Power distribution network power supply quality analysis method, apparatus and device, and storage medium

Abstract

The invention discloses a power distribution network power supply quality analysis method, apparatus and device, and a storage medium. The method comprises the steps of obtaining operation data of a power distribution network; determining the type of a power supply event of the power distribution network according to the operation data; wherein the types of the power supply events of the power distribution network comprise transient state events and steady state events; extracting corresponding features in the operation data, and selecting a plurality of target features for power supply quality analysis of the power distribution network from the extracted features according to the type of the power supply event of the power distribution network; performing empirical mode decomposition on the target features to obtain corresponding optimized feature components; and inputting the optimized characteristic component into a preset power distribution network power supply quality analysis model, so that the power distribution network power supply quality analysis model analyzes the power distribution network power supply quality problem according to the optimized characteristic component, and outputs the corresponding power distribution network power supply quality problem. According to the invention, the accuracy of power distribution network power supply quality problem analysis can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device, equipment and storage medium for analyzing power supply quality of a distribution network. Background Art

[0002] With the rapid progress of renewable energy and electric vehicle technology, the use of renewable energy such as distributed photovoltaic and wind power, and the popularization of electric vehicles as clean transportation have brought unprecedented challenges and opportunities to the power system. The large-scale access of distributed energy and loads has undoubtedly injected new vitality into the power system, but it has also caused many problems in the power supply quality of the distribution network, such as voltage deviation, harmonics and other power supply quality problems. Usually, these power supply quality problems can be divided into steady-state and transient according to the time scale of occurrence. Among them, steady-state power quality problems mainly include: voltage deviation, frequency deviation, three-phase voltage imbalance, harmonics, voltage fluctuations and voltage flicker. Transient power quality problems mainly include voltage swell, sag, interruption, transient oscillation and transient pulses.

[0003] However, existing power quality analysis methods are often based on fixed thresholds for analysis. When faced with complex and changing distribution network environments, these methods often cannot accurately identify the type and extent of power quality problems and have low accuracy. Especially in the case of large-scale access to distributed energy and electric vehicle loads, power quality problems in the power grid have become more complex and diverse, and traditional analysis methods are difficult to cope with. Summary of the invention

[0004] The present invention provides a power supply quality analysis method, device, equipment and storage medium for a distribution network, so as to solve the technical problem that the existing power supply quality analysis method is often based on a fixed threshold for analysis and has low accuracy.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for analyzing power supply quality of a distribution network, comprising:

[0006] Acquire a number of power supply events of the distribution network and operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network;

[0007] For each power supply event of the distribution network, the change amount of the corresponding operating data within the preset time period is calculated, and then the type of the power supply event of the distribution network is determined according to the change amount and the preset threshold value; wherein the types of power supply events of the distribution network include: transient events and steady-state events;

[0008] Extracting corresponding features from the operation data, and selecting a number of target features for power supply quality analysis of the distribution network from the extracted features according to the type of power supply events of the distribution network;

[0009] Performing empirical mode decomposition on the target features to obtain corresponding optimized feature components;

[0010] Inputting the optimized characteristic components into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the power supply quality problem of the distribution network power supply event according to the optimized characteristic components, and outputs the power supply quality problem corresponding to the distribution network power supply event;

[0011] The power supply quality analysis model of the distribution network is obtained by training a preset SVM model with the optimized feature components corresponding to the operation data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

[0012] As a preferred solution, before obtaining the changes in voltage and current of the distribution network within a preset time period according to the operation data, the method further includes:

[0013] The operation data is preprocessed; wherein the preprocessing includes: denoising and filtering.

[0014] As a preferred solution, the extracting corresponding features from the operation data includes:

[0015] Extract the time series before and after the power supply event of the distribution network;

[0016] Extracting time domain features of power supply events of the distribution network from the time series; wherein the time domain features include: duration of the event, amplitude change, and amplitude change rate;

[0017] Performing Fourier transformation on the time series, and then extracting frequency domain features of the power supply event of the distribution network from the transformed time series; wherein the frequency domain features include: the main frequency, frequency band energy and spectrum entropy of the event;

[0018] Performing statistical analysis on the time series to obtain statistical characteristics of power supply events in the distribution network; wherein the statistical characteristics include: mean, variance, skewness and kurtosis of the events;

[0019] The time domain features, frequency domain features and statistical features are used as features corresponding to the operating data.

