CN117970033A - Fault positioning method and device for power distribution network, storage medium and electronic device - Google Patents
Fault positioning method and device for power distribution network, storage medium and electronic device Download PDFInfo
- Publication number
- CN117970033A CN117970033A CN202410267993.2A CN202410267993A CN117970033A CN 117970033 A CN117970033 A CN 117970033A CN 202410267993 A CN202410267993 A CN 202410267993A CN 117970033 A CN117970033 A CN 117970033A
- Authority
- CN
- China
- Prior art keywords
- electric energy
- data
- power
- power distribution
- fault
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 99
- 239000011159 matrix material Substances 0.000 claims abstract description 72
- 238000007405 data analysis Methods 0.000 claims abstract description 35
- 238000004458 analytical method Methods 0.000 claims description 47
- 238000004422 calculation algorithm Methods 0.000 claims description 39
- 238000012545 processing Methods 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 24
- 230000008859 change Effects 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 21
- 238000006243 chemical reaction Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 18
- 238000013145 classification model Methods 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 14
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000003066 decision tree Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 8
- 238000001228 spectrum Methods 0.000 claims description 8
- 238000000546 chi-square test Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 230000003595 spectral effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 5
- 230000004807 localization Effects 0.000 claims 4
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000012544 monitoring process Methods 0.000 description 20
- 230000000694 effects Effects 0.000 description 9
- 238000010801 machine learning Methods 0.000 description 6
- 238000012423 maintenance Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 230000002596 correlated effect Effects 0.000 description 5
- 238000007637 random forest analysis Methods 0.000 description 5
- 238000007619 statistical method Methods 0.000 description 5
- 238000012806 monitoring device Methods 0.000 description 4
- 230000001052 transient effect Effects 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000011158 quantitative evaluation Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000007636 ensemble learning method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000000528 statistical test Methods 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application discloses a fault positioning method and device for a power distribution network, a storage medium and an electronic device, and relates to the technical field of power distribution network fault positioning. The method comprises the following steps: acquiring target electric energy data of a power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system; analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network; determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing the power quality disturbance features of different types; and identifying the feature matrix by adopting a fault positioning model to obtain the target fault position. The application solves the technical problems of weak correlation between the data characteristics and the specific positions of faults and inaccurate fault positioning in the related technology.
Description
Technical Field
The application relates to the technical field of fault location of power distribution networks, in particular to a fault location method and device of a power distribution network, a storage medium and an electronic device.
Background
In a power distribution network, disturbance factor monitoring of power quality is an important link for ensuring stable operation of a power grid. Disturbance factor monitoring of power quality refers to monitoring and analyzing power quality to detect causes and influencing factors of power quality problems. These perturbation factors may include voltage fluctuations, frequency offsets, harmonics, discontinuities, flicker, etc. Monitoring these disturbance factors can help identify the source of the power quality problem and take corresponding measures to improve the power quality and ensure stable and reliable operation of the power system.
In the power quality monitoring and analysis of a power distribution network, accurate identification and careful classification of power quality problems are required to facilitate understanding and solving of various disturbance sources, however, excessive attention to the details of classification may impair the ability to determine fault locations. The reason is that when design emphasis is placed on distinguishing and identifying a wide variety of power quality disturbance types, a large number of data features may be required to describe and distinguish these disturbances. These features may not be strongly correlated to the specific location of the fault, resulting in that the critical features needed to locate the fault may not be adequately learned when processing the disturbance classification task.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the application provides a fault positioning method, a fault positioning device, a storage medium and an electronic device for a power distribution network, which are used for at least solving the technical problems of weak correlation between data characteristics and specific positions of faults and inaccurate fault positioning in the related technology.
According to one embodiment of the application, a method for determining the end of a cable joint is provided. The method may include: acquiring target electric energy data of a power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system; analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network; determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing the power quality disturbance features of different types; and identifying the feature matrix by adopting a fault positioning model to obtain the target fault position.
Optionally, acquiring the target electrical energy data of the power distribution network includes: acquiring initial electric energy data of a power distribution network; and carrying out data preprocessing on the initial electric energy data to obtain target electric energy data.
Optionally, performing data preprocessing on the initial power data to obtain target power data includes: comparing the timestamp difference value of the initial electric energy data based on a preset acquisition interval, and determining a missing value in the initial electric energy data; filling the missing value in the initial electric energy data to obtain a first data sequence; determining an abnormal value in the first data sequence by adopting a standard deviation method, and processing the abnormal value by adopting a moving average method to obtain a second data sequence; and carrying out time zone calibration on the second data sequence, and carrying out unit conversion on the second data sequence after time zone calibration to obtain target electric energy data.
Optionally, the method further comprises: acquiring an electric energy data sample of a power distribution network, wherein the electric energy data sample is a data sample for transmitting, distributing and using electric energy in a power distribution system; determining a power quality characteristic based on the power data sample, wherein the power quality characteristic is used for representing the power quality of the power distribution network; and obtaining an electric energy data analysis model based on the decision tree algorithm and the quality feature training, wherein the electric energy data analysis model is used for identifying the electric energy quality disturbance feature.
Optionally, determining the feature matrix based on the power quality disturbance feature comprises: performing time-frequency analysis on the power quality disturbance characteristics to determine frequency spectrum characteristics and instantaneous frequency data; extracting preset characteristics of the power quality disturbance characteristics, and determining amplitude characteristics and a conversion process of the voltage under preset conditions; adopting a Hilbert-Huang transformation algorithm to analyze the disturbance characteristics of the electric energy quality and determining the phase angle change and the flicker degree; analyzing the power quality disturbance characteristics by adopting a harmonic analysis method, and determining the harmonic distortion degree in the power distribution system; analyzing time sequence information of the power quality disturbance characteristics, and determining disturbance duration; and determining disturbance information in response to the power quality disturbance characteristics meeting a preset analysis condition, wherein the disturbance information comprises at least one of the following: dynamic characteristics of power supply frequency variation, phase difference, frequency and amplitude of inter-harmonic or frequency and amplitude of subharmonic component; a feature matrix is constructed based on the spectral features, instantaneous frequency data, amplitude features, conversion process, phase angle change, flicker level, harmonic distortion level, disturbance duration and disturbance information.
Optionally, identifying the feature matrix by using a fault location model, and obtaining the target fault location includes: screening the feature matrix to determine a first feature, wherein the first feature is related to the fault position; processing the first feature based on a disturbance classification model to obtain a target disturbance classification type, wherein the disturbance classification model is obtained by training based on a multi-target optimization algorithm; and processing the target disturbance classification type by adopting a fault positioning model to obtain a target fault position.
Optionally, screening the feature matrix, and determining the first feature includes: screening the feature matrix by adopting a chi-square test method to obtain a second feature, wherein the second feature is related to the fault position; and performing redundancy processing on the second feature by adopting the mutual information to obtain the first feature.
Optionally, the method further comprises: acquiring a characteristic sample, topology information of a power distribution network and load distribution data of the power distribution network, wherein the characteristic sample is related to a fault position; and training based on a convolutional neural network algorithm, a characteristic sample, topology information and load distribution data to obtain a fault positioning model.
Optionally, the method further comprises: and adjusting the target fault position based on the topology information of the power distribution network to obtain an adjusted target fault position, wherein the position range of the adjusted target fault position is smaller than the position range of the target fault position before adjustment.
Optionally, the method further comprises: evaluating the fault positioning model based on the target fault position and the actual fault position to obtain an evaluation result; and updating the fault location model based on the evaluation result.
According to one embodiment of the application, a fault locating device of the power distribution network is also provided. The device comprises: the power distribution system comprises an acquisition module, a power distribution module and a power distribution module, wherein the acquisition module is used for acquiring target electric energy data of the power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in the power distribution system; the analysis module is used for analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network; the determining module is used for determining a characteristic matrix based on the power quality disturbance characteristics, wherein the characteristic matrix is used for dividing different types of power quality disturbance characteristics; and the identification module is used for identifying the feature matrix by adopting the fault positioning model to obtain the target fault position.
Optionally, the acquiring module is further configured to acquire initial electrical energy data of the power distribution network; and carrying out data preprocessing on the initial electric energy data to obtain target electric energy data.
