CN113342993A - Power failure map generation method - Google Patents

Abstract

A power failure map generation method relates to the technical field of power systems and comprises the steps of using power failure history wave recording data including phase current wave recording data and phase voltage wave recording data as samples, selecting sampling time points according to time sequences, converting relevant data into vectors, constructing a gradient image by using the vectors to serve as a power failure map, and detecting failure types by using the power failure map when power failures occur. The method provided by the invention is suitable for detecting the power failure in the power system.

Description

Power failure map generation method

Technical Field

The invention relates to the technology of a power system, in particular to the technology of a power failure map generation method.

Background

The power failure recording system is a system for recording power failure data (phase current, phase voltage and the like when a power failure occurs), and can find out the characteristics of the failure data through analyzing the power failure data, thereby judging the cause of the failure.

At present, a neural network is adopted to analyze power failure data, and the analysis method needs to extract failure features to match and classify failure types, but the existing algorithm for extracting the failure features is complex, and has the defects of low calculation speed and difficulty in detecting new failure types.

Disclosure of Invention

In view of the above-mentioned drawbacks in the prior art, the present invention provides a method for generating a power failure map, which can quickly detect the type of a failure when a power failure occurs.

In order to solve the technical problem, the method for generating the power failure map provided by the invention is characterized by comprising the following specific steps of:

1) acquiring power failure history recording data comprising phase current recording data and phase voltage recording data;

2) dividing phase current wave recording data in the power failure historical wave recording data into R phase current wave bands, wherein the duration of each phase current wave band is a cycle, and the starting point of each phase current wave band falls on the abscissa axis, wherein the cycle is a complete sinusoidal waveform of power frequency current;

3) taking L sampling points on each phase current wave band, defining a time point corresponding to each sampling point as a sampling time point, and forming each sampling time point into a matrix G with R rows and L columns;

in the matrix G, all sampling time points in the same row are sequentially arranged from left to right according to a time sequence from first to last, and the time sequence of the sampling time point in the upper row is prior to that of the sampling time point in the lower row;

each sampling time point has three values, the three values are a phase current per unit value I, a phase voltage per unit value U and a phase angle difference a of the phase current and the phase voltage at the sampling time point respectively, and the three values are obtained from phase current wave recording data and phase voltage wave recording data in power failure historical wave recording data;

4) setting a sampling point vector for each sampling time point, and calculating the vector amplitude and the vector angle of the sampling point vector of each sampling time point, wherein the calculation method comprises the following steps:

for each sampling point, constructing a triangle consisting of three sides S1, S2 and S3 for the sampling time point, taking the value of the phase current per unit value I of the sampling time point as the side length of S1, taking the value of the phase voltage per unit value U of the sampling time point as the side length of S2, and taking the phase angle difference a between the phase current and the phase voltage of the sampling time point as the included angle between S1 and S2; then calculating the length of S3 and the included angle beta between S1 and S3, setting the length of S3 as the vector amplitude of the sampling point vector of the sampling time point, and setting the included angle beta between S1 and S3 as the vector angle of the sampling point vector of the sampling time point;

5) forming a matrix M with R rows and L columns by using the sampling point vectors of all sampling time points, wherein the arrangement mode of all the sampling point vectors in the matrix M is consistent with the arrangement mode of the sampling time points to which all the sampling point vectors belong in the matrix G;

6) constructing a gradient image P consisting of R rows and L columns of pixel points, and calculating the gradient value and the gradient direction value of each pixel point in the gradient image P, wherein the calculation formula is as follows:

θ(x,y)＝int(8×Mβ(x,y)/2π)

in the formula, H (x, y) is a gradient value of a pixel point at the x-th row and the y-th column in the gradient image P, θ (x, y) is a gradient direction value of a pixel point at the x-th row and the y-th column in the gradient image P, Ml (x, y) is a vector magnitude of an element at the x-th row and the y-th column in the matrix M, M β (x, y) is a vector angle of an element at the x-th row and the y-th column in the matrix M, and int () is a rounding function;

7) and taking the gradient image P as a power failure map, and detecting the failure type by using the power failure map when the power failure occurs.

The power failure map generation method provided by the invention takes power failure history recording data as a sample, converts phase current recording data and phase voltage recording data into vectors, constructs a gradient image, takes the constructed gradient image as a power failure map, simulates data characteristic classification when a power failure occurs by using the power failure map, and can quickly detect the failure type according to the power failure map when the power failure occurs.

Detailed Description

The technical solution of the present invention is further described in detail with reference to the following specific embodiments, but the present invention is not limited thereto, and all similar structures and similar variations thereof adopting the present invention should be included in the protection scope of the present invention, wherein the pause numbers in the present invention all represent the relation of the sum, and the english letters in the present invention are distinguished by the case.

