CN-122017462-A - Island detection method, island detection device, island detection equipment, island detection medium and island detection product

CN122017462ACN 122017462 ACN122017462 ACN 122017462ACN-122017462-A

Abstract

The embodiment of the application provides an island detection method, an island detection device, island detection equipment, island detection media and island detection products, which are applied to the technical field of power grid fault detection and protection. The island judgment method comprises the steps of collecting electric signals of a target grid point based on preset frequency, carrying out signal processing and multi-time scale feature extraction based on the electric signals to obtain multi-dimensional feature parameters corresponding to each time scale, carrying out weighted fusion calculation on the multi-dimensional feature parameters based on preset timeliness weights corresponding to each time scale to obtain fusion feature values of each time scale, determining fusion coefficients of each time scale based on the current running state of the target grid, calculating fusion judgment values corresponding to a plurality of time scales based on the fusion coefficients and the fusion feature values of each time scale, and carrying out island judgment based on the fusion judgment values, the multi-dimensional feature parameters corresponding to each time scale and a preset non-island feature database to obtain island judgment results. The technical effect of improving island detection accuracy is achieved.

Inventors

JU HAO
ZOU PEIYUAN
LI JIAQI
CHEN SHIYI
YU XIONG
WANG FENG
LI QIUHUI
GUO ZHIJUN
LUN SHENGQUAN
LUO SHUQUAN
LIANG YONGCHANG
HE YIN

Assignees

广东电网有限责任公司东莞供电局

Dates

Publication Date: 20260512
Application Date: 20260303

Claims (10)

1. An island detection method, comprising: acquiring an electric signal of a target grid-connected point based on a preset frequency, wherein the electric signal comprises a three-phase current signal, a three-phase voltage signal, a zero-sequence current signal and a zero-sequence voltage signal; Performing signal processing and multi-time scale feature extraction based on the electrical signal to obtain multi-dimensional feature parameters corresponding to each time scale, wherein the multi-time scale comprises a millisecond-level time scale and a second-level time scale; for each time scale, carrying out weighted fusion calculation on the multidimensional characteristic parameters corresponding to the time scale based on the preset timeliness weight corresponding to the time scale to obtain fusion characteristic values corresponding to each time scale; Determining fusion coefficients corresponding to all time scales based on the current running state of the target power grid, and calculating fusion discrimination values corresponding to a plurality of time scales based on the fusion coefficients and the fusion characteristic values corresponding to all the time scales; and carrying out island judgment based on the fusion judgment value, the multidimensional characteristic parameters corresponding to each time scale and a preset non-island characteristic database to obtain an island judgment result.
2. The method according to claim 1, wherein the performing signal processing and multi-time scale feature extraction based on the electrical signal to obtain multi-dimensional feature parameters corresponding to each time scale includes: Generating the current harmonic total distortion rate of the grid-connected point based on the electrical signal; determining a filter window length based on the harmonic total distortion rate; Based on the filter window length, a preset interference compensation factor and a preset weight attenuation factor, performing sliding filter processing on the electric signal to obtain an interference-suppressed electric signal; Based on preset abnormal value judgment conditions, detecting abnormal data of the electrical signals with interference suppressed, and eliminating the abnormal data to obtain electrical signals with abnormal elimination; Performing linear interpolation processing on the electrical signals subjected to abnormal rejection based on a preset trend compensation factor to obtain standardized electrical signals; Performing feature extraction on the standardized electrical signal under each time scale to obtain a plurality of initial feature parameters corresponding to each time scale; performing parameter correction on the plurality of initial characteristic parameters aiming at each time scale to obtain a plurality of corrected standardized characteristic parameters; calculating a distinguishing coefficient of each standardized characteristic parameter based on a plurality of standardized characteristic parameters, and eliminating the standardized characteristic parameters with distinguishing degree lower than a first preset threshold value to obtain a plurality of target characteristic parameters; integrating the target characteristic parameters under each time scale according to a preset sequence, and mapping each target characteristic parameter into a preset interval to obtain the multidimensional characteristic parameter corresponding to each time scale.
3. The method of claim 2, wherein for each time scale, performing weighted fusion calculation on the multi-dimensional feature parameter corresponding to the time scale based on the preset timeliness weight corresponding to the time scale to obtain a fusion feature value corresponding to each time scale, and the method comprises: Aiming at the multi-dimensional characteristic parameters of each time scale, calculating to obtain the distinguishing degree weight of the characteristic parameters based on the distinguishing degree coefficient corresponding to each characteristic parameter in the multi-dimensional characteristic parameters; And carrying out weighted fusion calculation based on the distinguishing degree weight of each characteristic parameter, the parameter value of each characteristic parameter in the multidimensional characteristic parameters and the preset timeliness weight corresponding to the time scale to obtain a fusion characteristic value corresponding to the time scale.
4. The method of claim 3, wherein determining the fusion coefficient corresponding to each time scale based on the current operation state of the target power grid, and calculating the fusion discrimination values corresponding to a plurality of time scales based on the fusion coefficient and the fusion eigenvalue corresponding to each time scale, comprises: Calculating a fusion coefficient corresponding to each time scale based on the running state of the target power grid, wherein the running state comprises the current voltage change rate, the photovoltaic power change rate and an island type identification factor of the target power grid; and carrying out weighted fusion calculation on the fusion characteristic values corresponding to the time scales and a preset cross-time scale correction term based on the fusion coefficient to obtain the fusion discrimination value, wherein the preset cross-time scale correction term comprises a preset cross-time scale compensation term, a preset stability adjustment term and a preset error correction term.
5. The method of claim 4, wherein the performing island determination based on the fusion discrimination value, the multidimensional feature parameter corresponding to each time scale, and a preset non-island feature database to obtain an island determination result comprises: determining a target time scale for similarity matching based on the operational state of the target power grid; performing similarity matching based on the multidimensional feature parameters corresponding to the target time scale and the preset non-island feature database to obtain a similarity matching value; when the similarity matching value is larger than or equal to a second preset threshold value, determining that non-island disturbance exists in the current target power grid; Determining that an island effect exists at the grid-connected point when the similarity matching value is smaller than a second preset threshold value and the fusion judging value and island state data meet a first preset judging condition, wherein the first preset judging condition comprises that the fusion judging value is larger than or equal to the sum of a third preset threshold value and a preset judging correction term, the island effect duration is larger than a fourth preset threshold value, and the difference value between the island effect judging time and the triggering time is smaller than a fifth preset threshold value; When the similarity matching value is smaller than a second preset threshold value and the fusion judging value meets a second preset judging condition, determining that the target power grid normally operates, wherein the second preset judging condition comprises that the fusion judging value is smaller than a difference value between the third preset threshold value and a preset judging correction term; And when the similarity matching value is smaller than a second preset threshold value and the fusion judging value meets a third preset judging condition, determining that the target power grid is in an uncertain state, wherein the third preset judging condition comprises that the fusion judging value is larger than or equal to the difference value between the third preset threshold value and the preset judging correction term and smaller than the sum of the third preset threshold value and the preset judging correction term.
6. The method of claim 5, wherein after obtaining the islanding determination result, the method further comprises: Executing a locking protection action when the island judgment result indicates that the target power grid has non-island disturbance; When the island judgment result indicates that island effect exists in the target power grid, executing protection action; and when the island judgment result indicates that the target power grid is in an uncertain state, prolonging the monitoring time length and carrying out judgment again.
7. Island detection device, characterized by, include: the acquisition module is used for acquiring electric signals of a target grid connection point based on a preset frequency, wherein the electric signals comprise three-phase current signals, three-phase voltage signals, zero-sequence current signals and zero-sequence voltage signals; the first processing module is used for carrying out signal processing and multi-time scale feature extraction based on the electric signal to obtain multi-dimensional feature parameters corresponding to each time scale, wherein the multi-time scale comprises a millisecond-level time scale and a second-level time scale; The second processing module is used for carrying out weighted fusion calculation on the multidimensional characteristic parameters corresponding to the time scales based on the preset timeliness weights corresponding to the time scales aiming at each time scale to obtain fusion characteristic values corresponding to each time scale; The third processing module is used for determining fusion coefficients corresponding to all time scales based on the current running state of the target power grid, and calculating fusion discrimination values corresponding to a plurality of time scales based on the fusion coefficients and the fusion characteristic values corresponding to all the time scales; and the fourth processing module is used for carrying out island judgment based on the fusion judgment value, the multidimensional characteristic parameters corresponding to each time scale and a preset non-island characteristic database to obtain an island judgment result.
8. An electronic device is characterized by comprising a memory and a processor; The memory stores computer-executable instructions; The processor executing computer-executable instructions stored in the memory, causing the processor to perform the method of any one of claims 1-6.
9. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out the method of any one of claims 1-6.
10. A computer program product comprising a computer program which, when executed by a processor, implements the method of any of claims 1-6.