Optionally, the acquisition module is further configured to compare the timestamp difference value of the initial electrical energy data based on a preset acquisition interval, and determine a missing value in the initial electrical energy data; filling the missing value in the initial electric energy data to obtain a first data sequence; determining an abnormal value in the first data sequence by adopting a standard deviation method, and processing the abnormal value by adopting a moving average method to obtain a second data sequence; and carrying out time zone calibration on the second data sequence, and carrying out unit conversion on the second data sequence after time zone calibration to obtain target electric energy data.
Optionally, the apparatus further comprises: the analysis model module is used for acquiring an electric energy data sample of the power distribution network, wherein the electric energy data sample is a data sample for transmitting, distributing and using electric energy in a power distribution system; determining a power quality characteristic based on the power data sample, wherein the power quality characteristic is used for representing the power quality of the power distribution network; and obtaining an electric energy data analysis model based on the decision tree algorithm and the quality feature training, wherein the electric energy data analysis model is used for identifying the electric energy quality disturbance feature.
Optionally, the determining module is further configured to perform time-frequency analysis on the power quality disturbance characteristic, and determine a frequency spectrum characteristic and instantaneous frequency data; extracting preset characteristics of the power quality disturbance characteristics, and determining amplitude characteristics and a conversion process of the voltage under preset conditions; adopting a Hilbert-Huang transformation algorithm to analyze the disturbance characteristics of the electric energy quality and determining the phase angle change and the flicker degree; analyzing the power quality disturbance characteristics by adopting a harmonic analysis method, and determining the harmonic distortion degree in the power distribution system; analyzing time sequence information of the power quality disturbance characteristics, and determining disturbance duration; and determining disturbance information in response to the power quality disturbance characteristics meeting a preset analysis condition, wherein the disturbance information comprises at least one of the following: dynamic characteristics of power supply frequency variation, phase difference, frequency and amplitude of inter-harmonic or frequency and amplitude of subharmonic component; a feature matrix is constructed based on the spectral features, instantaneous frequency data, amplitude features, conversion process, phase angle change, flicker level, harmonic distortion level, disturbance duration and disturbance information.
Optionally, the identification module is further configured to screen the feature matrix to determine a first feature, where the first feature is related to the fault location; processing the first feature based on a disturbance classification model to obtain a target disturbance classification type, wherein the disturbance classification model is obtained by training based on a multi-target optimization algorithm; and processing the target disturbance classification type by adopting a fault positioning model to obtain a target fault position.
Optionally, the identification module is further configured to screen the feature matrix by using a chi-square inspection method to obtain a second feature, where the second feature is related to the fault location; and performing redundancy processing on the second feature by adopting the mutual information to obtain the first feature.
Optionally, the apparatus further comprises: the training module is used for acquiring a characteristic sample, topology information of the power distribution network and load distribution data of the power distribution network, wherein the characteristic sample is related to a fault position; and training based on a convolutional neural network algorithm, a characteristic sample, topology information and load distribution data to obtain a fault positioning model.
Optionally, the apparatus further comprises: the adjusting module is used for adjusting the target fault position based on the topology information of the power distribution network to obtain an adjusted target fault position, wherein the position range of the adjusted target fault position is smaller than the position range of the target fault position before adjustment.
Optionally, the apparatus further comprises: the updating module is used for evaluating the fault positioning model based on the target fault position and the actual fault position to obtain an evaluation result; and updating the fault location model based on the evaluation result.
According to one embodiment of the present application, there is further provided a computer readable storage medium, where the computer readable storage medium includes a stored executable program, and when the executable program runs, the device where the computer readable storage medium is located is controlled to execute the fault location determining method of the power distribution network in the embodiment of the present application.
According to one embodiment of the present application, there is further provided a computer program product, including a computer program, where the computer program when executed by a processor implements the fault location determining method of the power distribution network according to the embodiment of the present application.
According to an embodiment of the present application, there is further provided an electronic device including a memory and a processor, where the memory stores an executable program, and the processor is configured to run the program and configured to execute the method for determining fault location of a power distribution network according to the embodiment of the present application.
In the embodiment of the application, the target electric energy data of the power distribution network is obtained, wherein the target electric energy data is the data of transmitting, distributing and using electric energy in a power distribution system; analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network; determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing the power quality disturbance features of different types; and identifying the feature matrix by adopting a fault positioning model to obtain the target fault position. Therefore, the technical effects of efficiently processing the electric energy quality problem and positioning the power distribution network faults are achieved, the operation efficiency and reliability of the power distribution network are improved, an efficient method for processing the electric energy quality problem and positioning the power distribution network faults is provided for power grid operation and maintenance personnel, and the technical problems that the correlation between the data characteristics and the specific positions of the faults in the related technology is not strong and the fault positioning is inaccurate are solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a flow chart of a fault location method for a power distribution network according to an embodiment of the present application;
Fig. 2 is a block diagram of a fault location monitoring device for a power distribution network according to an embodiment of the present application.
Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In order to better understand the embodiments of the present application, technical terms related to the embodiments of the present application are explained as follows:
Decision tree algorithm: a machine learning algorithm for classification and prediction constructs a tree structure by partitioning and classifying data, each node representing an attribute or feature, each branch representing a possible value, and each leaf node representing a class or final prediction result. The decision tree algorithm can classify and predict according to the characteristics and attributes of the data, is an intuitive and easy-to-understand machine learning algorithm, and is widely applied to various fields.
And (3) time-frequency analysis: a signal processing technique for studying characteristics of a signal in both time and frequency domains. Common time-frequency analysis methods include short-time fourier transform (STFT), wavelet transform, wigner-Ville distribution, and the like. Time-frequency analysis may be applied in many fields including audio processing, image processing, radar signal processing, biomedical signal processing, and the like.
Support vector machine: a supervised learning algorithm for classification and regression analysis classifies data points of different categories by finding optimal hyperplanes in a feature space so that the data points can be separated to a maximum extent. SVM has a strong generalization ability and a processing ability for high-dimensional data, and thus is widely used in many fields including text classification, image recognition, bioinformatics, and the like.
And (5) chi-square test: a statistical method for verifying the existence of a correlation between two or more classified variables determines whether the relationship between the variables is significant by comparing the difference between the actual observed frequency and the expected frequency. Chi-square test is commonly used to analyze data in a list (also known as a cross-table) to determine if there is a statistically significant association between variables.
Random forest model: an ensemble learning method predicts by combining multiple decision trees. In random forests, each decision tree is trained based on a randomly selected data subset and feature subset, and the final predicted result is the average or voting result of a plurality of decision trees.
Pareto front: in Pareto optimal solution sets, the boundaries of all non-dominant solutions, typically used in a multi-objective optimization problem, represent that under a given resource constraint, one objective cannot be modified without compromising other objectives.
Mutual information: a statistic for measuring the correlation between two random variables is calculated by comparing the difference between the joint probability distribution of the two variables and the respective edge probability distribution. The larger the value of the mutual information, the higher the correlation between the two variables. Mutual information is commonly used in the fields of feature selection, cluster analysis, information retrieval, and the like.
Example 1
According to an embodiment of the present application, there is first provided a fault location determination method for a power distribution network, it being noted that the steps shown in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in an order different from that herein.
Fig. 1 is a flowchart of a fault locating method of a power distribution network according to an embodiment of the present application, as shown in fig. 1, the method at least includes steps S12-S18, where:
Step S12, target electric energy data of the power distribution network are obtained, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system.
The distribution network is a system for transmitting electric energy generated by a power plant to end users, and consists of a transmission network and a distribution network, wherein the transmission network is responsible for transmitting high-voltage electric energy from the power plant to substations in various areas, and the distribution network is used for transmitting low-voltage electric energy from the substations to buildings or equipment of the end users. Distribution networks typically include substations, power transmission and distribution lines, switching devices, and the like. The main function of the system is to realize distribution, control and protection of electric energy so as to meet the requirement of users on electric power.
For example, the fault location method of the power distribution network first obtains target power data of the power distribution network, where the target power data may be data of transmitting, distributing and using power in the power distribution system, and the data includes, but is not limited to, power consumption, power supply, power loss, power load curve, power load rate, and the like. The electric energy data can be used for monitoring and analyzing the operation condition of the power distribution network and optimizing the operation efficiency and stability of the power distribution network.
