Power quality monitoring system and method based on transformer area intelligent fusion terminal

Abstract

The invention provides an electric energy quality monitoring system and method based on a transformer area intelligent fusion terminal, and belongs to the technical field of electric energy quality monitoring. The system comprises an acquisition unit and a monitoring unit which are arranged in the intelligent fusion terminal. The acquisition module is used for acquiring voltage data of a transformer area, sequentially extracting voltage discrete data with a fixed length from the voltage data, and inputting the voltage discrete data extracted each time into the monitoring unit for electric energy quality judgment; the monitoring unit comprises a conversion module, a first processing module, a second processing module and a third processing module; the first processing module is used for carrying out feature extraction on the input voltage discrete data, the conversion module is used for converting the input voltage discrete data into a two-dimensional image, and the second processing module is used for carrying out feature extraction on the input two-dimensional image; and the third processing module is used for merging and classifying the feature data extracted by the first processing module and the second processing module to obtain an electric energy quality judgment result.

Description

Technical Field

[0001] The present invention belongs to the technical field of power quality monitoring, and in particular relates to a power quality monitoring system and method based on an intelligent fusion terminal in an area. Background Art

[0002] With the continuous development of smart grids, the integration of clean energy and nonlinear load electricity consumption may cause voltage sags, swells, flickers or frequency changes while generating a large number of harmonics, which will have a significant impact on the power quality of the power grid, directly affecting the safe operation of the power grid and the normal production of enterprises, causing a large amount of economic losses. Therefore, it is necessary to monitor and analyze the power quality.

[0003] The intelligent fusion terminal is an edge computing device of the "cloud-pipe-edge-end" architecture of the basic intelligent IoT system for building the low-voltage distribution IoT. It has functions such as edge IoT, equipment status information perception, power consumption data collection, distribution transformer monitoring, communication networking, and localized analysis and decision-making mechanism. It is necessary to use the edge computing capabilities of the intelligent fusion terminal to achieve accurate monitoring of power quality. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a power quality monitoring system and method based on an intelligent fusion terminal in an area in view of the deficiencies of the prior art.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A power quality monitoring system based on an intelligent fusion terminal in a substation area includes a collection unit and a monitoring unit arranged in the intelligent fusion terminal;

[0007] The acquisition module is used to acquire voltage data of the substation, to sequentially extract voltage discrete data of fixed length from the voltage data, and to input the extracted voltage discrete data into the monitoring unit for power quality determination.

[0008] The monitoring unit includes a conversion module, a first processing module, a second processing module and a third processing module; the first processing module is used to extract features from the input discrete voltage data, the conversion module is used to convert the input discrete voltage data into a two-dimensional image, the second processing module is used to extract features from the input two-dimensional image, and the third processing module is used to merge and classify the feature data extracted by the first processing module and the second processing module to obtain a power quality judgment result.

[0009] Furthermore, the conversion module is used to obtain a distance matrix of the voltage discrete data, and to generate a two-dimensional image of the voltage discrete data according to the distance matrix.

[0010] Furthermore, the size of the distance matrix is ​​N×N, where N is the length of the voltage discrete data, and the element d in the i-th row and j-th column of the distance matrix is i,j =‖x i ―x j ‖ 2 , ‖*‖ is the binary norm.

[0011] Furthermore, the size of the two-dimensional image is N×N, N is the length of the voltage discrete data, and the R, G, and B color values ​​of the pixel at the coordinate (m, n) are D represents the distance matrix, d i,j =‖x i ―x j ‖ 2 , ‖*‖ is the binary norm.

[0012] Furthermore, a plurality of processing structures are provided in the first processing module; each processing structure of the first processing module is composed of a one-dimensional convolution layer, a pooling layer and a batch normalization layer.

[0013] Furthermore, multiple processing structures are arranged in the second processing module; the last processing structure of the second processing module is composed of a two-dimensional convolution layer and a batch normalization layer, and the other processing structures are composed of a two-dimensional convolution layer, a pooling layer and a batch normalization layer.

[0014] Furthermore, the third processing module consists of multiple fully connected layers and a Softmax layer.

