A harmonic detection method and system for distribution network

Through the combination of adaptive filter and fine filter, the problem of inability to detect harmonics of the power system in the prior art is solved, and high-precision harmonic detection is realized, which improves detection accuracy and robustness, ensuring the system safe operation.

CN118503666BActive Publication Date: 2024-11-05STATE GRID GANSU ELECTRIC POWER CO JIUQUAN POWER SUPPLY CO +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410951367.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-05
Estimated Expiration
2044-07-16

AI Technical Summary

Technical Problem

The prior art cannot finely detect harmonics in power systems, resulting in increased equipment loss, heating, shortened life, electromagnetic interference and malfunction of protection devices.

Method used

Using the combination method of adaptive filter and fine filter, the state vector and measurement vector are defined by collecting voltage signals, removing DC components and high-frequency noise, initializing the state estimation and covariance matrix, predicting and updating the state and covariance, and calculating System gain, the main harmonic components are gradually extracted and finely processed.

Benefits of technology

It realizes high-precision harmonic detection, improves the accuracy and robustness of the detection, can respond in real time and effectively filter out noise, ensuring safe operation of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118503666B_ABST
    Figure CN118503666B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of harmonic detection, and provides a harmonic detection method and system for a distribution network, comprising: collecting voltage signals of a power system to obtain an original signal, removing DC components and high-frequency noise from the original signal to obtain a pre-processed signal; performing adaptive filter processing on the pre-processed signal to extract the main harmonic components; performing fine filter processing on the main harmonic components to extract the harmonic components. The present invention can detect harmonics accurately and timely.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of harmonic detection, and in particular relates to a harmonic detection method and system for a distribution network. Background Art

[0002] Harmonics refer to the components in the current or voltage waveform in the power system whose frequencies are integer multiples of the fundamental frequency. The sources of harmonics mainly include: nonlinear loads: such as power electronic equipment (inverters, rectifiers), computers, fluorescent lamps, etc.; inside the power system: such as transformer magnetic saturation, resonance, etc.

[0003] Harmonics will increase equipment losses, cause heat and shorten life, affect the quality of voltage waveforms, generate electromagnetic interference, affect other electronic equipment, and further cause protective devices to malfunction, affecting the safe operation of the system.

[0004] In the prior art, Fourier Transform (FT), Fast Fourier Transform (FFT), Wavelet Transform (WT) and the like are often used to detect harmonics in power systems. However, the above methods all have the problem of being unable to detect fine harmonics. Summary of the invention

[0005] In order to solve the problems in the prior art, the present invention provides a harmonic detection method for a distribution network, comprising the following steps:

[0006] Collecting a voltage signal of the power system to obtain an original signal, and preprocessing the original signal to obtain a preprocessed signal, wherein the preprocessing includes removing a DC component and high-frequency noise;

[0007] Performing adaptive filter processing on the preprocessed signal to extract the main harmonic components;

[0008] Performing fine filter processing on the main harmonic components to extract the harmonic components;

[0009] Performing fine filter processing on the main harmonic component to extract the harmonic component includes:

[0010] S1 defines the state vector;

[0011] S2 measurement vector;

[0012] S3 maps the state vector to the measurement space;

[0013] S4 initializes state estimation;

[0014] S5 initializes the covariance matrix;

[0015] S6 then predicts the state;

[0016] S7 prediction covariance;

[0017] S8 calculates the system gain;

[0018] S9 updates the state estimate;

[0019] S10 updates the covariance matrix.

[0020] Further, the performing of adaptive filter processing on the preprocessed signal to extract the main harmonic components comprises:

[0021] The filter weights are initialized to a zero vector:

[0022] ,

[0023] in is the filter order, is the initial weight vector of the filter, is the zero vector, is the initial weight value of the 0th order filter, is the initial weight value of the first-order filter, For the The initial weight value of the filter of order;

[0024] At every point in time , construct the input signal vector ;

[0025] ,

[0026] Represents the input signal vector at the current time point k, including the current and previous The signal of the sampling point, represents the signal at the current time point k, Indicates k before the current time point Signals at time points;

[0027] Filter output :

[0028]

[0029] is the output signal of the filter, which is the inner product of the current weight vector and the input signal vector. Indicates the main harmonic components initially extracted;

[0030] represents the transpose of the weight vector at the current time point k; represents the i-th weight at the current time point k;

[0031] The calculation error is :

[0032] ,

[0033] in, It is a signal of expectation;

[0034] Update filter weights :

[0035] ,

[0036] is the step size factor

[0037] Represents the weight vector at the next time point k+1, updated by the current weight, error and step factor.

