CN-122001079-A - Distributed photovoltaic power quality monitoring and analyzing method, device, equipment and medium

CN122001079ACN 122001079 ACN122001079 ACN 122001079ACN-122001079-A

Abstract

The application provides a distributed photovoltaic power quality monitoring and analyzing method, device, equipment and medium, and relates to the technical field of power quality assessment. The method comprises the steps of monitoring electric energy quality data of key nodes in a distributed photovoltaic power station, preprocessing the data to obtain current characteristic data of a plurality of set characteristic indexes, distributing weights for all the set characteristic indexes by adopting an entropy weight method based on the current characteristic data and historical characteristic data in set time periods to reflect the influence degree of different characteristics on the electric energy quality, determining corrected characteristic data corresponding to the set characteristic indexes according to the weights and the current characteristic data, and determining an electric energy quality analysis result of the distributed photovoltaic power station node according to the corrected characteristic data. According to the application, the actual running state and potential problems of the power station can be reflected more accurately by comprehensively considering a plurality of set characteristic indexes and dynamically distributing weights by utilizing an entropy weight method, so that the accuracy of the electric energy quality analysis result is effectively improved.

Inventors

LIN YANGBO
Zhu Murui
HUANG WEIDA
YE DELIANG
Zheng Yizhang
HUANG SHOUYE
YANG HUIBO
Lan Yingbin
CHEN TIANYU
XIAO YUXI

Assignees

广东电网有限责任公司汕尾供电局

Dates

Publication Date: 20260508
Application Date: 20260112

Claims (10)

1. The distributed photovoltaic power quality monitoring and analyzing method is characterized by comprising the following steps of: Monitoring electric energy quality data of key nodes in a distributed photovoltaic power station, wherein the key nodes comprise inverters, branches and junction boxes; Preprocessing the power quality data to obtain current characteristic data of the power quality data corresponding to a plurality of set characteristic indexes; Based on the current characteristic data and the historical characteristic data of the plurality of set characteristic indexes in set time periods respectively, weighting is distributed to each set characteristic index by adopting an entropy weighting method, and the weights respectively corresponding to the set characteristic indexes are obtained, wherein the weights represent the influence degree of the corresponding set characteristic indexes on the power quality; For each set characteristic index, determining corrected characteristic data corresponding to the set characteristic index according to the weight corresponding to the set characteristic index and the current characteristic data; And determining the power quality analysis result of the distributed photovoltaic power station node according to the corrected characteristic data respectively corresponding to the plurality of set characteristic indexes.
2. The method according to claim 1, wherein determining the power quality analysis result of the distributed photovoltaic power station node according to the corrected feature data corresponding to the plurality of set feature indexes, respectively, includes: Inputting the corrected characteristic data corresponding to the plurality of set characteristic indexes and the corrected characteristic data corresponding to the historical characteristic data into a pre-trained long-period and short-period memory neural network model to obtain a power quality trend prediction result of the distributed photovoltaic power station node; Inputting corrected characteristic data corresponding to the plurality of set characteristic indexes and corrected characteristic data corresponding to the historical characteristic data into a pre-trained random forest model to obtain an abnormality diagnosis result of the power quality of the distributed photovoltaic power station node, wherein the abnormality diagnosis result comprises an abnormality type and an abnormality occurrence probability; and determining the power quality analysis result of the distributed photovoltaic power station node based on the corrected characteristic data, the power quality trend prediction result and the abnormality diagnosis result which are respectively corresponding to the plurality of set characteristic indexes.
3. The method according to claim 2, wherein determining the power quality analysis result of the distributed photovoltaic power station node based on the corrected feature data, the power quality trend prediction result, and the abnormality diagnosis result, respectively, corresponding to the plurality of set feature indexes, comprises: Based on the corrected characteristic data respectively corresponding to the plurality of set characteristic indexes, generating the electric energy quality fraction of the distributed photovoltaic power station node by adopting a weighted scoring model, wherein the electric energy quality fraction reflects the health degree of the running state of the distributed photovoltaic power station node; dividing the risk level of the power quality of the distributed photovoltaic power station node according to the power quality fraction and the power quality trend prediction result, and generating trend early warning information of the power quality; Generating a regulation suggestion for the distributed photovoltaic power station node according to the abnormality diagnosis result; and generating an electric energy quality analysis result of the distributed photovoltaic power station node based on the electric energy quality fraction, the trend early warning information and the regulation suggestion.
4. The distributed photovoltaic power quality monitoring analysis method according to claim 3, further comprising: if the power quality score is in a first threshold range, executing a first regulation strategy, wherein the first regulation strategy is used for monitoring and early warning the power quality of the distributed photovoltaic power station node; If the power quality score is within a second threshold range, executing a second regulation strategy, wherein the second regulation strategy is used for triggering trend analysis so as to evaluate whether the power quality of the distributed photovoltaic power station node is further deteriorated; and if the power quality fraction is in a third threshold range, executing a third regulation strategy, wherein the third regulation strategy is used for triggering a regulation mechanism, and the regulation mechanism comprises the steps of reducing the output power of the inverter, switching a standby power supply or adjusting a reactive compensation device.
5. The method according to any one of claims 1 to 4, wherein the step of assigning weights to the set feature indicators by an entropy weight method based on the current feature data and the historical feature data of the plurality of set feature indicators within a set time period, respectively, to obtain weights corresponding to the set feature indicators, respectively, includes: based on the historical characteristic data, carrying out normalization processing on the current characteristic data to obtain normalized characteristic data; For each set characteristic index, determining the information entropy corresponding to the set characteristic index according to the normalized characteristic data corresponding to the set characteristic index; determining the initial weight of the set characteristic index according to the information entropy by utilizing an entropy weight method; and adjusting the initial weights corresponding to the set characteristic indexes by combining with environmental factors to obtain weights corresponding to the set characteristic indexes respectively.
6. The method of any one of claims 1 to 4, wherein monitoring the power quality data of key nodes in a distributed photovoltaic power plant comprises: And acquiring the electric energy quality data of the key nodes in the distributed photovoltaic power station by adopting a set sampling frequency, wherein the set sampling frequency is more than or equal to 200kHz.
7. The method according to any one of claims 1 to 4, wherein the power quality data includes voltage, current, frequency, harmonic, voltage fluctuation and sag data, the plurality of set characteristic indexes include a total harmonic distortion rate, a voltage fluctuation frequency and a sag duration, the preprocessing the power quality data to obtain current characteristic data of the power quality data corresponding to the plurality of set characteristic indexes includes: Performing data cleaning on the electric energy quality data to obtain cleaned electric energy quality data; performing frequency domain analysis on the cleaned voltage time sequence data and current time sequence data by adopting fast Fourier transform to obtain current characteristic data corresponding to the total harmonic distortion rate; And respectively analyzing the cleaned voltage and the cleaned dip data through wavelet transformation to obtain current characteristic data respectively corresponding to the voltage fluctuation frequency and the dip duration.
8. A distributed photovoltaic power quality monitoring and analysis device, comprising: The data acquisition module is used for monitoring the electric energy quality data of key nodes in the distributed photovoltaic power station, wherein the key nodes comprise an inverter, a branch and a combiner box; The preprocessing module is used for preprocessing the power quality data to obtain current characteristic data of the power quality data corresponding to a plurality of set characteristic indexes; The weight distribution module is used for distributing weights to the set characteristic indexes by adopting an entropy weight method based on the current characteristic data and the historical characteristic data of the plurality of set characteristic indexes in set time periods respectively to obtain weights respectively corresponding to the set characteristic indexes, wherein the weights represent the influence degree of the corresponding set characteristic indexes on the electric energy quality; And the power quality analysis module is used for determining a power quality analysis result of the distributed photovoltaic power station node according to the corrected characteristic data respectively corresponding to the plurality of set characteristic indexes.
9. An electronic device comprising a processor and a memory communicatively coupled to the processor; the memory is used for storing computer execution instructions; the processor for executing the computer-executable instructions to implement the method of any one of claims 1 to 7.
10. A computer readable storage medium having stored therein computer executable instructions which when executed are adapted to implement the method of any one of claims 1 to 7.