[0015] Furthermore, the monitoring unit completes training using a disturbance data set; the disturbance data set is constructed using a mathematical model of power quality disturbance.

[0016] Furthermore, the power quality disturbance includes single disturbance and compound disturbance; the single disturbance includes voltage swell, voltage sag, voltage interruption, voltage flicker, transient pulse, transient oscillation, and harmonics; the compound disturbance includes voltage interruption + harmonics, voltage sag + harmonics, voltage swell + harmonics, voltage swell + voltage flicker, voltage sag + voltage flicker, harmonic + voltage flicker.

[0017] A power quality monitoring method based on a substation intelligent fusion terminal comprises: collecting voltage data of the substation, sequentially extracting voltage discrete data of fixed length from the voltage data, and performing power quality judgment on the voltage discrete data after each extraction;

[0018] The method for distinguishing power quality of discrete voltage data includes: converting the input discrete voltage data into a two-dimensional image; performing feature extraction on the input discrete voltage data to obtain a first feature; performing feature extraction on the two-dimensional image to obtain a second feature; merging and classifying the first feature and the second feature to obtain a recognition result of power quality;

[0019] The method for converting discrete voltage data into a two-dimensional image includes: acquiring a distance matrix of the discrete voltage data, and generating a two-dimensional image of the discrete voltage data according to the distance matrix.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] The present invention extracts the characteristics of the voltage signal to obtain the signal characteristics, extracts the characteristics of the two-dimensional image corresponding to the voltage signal to obtain the image characteristics, and finally analyzes the fused signal characteristics and image characteristics to obtain the judgment result of the power quality. Since the two characteristic data are fused, the judgment of the power quality will be more accurate.

[0022] The present invention completes power quality monitoring in a local intelligent fusion terminal, and power quality monitoring does not need to be performed in the cloud. The present invention helps to timely discover and classify power quality disturbance events, and provides a reference for subsequent power quality management. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0024] Figure 1 : Schematic diagram of embodiment 1 of the present invention. DETAILED DESCRIPTION

[0025] In order to better understand the present invention, the content of the present invention is further clearly described below in conjunction with the embodiments and the accompanying drawings, but the protection content of the present invention is not limited to the following embodiments. In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details.

[0026] Example 1: See Figure 1 The purpose of this embodiment is to provide a power quality monitoring system based on an intelligent fusion terminal in an area, including a collection unit and a monitoring unit arranged in the intelligent fusion terminal.

[0027] The acquisition module is used to acquire voltage data of the substation, to sequentially extract voltage discrete data of fixed length from the voltage data, and to input the extracted voltage discrete data into the monitoring unit for power quality determination.

[0028] The monitoring unit includes a conversion module, a first processing module, a second processing module and a third processing module. The input data of the conversion module and the first processing module are both the voltage discrete data, the first processing module is used to extract features from the input voltage discrete data, the conversion module is used to convert the input voltage discrete data into a two-dimensional image, the second processing module is used to extract features from the input two-dimensional image, and the third processing module is used to merge and classify the feature data extracted by the first processing module and the second processing module to obtain a power quality discrimination result.

[0029] The conversion module is used to obtain the distance matrix of the voltage discrete data and to generate a two-dimensional image of the voltage discrete data according to the distance matrix. The size of the distance matrix is ​​N×N, where N is the length of the voltage discrete data. The element d in the i-th row and j-th column of the distance matrix i,j =‖x i ―x j ‖ 2 , represents the sampling point x i With x j The size of the two-dimensional image is N×N, and the RGB color value of the pixel at the coordinate (i, j) is (R i,j ,G i,j ,B i,j ),but D represents the distance matrix.

[0030] The structures of the first processing module and the second processing module refer to Figure 1 , Figure 1 In the above, 1D Conv represents a one-dimensional convolution layer, 2D Conv represents a two-dimensional convolution layer, Pool represents a pooling layer, BN represents a batch normalization layer, and FC represents a fully connected layer. The first processing module is provided with multiple processing structures, each of which is composed of a one-dimensional convolution layer, a pooling layer, and a batch normalization layer; the second processing module is also provided with multiple processing structures, the last processing structure is composed of a two-dimensional convolution layer and a batch normalization layer, and the other processing structures are composed of a two-dimensional convolution layer, a pooling layer, and a batch normalization layer; the third processing module is composed of multiple fully connected layers and a Softmax layer. The first processing module directly extracts features from the original voltage discrete data, and the second processing module extracts spatial features from the converted two-dimensional image. The features extracted by the first processing module and the second processing module are input into the third processing module, and the third processing module merges and classifies the input features, and finally outputs the recognition result of the power quality.