[0020] As a preferred solution, performing empirical mode decomposition on the target feature to obtain the corresponding optimized feature component includes:

[0021] When the power supply event of the distribution network is a steady-state event, obtaining a preset number of extracted optimized characteristic components;

[0022] Performing empirical mode decomposition on the target feature according to the extraction quantity, to obtain optimized feature components corresponding to the extraction quantity and a residual component.

[0023] As a preferred solution, performing empirical mode decomposition on the target feature to obtain the corresponding optimized feature component also includes:

[0024] When the power supply event of the distribution network is a transient event, obtaining a preset number of extracted optimized characteristic components;

[0025] According to the extracted quantity, empirical mode decomposition is performed on the time series before and after the power supply event of the distribution network occurs, so as to obtain optimized characteristic components corresponding to the extracted quantity and a residual component.

[0026] As a preferred solution, the process of empirical mode decomposition includes:

[0027] According to the target feature or the time series before and after the power supply event of the distribution network occurs, an initial residual component is obtained;

[0028] Performing an optimization feature extraction operation on the initial residual component to extract the corresponding optimization feature component;

[0029] The difference between the target feature or the time series before and after the power supply event of the distribution network occurs and the extracted optimized feature component is used as a new residual component, and the optimized feature extraction operation is continued to be performed on the feature component until the number of extracted feature components meets the extraction number;

[0030] Wherein, the optimization feature extraction operation includes:

[0031] Find all the maximum and minimum points of the initial residual component, and then fit the maximum points according to the cubic spline interpolation method to obtain the corresponding upper envelope, and fit the minimum points according to the cubic spline interpolation method to obtain the corresponding lower envelope;

[0032] Calculating the average value of the upper envelope and the lower envelope;

[0033] It is determined whether the difference between the initial residual component and the average value satisfies the preset condition of the optimized characteristic component. If so, the difference is used as the optimized characteristic component.

[0034] As a preferred solution, the SVM model is specifically:

[0035]

[0036] Where J is the loss value of the fitting, w and b are the slope and intercept of the hyperplane respectively, κ is the penalty coefficient, ξq is the slack variable, σ is the parameter of the radial basis kernel function φ(x), T is the time or the set of time points, N is the number of iterations, st is the constraint condition, and y is the number of iterations. q is the label or category of the sample, representing different types of power quality problems.

[0037] Based on the above embodiment, another embodiment of the present invention provides a power supply quality analysis device for a distribution network, comprising: a data acquisition module, an event type determination module, a feature extraction module, an empirical mode decomposition module, and a power supply quality problem identification module;

[0038] The data acquisition module is used to acquire a number of power supply events of the distribution network and the operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network;

[0039] The event type determination module is used to calculate the change amount of the corresponding operating data within a preset time period for each power supply event of the distribution network, and then determine the type of the power supply event of the distribution network according to the change amount and a preset threshold; wherein the types of power supply events of the distribution network include: transient events and steady-state events;

[0040] The feature extraction module is used to extract the corresponding features in the operation data, and select a number of target features for power supply quality analysis of the distribution network from the extracted features according to the type of power supply events of the distribution network;

[0041] The empirical mode decomposition module is used to perform empirical mode decomposition on the target feature to obtain corresponding optimized feature components;

[0042] The power supply quality problem identification module is used to input the optimized characteristic components into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the power supply quality problems of the distribution network power supply events according to the optimized characteristic components, and outputs the power supply quality problems corresponding to the distribution network power supply events; wherein the distribution network power supply quality analysis model is obtained by training a preset SVM model with the optimized characteristic components corresponding to the operating data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

[0043] Based on the above embodiments, another embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, the distribution network power supply quality analysis method described in the above invention embodiment is implemented.

[0044] Based on the above embodiments, another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the distribution network power supply quality analysis method described in the above invention embodiment.