And S14, analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network.
Among these, the power data analysis models are mathematical and statistical models for analyzing power data, which can be constructed based on conventional statistical methods, machine learning algorithms, or deep learning techniques to better understand and utilize power data. The power quality disturbance features are specific features of various power quality problems occurring in the power system, including: wave motion, harmonics, discontinuities, asymmetry, and flicker. Fluctuations, i.e. power quality disturbances, lead to fluctuations in voltage and current, which are manifested as waveform instability or frequency variation. Harmonics, i.e. power quality disturbances, cause the generation of harmonics in the power grid, resulting in distortion of the voltage and current waveforms. Interruption, i.e. disturbance of the power quality, may lead to interruption of the power supply, manifested as a short power outage or a sudden drop in the power quality. An asymmetry, i.e. a disturbance of the power quality, may lead to an asymmetry of the voltage and current, i.e. an imbalance of the three-phase voltage and current. Flicker, i.e. disturbances in the quality of the electrical energy, may also lead to momentary changes in the voltage, which may be manifested as flickering or short voltage fluctuations. These features are a common manifestation of power quality disturbances, which can have an impact on the normal operation of the power system and on the safety of power usage.
The electric energy data analysis model is used for analyzing the target electric energy data, the electric energy data analysis model is used for analyzing the electric energy data of the power distribution network, so that specific disturbance characteristics of electric energy quality in the electric power system are obtained, the disturbance characteristics of the electric energy quality are used for reflecting the stability degree of the electric energy quality of the power distribution network, the disturbance characteristics of the electric energy quality are monitored and analyzed, operators and users of the electric power system can be helped to find problems in time, and corresponding measures are taken to improve the electric energy quality.
And S16, determining a feature matrix based on the power quality disturbance characteristics, wherein the feature matrix is used for dividing different types of power quality disturbance characteristics.
For example, based on features related to power quality disturbances, a feature matrix may be generated for disturbance classification that helps identify and classify different types of power quality disturbances, such as, for example, transient power quality disturbances: including transient voltage fluctuations, transient voltage dips, and the like; short-term power quality disturbance: including short-time voltage fluctuation, short-time voltage interruption, and the like; long-term power quality disturbance: including long-time voltage sag, and the like; harmonic wave: the voltage or the current contains frequency components with the frequency being integral multiples of the fundamental wave frequency; power quality discontinuity: including voltage dips, voltage phase breaks, voltage imbalances, and the like. Through the feature matrix, the accuracy and reliability of disturbance classification can be further improved.
And S18, identifying the feature matrix by adopting a fault positioning model to obtain a target fault position.
The fault positioning model is a model for identifying and positioning faults of a system or equipment, and can identify the cause possibly causing the faults by analyzing the operation data and the behavior mode of the system, determine the specific position of the faults, and quickly and accurately position and solve the faults by commonly utilizing the technologies of data mining, machine learning, statistical analysis and the like, thereby improving the reliability and the stability of the system. The target fault location may be a fault or point of failure in the power system, such as a fault location of a wire, cable, switchgear, transformer, etc. These fault locations may cause problems with power system interruption, short circuits, overloads, etc., affecting the proper operation of the power supply.
The fault location model is used for analyzing the monitoring data after dividing the disturbance characteristics of the power quality of different types, and the feature matrix is identified, so that the target fault position is obtained, and the fault result is located.
Based on the steps, target electric energy data of the power distribution network are obtained, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system; analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network; determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing the power quality disturbance features of different types; and identifying the feature matrix by adopting a fault positioning model to obtain the target fault position. Therefore, the technical effects of efficiently processing the electric energy quality problem and positioning the power distribution network faults are achieved, the operation efficiency and reliability of the power distribution network are improved, an efficient method for processing the electric energy quality problem and positioning the power distribution network faults is provided for power grid operation and maintenance personnel, and the technical problems that the correlation between the data characteristics and the specific positions of the faults in the related technology is not strong and the fault positioning is inaccurate are solved.
Optionally, in step S12, acquiring the target electrical energy data of the power distribution network includes performing the following steps:
Step S121, initial electric energy data of a power distribution network are obtained;
Step S122, data preprocessing is performed on the initial electric energy data to obtain target electric energy data.
For example, the power quality monitoring device is first configured to acquire the target power data of the power distribution network, and the power quality monitoring device includes a smart meter and a power quality analyzer to acquire real-time power data of the power distribution network, that is, initial power data of the power distribution network, and further perform preprocessing on the initial power data, where the purpose of data preprocessing is to clean, convert and prepare the data for further analysis and modeling. Through data preprocessing, the error value can be removed, missing data can be processed, standardized data can be processed, and operations such as feature selection, dimension reduction and the like can be performed on the data, so that the quality and usability of the data and the accuracy and stability of a model are improved, the time and resource consumption of model training are reduced, and target electric energy data are obtained.
Optionally, in step S122, performing data preprocessing on the initial power data to obtain target power data includes performing the following steps:
step S1221, comparing the time stamp difference values of the initial power data based on a preset acquisition interval, and determining a missing value in the initial power data;
step S1222, filling the missing value in the initial electric energy data to obtain a first data sequence;
step S1223, determining an outlier in the first data sequence by using a standard deviation method, and processing the outlier by using a moving average method to obtain a second data sequence;
Step S1224, time zone calibration is performed on the second data sequence, and unit conversion is performed on the second data sequence after time zone calibration to obtain target electric energy data.
The preset acquisition interval is a preset time interval for acquiring data, and the setting of the time interval is determined according to practical situations and is not limited here. A time stamp is a number or character that indicates a particular point in time, and the time stamp difference may be used to calculate the time interval between two events or points in time to determine a missing value in the power data, i.e., data that has not been recorded or collected for a certain period of time or a particular point in time when the power consumption or the generation of data was recorded, possibly due to equipment failure, data collection errors, communication problems, or other reasons. Missing values can affect the integrity and accuracy of the data, requiring data cleansing and padding processes.
Illustratively, the data preprocessing for the power data of the power distribution network includes: obtaining a missing value by comparing the timestamp difference value with a preset acquisition interval; filling the missing value by an interpolation method or based on the data of adjacent time points to obtain a complete data sequence, namely a first data sequence; then determining an abnormal value in the first data sequence by using a standard deviation method, and processing the abnormal value by using a moving average method to obtain a processed data sequence, namely a second data sequence, so as to reduce the interference of short-term fluctuation on long-term trend analysis; and finally, carrying out time zone calibration and unit conversion on the power data of the power distribution network, determining that the power data of the power distribution network all conform to the same standard and format, namely carrying out time zone calibration on the second data sequence, and carrying out unit conversion on the second data sequence after time zone calibration to obtain target power data.
Illustratively, if the preset acquisition interval of the smart meter is 1 minute, but two consecutive timestamps are found to be 3 minutes apart in a certain examination, this means that there is a2 minute data loss in the middle. The two data points may be padded by a linear interpolation method. If the missing timestamps are 12:01 and 12:04, then the 12:01 and 12:02 power values can be estimated using the 12:00 and 12:05 data. In the continuously monitored voltage data, the voltage value at most time points fluctuates around 220V (volts), but an abnormal point of 300V voltage value suddenly appears at a certain time point. By calculating the standard deviation of the voltage data, it was found that the abnormal value exceeded the range of the average value plus three times the standard deviation. To smooth this outlier, a 5-point moving average method may be used to replace the voltage value at that point with the average of its two data points before and after it. If the monitoring devices are distributed in different time zones, all data needs to be calibrated to a uniform reference time zone for data consistency. As in device a in the eastern eight sector, the recorded power consumption was 10:00, 10kWh. Device B was in the eastern nine zone with a recorded power consumption of 11:00, 12kWh. For data consistency, the recording time of device B was adjusted to 10:00 to remain consistent with device a. Meanwhile, if the power units are different, if some of the devices record kilowatt-hours and some of the units are joules, all the data need to be converted into the same unit, for example, 1 kwh=6×10 6 J. If the frequency fluctuation characteristic in one period is selected. In one month of data recording, the normal grid frequency should be kept around 50Hz, but the frequency is found to be relatively high in fluctuation within a certain period of days when the data are analyzed, and the frequency is varied from 48Hz to 52 Hz.