[0031] The monitoring unit completes training using a disturbance data set. The disturbance data set is constructed according to a mathematical model of power quality disturbances. Power quality disturbances include single disturbances and compound disturbances.

[0032] First, determine the mathematical model of the standard sinusoidal signal, the mathematical model of single disturbances such as voltage swell, voltage sag, voltage interruption, voltage flicker, transient pulse, transient oscillation, and harmonics, as well as the mathematical model of composite disturbances such as voltage interruption + harmonics, voltage sag + harmonics, voltage swell + harmonics, voltage swell + voltage flicker, voltage sag + voltage flicker, and harmonics + voltage flicker.

[0033] For example, the mathematical model of a single disturbance is: The mathematical model of a standard sinusoidal signal is Where ω=2πf, is the initial phase, f is the frequency;

[0034] The mathematical model of voltage swell is: Where α is the temporary rise coefficient, 0.1≤α≤0.8, 0.5T≤t 2 -t 1 ≤30T, u() is a step function, T represents the signal period, t 1 and t 2 Respectively represent the start and end time, t 2 -t 1 represents the duration of disturbance;

[0035] The mathematical model of voltage sag is: Where α is the temporary sag coefficient, 0.1≤α≤0.9, 0.5T≤t 2 -t 1 ≤30T;

[0036] The mathematical model of voltage interruption is: Where α is the interruption coefficient, 0.9≤α≤1, 0.5T≤t 2 -t 1 ≤30T;

[0037] The mathematical model of voltage flicker is: Where α is the flicker coefficient, 0.05≤α≤0.25; β is the fluctuation frequency, 5Hz≤β≤20Hz.

[0038] The mathematical model of transient pulse is V(t)={sin(ωt)+α[u(t―t 1 )―u(t―t 2 )]}, where α is the pulse coefficient, 0≤α≤2, 1ms≤t 2 -t 1 ≤3ms;

[0039] The mathematical model of transient oscillation is: α is the oscillation coefficient, 0.1≤α≤0.8; 0.5T≤t 2 -t 1 ≤30T,w n =2πf n ; 300Hz≤f n ≤900Hz, 8ms≤τ≤40ms;

[0040] The mathematical model of harmonics (taking the 3rd, 5th, and 7th harmonics as an example) is V(t)=sin(wt)+α 3 sin(3wt)+α 5 sin(5wt)+α 7 sin(7wt), where α h is the amplitude of the hth harmonic, 0.05≤α h ≤0.25, h=3,5,7.

[0041] The mathematical model of a composite disturbance is the sum of the mathematical models corresponding to single disturbances.

[0042] Then, according to the mathematical model of power quality disturbance, a disturbance data set containing various power quality disturbances was constructed using MATLAB as the modeling software. The data type in the disturbance data set is voltage discrete data. Finally, the disturbance data set is used to complete the training of the monitoring unit.

[0043] Embodiment 2: The purpose of this embodiment is to provide a method for monitoring power quality based on a substation intelligent fusion terminal, which is implemented using the system described in Embodiment 1. The method includes: collecting voltage data of the substation, sequentially extracting voltage discrete data of fixed length from the voltage data, and performing power quality judgment on the voltage discrete data after each extraction.

[0044] The method for distinguishing power quality of voltage discrete data includes: converting the input voltage discrete data into a two-dimensional image; performing feature extraction on the input voltage discrete data to obtain a first feature; performing feature extraction on the two-dimensional image to obtain a second feature; merging and classifying the first feature and the second feature to obtain a power quality recognition result.

[0045] The method for converting discrete voltage data into a two-dimensional image includes: acquiring a distance matrix of the discrete voltage data, and generating a two-dimensional image of the discrete voltage data according to the distance matrix.

[0046] The monitoring unit is used to distinguish the power quality of discrete voltage data. The training method of the monitoring unit includes: generating a disturbance data set according to a mathematical model of power quality disturbance, and completing the training of the monitoring unit using the disturbance data set.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A power quality monitoring system based on an intelligent fusion terminal in an area, characterized in that: It includes a collection unit and a monitoring unit arranged in an intelligent fusion terminal; The acquisition module is used to acquire voltage data of the substation, to sequentially extract voltage discrete data of fixed length from the voltage data, and to input the extracted voltage discrete data into the monitoring unit for power quality determination. The monitoring unit includes a conversion module, a first processing module, a second processing module and a third processing module; the first processing module is used to extract features from the input discrete voltage data, the conversion module is used to convert the input discrete voltage data into a two-dimensional image, the second processing module is used to extract features from the input two-dimensional image, and the third processing module is used to merge and classify the feature data extracted by the first processing module and the second processing module to obtain a power quality judgment result.

2. The power quality monitoring system based on the intelligent fusion terminal of the substation according to claim 1 is characterized in that: The conversion module is used to obtain a distance matrix of the voltage discrete data, and to generate a two-dimensional image of the voltage discrete data according to the distance matrix.

3. The power quality monitoring system based on the intelligent fusion terminal of the substation according to claim 2 is characterized in that: The size of the distance matrix is ​​N×N, where N is the length of the voltage discrete data, and the element d in the i-th row and j-th column of the distance matrix i,j =‖x i ―x j ‖ 2 , ‖*‖ is the binary norm.

4. The power quality monitoring system based on the intelligent fusion terminal of the substation according to claim 2 is characterized in that: The size of the two-dimensional image is N×N, where N is the length of the voltage discrete data, and the R, G, and B color values ​​of the pixel at the coordinate (m, n) are D represents the distance matrix, d i,j =‖x i ―x j ‖ 2 , ‖*‖ is the binary norm.

5. The power quality monitoring system based on the intelligent fusion terminal of the substation according to claim 1 is characterized in that: A plurality of processing structures are arranged in the first processing module; each processing structure of the first processing module is composed of a one-dimensional convolution layer, a pooling layer and a batch normalization layer.

6. The power quality monitoring system based on the intelligent fusion terminal of the substation according to claim 1 is characterized in that: The second processing module is provided with a plurality of processing structures; the last processing structure of the second processing module is composed of a two-dimensional convolution layer and a batch normalization layer, and the other processing structures are composed of a two-dimensional convolution layer, a pooling layer and a batch normalization layer.

7. The power quality monitoring system based on the intelligent fusion terminal of the substation according to claim 1 is characterized in that: The third processing module consists of multiple fully connected layers and a Softmax layer.

8. The power quality monitoring system based on the area intelligent fusion terminal according to claim 1 is characterized in that: The monitoring unit completes training using a disturbance data set; the disturbance data set is constructed using a mathematical model of power quality disturbance.

9. The power quality monitoring system based on the intelligent fusion terminal of the substation area according to claim 8 is characterized in that: The power quality disturbance includes single disturbance and compound disturbance; the single disturbance includes voltage swell, voltage sag, voltage interruption, voltage flicker, transient pulse, transient oscillation and harmonic; the compound disturbance includes voltage interruption + harmonic, voltage sag + harmonic, voltage swell + harmonic, voltage swell + voltage flicker, voltage sag + voltage flicker and harmonic + voltage flicker.

10. A method for monitoring power quality based on an intelligent fusion terminal in an area, characterized in that: include: Collecting voltage data of the substation area, sequentially extracting voltage discrete data of fixed length from the voltage data, and performing power quality judgment on the voltage discrete data after each extraction; The method for distinguishing power quality of discrete voltage data includes: converting the input discrete voltage data into a two-dimensional image; performing feature extraction on the input discrete voltage data to obtain a first feature; performing feature extraction on the two-dimensional image to obtain a second feature; merging and classifying the first feature and the second feature to obtain a recognition result of power quality; The method for converting discrete voltage data into a two-dimensional image includes: acquiring a distance matrix of the discrete voltage data, and generating a two-dimensional image of the discrete voltage data according to the distance matrix.

Images