[0038] Furthermore, the expected signal is obtained by the following method:

[0039] Assuming that there is a 50 Hz fundamental wave and a 150 Hz third harmonic in the signal, construct the expected signal:

[0040]

[0041] in Hz, Hz, is the sampling period, and are the amplitudes of the fundamental and harmonics respectively.

[0042] Further, the main harmonic component is subjected to fine filter processing, and the harmonic component extraction includes:

[0043] The state vector is defined as :

[0044]

[0045] represents the state vector at time k; represents the i-th state variable at time k;

[0046] The measurement vector is , corresponding to the output of the adaptive filter ;

[0047]

[0048] in, represents the measurement matrix, which maps the state vector to the measurement space, represents the measurement noise at time k;

[0049] The state transfer equation is:

[0050]

[0051] in, represents the state transition matrix, which describes how the system state is transferred from time k to time k+1. represents the process noise at time k;

[0052] The initialization state is estimated to be ;

[0053] The initialization covariance matrix is ;

[0054] The forecast states:

[0055]

[0056] in, Indicates that at time k based on Predicted state estimation based on time information

[0057] express The state of the moment;

[0058] The prediction covariance is:

[0059]

[0060] in, Indicates that at time k based on The prediction covariance matrix of the moment information, express The covariance matrix of the time instant; represents the process noise covariance matrix, describing the statistical characteristics of process noise;

[0061] The calculated system gain is:

[0062]

[0063] in, represents the Kalman gain at time k, weighing the weight between prediction and measurement, represents the measurement noise covariance matrix, describing the statistical characteristics of the measurement noise;

[0064] The update status is estimated to be:

[0065]

[0066] in, represents the updated state estimate at time k, which is a more accurate harmonic component extracted after further fine processing by the fine filter;

[0067] The updated covariance matrix is:

[0068]

[0069] in, Represents the updated covariance matrix at time k, reflecting the change in uncertainty of state estimation after the update.

[0070] Furthermore, the state vector is obtained in the following manner:

[0071] For the fundamental and the first two harmonics, the state vector is expressed as:

[0072]

[0073] in, is the amplitude of the ith harmonic, is the phase angle of the ith harmonic, represents the i-th state variable at time k.

[0074] Furthermore, if there is no prior knowledge of the system state, the initial state estimate is set to the zero vector.

[0075] Furthermore, if there is prior knowledge of the system state, the prior estimate is used as the initial state.

[0076] Furthermore, if there is no specific information about the uncertainty of the initial state, the covariance matrix is ​​initialized using the identity matrix multiplied by a constant.

[0077] Furthermore, if there is prior knowledge, the prior covariance matrix is ​​used.

[0078] Another aspect of the present invention further provides a harmonic detection system for a power distribution network, comprising the following modules:

[0079] The acquisition module is used to acquire the voltage signal of the power system to obtain the original signal, remove the DC component and high-frequency noise from the original signal, and obtain the pre-processed signal;

[0080] The initial processing module is used to perform adaptive filter processing on the pre-processed signal to extract the main harmonic components;

[0081] A fine processing module for performing fine filter processing on the main harmonic components to extract the harmonic components;

[0082] Performing fine filter processing on the main harmonic component to extract the harmonic component includes:

[0083] S1 defines the state vector;

[0084] S2 measurement vector;

[0085] S3 maps the state vector to the measurement space;

[0086] S4 initializes state estimation;

[0087] S5 initializes the covariance matrix;

[0088] S6 then predicts the state;

[0089] S7 prediction covariance;

[0090] S8 calculates the system gain;

[0091] S9 updates the state estimate;

[0092] S10 updates the covariance matrix.

[0093] The present invention can produce the following beneficial effects through the above technical solution:

[0094] The harmonic detection method combining adaptive filter and fine filter, the adaptive filter is used to initially extract the main harmonic components, and then the fine filter is used to perform fine processing to obtain a more accurate state estimation, which can achieve high-precision harmonic detection in the distribution network. The adaptive filter has the characteristics of real-time and fast response, while the fine filter provides accurate state estimation and noise filtering capabilities. The combination of the two can effectively improve the accuracy and robustness of harmonic detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0096] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0097] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0098] This embodiment solves the above problem through the following steps:

[0099] In one embodiment, reference Figure 1The present invention provides a harmonic detection method for a distribution network, comprising the following steps:

[0100] 1. Data Collection and Preprocessing

[0101] Collect the voltage signal of the power system and obtain the original signal . The DC component refers to the average value or DC offset of the signal. In the power system, the DC component in the signal will not provide useful information for the analysis of AC power, but will interfere with the analysis of harmonics and other AC components. High-frequency noise refers to the high-frequency components in the signal that exceed the frequency range of interest. These high-frequency components are usually random noise caused by electromagnetic interference, switching actions, etc., which will interfere with the analysis of the actual signal. Therefore, it is necessary to preprocess the signal, including removing the DC component and high-frequency noise, to obtain the preprocessed signal .