[0045] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0046] The present invention provides a method for analyzing the power supply quality of a distribution network, which obtains a plurality of power supply events of the distribution network and operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network; for each power supply event of the distribution network, the change amount of the corresponding operation data within a preset time period is calculated, and then the type of the power supply event of the distribution network is determined according to the change amount and a preset threshold; wherein the types of the power supply events of the distribution network include: transient events and steady-state events; the corresponding features in the operation data are extracted, and according to the types of the power supply events of the distribution network, a plurality of target features for analyzing the power supply quality of the distribution network are selected from the extracted features; the target features are subjected to empirical mode decomposition to obtain corresponding optimized feature components; the optimized feature components are input into a preset power supply quality analysis model of the distribution network, so that the power supply quality analysis model of the distribution network analyzes the power supply quality problems of the power supply events of the distribution network according to the optimized feature components, and outputs the power supply quality problems corresponding to the power supply events of the distribution network; wherein the power supply quality analysis model of the distribution network is obtained by training a preset SVM model with the optimized feature components corresponding to the operation data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

[0047] Compared with the traditional threshold comparison method, the present invention provides a distribution network power supply quality analysis method based on the empirical mode decomposition algorithm and the SVM algorithm. The empirical mode decomposition algorithm is used to decompose the features extracted from the distribution network operation data, which can effectively extract the key feature information in the voltage and current waveforms, providing strong support for subsequent analysis. And the combination of the empirical mode decomposition algorithm and the SVM algorithm can realize the accurate identification and classification of steady-state and transient power supply quality problems in the distribution network system, improve the accuracy of the analysis of the distribution network power supply quality problems, and provide strong guarantee for the stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of a method for analyzing power supply quality of a distribution network provided by an embodiment of the present invention;

[0049] Figure 2 It is a structural schematic diagram of a distribution network power supply quality analysis device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0052] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0053] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0054] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0055] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0056] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0057] Embodiment 1

[0058] Please refer to Figure 1 , is a flow chart of a method for analyzing power quality of a distribution network provided by an embodiment of the present invention, comprising the following specific steps:

[0059] S1. Acquire a number of power supply events of the distribution network and operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network;

[0060] Preferably, before obtaining the changes in voltage and current of the distribution network within a preset time period based on the operating data, the method further includes: preprocessing the operating data; wherein the preprocessing includes: denoising and filtering.

[0061] Specifically, the present invention provides a power supply quality analysis method for a distribution network system based on an empirical mode decomposition (EMD) algorithm and a support vector machine (SVM) algorithm, and the power supply quality analysis process of the distribution network is as follows:

[0062] (1) Data acquisition and preprocessing: Real-time data collection of power supply events in the distribution network, as well as the operating data corresponding to each power supply event in the distribution network—voltage and current waveform data, and necessary preprocessing, such as denoising and filtering, is performed.

[0063] S2. For each power supply event of the distribution network, calculate the change amount of the corresponding operating data within a preset time period, and then determine the type of the power supply event of the distribution network according to the change amount and a preset threshold; wherein the types of power supply events of the distribution network include: transient events and steady-state events;

[0064] (2) Identification of steady-state and transient problems: Power supply quality problems are divided into steady-state and transient problems based on the time scale of occurrence, mainly based on the operating characteristics and changing states of the power system at different time scales.

[0065] Steady state refers to the equilibrium state of the power system under normal operation, when the voltage, current and power in the system are maintained at a certain stable value. Steady state analysis mainly studies the operating characteristics of the power system under steady state, such as voltage stability, frequency stability, etc. Steady state power supply quality problems are mainly characterized by waveform distortion, which generally lasts for a long time and is an abnormal power supply quality that occurs over a period of time (usually more than 1 minute).

[0066] Transient refers to the process of power system transitioning from one steady state to another, which is usually caused by load changes, fault removal or power supply input. The transient process usually lasts for a short time, but may cause adverse effects such as voltage reduction and current increase. Transient analysis mainly studies the dynamic response characteristics of the power system during the transient process, such as voltage recovery time, current decay time, etc. The frequency of transient power quality problems is low, but the consequences are more serious. The event characteristic description quantity is generally the voltage "effective value" of the event process and its duration (sometimes considering its rise and fall rate, etc.).

[0067] Event detection: Use the threshold method or sliding window method to detect the occurrence of steady-state events and transient events. Specifically, set reasonable thresholds or window parameters. When the change in voltage or current exceeds the threshold or the statistical characteristics (such as mean, variance) in the window change significantly, it is considered that a transient event has occurred; otherwise, it is considered that a steady-state event has occurred.

