresponding to the Nth support vector machine or n is greater than N; The third determination step, when the classification result is the type corresponding to the Nth support vector machine, determines that the type corresponding to the Nth support vector machine is the type of the power quality interference signal, and when n is greater than N and the classification result is not the type corresponding to the Nth support vector machine, determines that the non-identification type is the type of the power quality interference signal, and the non-identification type is a type other than the types corresponding to all the support vector machines.

5. The method according to claim 3, characterized in that: Before inputting the denoised power quality interference signal into a type identification model to obtain the type of the power quality interference signal, the method further includes: An optimization step, taking the maximum classification accuracy of the target support vector machine as the goal, using the IGWO algorithm to optimize the parameters of the target support vector machine, the parameters including: a penalty factor and a kernel function parameter, the target support vector machine being one of the support vector machines in the classification structure; The third repetition step is to repeat the optimization step until the optimization of all the support vector machines in the classification structure is completed.

6. The method according to claim 5, characterized in that The optimization step, with the classification accuracy of the target support vector machine as the maximum, uses the IGWO algorithm to optimize the parameters of the target support vector machine, including: An acquisition step, acquiring a data set, and randomly dividing the data set into a training set and a validation set according to a ratio of a:b, wherein the data set, the training set and the validation set all include multiple groups of the training data; An initialization step, initializing the size of the gray wolf population, the positions of the gray wolves in the gray wolf population, the optimization interval of the positions of the gray wolves and the maximum number of iterations, wherein the position of the gray wolf includes a value of the parameter; A training determination step, in which the target support vector machine is trained using the training set when the values ​​of the parameters of the target support vector machine are the positions of the gray wolves respectively, and the fitness value of each gray wolf is determined based on the validation set and the trained target support vector machine, wherein the fitness value is the classification accuracy of the target support vector machine when the values ​​of the parameters of the target support vector machine are the positions of the gray wolves; The fourth determination step is to determine the gray wolves with the top three fitness values ​​in the gray wolf population as the gray wolves of the first level, the gray wolves of the second level, and the gray wolves of the third level, respectively, in the order of the fitness values ​​of the gray wolves from large to small; Update steps, according to and Update the position of each low-level gray wolf, where A = 2a·r1-a, C = 2·r2, is the position of the gray wolf at the first level during the t-th iteration, is the position of the gray wolf at the second level during the t-th iteration, is the position of the gray wolf at the third level during the t-th iteration, is the position of the i-th low-level gray wolf in the t-th iteration, is the position of the i-th gray wolf of the low level in the t+1-th iteration process, r1, r2 and r3 are all random numbers in the range [0,1], and the gray wolf of the low level is the gray wolf in the gray wolf population except the gray wolf of the first level, the gray wolf of the second level and the gray wolf of the third level; The fourth repetitive step increases the value of t in sequence, repeats the training determination step, the fourth determination step and the update step, until the value of t is equal to or greater than the maximum number of iterations, and outputs the best position, which is the position of the gray wolf with the largest fitness value in the gray wolf population.

7. The method according to any one of claims 1 to 6, characterized in that: The types include at least: normal, transient rise, transient drop, interruption, pulse, and harmonic.

8. A device for detecting the type of power quality interference signal, characterized in that: The device comprises: A decomposition unit, used to decompose the power quality interference signal by using a VMD algorithm to obtain target intrinsic mode data, wherein the target intrinsic mode data includes: a plurality of intrinsic mode components and a center frequency of each of the intrinsic mode components; An extraction unit, configured to extract a plurality of denoised eigenmode components from the plurality of eigenmode components, wherein the denoised eigenmode components are the eigenmode components satisfying a predetermined condition, wherein the predetermined condition is that the center frequency is less than or equal to a predetermined frequency; A reconstruction unit, used for reconstructing the denoised intrinsic mode component to obtain a denoised power quality interference signal; A determination unit is used to determine the type of the power quality interference signal according to the denoised power quality interference signal.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the power quality interference signal type detection method according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the method for detecting the type of power quality interference signal described in any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Generalized S transformation and SVM electric energy quality disturbance efficient identification method

    CN111368892A

  • High-frequency power supply welding arc voltage signal denoising method and device and medium

    CN117828273A

Cited By

  • Surge performance test method for silicon carbide MOSFET

    CN119758015A