The embodiment of the invention provides a power failure map generation method which is characterized by comprising the following specific steps:

1) acquiring power failure history recording data comprising phase current recording data and phase voltage recording data;

the historical wave recording data of the power failure can be obtained by the existing power failure wave recording system, the phase current wave recording data and the phase voltage wave recording data are waveform data represented by two-dimensional rectangular coordinates, the abscissa axes of the phase current wave recording data and the phase voltage wave recording data are time axes, the ordinate axis of the phase current wave recording data is a phase current numerical axis, and the ordinate axis of the phase voltage wave recording data is a phase voltage numerical axis;

2) dividing phase current wave recording data in the power failure historical wave recording data into R phase current wave bands, wherein the duration of each phase current wave band is a cycle, and the starting point of each phase current wave band falls on the abscissa axis, wherein the cycle is a complete sinusoidal waveform of power frequency current;

3) taking L sampling points on each phase current wave band, defining a time point corresponding to each sampling point as a sampling time point, and forming each sampling time point into a matrix G with R rows and L columns;

in the matrix G, all sampling time points in the same row are sequentially arranged from left to right according to a time sequence from first to last, and the time sequence of the sampling time point in the upper row is prior to that of the sampling time point in the lower row;

each sampling time point has three values, the three values are a phase current per unit value I, a phase voltage per unit value U and a phase angle difference a of the phase current and the phase voltage at the sampling time point respectively, and the three values are obtained from phase current wave recording data and phase voltage wave recording data in power failure historical wave recording data;

4) setting a sampling point vector for each sampling time point, and calculating the vector amplitude and the vector angle of the sampling point vector of each sampling time point, wherein the calculation method comprises the following steps:

for each sampling point, constructing a triangle consisting of three sides S1, S2 and S3 for the sampling time point, taking the value of the phase current per unit value I of the sampling time point as the side length of S1, taking the value of the phase voltage per unit value U of the sampling time point as the side length of S2, and taking the phase angle difference a between the phase current and the phase voltage of the sampling time point as the included angle between S1 and S2; then calculating the length of S3 and the included angle beta between S1 and S3, setting the length of S3 as the vector amplitude of the sampling point vector of the sampling time point, and setting the included angle beta between S1 and S3 as the vector angle of the sampling point vector of the sampling time point;

5) forming a matrix M with R rows and L columns by using the sampling point vectors of all sampling time points, wherein the arrangement mode of all the sampling point vectors in the matrix M is consistent with the arrangement mode of the sampling time points to which all the sampling point vectors belong in the matrix G;

6) constructing a gradient image P consisting of R rows and L columns of pixel points, and calculating the gradient value and the gradient direction value of each pixel point in the gradient image P, wherein the calculation formula is as follows:

θ(x,y)＝int(8×Mβ(x,y)/2π)

in the formula, H (x, y) is a gradient value of a pixel point at the x-th row and the y-th column in the gradient image P, θ (x, y) is a gradient direction value of a pixel point at the x-th row and the y-th column in the gradient image P, Ml (x, y) is a vector magnitude of an element at the x-th row and the y-th column in the matrix M, M β (x, y) is a vector angle of an element at the x-th row and the y-th column in the matrix M, and int () is a rounding function;

7) and taking the gradient image P as a power failure map, and detecting the failure type by using the power failure map when the power failure occurs.

Claims (1)

1. A power failure map generation method is characterized by comprising the following specific steps:

1) acquiring power failure history recording data comprising phase current recording data and phase voltage recording data;

2) dividing phase current wave recording data in the power failure historical wave recording data into R phase current wave bands, wherein the duration of each phase current wave band is a cycle, and the starting point of each phase current wave band falls on the abscissa axis, wherein the cycle is a complete sinusoidal waveform of power frequency current;

3) taking L sampling points on each phase current wave band, defining a time point corresponding to each sampling point as a sampling time point, and forming each sampling time point into a matrix G with R rows and L columns;

in the matrix G, all sampling time points in the same row are sequentially arranged from left to right according to a time sequence from first to last, and the time sequence of the sampling time point in the upper row is prior to that of the sampling time point in the lower row;

each sampling time point has three values, the three values are a phase current per unit value I, a phase voltage per unit value U and a phase angle difference a of the phase current and the phase voltage at the sampling time point respectively, and the three values are obtained from phase current wave recording data and phase voltage wave recording data in power failure historical wave recording data;

4) setting a sampling point vector for each sampling time point, and calculating the vector amplitude and the vector angle of the sampling point vector of each sampling time point, wherein the calculation method comprises the following steps:

for each sampling point, constructing a triangle consisting of three sides S1, S2 and S3 for the sampling time point, taking the value of the phase current per unit value I of the sampling time point as the side length of S1, taking the value of the phase voltage per unit value U of the sampling time point as the side length of S2, and taking the phase angle difference a between the phase current and the phase voltage of the sampling time point as the included angle between S1 and S2; then calculating the length of S3 and the included angle beta between S1 and S3, setting the length of S3 as the vector amplitude of the sampling point vector of the sampling time point, and setting the included angle beta between S1 and S3 as the vector angle of the sampling point vector of the sampling time point;

5) forming a matrix M with R rows and L columns by using the sampling point vectors of all sampling time points, wherein the arrangement mode of all the sampling point vectors in the matrix M is consistent with the arrangement mode of the sampling time points to which all the sampling point vectors belong in the matrix G;

6) constructing a gradient image P consisting of R rows and L columns of pixel points, and calculating the gradient value and the gradient direction value of each pixel point in the gradient image P, wherein the calculation formula is as follows:

θ(x,y)＝int(8×Mβ(x,y)/2π)

in the formula, H (x, y) is a gradient value of a pixel point at the x-th row and the y-th column in the gradient image P, θ (x, y) is a gradient direction value of a pixel point at the x-th row and the y-th column in the gradient image P, Ml (x, y) is a vector magnitude of an element at the x-th row and the y-th column in the matrix M, M β (x, y) is a vector angle of an element at the x-th row and the y-th column in the matrix M, and int () is a rounding function;

7) and taking the gradient image P as a power failure map, and detecting the failure type by using the power failure map when the power failure occurs.