[0068] S3, extracting corresponding features from the operation data, and selecting a number of target features for power supply quality analysis of the distribution network from the extracted features according to the type of power supply events of the distribution network;

[0069] Preferably, the extracting corresponding features from the operating data includes: extracting the time series before and after the distribution network power supply event occurs; extracting the time domain features of the distribution network power supply event from the time series; wherein the time domain features include: the duration of the event, the amplitude change and the amplitude change rate; performing Fourier transform on the time series, and then extracting the frequency domain features of the distribution network power supply event from the transformed time series; wherein the frequency domain features include: the main frequency, frequency band energy and spectrum entropy of the event; performing statistical analysis on the time series to obtain the statistical features of the distribution network power supply event; wherein the statistical features include: the mean, variance, skewness and kurtosis of the event; and using the time domain features, frequency domain features and statistical features as the features corresponding to the operating data.

[0070] (3) Feature extraction:

[0071] First, the time series analysis method is used to extract the features of transient voltage and current waveform data. Then, according to the specific requirements of the problem and the characteristics of the data, appropriate features are selected for subsequent analysis. The specific steps are as follows:

[0072] a) Time series segmentation: extract a time series before and after the detected event for further analysis and feature extraction;

[0073] b) Time domain feature extraction: extracting the time domain features of an event in the time domain, such as the duration, amplitude change, change rate, etc. of the event. The time domain features can directly reflect the intensity and duration of the event;

[0074] c) Frequency domain feature extraction: Perform Fourier transform (FFT) or wavelet transform (WT) on the time series of the event, convert it to the frequency domain for analysis, and extract the frequency domain features of the event, including the main frequency, frequency band energy, spectrum entropy, etc. The frequency domain features can reflect the frequency components and distribution of the event;

[0075] d) Statistical feature extraction: Statistical analysis is performed on the time series of events to extract statistical features, such as the mean, variance, skewness, kurtosis, etc. Statistical features can reflect the distribution characteristics and statistical laws of events;

[0076] e) Feature selection and optimization: According to the specific requirements of the problem and the characteristics of the data, select appropriate features for subsequent analysis and classification, and use feature selection algorithms (such as principal component analysis PCA, recursive feature elimination RFE, etc.) to optimize the feature set to improve the performance and generalization ability of the classifier;

[0077] Specific examples:

[0078] Case 1: Voltage sag identification

[0079] Problem Requirements: Identify voltage sag events in power systems.

[0080] Data characteristics: voltage waveform data, which may contain noise, rapid decline and recovery processes.

[0081] Selected Features:

[0082] Sag amplitude: the maximum value of voltage drop.

[0083] Sag duration: The length of time the voltage is below a certain threshold.

[0084] Energy loss: The reduction in energy in the voltage waveform during a sag.

[0085] Recovery time: The time from the end of a sag to the return of the voltage to normal levels.

[0086] Case 2: Power load forecasting

[0087] Problem requirement: Predict the power load in the future.

[0088] Data characteristics: historical load data, weather data, date type (weekdays / weekends / holidays), etc.

[0089] Selected Features:

[0090] Historical load: load data for the past few hours or days.

[0091] Temperature: A key factor affecting power consumption.

[0092] Humidity: Relevant to certain types of loads such as air conditioning.

[0093] Day Type: Load patterns are usually different on weekdays and weekends.

[0094] Holidays: The load pattern during holidays is different from that during normal days.

[0095] S4, performing empirical mode decomposition on the target feature to obtain corresponding optimized feature components;

[0096] Preferably, performing empirical mode decomposition on the target feature to obtain corresponding optimized feature components includes: when the power supply event of the distribution network is a steady-state event, obtaining a preset number of extracted optimized feature components; performing empirical mode decomposition on the target feature according to the extracted number to obtain the optimized feature components corresponding to the extracted number and a residual component.

[0097] Preferably, performing empirical mode decomposition on the target features to obtain corresponding optimized feature components also includes: when the power supply event in the distribution network is a transient event, obtaining a preset number of extracted optimized feature components; based on the extracted number, performing empirical mode decomposition on the time series before and after the power supply event in the distribution network occurs to obtain the corresponding extracted number of optimized feature components and a residual component.

[0098] Preferably, the process of empirical mode decomposition includes: obtaining an initial residual component according to the time series before and after the target feature or the power supply event of the distribution network occurs; performing an optimized feature extraction operation on the initial residual component to extract the corresponding optimized feature component; taking the difference between the time series before and after the target feature or the power supply event of the distribution network occurs and the extracted optimized feature component as a new residual component, and continuing to perform the optimized feature extraction operation on the feature component until the number of extracted feature components meets the extraction number; wherein the optimized feature extraction operation includes: finding all the maximum and minimum points of the initial residual component, and then fitting the maximum points according to the cubic spline interpolation method to obtain the corresponding upper envelope, and fitting the minimum points according to the cubic spline interpolation method to obtain the corresponding and lower envelope; calculating the average values ​​of the upper envelope and the lower envelope; judging whether the difference between the initial residual component and the average value meets the preset conditions of the optimized feature component, and if so, taking the difference as the optimized feature component.