Optionally, the method further comprises performing the steps of:
Step S131, acquiring an electric energy data sample of the power distribution network, wherein the electric energy data sample is a data sample for transmitting, distributing and using electric energy in a power distribution system;
step S132, determining a power quality characteristic based on the power data sample, wherein the power quality characteristic is used for representing the power quality of the power distribution network;
And step S133, training based on a decision tree algorithm and quality characteristics to obtain an electric energy data analysis model, wherein the electric energy data analysis model is used for identifying electric energy quality disturbance characteristics.
The fault location method of the power distribution network further comprises the step of obtaining a power data sample of the power distribution network, wherein the power data sample is a data sample of power transmitted, distributed and used in the power distribution system, and the power data sample comprises, but is not limited to, power consumption, power load, power loss, power quality and the like. Further, a power quality characteristic is determined based on the power data samples, wherein the power quality characteristic is used for representing the power quality of the power distribution network, namely, the characteristic describing the power quality is extracted from the power data of the power distribution network, and the characteristic describing the power quality includes, but is not limited to, voltage and current, frequency, waveform distortion, switching operation, weather factors, load changes, equipment aging, equipment maintenance records and the like.
Finally, an electric energy data analysis model is obtained based on a decision tree algorithm and quality feature training, wherein the electric energy data analysis model is used for identifying electric energy quality disturbance features and is used for analyzing electric energy data, and data mining, statistical analysis, machine learning and other technologies are generally utilized for processing and analyzing the electric energy data so as to provide useful insight and decision support and improve energy utilization efficiency. And training an electric energy data analysis model by using a decision tree algorithm based on the characteristics describing the electric energy quality to identify the characteristics related to the electric energy quality disturbance, and further applying the trained electric energy data analysis model to new electric energy data for real-time monitoring and analysis to evaluate the performance of the model. And according to the analysis result, optimizing and managing the power quality of the power distribution network, and updating and optimizing the model periodically so as to adapt to different power quality problems and scenes. The electric energy quality of the power distribution network is monitored in real time through an electric energy data analysis model, so that the electric energy quality problem can be found and solved in time; and the electric energy data and the analysis result of the power distribution network are displayed in a visual form, so that a user can intuitively know the electric energy quality condition and the quality problem.
Illustratively, a power data set comprising the following features describing power quality: voltage (V), current (a), frequency (Hz), total Harmonic Distortion (THD), number of switching operations, air temperature (c), rate of load change (%), age of transformer and last maintenance date of equipment. By preliminary analysis of the electrical energy data set, it was found that the standard voltage was 220V at a voltage value below 217V, and the total harmonic distortion rate reached 7%, exceeding the prescribed 5% standard, indicating the waveform distortion problem. The power data is trained using a decision tree algorithm that can efficiently process both classification and numerical features. In the cross-validation process, a model with an accuracy of 95% is obtained, which indicates that the model can well identify key features of the power quality disturbance. And applying the trained electric energy data analysis model to a new data set for real-time monitoring. The model detects that after a sudden increase in load from 50kW to 55kW, by 10%, the frequency drops to 48Hz, well below the normal 50Hz, revealing a severe impact of load variations on the power quality. Statistical analysis results show that in winter, the temperature is reduced by-5 ℃, the efficiency of the transformer is reduced, and the problem of electric energy quality is frequently caused. During the 12 month to 2 month statistics, the power quality disturbance event increased by 20% over summer. Based on these analysis results, maintenance teams of the distribution network can be optimized and managed in a targeted manner, for example by adding winter transformer insulation measures, or by starting up backup power to stabilize the frequency in case of sudden load increases. At the same time, the model is updated periodically, such as once a quarter and optimized to accommodate new power quality conditions and scenarios. The electric energy data and analysis results of the distribution network can be displayed through a visual instrument panel. Such as a graph showing voltage, frequency and harmonic distortion in real time, and a red warning signal to indicate any indicators outside the normal range to help the user to intuitively understand the current power quality condition.
Optionally, in step S16, determining the feature matrix based on the power quality disturbance characteristics includes performing the steps of:
step S161, performing time-frequency analysis on the power quality disturbance characteristics to determine frequency spectrum characteristics and instantaneous frequency data;
Step S162, extracting preset characteristics of the power quality disturbance characteristics, and determining amplitude characteristics and conversion processes of the voltage under preset conditions;
Step S163, adopting a Hilbert-Huang transform algorithm to analyze the power quality disturbance characteristics and determining the phase angle change and the flicker degree;
step S164, analyzing the power quality disturbance characteristics by adopting a harmonic analysis method, and determining the harmonic distortion degree in the power distribution system;
step S165, analyzing time sequence information of the power quality disturbance characteristics to determine disturbance duration;
step S166, determining disturbance information in response to the power quality disturbance characteristics meeting a preset analysis condition, wherein the disturbance information comprises at least one of the following: dynamic characteristics of power supply frequency variation, phase difference, frequency and amplitude of inter-harmonic or frequency and amplitude of subharmonic component;
Step S167, constructing a feature matrix based on the spectral feature, the instantaneous frequency data, the amplitude feature, the conversion process, the phase angle change, the flicker level, the harmonic distortion level, the disturbance duration and the disturbance information.
Illustratively, the feature matrix is determined based on the power quality disturbance feature, and the power quality disturbance feature is first subjected to time-frequency analysis to determine the frequency spectrum feature and the instantaneous frequency data, where the frequency spectrum feature is the feature of the signal in the frequency domain, such as the distribution of frequency components, the intensity of the frequency components, and so on. Instantaneous frequency data is the frequency variation of the signal over time, i.e. the frequency values at different points in time.
And further, extracting preset characteristics of the power quality disturbance characteristics, and determining amplitude characteristics and conversion processes of the voltage under preset conditions. The preset condition may be preset amplitude characteristic and conversion process of the voltage, which is the case of voltage sag, sag rise or short interruption in the present application.
And then, adopting a Hilbert-Huang transform algorithm to analyze the disturbance characteristics of the electric energy quality, namely, nonlinear and non-stable electric energy signals, and determining the phase angle change and the flicker degree. Phase angle variation refers to the amount of phase change of a signal over a period, typically expressed in terms of angle or radians. The degree of flicker refers to the degree of rapid change in frequency or phase of a signal over a period of time, i.e., the instantaneous frequency change of the signal. In signal processing, the phase angle change and the degree of flicker can be used to describe the frequency characteristics and dynamic characteristics of the signal.
Meanwhile, the power quality disturbance characteristics are analyzed by adopting a harmonic analysis method, and the total harmonic distortion degree, which is an index for describing the distortion degree of all harmonic components in the signal, is the square root of the ratio of the effective values of all harmonic components to the effective value of the fundamental component. The smaller the total harmonic distortion, the less harmonic components in the signal, and the smaller the distortion. The degree of harmonic distortion in the power distribution system is determined by analysis. The degree of harmonic distortion refers to the degree of non-fundamental wave (i.e., harmonic) components contained in the waveform output from the power supply or circuit. The degree of harmonic distortion is typically expressed by the harmonic distortion ratio (THD), i.e., the ratio of the effective value of the non-fundamental voltage or current to the effective value of the fundamental voltage or current.
In addition, the time sequence information analysis is carried out on the power quality disturbance characteristics, namely, the time sequence information of the power quality disturbance event is obtained, and the disturbance duration time, namely, the duration time of the event is determined through the analysis on the starting time and the ending time of the event.
And determining disturbance information in response to the power quality disturbance characteristics meeting preset analysis conditions. The predetermined analysis condition may be that the power quality disturbance signal exhibits a frequency deviation exceeding a predetermined range/if the power disturbance event exhibits phase asymmetry/if an inter-harmonic or sub-harmonic component is detected in the power quality disturbance signal. Wherein the disturbance information includes at least one of: dynamic characteristics of power supply frequency variation, phase difference, frequency and amplitude of inter-harmonics or frequency and amplitude of sub-harmonic components. That is, if the power quality disturbance signal shows a frequency deviation exceeding a predetermined range, the rate of change of the frequency is analyzed to determine the dynamic characteristics of the power supply frequency change. If the electric energy disturbance event presents phase asymmetry, calculating the phase difference between the phases. If an inter-harmonic or sub-harmonic component is detected in the power quality disturbance signal, the frequency and the amplitude of the inter-harmonic or sub-harmonic component are positioned through spectrum analysis so as to evaluate the influence on the power quality.