Description

Island detection method, island detection device, island detection equipment, island detection medium and island detection product Technical Field The application relates to the technical field of power grid fault detection and protection, in particular to an island detection method, an island detection device, island detection equipment, island detection media and island detection products. Background In a distributed energy grid-connected power generation system such as photovoltaic and wind power, in order to realize anti-islanding protection, islanding detection needs to be carried out at grid connection points of a distributed energy and a public power grid, and the islanding effect can be timely identified when the distributed energy and a local load form an islanding effect due to the occurrence of faults of the power grid through monitoring the power grid operation state of the grid connection points in real time so as to rapidly cut off, so that the safety of power grid equipment and operators is ensured. In the prior art, the island detection mode mainly comprises the steps of extracting voltage characteristics or current characteristics of the grid-connected point in a certain time scale, and carrying out model detection or rule base detection according to the extracted characteristics so as to determine whether an island effect exists currently. Because the island detection mode in the prior art depends on feature extraction of single dimension and single time scale, multiple island characteristics cannot be considered, and therefore the technical problem of low island detection accuracy in the prior art is caused. Disclosure of Invention The embodiment of the application provides an island detection method, an island detection device, island detection equipment, island detection media and island detection products, which are used for achieving the technical effect of improving island detection accuracy. In a first aspect, an embodiment of the present application provides an island detection method, including: acquiring electric signals of a target grid connection point based on a preset frequency, wherein the electric signals comprise three-phase current signals, three-phase voltage signals, zero-sequence current signals and zero-sequence voltage signals; performing signal processing and multi-time scale feature extraction based on the electrical signals to obtain multi-dimensional feature parameters corresponding to each time scale, wherein the multi-time scale comprises a millisecond-level time scale and a second-level time scale; For each time scale, carrying out weighted fusion calculation on the multidimensional characteristic parameters corresponding to the time scale based on the preset timeliness weight corresponding to the time scale to obtain fusion characteristic values corresponding to each time scale; Determining fusion coefficients corresponding to all time scales based on the current running state of a target power grid, and calculating fusion discrimination values corresponding to a plurality of time scales based on the fusion coefficients and fusion characteristic values corresponding to all the time scales; and carrying out island judgment based on the fusion judgment value, the multidimensional characteristic parameters corresponding to each time scale and a preset non-island characteristic database to obtain an island judgment result. In one possible implementation manner, signal processing and multi-time scale feature extraction are performed based on the electrical signal, so as to obtain multi-dimensional feature parameters corresponding to each time scale, including: generating the current harmonic total distortion rate of the grid-connected point based on the electrical signal; determining a filter window length based on the harmonic total distortion rate; sliding filtering is carried out on the electric signal based on the length of the filtering window, a preset interference compensation factor and a preset weight attenuation factor, so that an interference-suppressed electric signal is obtained; based on preset abnormal value judgment conditions, detecting abnormal data of the electrical signals with interference suppressed, and eliminating the abnormal data to obtain electrical signals with abnormal elimination; Performing linear interpolation processing on the electrical signals subjected to abnormal rejection based on a preset trend compensation factor to obtain standardized electrical signals; Carrying out feature extraction on the standardized electrical signal under each time scale to obtain a plurality of initial feature parameters corresponding to each time scale; Carrying out parameter correction on a plurality of initial characteristic parameters aiming at each time scale to obtain a plurality of corrected standardized characteristic parameters; Calculating a distinguishing coefficient of each standardized characteristic parameter based on the plurality of