[0099] (4) Feature extraction and optimization module: The EMD algorithm is used to decompose the extracted steady-state and transient problem features to obtain multiple optimized feature components (IMFs). The optimized feature components contain key information in the waveform, which can effectively characterize the characteristics of power supply quality problems and provide strong support for subsequent identification and classification.

[0100] Among them, for steady-state characteristics (such as voltage deviation, frequency deviation, etc.), EMD or optimized EMD algorithm can be directly applied for decomposition, including the following steps:

[0101] a) Initialization: setting the number of IMFs to be extracted or the stop condition (such as the energy threshold of the remaining components);

[0102] b) Decomposition: Apply EMD or EEMD / CEEMDAN algorithm to decompose the steady-state characteristics to obtain a series of IMF components and a residual component;

[0103] For transient characteristics (such as voltage sag, swell, etc.), special attention should be paid due to their short duration and rapid changes, including the following steps:

[0104] a) Event detection and segmentation: First, use the time series analysis method to detect transient events, and extract the time series before and after the event for segmentation;

[0105] b) Local EMD decomposition: Apply EMD or optimized EMD algorithm to each detected transient event segment for local decomposition. Since transient events usually last only for a short time, applying EMD to the entire time series may not be the most effective.

[0106] Among them, the EMD algorithm formula is as follows:

[0107] a) Initialization: Given a signal x(t), initialize the residual component r(t) = x(t).

[0108] b) Extract IMF:

[0109] Find all the maximum and minimum points of r(t).

[0110] The cubic spline interpolation method is used to fit the maximum and minimum points respectively to obtain the upper envelope u(t) and the lower envelope l(t).

[0111] Calculate the average value of the upper and lower envelopes m(t)=[u(t)+l(t)] / 2.

[0112] Subtracting the mean value from the residual component yields a new residual component r'(t) = r(t) - m(t).

[0113] Determine whether r'(t) satisfies the two conditions of IMF (i.e., the number of local extreme points and zero-crossing points is equal or differs by at most one, and the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is zero): If so, r'(t) is the first IMF; otherwise, take r'(t) as the new residual component r(t) and repeat the above steps.

[0114] c) Update the residual component: Subtract the extracted IMF from the original signal to obtain a new residual component r(t)=x(t)-IMF(t).

[0115] d) Iteration: Repeat steps 2 and 3 until the remaining component r(t) meets the stopping condition (such as a preset number of IMFs or an energy threshold of the remaining component).

[0116] Optimized EMD algorithms (such as EEMD or CEEMDAN), EEMD (Ensemble Empirical Mode Decomposition) and CEEMDAN (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) are improved versions of EMD that reduce mode aliasing by adding noise.

[0117] S5. Input the optimized characteristic components into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the power supply quality problems of the distribution network power supply events according to the optimized characteristic components, and outputs the power supply quality problems corresponding to the distribution network power supply events; wherein the distribution network power supply quality analysis model is obtained by training a preset SVM model with the optimized characteristic components corresponding to the operating data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

[0118] Preferably, the SVM model is specifically:

[0119]

[0120] Where J is the loss value of the fitting, w and b are the slope and intercept of the hyperplane respectively, κ is the penalty coefficient, ξq is the slack variable, σ is the parameter of the radial basis kernel function φ(x), T is the time or the set of time points, N is the number of iterations, st is the constraint condition, and y is the number of iterations. q is the label or category of the sample, representing different types of power quality problems.

[0121] (5) Identification and classification: The IMF obtained by feature extraction optimization is used as input data and input into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the distribution network power supply quality problem according to the optimized feature component and outputs the corresponding distribution network power supply quality problem. The distribution network power supply quality analysis model is an SVM model, which can automatically extract the internal correlation and rules in the IMF through training and learning, thereby realizing accurate identification and classification of power supply quality problems.

[0122] The core of the SVM algorithm is to transform the nonlinear data classification problem into a linear data classification problem in a high-dimensional space through the kernel function φ(x). The decision-making process of the algorithm can be equivalently described as the following mathematical programming problem:

[0123]

[0124] Where: J is the loss value of the fitting; w and b are the slope and intercept of the hyperplane respectively; κ is the penalty coefficient; ξq is the slack variable. The expression of φ(x) is:

[0125]

[0126] Where: σ is the parameter of the radial basis kernel function.

[0127] It can be seen that the present invention provides a method for analyzing power supply quality of a distribution network. Through the present invention, the following beneficial effects can be achieved:

[0128] (1) Decomposing the voltage and current waveform data through the EMD algorithm can effectively extract the key feature information in the waveform, providing strong support for subsequent identification and classification;

[0129] (2) Combining EMD and SVM algorithms, it can accurately identify and classify steady-state and transient power supply quality problems in new power systems, improve analysis accuracy and speed, and provide strong guarantees for the stable operation of power systems;

[0130] (3) The present invention can also customize feature extraction methods and classifier models according to different types of power supply quality problems to achieve flexible and diverse power supply quality identification and analysis requirements.

[0131] Embodiment 2

[0132] Please refer to Figure 2 , is a structural schematic diagram of a power supply quality analysis device for a distribution network provided by an embodiment of the present invention, the device comprising: a data acquisition module, an event type determination module, a feature extraction module, an empirical mode decomposition module and a power supply quality problem identification module;

[0133] The data acquisition module is used to acquire a number of power supply events of the distribution network and the operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network;

[0134] The event type determination module is used to calculate the change amount of the corresponding operating data within a preset time period for each power supply event of the distribution network, and then determine the type of the power supply event of the distribution network according to the change amount and a preset threshold; wherein the types of power supply events of the distribution network include: transient events and steady-state events;

[0135] The feature extraction module is used to extract the corresponding features in the operation data, and select a number of target features for power supply quality analysis of the distribution network from the extracted features according to the type of power supply events of the distribution network;

[0136] The empirical mode decomposition module is used to perform empirical mode decomposition on the target feature to obtain corresponding optimized feature components;

[0137] The power supply quality problem identification module is used to input the optimized characteristic components into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the power supply quality problems of the distribution network power supply events according to the optimized characteristic components, and outputs the power supply quality problems corresponding to the distribution network power supply events; wherein the distribution network power supply quality analysis model is obtained by training a preset SVM model with the optimized characteristic components corresponding to the operating data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

[0138] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0139] Those skilled in the art can clearly understand that, for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0140] Embodiment 3

[0141] Accordingly, an embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the distribution network power supply quality analysis method described in the above-mentioned embodiment of the invention when executing the computer program.

[0142] The electronic device may be a computing device such as a desktop computer, a notebook, a palmtop computer, a cloud server, etc. The device may include, but is not limited to, a processor and a memory.

[0143] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the device, and various interfaces and lines are used to connect various parts of the entire device.

[0144] Embodiment 4

[0145] Accordingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the distribution network power supply quality analysis method described in the above-mentioned embodiment of the invention.

[0146] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0147] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0148] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for analyzing power supply quality of a distribution network, characterized in that: include: Acquire a number of power supply events of the distribution network and operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network; For each power supply event of the distribution network, the change amount of the corresponding operating data within the preset time period is calculated, and then the type of the power supply event of the distribution network is determined according to the change amount and the preset threshold value; wherein the types of power supply events of the distribution network include: transient events and steady-state events; Extracting corresponding features from the operation data, and selecting a number of target features for power supply quality analysis of the distribution network from the extracted features according to the type of power supply events of the distribution network; Performing empirical mode decomposition on the target features to obtain corresponding optimized feature components; Inputting the optimized characteristic components into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the power supply quality problem of the distribution network power supply event according to the optimized characteristic components, and outputs the power supply quality problem corresponding to the distribution network power supply event; The power supply quality analysis model of the distribution network is obtained by training a preset SVM model with the optimized feature components corresponding to the operation data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

2. The method for analyzing power supply quality of a distribution network according to claim 1, characterized in that: Before obtaining the changes of the voltage and current of the distribution network within a preset time period according to the operation data, the method further includes: The operation data is preprocessed; wherein the preprocessing includes: denoising and filtering.