[0102] 2. Adaptive filter processing to extract the main harmonic components.

[0103] The filter weights are initialized to zero vectors

[0104]

[0105] in is the filter order, is the initial weight vector of the filter, a zero vector, and its dimension is , is the initial weight value of the 0th order filter, is the initial weight value of the first-order filter, For the The initial weight value of the filter of order;

[0106] At each time point k, construct the input signal vector

[0107]

[0108] Represents the input signal vector at the current time point k, including the current and previous The signal of the sampling point, represents the signal at the current time point k, Indicates k before the current time point Signals at time points;

[0109] Filter output

[0110]

[0111] The output signal of the filter is the inner product of the current weight vector and the input signal vector. Output signal Represents the initially extracted harmonic components.

[0112] represents the transpose of the weight vector at the current time point k; represents the i-th weight at the current time point k;

[0113] Calculation error

[0114]

[0115] in, is the desired signal. For example, assuming that there is a 50 Hz fundamental wave and a 150 Hz third harmonic in the signal, a desired signal can be constructed:

[0116]

[0117] in Hz, Hz, is the sampling period, and They are the amplitudes of the fundamental wave and the harmonic wave respectively. Similarly, other desired signals can be constructed to collect harmonic signals of different multiple frequencies.

[0118] Update filter weights

[0119]

[0120] Represents the weight vector at the next time point k+1, updated by the current weight, error and step factor.

[0121] Through the above steps, the output of the adaptive filter The main harmonic components of the signal are extracted.

[0122] In the above steps, the adaptive filter dynamically adjusts the weights so that the output signal It can accurately track specific frequency components (i.e. harmonic components) in the input signal. In the process of continuously updating the weights, the filter gradually adapts to the harmonic frequencies in the input signal, so that the output signal mainly contains these harmonic components.

[0123] The goal of an adaptive filter is to minimize the error between the output signal and the desired signal. When the input signal contains significant harmonic components, the filter adjusts the weights so that the output signal mainly contains these harmonic components to minimize the error.

[0124] Since the adaptive filter can identify and track specific frequency components (such as harmonics) in the input signal, the output signal will mainly reflect these frequency components. The process of adjusting the filter weights is actually optimizing the filter's response to the harmonic components in the input signal.

[0125] 3. Fine filtering of main harmonic components

[0126] Define the state vector

[0127]

[0128] Represents the state vector at time k, which contains all the state variables of the system. The state variables are used to describe the harmonic characteristics and dynamic behavior of the power signal in the system. The specific choice of state variables depends on the complexity of the model and the required harmonic components.

[0129] Exemplarily, for the fundamental wave and the first two harmonics (fundamental wave, third harmonic and fifth harmonic), the state vector can be expressed as:

[0130]

[0131] in, is the amplitude of the ith harmonic, is the phase angle of the ith harmonic. represents the i-th state variable at time k.

[0132] Measuring vector , which is the output of the adaptive filter ;

[0133]

[0134] Represents the measurement matrix, which maps the state vector to the measurement space.

[0135] represents the measurement noise at time k, which is set to Gaussian white noise in this embodiment.

[0136] State transfer equation:

[0137]

[0138] represents the state transition matrix, which describes how the system state is transferred from time k to time k+1;

[0139] represents the process noise at time k, which is set to Gaussian white noise in this embodiment.

[0140] Initialize state estimate If there is no prior knowledge about the system state, the initial state estimate can be set to the zero vector. If there is some prior knowledge about the system state, the prior estimate can be used as the initial state, and historical data can also be used for estimation to obtain the initial state.

[0141] Initialize the covariance matrix If you do not have specific information about the uncertainty of the initial state, you can use the identity matrix multiplied by a large constant to initialize. If you have prior knowledge, you can use the prior covariance matrix. You can also use the variance estimate of historical data to initialize.