And finally, constructing a feature matrix based on the frequency spectrum features, the instantaneous frequency data, the amplitude features, the conversion process, the phase angle change, the flicker degree, the harmonic distortion degree, the disturbance duration and the disturbance information, namely constructing a multi-dimensional feature matrix based on the extracted features, and classifying the power quality disturbance of different types.
Illustratively, the grid signal has a power quality disturbance event at 3 seconds. Analysis showed that near the fundamental frequency of 50Hz, a disturbance with an instantaneous frequency of 52Hz occurs, with an amplitude of 8pu (numerical marking method, representing the relative values of the physical quantities and parameters). This indicates the spectral characteristics of the disturbance and the instantaneous frequency information. Wavelet transformation was applied to the same signal and found to have a voltage dip event on the time scale of 1-2 seconds, with the amplitude falling from 1pu to 7pu and recovering within 5 seconds. Hilbert-Huang transform analysis shows that the phase angle at which the disturbance occurs varies by 30 degrees and the degree of flicker is 05, which can help determine the non-linear and non-stationary nature of the power quality disturbance. And the total harmonic distortion degree is calculated, so that the harmonic distortion existing in the system is determined to be 5% and exceeds the recommended value of International standard IEC61000-2-2 by 2%. By analyzing the time sequence information of the event, the voltage sag event is determined to start at 00 seconds, end at 50 seconds and last for 50 seconds. The disturbance signal shows that the frequency deviation is reduced from 50Hz to 48Hz, the frequency deviation is restored to 50Hz within 10 seconds, and the change rate is 2 Hz/second, so that the slow frequency restoration characteristic of the power grid is shown. In a three-phase system, the difference between phase a and phase B is 120 degrees, the difference between phase B and phase C is 125 degrees, the difference between phase C and phase a is 115 degrees, and a phase asymmetry of 5 degrees is exhibited. Spectral analysis showed that an inter-harmonic component was detected at 100Hz with an amplitude of 1pu and a sub-harmonic component was detected at 25Hz with an amplitude of 05pu, indicating that these components may have some impact on the power quality. Based on the extracted characteristics, a multidimensional characteristic matrix is constructed, and different types of power quality disturbance are classified by using a support vector machine, so that 95% of accurate diagnosis rate is realized.
Optionally, in step S18, identifying the feature matrix using the fault location model, and obtaining the target fault location includes performing the following steps:
Step S181, screening the characteristic matrix to determine a first characteristic, wherein the first characteristic is related to the fault position;
s182, processing the first feature based on a disturbance classification model to obtain a target disturbance classification type, wherein the disturbance classification model is obtained based on training of a multi-target optimization algorithm;
And step S183, processing the target disturbance classification type by adopting a fault positioning model to obtain a target fault position.
The feature matrix is screened to determine first features, wherein the first features can be understood as positive correlation or negative correlation features of the screened fault positions after the acquired features are ranked according to the contribution degree of the fault recognition capability. The disturbance classification accuracy and the fault location accuracy are used as optimization targets. And further processing the first feature based on the disturbance classification model to obtain a target disturbance classification type, wherein the disturbance classification model is obtained based on training of a multi-target optimization algorithm. The first characteristic is used as input, a multi-objective optimization algorithm is selected to establish a disturbance classification model, configuration algorithm parameters comprise population size, cross rate and variation rate, and disturbance classification types are used as output. And continuously updating the population by the iterative disturbance classification model until the preset iterative times are reached or the stopping condition is met. In each iteration process, the classification result is evaluated and selected according to the performance of disturbance classification accuracy and fault location accuracy.
And finally, processing the target disturbance classification type by adopting a fault positioning model to obtain a target fault position. The method comprises the steps of carrying out multi-level classification on characteristics positively or negatively related to fault positions, constructing a fault positioning model, carrying out fault positioning by using a corresponding fault positioning method based on each subdivision category, and determining a target fault position, namely a fault point of the power distribution network.
Illustratively, first, mutual information is used to screen out 30 features strongly related to the fault location from the original 100 features. Among these features, it was found that using a mutual information value greater than 8 as a threshold could remove 5 redundant features and another 5 uncorrelated features, ultimately preserving 20 features. Next, a perturbation classification model is built using a genetic algorithm-based multi-objective optimization framework. The algorithm parameters are set to 100 groups, 7 crossover rates and 01 mutation rates. The output of the model is the disturbance classification type. And starting an iteration process, and setting the maximum iteration number to be 1000. At iteration 100, a specific solution was observed with a classification accuracy of 94% and a fault location accuracy of 88% at the perturbation. However, in the same iteration, the disturbance classification accuracy of the other solution is 90%, but the fault location accuracy reaches 93%. By evaluating these solutions and comparing their performance on two targets, a more optimal solution can be selected. This selection process is based on the concept of Pareto fronts, i.e. none of the solutions is better than the other solutions on all targets. During the iteration, the population gradually converges and after the 800 th iteration, the improvement is observed to become slow, where the disturbance classification accuracy of the optimal solution is improved to 96% while the fault location accuracy remains at 92%. This solution is considered to be an optimal compromise in the Pareto frontier currently found. By analyzing the entire Pareto front solution set, a compromise between disturbance classification accuracy and fault location accuracy can be obtained. It may be found that up to 98% classification accuracy may be combined with 85% fault location accuracy, while a relatively lower 91% classification accuracy may be combined with up to 95% fault location accuracy.
Optionally, in step S181, filtering the feature matrix, and determining the first feature includes performing the following steps:
step S1811, screening the feature matrix by adopting a chi-square test method to obtain a second feature, wherein the second feature is related to the fault position;
In step S1812, the second feature is subjected to redundancy processing using the mutual information, so as to obtain the first feature.
For example, the obtained feature matrix is screened out features positively or negatively correlated with the fault location. And carrying out quantitative evaluation on the correlation between each feature and the fault position through chi-square test to obtain a correlation index between the feature and the fault position, namely a second feature, wherein the second feature is related to the fault position. If the statistical test shows that a feature has positive correlation or negative correlation with the fault position, the mutual information calculation is utilized to judge the information sharing degree of the feature and the fault position. And taking a correlation index between the features and the fault positions as input, adopting a random forest model, evaluating the contribution degree of each feature to the fault diagnosis model, and screening out the features with the greatest influence on the prediction capability of the model. And sorting the screened features according to the contribution degree of the features to the fault location identification capability through feature importance ranking, and determining the features positively correlated or negatively correlated with the fault location after screening. Meanwhile, based on the characteristics positively or negatively related to the fault position after screening, the redundant and uncorrelated characteristics in the second characteristics are removed by adopting mutual information, so that the first characteristics are obtained.
Illustratively, for a set of sensor data, the original value is in the range of 0 to 1000, which can be normalized to the range of 0 to 1 by a min-max normalization method, calculated as (x-min)/(max-min). One particular sensor reading is 500, and the normalized value will be (500-0)/(1000-0) =5. A list of features and fault locations, wherein the observed and expected values for one feature are as follows, respectively: [20,30,50], expected values: [25,25,50] the chi-square statistic is Σ ((observed value-expected value) 2/expected value), by which the following calculation can be obtained: (20-25) 2/25+(30-25)2/25+(50-50)2/50=1+1+0=2; this value can be used to determine if the correlation of the feature to the fault location is significant. The mutual information of the same feature and the fault location is 8, which indicates that a great amount of information is shared between the feature and the fault location, and is a better feature. A list of feature importance may be obtained after training the random forest model by inputting a list of features and fault locations, [3,15,55]. The list shows that the third feature contributes most to the model predictive power. After several iterations, a subset of features [ feature 1, feature 3] is obtained, which means that both features are most useful for fault location identification, applying a recursive feature elimination technique. According to the importance of features provided by random forests, it may be decided to preserve feature 3, which has the highest contribution 55, while other features may be discarded or be preserved with a lower priority.