3. The method for analyzing power supply quality of a distribution network according to claim 1, characterized in that: The extracting corresponding features from the operation data includes: Extract the time series before and after the power supply event of the distribution network; Extracting time domain features of power supply events of the distribution network from the time series; wherein the time domain features include: duration of the event, amplitude change, and amplitude change rate; Performing Fourier transformation on the time series, and then extracting frequency domain features of the power supply event of the distribution network from the transformed time series; wherein the frequency domain features include: the main frequency, frequency band energy and spectrum entropy of the event; Performing statistical analysis on the time series to obtain statistical characteristics of power supply events in the distribution network; wherein the statistical characteristics include: mean, variance, skewness and kurtosis of the events; The time domain features, frequency domain features and statistical features are used as features corresponding to the operating data.

4. The method for analyzing power supply quality of a distribution network according to claim 3, characterized in that: The performing of empirical mode decomposition on the target feature to obtain the corresponding optimized feature component includes: When the power supply event of the distribution network is a steady-state event, obtaining a preset number of extracted optimized characteristic components; Performing empirical mode decomposition on the target feature according to the extraction quantity, to obtain optimized feature components corresponding to the extraction quantity and a residual component.

5. The method for analyzing power supply quality of a distribution network according to claim 4, characterized in that: The step of performing empirical mode decomposition on the target feature to obtain a corresponding optimized feature component further includes: When the power supply event of the distribution network is a transient event, obtaining a preset number of extracted optimized characteristic components; According to the extracted quantity, empirical mode decomposition is performed on the time series before and after the power supply event of the distribution network occurs, so as to obtain optimized characteristic components corresponding to the extracted quantity and a residual component.

6. The method for analyzing power supply quality of a distribution network according to claim 5, characterized in that: The process of the empirical mode decomposition includes: According to the target feature or the time series before and after the power supply event of the distribution network occurs, an initial residual component is obtained; Performing an optimization feature extraction operation on the initial residual component to extract the corresponding optimization feature component; The difference between the target feature or the time series before and after the power supply event of the distribution network occurs and the extracted optimized feature component is used as a new residual component, and the optimized feature extraction operation is continued to be performed on the feature component until the number of extracted feature components meets the extraction number; Wherein, the optimization feature extraction operation includes: Find all the maximum and minimum points of the initial residual component, and then fit the maximum points according to the cubic spline interpolation method to obtain the corresponding upper envelope, and fit the minimum points according to the cubic spline interpolation method to obtain the corresponding lower envelope; Calculating the average value of the upper envelope and the lower envelope; It is determined whether the difference between the initial residual component and the average value satisfies the preset condition of the optimized characteristic component. If so, the difference is used as the optimized characteristic component.

7. The method for analyzing power supply quality of a distribution network according to claim 1, characterized in that: The SVM model is specifically: Where J is the loss value of the fitting, w and b are the slope and intercept of the hyperplane respectively, κ is the penalty coefficient, ξq is the slack variable, σ is the parameter of the radial basis kernel function φ(x), T is the time or the set of time points, N is the number of iterations, st is the constraint condition, and y is the number of iterations. q is the label or category of the sample, representing different types of power quality problems.

8. A power supply quality analysis device for a distribution network, characterized in that: include: Data acquisition module, event type determination module, feature extraction module, empirical mode decomposition module and power supply quality problem identification module; The data acquisition module is used to acquire a number of power supply events of the distribution network and the operation data corresponding to each power supply event of the distribution network; wherein the operation data includes: voltage waveform data and current waveform data of the distribution network; The event type determination module is used to calculate the change amount of the corresponding operating data within a preset time period for each power supply event of the distribution network, and then determine the type of the power supply event of the distribution network according to the change amount and a preset threshold; wherein the types of power supply events of the distribution network include: transient events and steady-state events; The feature extraction module is used to extract the corresponding features in the operation data, and select a number of target features for power supply quality analysis of the distribution network from the extracted features according to the type of power supply events of the distribution network; The empirical mode decomposition module is used to perform empirical mode decomposition on the target feature to obtain corresponding optimized feature components; The power supply quality problem identification module is used to input the optimized characteristic components into a preset distribution network power supply quality analysis model, so that the distribution network power supply quality analysis model analyzes the power supply quality problems of the distribution network power supply events according to the optimized characteristic components, and outputs the power supply quality problems corresponding to the distribution network power supply events; wherein the distribution network power supply quality analysis model is obtained by training a preset SVM model with the optimized characteristic components corresponding to the operating data of the historical power supply events of the distribution network as input and the corresponding power supply quality problems as output.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for analyzing power supply quality of a distribution network as claimed in any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute the distribution network power supply quality analysis method according to any one of claims 1 to 7.

Images