[0142] Prediction status:

[0143]

[0144] Indicates that at time k based on Predicted state estimation of moment information;

[0145] represents the state at time k-1, It is the previous time slot, which has already occurred, so this is the real state. In order to facilitate the consistency of program implementation and mathematical expression, it is written as .

[0146] Prediction covariance:

[0147]

[0148] Indicates that at time k based on The prediction covariance matrix of the moment information;

[0149] express The covariance matrix at the moment is similar to the state. This expression is for the convenience of program implementation and consistency of mathematical expression.

[0150] represents the process noise covariance matrix, describing the statistical characteristics of process noise;

[0151] Calculate the system gain:

[0152]

[0153] represents the Kalman gain at time k, weighing the weight between prediction and measurement;

[0154] represents the measurement noise covariance matrix, which describes the statistical characteristics of the measurement noise.

[0155] Update state estimate

[0156]

[0157] represents the updated state estimate at time k, which is corrected by combining the prediction and measurement information. It is a more accurate harmonic component extracted after further fine processing by the fine filter.

[0158] Update the covariance matrix:

[0159]

[0160] Represents the updated covariance matrix at time k, reflecting the change in uncertainty of state estimation after the update.

[0161] The harmonic detection method combining adaptive filter and fine filter, the adaptive filter is used to initially extract the main harmonic components, and then the fine filter is used to perform fine processing to obtain a more accurate state estimation, which can achieve high-precision harmonic detection in the distribution network. The adaptive filter has the characteristics of real-time and fast response, while the fine filter provides accurate state estimation and noise filtering capabilities. The combination of the two can effectively improve the accuracy and robustness of harmonic detection.

[0162] On the other hand, the present invention also provides a harmonic detection system for a power distribution network, comprising:

[0163] The acquisition module is used to acquire the voltage signal of the power system to obtain the original signal, remove the DC component and high-frequency noise from the original signal, and obtain the pre-processed signal;

[0164] The initial processing module is used to perform adaptive filter processing on the pre-processed signal to extract the main harmonic components;

[0165] A fine processing module for performing fine filter processing on the main harmonic components to extract the harmonic components;

[0166] Performing fine filter processing on the main harmonic component to extract the harmonic component includes:

[0167] S1 defines the state vector;

[0168] S2 measurement vector;

[0169] S3 maps the state vector to the measurement space;

[0170] S4 initializes state estimation;

[0171] S5 initializes the covariance matrix;

[0172] S6 then predicts the state;

[0173] S7 prediction covariance;

[0174] S8 calculates the system gain;

[0175] S9 updates the state estimate;

[0176] S10 updates the covariance matrix.

[0177] Furthermore, the specific implementation methods of the harmonic detection system for a distribution network described above are the same as a harmonic detection method for a distribution network, and all further technical solutions in a harmonic detection method for a distribution network are completely introduced into a harmonic detection system for a distribution network.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

[0179] The present invention does not particularly specify the structure of some modules, which shall be subject to the contents recorded in the prior art. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be used as part of the present invention to understand the meaning of some technical features or parameters. The scope of protection of the present invention shall be subject to the contents actually recorded in the claims.

Claims

1. A harmonic detection method for a distribution network, It is characterized in that The method comprises the following steps: Collecting a voltage signal of the power system to obtain an original signal, and preprocessing the original signal to obtain a preprocessed signal, wherein the preprocessing includes removing a DC component and high-frequency noise; Performing adaptive filter processing on the preprocessed signal to extract main harmonic components; Performing fine filter processing on the main harmonic components to extract the harmonic components; Performing fine filter processing on the main harmonic component to extract the harmonic component includes: S1 defines the state vector; S2 measurement vector; S3 maps the state vector to the measurement space; S4 initializes state estimation; S5 initializes the covariance matrix; S6 then predicts the state; S7 prediction covariance; S8 calculates the system gain; S9 updates the state estimate; S10 updates the covariance matrix; The performing of adaptive filter processing on the preprocessed signal to extract the main harmonic components comprises: The filter weights are initialized to a zero vector: , in is the filter order, is the initial weight vector of the filter, is the zero vector, is the initial weight value of the 0th order filter, is the initial weight value of the first-order filter, is the initial weight value of the N-1th order filter; At each time point k, construct the input signal vector : , Represents the input signal vector at the current time point k, including the current and previous The signal of the sampling point, represents the signal at the current time point k, Indicates k before the current time point Signals at time points; Filter output : , is the output signal of the filter, that is, the inner product of the current weight vector and the input signal vector. represents the main harmonic component extracted initially, represents the transpose of the weight vector at the current time point k; represents the i-th weight at the current time point k; Calculation error : , in It is a signal of expectation; Update filter weights : , is the step size factor; Represents the weight vector at the next time point k+1, which is updated by the current weight, error and step factor.

2. A method for detecting harmonics in a distribution network according to claim 1, It is characterized in that The expected signal is obtained by the following method: Assuming that there is a 50 Hz fundamental wave and a 150 Hz third harmonic in the signal, construct the expected signal: , in , , is the sampling period, and are the amplitudes of the fundamental and harmonics respectively.

3. A method for detecting harmonics in a distribution network according to claim 1, It is characterized in that Performing fine filter processing on the main harmonic component to extract the harmonic component includes: The state vector is defined as : , represents the state vector at time k; represents the i-th state variable at time k; The measurement vector is y(k), which is the output y(k) of the adaptive filter: , in, represents the measurement matrix, which maps the state vector to the measurement space, represents the measurement noise at time k; The state transfer equation is: , in, represents the state transition matrix, which describes how the system state is transferred from time k to time k+1. represents the process noise at time k; The initialization state is estimated to be ; The initialization covariance matrix is ; The forecast states: , in, Indicates that at time k based on The predicted state estimation of the time information, express The state of the moment; The prediction covariance is: , in, Indicates that at time k based on The prediction covariance matrix of the moment information, express The covariance matrix of the moment; represents the process noise covariance matrix, describing the statistical characteristics of process noise; The calculated system gain is: , in, represents the Kalman gain at time k, weighing the weight between prediction and measurement, represents the measurement noise covariance matrix, describing the statistical characteristics of the measurement noise; The update status is estimated to be: , in, represents the updated state estimate at time k, which is a more accurate harmonic component extracted after further fine processing by the fine filter; The updated covariance matrix is: , in, Represents the updated covariance matrix at time k, reflecting the change in the uncertainty of the state estimate after the update.

4. A method for detecting harmonics in a distribution network according to claim 3, It is characterized in that The state vector is obtained in the following way: For the fundamental and the first two harmonics, the state vector is expressed as: , in, is the amplitude of the ith harmonic, is the phase angle of the ith harmonic, represents the state variable at time k.

5. A method for detecting harmonics in a distribution network according to claim 3, Features: If there is no a priori knowledge of the system state, set the initial state estimate to the zero vector.

6. A method for detecting harmonics in a distribution network according to claim 3, Features: If the system state has prior knowledge, use the prior estimate as the initial state.

7. A method for detecting harmonics in a distribution network according to claim 3, Features: If there is no specific information about the uncertainty of the initial state, the covariance matrix is ​​initialized using the identity matrix multiplied by a constant.

8. A method for detecting harmonics in a distribution network according to claim 3, Features: If prior knowledge is available, the prior covariance matrix is ​​used.

9. A harmonic detection system for a distribution network, Features The system includes the following modules: The acquisition module is used to acquire the voltage signal of the power system to obtain the original signal, remove the DC component and high-frequency noise from the original signal, and obtain the pre-processed signal; The initial processing module is used to perform adaptive filter processing on the pre-processed signal to extract the main harmonic components; A fine processing module for performing fine filter processing on the main harmonic components to extract the harmonic components; Performing fine filter processing on the main harmonic component to extract the harmonic component includes: S1 defines the state vector; S2 measurement vector; S3 maps the state vector to the measurement space; S4 initializes state estimation; S5 initializes the covariance matrix; S6 then predicts the state; S7 prediction covariance; S8 calculates the system gain; S9 updates the state estimate; S10 updates the covariance matrix; The performing of adaptive filter processing on the preprocessed signal to extract the main harmonic components comprises: The filter weights are initialized to a zero vector: The filter weights are initialized to a zero vector: , in is the filter order, is the initial weight vector of the filter, is the zero vector, is the initial weight value of the 0th order filter, is the initial weight value of the first-order filter, is the initial weight value of the N-1th order filter; At each time point k, construct the input signal vector : , Represents the input signal vector at the current time point k, including the current and previous The signal of the sampling point, represents the signal at the current time point k, Indicates k before the current time point Signals at time points; Filter output : , is the output signal of the filter, that is, the inner product of the current weight vector and the input signal vector. represents the main harmonic component extracted initially, represents the transpose of the weight vector at the current time point k; represents the i-th weight at the current time point k; Calculation error : , in It is a signal of expectation; Update filter weights : , is the step size factor; Represents the weight vector at the next time point k+1, which is updated by the current weight, error and step factor.