Optionally, the fault locating method of the power distribution network further comprises the following steps:
Step S191, obtaining a characteristic sample, topology information of a power distribution network and load distribution data of the power distribution network, wherein the characteristic sample is related to a fault position;
And step S192, training to obtain a fault positioning model based on a convolutional neural network algorithm, the characteristic samples, the topology information and the load distribution data.
By way of example, acquiring sample data may be understood acquiring features that are positively or negatively correlated with the location of the fault, as well as acquiring grid topology information and load distribution data, determining whether the disturbance is caused by an internal fault or an external factor. The power grid topology information refers to connection relations and layout conditions among various power equipment (such as transformer substations, transmission lines, distribution equipment and the like) in the power system, and comprises information such as connection modes among nodes, lengths and capacities of the lines, types and positions of the equipment and the like. The load distribution data refer to the power consumption condition of each node or region in the power system, and include information such as load size, load type, load change rule and the like. And then, the characteristics, the power grid topology information and the load distribution data which are positively or negatively related to the fault position are taken as input, a fault positioning model is constructed by utilizing a convolutional neural network algorithm, the fault position is taken as output, and the fault positioning model is obtained through training.
Illustratively, the voltage sensor displays a voltage value of 220V, the current sensor displays a current of 10A, a frequency of 50Hz, and a phase angle of 30. Time-frequency analysis is performed on the acquired data using a Fast Fourier Transform (FFT). FFT analysis shows that within 60 seconds, there is a prominent harmonic component in the frequency component, which has a frequency of 150Hz. The amplitude of the analyzed power quality disturbance was 100V, the frequency was 150Hz, and the duration was 5 seconds, classified as a transient disturbance. Line a is observed to connect substations S1 and S2, while line B connects S2 and S3. At the same time, load profile data was also obtained, and it was found that the load at the S2 station suddenly increased by 1500kW (kW) when a fault occurred. And taking disturbance characteristic parameters, power grid topology information and load distribution data as input, and constructing a fault location model by using a convolutional neural network algorithm. By training, the model can identify the disturbance mode and take the fault position as output. In one test case, the model correctly identified the fault location on line a, approximately 3 km from substation S1. If the fault point is initially located on a section of the line a, further acquiring monitoring data of the area. The current measured at a series of monitoring points on line a drops sharply at the moment of failure, the current recorded at the first monitoring point drops from 1000A (amperes) to 200A, the second monitoring point drops from 900A to 300A, and the current at the third monitoring point drops from 800A to 400A. By detailed analysis of the monitoring data, fault location results can be refined, and finally, the fault point is determined to be located between the first monitoring point and the second monitoring point and is located at a distance of about 5 km from the first monitoring point. This accurate fault location information facilitates quick and efficient fault repair operations.
Optionally, the fault locating method of the power distribution network further includes: and adjusting the target fault position based on the topology information of the power distribution network to obtain an adjusted target fault position, wherein the position range of the adjusted target fault position is smaller than the position range of the target fault position before adjustment.
By way of example, the fault location result and the topology information of the power distribution network are combined, the target fault location is adjusted based on the topology information of the power distribution network, and the fault location is accurate, so that the interaction relation between the fault and the power distribution network is obtained. And acquiring structural data of the power distribution network, and mapping geographic positions and connection relations of the transformer, the switch, the circuit and the bus element through a geographic information system to obtain topology information of the power distribution network. And acquiring fault indication information and abnormal change data of current, voltage and frequency. And analyzing a preliminary region where the fault occurs through the fault indication information, comparing the preliminary region with the topology information of the power distribution network, and narrowing the range of potential fault positions, namely, the position range of the adjusted target fault position is smaller than the position range of the target fault position before adjustment. And judging the fault type, including short circuit or grounding, by combining the fault indication information with the power distribution network, and determining the fault position. And monitoring fault position data, integrating the fault position data monitored in real time with time stamp information, and identifying the specific position of the fault by comparing the data change before and after the fault occurs to obtain the adjusted target fault position. And analyzing the historical data and the fault records, and determining the characteristics of the current fault by comparing the past fault conditions. And recovering the network state when the fault occurs according to the operation record of the power distribution network. If the operation record shows that the switch operation is displayed, judging whether the switch operation is related to the fault or not. If the analysis shows that the fault rate is improved when the extreme weather occurs, the influence of weather factors on the fault position is estimated by combining the current weather condition.
The distribution network structure data are obtained, and the geographic position and the connection relation of the elements are mapped through a geographic information system. The geographic information system shows transformer T1 at coordinates (312345, -1254321) and switch S1 at coordinates (312375, -1254355), with line L1 connecting the two elements. And collecting fault indication information and analyzing a primary area where the fault occurs. The current monitor records a sudden increase in current from 10A to 150A at transformer T1 while the voltage drops from 220V to 180V, suggesting that a fault may occur in the region of transformer T1. And analyzing the action information of the protection device to further determine the fault position. The overcurrent protection device of the power distribution network is triggered at a time stamp 13:05:22, and a short circuit occurs on a display line L1. And integrating the real-time monitoring data with the time stamp to identify the specific position of the fault point. The voltage of the monitored data display area A of 13:05:20 was 220V, while the data display voltage of 13:05:24 was reduced to 180V, indicating that the fault occurred within this time window. And analyzing the historical data and fault records, and predicting possible fault reasons. Historical data shows that line L1 has failed over three times in the past six months, pointing to the potential cause of insulation aging. And recovering the network state when the fault occurs according to the operation record of the power distribution network. The operation record shows that the switch S1 is switched on within five minutes before the fault, and an overload may be generated thereby to cause a short circuit. The influence of weather and environmental information on the fault location is evaluated in consideration of the weather and the environmental information. If the weather station records show that storm wind exists on the day of the fault, the wind speed reaches 15m/s, and the line breakage caused by wind power can be considered as a fault reason.
Optionally, the fault locating method of the power distribution network further includes: evaluating the fault positioning model based on the target fault position and the actual fault position to obtain an evaluation result; and updating the fault location model based on the evaluation result.
Illustratively, the fault location effect is evaluated, the fault location model is optimized according to the location effect evaluation result, and meanwhile, new power quality disturbance data are periodically used for learning and adjusting the power data analysis model so as to adapt to the evolution of the power grid and the newly-appearing disturbance type. When the positioning effect is evaluated, a quantitative evaluation index is obtained by comparing the positioning result with the actual fault location, wherein the quantitative evaluation index comprises an accuracy rate and a recall rate and is compared with expert opinion. If the positioning effect does not meet the expectation, adjusting a fault positioning model according to the evaluation result to optimize the fault positioning precision; by using the newly collected power quality disturbance data for retraining of the power data analysis model, it is ensured that the model can adapt to the evolution of the power grid and the newly appearing disturbance characteristics. And identifying an unknown disturbance mode, incorporating the new disturbance type into a disturbance library, and updating the disturbance type library to improve the response capability to the new disturbance. And displaying the positioning result to an operator in a visual mode through a user interface display platform, and generating a disturbance event report through the report for a decision maker to analyze and refer to.
Illustratively, the statistical fault location model correctly locates 270 out of 300 historical disturbance events, reaching 90% accuracy; the fault locating model achieves 94% accuracy and 282 events are correctly located. When the positioning effect is evaluated, the positioning accuracy of the machine learning algorithm is found to be 94% and the recall is found to be 91% by comparing the fault positioning result of the fault positioning model with the actual fault location, which means that most of the fault positions are correctly identified and most of the actual faults are detected. The evaluation result is consistent with the opinion of the power grid expert, and the expert agrees that the reliability and stability of the algorithm are higher. If the positioning effect does not meet the expectation, if a specific feature such as frequency change trend is high in weight in the model, but the feature is found to contribute little to the positioning accuracy in the evaluation, the accuracy is improved by only 2%. Then it may be considered to reject this feature or introduce new features, such as maximum load increase rate over time, to optimize fault location accuracy. The newly collected power quality disturbance data are continuously used for retraining the power data analysis model, so that the adaptability of the fault location model is improved. As a new pattern of disturbances is identified in the month of february, which occurs during the afternoon load surge period of the day and is caused by a new start-up. The new disturbance mode is added into a disturbance type library, and after the fault location model is retrained, the accuracy of identifying the disturbance of the type is improved from the initial 70% to 95%. Through the data interface, this information can be received within 30 seconds after the real-time positioning result appears, and the response flow is started. At the same time, the data stream and analysis results can be updated once per minute to ensure that the operator obtains up-to-date information. Through the user interface display platform, an operator can intuitively see a real-time chart of voltage fluctuation and harmonic content, and automatically generate reports containing key parameters and analysis results after each disturbance event, for example, one report shows that in power quality disturbance of 14 days of 2 months, the standard deviation of the voltage fluctuation is 8%, and the third harmonic current accounts for 25% of the fundamental current. The symmetric encryption algorithm-256 bits are adopted to encrypt the data in transmission, and access control based on roles is set so as to ensure that only authorized personnel can access the sensitive electric energy data.