Description

Distributed photovoltaic power quality monitoring and analyzing method, device, equipment and medium Technical Field The application relates to the technical field of electric energy quality assessment, in particular to a distributed photovoltaic electric energy quality monitoring and analyzing method, device, equipment and medium. Background Under the background of modern energy transformation, photovoltaic power generation is becoming an important choice for replacing non-renewable energy due to the characteristics of green, high efficiency and environmental protection. However, with the wide application of the distributed photovoltaic power station, the random access of a large number of power electronic devices and nonlinear loads causes increasingly prominent electric energy quality problems such as harmonic waves, voltage fluctuation, frequency deviation and the like in the power grid, and the stable operation and the safety and the reliability of the power grid are extremely challenging. In the related art, the total harmonic distortion (Total Harmonic Distortion, abbreviated as THD) of a photovoltaic power station is generally determined according to the harmonic content by monitoring the harmonic content of the photovoltaic power station, and the overall power quality of the photovoltaic power station is evaluated according to the relationship between the THD and the harmonic distortion threshold. The overall power quality obtained in this way has the problem of low accuracy. Disclosure of Invention The application provides a distributed photovoltaic power quality monitoring and analyzing method, device, equipment and medium, which are used for solving the problem of low accuracy of power quality obtained by related technologies. In a first aspect, the present application provides a distributed photovoltaic power quality monitoring and analysis method, including: Monitoring electric energy quality data of key nodes in a distributed photovoltaic power station, wherein the key nodes comprise an inverter, a branch circuit and a combiner box; preprocessing the power quality data to obtain current characteristic data of the power quality data corresponding to a plurality of set characteristic indexes; based on the current characteristic data and historical characteristic data of a plurality of set characteristic indexes in set time periods respectively, weighting is distributed to each set characteristic index by adopting an entropy weighting method, and weights respectively corresponding to each set characteristic index are obtained, wherein the weights represent the influence degree of the corresponding set characteristic index on the power quality; For each set characteristic index, determining corrected characteristic data corresponding to the set characteristic index according to the weight corresponding to the set characteristic index and the current characteristic data; and determining the electric energy quality analysis result of the distributed photovoltaic power station node according to the corrected characteristic data respectively corresponding to the set characteristic indexes. In a possible implementation manner, according to the corrected feature data corresponding to the set feature indexes, the power quality analysis result of the distributed photovoltaic power station node is determined, wherein the power quality analysis result comprises the steps of inputting the corrected feature data corresponding to the set feature indexes and the corrected feature data corresponding to the historical feature data into a Long Short-term memory neural network (LSTM) model which is trained in advance to obtain a power quality trend prediction result of the distributed photovoltaic power station node, inputting the corrected feature data corresponding to the set feature indexes and the corrected feature data corresponding to the historical feature data into a random forest model which is trained in advance to obtain an abnormal diagnosis result of the power quality of the distributed photovoltaic power station node, wherein the abnormal diagnosis result comprises an abnormal type and an abnormal occurrence probability, and determining the power quality analysis result of the distributed photovoltaic power station node based on the corrected feature data, the power quality trend prediction result and the abnormal diagnosis result corresponding to the set feature indexes. In a possible implementation manner, the power quality analysis result of the distributed photovoltaic power station node is determined based on the corrected feature data, the power quality trend prediction result and the abnormality diagnosis result which correspond to the set feature indexes respectively, the power quality analysis result of the distributed photovoltaic power station node is determined based on the corrected feature data, the power quality score of the distributed photovoltaic power station node is generated by a