In summary, in the embodiment of the present application, the target electric energy data of the power distribution network is obtained, where the target electric energy data is data of electric energy transmitted, distributed and used in the power distribution system; analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network; determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing the power quality disturbance features of different types; and identifying the feature matrix by adopting a fault positioning model to obtain the target fault position. Therefore, the technical effects of efficiently processing the electric energy quality problem and positioning the power distribution network faults are achieved, the operation efficiency and reliability of the power distribution network are improved, an efficient method for processing the electric energy quality problem and positioning the power distribution network faults is provided for power grid operation and maintenance personnel, and the technical problems that the correlation between the data characteristics and the specific positions of the faults in the related technology is not strong and the fault positioning is inaccurate are solved.
Example 2
According to an embodiment of the present application, there is also provided a fault locating device for a power distribution network for implementing the method in embodiment 1. Fig. 2 is a block diagram of a fault location device of an electric distribution network according to an embodiment of the present application, and as shown in fig. 2, the fault location device 20 of an electric distribution network includes: the acquiring module 22 is configured to acquire target electric energy data of the power distribution network, where the target electric energy data is data of transmitting, distributing and using electric energy in the power distribution system; the analysis module 24 is configured to analyze the target electrical energy data by using an electrical energy data analysis model, so as to obtain an electrical energy quality disturbance feature, where the electrical energy quality disturbance feature is used for reflecting a stability degree of electrical energy quality of the power distribution network; a determining module 26, configured to determine a feature matrix based on the power quality disturbance characteristics, where the feature matrix is used to divide the power quality disturbance characteristics of different types; and the identification module 28 is used for identifying the feature matrix by adopting the fault location model to obtain the target fault position.
Optionally, the obtaining module 22 is further configured to obtain initial electrical energy data of the power distribution network; and carrying out data preprocessing on the initial electric energy data to obtain target electric energy data.
Optionally, the obtaining module 22 is further configured to compare the timestamp difference of the initial power data based on a preset collection interval, and determine a missing value in the initial power data; filling the missing value in the initial electric energy data to obtain a first data sequence; determining an abnormal value in the first data sequence by adopting a standard deviation method, and processing the abnormal value by adopting a moving average method to obtain a second data sequence; and carrying out time zone calibration on the second data sequence, and carrying out unit conversion on the second data sequence after time zone calibration to obtain target electric energy data.
Optionally, the apparatus further comprises: the analysis model module is used for acquiring an electric energy data sample of the power distribution network, wherein the electric energy data sample is a data sample for transmitting, distributing and using electric energy in a power distribution system; determining a power quality characteristic based on the power data sample, wherein the power quality characteristic is used for representing the power quality of the power distribution network; and obtaining an electric energy data analysis model based on the decision tree algorithm and the quality feature training, wherein the electric energy data analysis model is used for identifying the electric energy quality disturbance feature.
Optionally, the determining module 26 is further configured to perform time-frequency analysis on the power quality disturbance characteristic, and determine a frequency spectrum characteristic and instantaneous frequency data; extracting preset characteristics of the power quality disturbance characteristics, and determining amplitude characteristics and a conversion process of the voltage under preset conditions; adopting a Hilbert-Huang transformation algorithm to analyze the disturbance characteristics of the electric energy quality and determining the phase angle change and the flicker degree; analyzing the power quality disturbance characteristics by adopting a harmonic analysis method, and determining the harmonic distortion degree in the power distribution system; analyzing time sequence information of the power quality disturbance characteristics, and determining disturbance duration; and determining disturbance information in response to the power quality disturbance characteristics meeting a preset analysis condition, wherein the disturbance information comprises at least one of the following: dynamic characteristics of power supply frequency variation, phase difference, frequency and amplitude of inter-harmonic or frequency and amplitude of subharmonic component; a feature matrix is constructed based on the spectral features, instantaneous frequency data, amplitude features, conversion process, phase angle change, flicker level, harmonic distortion level, disturbance duration and disturbance information.
Optionally, the identification module 28 is further configured to screen the feature matrix to determine a first feature, where the first feature is related to the fault location; processing the first feature based on a disturbance classification model to obtain a target disturbance classification type, wherein the disturbance classification model is obtained by training based on a multi-target optimization algorithm; and processing the target disturbance classification type by adopting a fault positioning model to obtain a target fault position.
Optionally, the identification module 28 is further configured to screen the feature matrix by using a chi-square test method to obtain a second feature, where the second feature is related to the fault location; and performing redundancy processing on the second feature by adopting the mutual information to obtain the first feature.
Optionally, the apparatus further comprises: the training module is used for acquiring a characteristic sample, topology information of the power distribution network and load distribution data of the power distribution network, wherein the characteristic sample is related to a fault position; and training based on a convolutional neural network algorithm, a characteristic sample, topology information and load distribution data to obtain a fault positioning model.
Optionally, the apparatus further comprises: the adjusting module is used for adjusting the target fault position based on the topology information of the power distribution network to obtain an adjusted target fault position, wherein the position range of the adjusted target fault position is smaller than the position range of the target fault position before adjustment.
Optionally, the apparatus further comprises: the updating module is used for evaluating the fault positioning model based on the target fault position and the actual fault position to obtain an evaluation result; and updating the fault location model based on the evaluation result.
It should be noted that, each module in the fault locating device of the power distribution network in the embodiment of the present application corresponds to each implementation step of the fault locating method of the power distribution network in embodiment 1 one by one, and since the detailed description has been already made in embodiment 1, part of details not shown in this embodiment may refer to embodiment 1, and will not be repeated here.
Example 3
According to an embodiment of the present application, there is further provided a nonvolatile storage medium, where the nonvolatile storage medium includes a stored computer program, and a device where the nonvolatile storage medium is located executes the fault locating method of the power distribution network in embodiment 1 by running the computer program.
The non-volatile storage medium is embodied in a device that executes the computer program to perform the steps of:
Step S12, acquiring target electric energy data of the power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system;
s14, analyzing target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network;
Step S16, determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing different types of power quality disturbance features;
and S18, identifying the feature matrix by adopting a fault positioning model to obtain a target fault position.
According to an embodiment of the present application, there is also provided a computer program product, including a computer program, which when executed by a processor implements the fault locating method of the power distribution network in embodiment 1 of the present application.
Illustratively, the computer program execution when run performs the steps of:
Step S12, acquiring target electric energy data of the power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system;
s14, analyzing target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network;
Step S16, determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing different types of power quality disturbance features;
and S18, identifying the feature matrix by adopting a fault positioning model to obtain a target fault position.
According to an embodiment of the present application, there is also provided an electronic apparatus including: a memory and a processor, wherein the memory stores a computer program, the processor configured to execute the fault location method of the power distribution network in embodiment 1 by the computer program.
Illustratively, the processor is configured to implement the following steps by computer program execution:
Step S12, acquiring target electric energy data of the power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system;
s14, analyzing target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network;
Step S16, determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing different types of power quality disturbance features;
and S18, identifying the feature matrix by adopting a fault positioning model to obtain a target fault position.
The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the related art or all or part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application, which are intended to be comprehended within the scope of the present application.
Claims (14)
1. A method for locating faults in a power distribution network, the method comprising:
Acquiring target electric energy data of a power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in a power distribution system;
Analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network;
determining a feature matrix based on the power quality disturbance features, wherein the feature matrix is used for dividing the power quality disturbance features of different types;
And identifying the feature matrix by adopting a fault positioning model to obtain a target fault position.
2. The method of claim 1, wherein the obtaining target electrical energy data for the power distribution network comprises:
acquiring initial electric energy data of the power distribution network;
And carrying out data preprocessing on the initial electric energy data to obtain the target electric energy data.
3. The method of claim 2, wherein the performing data preprocessing on the initial power data to obtain the target power data comprises:
Comparing the timestamp difference value of the initial electric energy data based on a preset acquisition interval, and determining a missing value in the initial electric energy data;
filling the missing value in the initial electric energy data to obtain a first data sequence;
Determining an abnormal value in the first data sequence by adopting a standard deviation method, and processing the abnormal value by adopting a moving average method to obtain a second data sequence;
And carrying out time zone calibration on the second data sequence, and carrying out unit conversion on the second data sequence after time zone calibration to obtain the target electric energy data.
4. The method according to claim 1, wherein the method further comprises:
Acquiring an electric energy data sample of a power distribution network, wherein the electric energy data sample is a data sample for transmitting, distributing and using electric energy in a power distribution system;
determining a power quality feature based on the power data sample, wherein the power quality feature is used for representing the power quality of the power distribution network;
And training based on a decision tree algorithm and the quality features to obtain an electric energy data analysis model, wherein the electric energy data analysis model is used for identifying electric energy quality disturbance features.
5. The method of claim 1, wherein the determining a feature matrix based on the power quality disturbance characteristics comprises:
performing time-frequency analysis on the power quality disturbance characteristics to determine frequency spectrum characteristics and instantaneous frequency data;
Extracting preset characteristics of the power quality disturbance characteristics, and determining amplitude characteristics and a conversion process of the voltage under preset conditions;
analyzing the power quality disturbance characteristics by adopting a Hilbert-Huang transformation algorithm, and determining the phase angle change and the flicker degree;
Analyzing the power quality disturbance characteristics by adopting a harmonic analysis method, and determining the harmonic distortion degree in the power distribution system;
analyzing the time sequence information of the power quality disturbance characteristics to determine the disturbance duration;
And determining disturbance information in response to the power quality disturbance characteristic meeting a preset analysis condition, wherein the disturbance information comprises at least one of the following: dynamic characteristics of power supply frequency variation, phase difference, frequency and amplitude of inter-harmonic or frequency and amplitude of subharmonic component;
The feature matrix is constructed based on the spectral features, the instantaneous frequency data, the amplitude features, the conversion process, the phase angle variation, the flicker level, the harmonic distortion level, the disturbance duration, and the disturbance information.
6. The method of claim 1, wherein identifying the feature matrix using a fault localization model comprises:
screening the feature matrix to determine a first feature, wherein the first feature is related to a fault position;
processing the first feature based on a disturbance classification model to obtain a target disturbance classification type, wherein the disturbance classification model is obtained based on training of a multi-target optimization algorithm;
And processing the target disturbance classification type by adopting the fault positioning model to obtain the target fault position.
7. The method of claim 6, wherein the screening the feature matrix to determine the first feature comprises:
Screening the feature matrix by adopting a chi-square test method to obtain a second feature, wherein the second feature is related to the fault position;
and performing redundancy processing on the second characteristic by adopting mutual information to obtain the first characteristic.
8. The method according to claim 1, wherein the method further comprises:
Acquiring a characteristic sample, topology information of the power distribution network and load distribution data of the power distribution network, wherein the characteristic sample is related to a fault position;
And training to obtain the fault positioning model based on a convolutional neural network algorithm, the characteristic samples, the topology information and the load distribution data.
9. The method according to any one of claims 1-8, further comprising:
and adjusting the target fault position based on the topology information of the power distribution network to obtain an adjusted target fault position, wherein the position range of the adjusted target fault position is smaller than the position range of the target fault position before adjustment.
10. The method according to any one of claims 1-8, further comprising:
Evaluating the fault positioning model based on the target fault position and the actual fault position to obtain an evaluation result;
and updating the fault location model based on the evaluation result.
11. A fault locating device for an electrical distribution network, the device comprising:
The power distribution system comprises an acquisition module, a power distribution module and a power distribution module, wherein the acquisition module is used for acquiring target electric energy data of a power distribution network, wherein the target electric energy data are data for transmitting, distributing and using electric energy in the power distribution system;
The analysis module is used for analyzing the target electric energy data by adopting an electric energy data analysis model to obtain electric energy quality disturbance characteristics, wherein the electric energy quality disturbance characteristics are used for reflecting the stability degree of the electric energy quality of the power distribution network;
The determining module is used for determining a characteristic matrix based on the power quality disturbance characteristics, wherein the characteristic matrix is used for dividing different types of power quality disturbance characteristics;
and the identification module is used for identifying the feature matrix by adopting a fault positioning model to obtain a target fault position.
12. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program, wherein the computer program is arranged to perform the fault localization method of the power distribution network according to any of the preceding claims 1 to 10 when run on a computer or processor.
13. A computer program product comprising a computer program which, when executed by a processor, implements the fault localization method of the power distribution network according to any one of claims 1to 10.
14. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the fault localization method of the power distribution network as claimed in any of the preceding claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410267993.2A CN117970033A (en) | 2024-03-08 | 2024-03-08 | Fault positioning method and device for power distribution network, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410267993.2A CN117970033A (en) | 2024-03-08 | 2024-03-08 | Fault positioning method and device for power distribution network, storage medium and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117970033A true CN117970033A (en) | 2024-05-03 |
Family
ID=90862639
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410267993.2A Pending CN117970033A (en) | 2024-03-08 | 2024-03-08 | Fault positioning method and device for power distribution network, storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117970033A (en) |
-
2024
- 2024-03-08 CN CN202410267993.2A patent/CN117970033A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10324132B2 (en) | Systems and methods for power line event zone identification | |
| Lin et al. | Grey clustering analysis for incipient fault diagnosis in oil-immersed transformers | |
| JP2000512766A (en) | Statistical pattern analysis method for partial discharge measurement in high voltage insulation | |
| US10802057B2 (en) | Systems and methods for monitoring a power system | |
| US9336491B2 (en) | Identification of power system events using fuzzy logic | |
| Jia et al. | Defect prediction of relay protection systems based on LSSVM-BNDT | |
| Jiang et al. | Automated distribution network fault cause identification with advanced similarity metrics | |
| Dey et al. | Solar farm voltage anomaly detection using high-resolution μPMU data-driven unsupervised machine learning | |
| CN110647924A (en) | GIS equipment state evaluation method based on support vector description and K-nearest neighbor algorithm | |
| Dong | Combining unsupervised and supervised learning for asset class failure prediction in power systems | |
| CN117349624A (en) | Electric power energy monitoring method, system, terminal equipment and storage medium | |
| CN111044100A (en) | Sensor device for electric power metering and control method | |
| CN117970033A (en) | Fault positioning method and device for power distribution network, storage medium and electronic device | |
| Radhakrishnan et al. | Quality assessment of smart grid data | |
| Gorjani et al. | Application of optimized deterministic methods in long-term power quality | |
| de Oliveira et al. | Voltage dips caused by faults in a transmission system: a monitoring case study of a sensitive industrial consumer | |
| Sarquis Filho et al. | Practical recommendations for the design of automatic fault detection algorithms based on experiments with field monitoring data | |
| Ajayi | Explainable AI (XAI) for Fault Detection and Classification in Microgrids Using a Real-Time Simulation Framework | |
| Galbau et al. | Characterization of the power quality in the electric distribution networks using data mining with K-means clustering | |
| Amoateng et al. | An Intelligent Event Detection Framework To Improve Situational Awareness In PMU Power Distribution Networks | |
| Men et al. | Study of the Bus Voltage Magnitude to Monitor Power Quality in the Distribution System | |
| Wang et al. | An Efficient Power System Risk Identification Method Considering PMU Bad Data | |
| Jiang et al. | Automated fault analysis and diagnosis using high-frequency and maintenance data from distribution networks | |
| Han et al. | A State Assessment Algorithm of Power Cable Based on Association Rules Data Mining Technology | |
| Li et al. | Switchgear Insulation Condition Assessment with Multi-Dimensional Feature Database Based on an Optimal Grade Clustering Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication |