CN117420345B - Power grid operation abnormal state monitoring system based on data driving - Google Patents
Power grid operation abnormal state monitoring system based on data driving Download PDFInfo
- Publication number
- CN117420345B CN117420345B CN202311733726.1A CN202311733726A CN117420345B CN 117420345 B CN117420345 B CN 117420345B CN 202311733726 A CN202311733726 A CN 202311733726A CN 117420345 B CN117420345 B CN 117420345B
- Authority
- CN
- China
- Prior art keywords
- current data
- power grid
- data set
- grid current
- current
- 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.)
- Active
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 69
- 238000012544 monitoring process Methods 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 55
- 238000012216 screening Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012550 audit Methods 0.000 claims description 2
- 230000005856 abnormality Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/10—Measuring sum, difference or ratio
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Power Engineering (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the technical field of data processing for power supply circuits, in particular to a power grid operation abnormal state monitoring system based on data driving, which comprises the following components: the system comprises a current data and voltage data acquisition module, an expandable power grid current data set and unexpanded power grid current data set acquisition module, an addition feasibility acquisition module and an abnormal current data monitoring module; collecting current data and voltage data; obtaining a power grid current data set, a current difference amount and a voltage difference amount according to the current data and the voltage data; acquiring the set extensibility of a power grid current data set; obtaining current-voltage affinity according to the current difference and the voltage difference; obtaining addition feasibility according to the current voltage affinity; and obtaining a monitorable power grid current data set according to the adding feasibility and the set expandability degree, and performing anomaly monitoring. The invention reduces the interference of zero drift on current data and improves the accuracy of abnormal monitoring results.
Description
Technical Field
The invention relates to the technical field of data processing for power supply circuits, in particular to a power grid operation abnormal state monitoring system based on data driving.
Background
The condition that the electric network possibly has current mutation in the process of supplying electric energy, the current mutation can lead to the circuit and the power equipment to receive extra stress impact, can lead to the steady state of the electric network to produce certain influence, in order to guarantee the safe operation of the electric network, the abnormal state monitoring is needed to be carried out on the current data in the operation process of the electric network.
The traditional method generally utilizes a Hall current sensor to collect current data of a power grid, and then utilizes an isolated forest algorithm to monitor the abnormal state of the collected current data; however, because the zero drift phenomenon exists in the current data collected by the Hall current sensor, the originally collected current data has certain offset, and the detected abnormal data is not necessarily the actual abnormal current data.
Disclosure of Invention
The invention provides a power grid operation abnormal state monitoring system based on data driving, which aims to solve the existing problems: the phenomenon of zero drift exists in the current data collected by the Hall current sensor, so that the originally collected current data has certain offset, and the monitored abnormal data is not necessarily the real abnormal current data.
The power grid operation abnormal state monitoring system based on data driving adopts the following technical scheme:
the method comprises the following modules:
the system comprises a current data and voltage data acquisition module, a power supply module and a power supply module, wherein the current data and voltage data acquisition module is used for acquiring a plurality of current data of a power grid and voltage data corresponding to each current data;
the system comprises an expandable power grid current data set and a non-expandable power grid current data set acquisition module, wherein the expandable power grid current data set and the non-expandable power grid current data set acquisition module are used for recording a sequence of all current data sequenced according to the sequence from the early to the late of the acquisition time as a power grid current data sequence; dividing a power grid current data sequence into a plurality of power grid current data sets; obtaining the set extensibility of each power grid current data set according to the duty ratio difference between the current data in the power grid current data sets; screening a plurality of expandable power grid current data sets and a plurality of unexpanded power grid current data sets from the power grid current data sets according to the degree of expansion of the set;
the adding feasibility obtaining module is used for obtaining the current difference quantity and the voltage difference quantity of each current data according to the difference between adjacent current data and the difference between corresponding voltage data in the power grid current data sequence; obtaining the current voltage affinity of each current data according to the comparison difference between the current difference and the voltage difference; screening a plurality of current data to be audited from the current data on two sides of the current data set of the expandable power grid; obtaining the final adding necessity of each current data to be audited according to the difference of the current voltage affinities between the current data to be audited and the difference between the current data in the current data set of the scalable power grid; obtaining the adding feasibility of each piece of to-be-examined nuclear power flow data according to the final adding necessity and the set expandability, wherein the adding feasibility is used for describing the probability that the current data belong to abnormal current data;
the abnormal current data monitoring module is used for adding the to-be-inspected nuclear power flow data into the expandable power grid current data set according to the adding feasibility and the inextensible power grid current data set to obtain a plurality of monitorable power grid current data sets, and carrying out abnormal monitoring on the monitorable power grid current data sets.
Preferably, the dividing the grid current data sequence into a plurality of grid current data sets includes the following specific steps:
presetting a current data quantity T1; and (3) carrying out simple random sampling on the power grid current data sequence to obtain T1 current data, taking the T1 current data as demarcation points, and recording a set formed by all current data between two adjacent demarcation points as a power grid current data set.
Preferably, the method for obtaining the set extensibility degree of each power grid current data set by the duty ratio difference between the current data in the power grid current data sets includes the following specific steps:
for any one power grid current data set, the median of all current data in the power grid current data set is recorded as a first median, the current data which are larger than the first median in the power grid current data set are recorded as high current data, and the current data which are smaller than the first median in the power grid current data set are recorded as low current data; obtaining the expansibility of the power grid current data set according to all high current data and all low current data in the power grid current data set; the method for calculating the scalability degree of the power grid current data set comprises the following steps:
in the method, in the process of the invention,representing the degree of scalability of the grid current dataset; />Representing the average value of all high-current data in the power grid current data set; />Representing the average value of all low current data in the power grid current data set; />Representing preset super parameters; />Representing the number of all current data in the grid current data set; />Representing the number of all grid current datasets; />Indicate->The number of all current data in the individual grid current data sets; />The representation takes absolute value;
and acquiring the expansibility of all the power grid current data sets, carrying out linear normalization processing on all the expansibility, and recording each normalized expansibility as a set expansibility.
Preferably, the screening of the plurality of expandable power grid current data sets and the plurality of unexpanded power grid current data sets from the power grid current data sets according to the set expandability degree includes the following specific methods:
presetting a set extensible degree threshold T2, and recording a power grid current data set with the set extensible degree larger than T2 as an extensible power grid current data set; the grid current data set with the set extensibility less than or equal to T2 is noted as a non-extensible grid current data set.
Preferably, the method for obtaining the current difference and the voltage difference of each current data from the difference between adjacent current data and the difference between corresponding voltage data in the power grid current data sequence includes the following specific steps:
for any one current data in the power grid current data sequence, recording the previous current data of the current data in the power grid current data sequence as the front current data of the current data; recording the absolute value of the difference value between the current data and the front current data of the current data as the current difference value of the current data; and recording the absolute value of the difference value between the voltage data corresponding to the current data and the voltage data corresponding to the front current data as the voltage difference value of the current data.
Preferably, the current-voltage affinity of each current data is obtained according to the comparison difference between the current difference and the voltage difference, which comprises the following specific steps:
in the method, in the process of the invention,the current voltage affinity of any one current data is represented; />A voltage difference amount representing current data; />A current difference amount representing current data; />Representing preset super parameters; />Representing the average value of the voltage difference amounts of all the current data; />The average value of the current difference amounts of all the current data is represented.
Preferably, the method for screening a plurality of current data to be audited from the current data on two sides of the expandable power grid current data set includes the following specific steps:
and presetting a current data quantity T3 for any one expandable power grid current data set, and recording the front T3 current data of the expandable power grid current data set and the rear T3 current data of the expandable power grid current data set in a power grid current data sequence as current data to be audited of the expandable power grid current data set.
Preferably, the final adding necessity of each current data to be audited is obtained according to the difference of the current voltage affinities between the current data to be audited and the difference between the current data in the current data set of the scalable power grid, and the specific method comprises the following steps:
for any one expandable power grid current data set, sequentially taking the maximum value, the minimum value, the median and the mode of all current data in the expandable power grid current data set as reference characteristic current data in the expandable power grid current data set; for any one auditing current data, obtaining the adding necessity of the current data to be audited according to the reference characteristic current data of the expandable power grid current data set and the current voltage affinities of all the current data to be audited; the calculation method for the adding necessity of the current data to be audited comprises the following steps:
in the method, in the process of the invention,indicating the necessity of adding current data to be audited; />Representing the number of all reference characteristic current data within the scalable grid current dataset; />The current voltage affinity of the current data to be audited is represented; />Representing +.>Current voltage affinities of the individual reference characteristic current data; />Representing +.>Reference characteristic current data; />Representing the mean value of all reference characteristic current data in the expandable grid current data set; />Representing preset super parameters; />Representing the number of all current data within the scalable grid current dataset; />Representing current data to be audited; />Representing +.>-individual current data; />The representation takes absolute value;
obtaining the adding necessity of all current data to be audited of the expandable power grid current data set, carrying out linear normalization on all the adding necessity, and recording the adding necessity after normalization as the final adding necessity.
Preferably, the obtaining the adding feasibility of each piece of pending nuclear power stream data according to the final adding necessity and the aggregate scalability degree includes the following specific methods:
in the method, in the process of the invention,representing the addition feasibility of any one audit current data of any one extensible power grid current data set; />Representing a set scalability of the scalable grid current dataset; />Indicating the final added necessity of the current data to be audited.
Preferably, according to the adding feasibility and the inextensible power grid current data set, adding the to-be-inspected nuclear power flow data into the extensible power grid current data set to obtain a plurality of monitorable power grid current data sets, and performing anomaly monitoring on the monitorable power grid current data sets, including the specific method:
for any one expandable power grid current data set, presetting an adding feasibility threshold T4, recording current data to be audited with adding feasibility larger than T4 as current data to be combined in all current data to be audited of the expandable power grid current data set, recording the expandable power grid current data set as an initial monitorable power grid current data set after adding all current data to be combined of the expandable power grid current data set into the expandable power grid current data set, and acquiring all initial monitorable power grid current data sets;
all the initial monitorable grid current data sets and the unexpanded grid current data sets are recorded as one monitorable grid current data set; constructing an isolated tree for each monitorable power grid current data set, and acquiring an abnormal score of each current data through the isolated tree; if a plurality of abnormal scores exist in one piece of current data, taking the average value of all abnormal scores of the current data as the final abnormal score of the current data;
an abnormal scoring threshold T5 is preset, and current data with abnormal scores greater than T5 are recorded as abnormal current data.
The technical scheme of the invention has the beneficial effects that: obtaining a set extensible degree according to the current data, obtaining an extensible power grid current data set and a non-extensible power grid current data set according to the set extensible degree, obtaining current voltage affinities of the current data according to the current data and the voltage data, obtaining final addition necessity of the current data to be audited according to the current voltage affinities, obtaining addition feasibility of the current data to be audited according to the final addition necessity and the set extensible degree, and carrying out anomaly detection according to the addition feasibility; the set expandability of the invention reflects the relevance among all current data in the power grid current data set after initial judgment, the current voltage affinity reflects the relevance of the current data and the voltage data corresponding to the current data, the final addition necessity reflects the degree to which the current data to be audited can be added into the expandable power grid current data set, and the addition feasibility reflects the accuracy of abnormal monitoring; the interference of zero drift on current data is reduced, and the accuracy of abnormal monitoring results is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of a system for monitoring abnormal operation of a power grid based on data driving.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the power grid abnormal operation state monitoring system based on data driving according to the invention with reference to the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the power grid abnormal operation state monitoring system based on data driving provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of a system for monitoring abnormal operation of a power grid based on data driving according to an embodiment of the present invention is shown, where the system includes the following modules:
the current data and voltage data acquisition module 101 acquires a plurality of current data of a power grid and voltage data corresponding to each current data.
It should be noted that, in the conventional method, the hall current sensor is generally used to collect current data of the power grid, and then the isolated forest algorithm is used to monitor the abnormal state of the collected current data; however, because the zero drift phenomenon exists in the current data collected by the Hall current sensor, the originally collected current data has certain offset, and the detected abnormal data is not necessarily the actual abnormal current data. For this reason, the embodiment proposes a system for monitoring abnormal operation state of a power grid based on data driving.
In order to realize the abnormal state monitoring system for power grid operation based on data driving, which is provided by the embodiment, current data and voltage data need to be collected at first, and the specific process is as follows: the Hall current sensor and the voltage sensor are connected to a lead of a power grid, and current data displayed on the Hall current sensor and voltage data displayed on the voltage sensor are collected once every 1 second, and the total collection time is one hour. The number of the current data is equal to the number of the voltage data, and each current data corresponds to one voltage data.
So far, a plurality of current data of the power grid and voltage data corresponding to each current data are obtained through the method.
An expandable grid current dataset and non-expandable grid current dataset acquisition module 102 for acquiring a grid current data sequence from the current data; dividing a power grid current data sequence into a plurality of power grid current data sets; obtaining the set extensibility of each power grid current data set according to the duty ratio difference between the current data in the power grid current data sets; and screening a plurality of expandable power grid current data sets and a plurality of unexpanded power grid current data sets from the power grid current data sets according to the degree of the expansion of the set.
It should be noted that, the phenomenon of zero drift existing in the current data collected by the hall current sensor means that the original real current data will fluctuate to a certain extent on the basis of the value thereof, so that the finally collected current data is not the real current data in the power grid; because of randomness of numerical value fluctuation of real current data to a certain extent, the difference between partial current data is larger, so that the traditional isolated forest algorithm can incorrectly identify partial original normal current data as abnormal current data.
It should be further noted that, in the conventional isolated forest algorithm, current data is completely and randomly divided into a plurality of sample sets, wherein a plurality of sample sets exist in the same current data, and abnormal current data is obtained by constructing an isolated tree for the sample sets; compared with real current data, the acquired current data has certain numerical fluctuation, the current data can continuously change with the passage of time to a certain extent, and certain change trends exist, so that the current data between different time points can have relevance of different degrees; for any one current data, if the correlation between the surrounding current data and the current data is weak, it is indicated that the current data does not conform to the trend of the surrounding current data, and the possibility of abnormality of the current data is high; in the conventional completely randomly acquired sample set, the relevance between the current data contained in part of the sample sets is lower, the accuracy of the finally acquired abnormal result is lower, and in order to improve the accuracy of the abnormal monitoring result, the current data is randomly divided into a plurality of continuous data sets, each data set contains a plurality of time continuous current data, and the abnormal current data is acquired by constructing an isolated tree through the data sets.
It should be further noted that, in the process of constructing the isolated tree according to the data sets, the data sets are randomly acquired according to time, and the correlations between the current data contained in different data sets are different; for any one data set, if the relevance among all current data in the data set is weaker, the situation that the relevance among local current data in the data set is stronger and more current data with relatively weaker relevance are included is indicated, and meanwhile, the accuracy of the abnormal monitoring result obtained through the data set is higher because only the current data with stronger relevance in the data set has larger influence on the abnormal monitoring result; if the correlation between all the current data in the data set is strong, it is indicated that there may be current data with strong correlation with the data set outside the data set, and in order to ensure the accuracy of the anomaly monitoring result, it is necessary to incorporate the current data with strong correlation with all the current data in the data set outside the data set into the data set as much as possible.
Specifically, a sequence obtained by sequencing all collected current data according to the sequence from the early to the late of the collection time is recorded as a power grid current data sequence; presetting a current data amount T1, wherein the present embodiment is described by taking t1=50 as an example, and the present embodiment is not particularly limited, wherein T1 may be determined according to the specific implementation situation; the method comprises the steps of performing simple random sampling on a power grid current data sequence to obtain T1 current data, taking the T1 current data as demarcation points, and recording a set formed by all current data between two adjacent demarcation points as a power grid current data set; taking any one power grid current data set as an example, recording the median of all current data in the power grid current data set as a first median, recording the current data which are larger than the first median in the power grid current data set as high current data, and recording the current data which are smaller than the first median in the power grid current data set as low current data; and obtaining the expandability of the power grid current data set according to all the high current data and all the low current data in the power grid current data set. Each power grid current data set contains a plurality of high current data and a plurality of low current data, and each power grid current data set does not contain simple random sampling of demarcation points, which is a known technique and will not be described in detail in this embodiment; in addition, the calculation method of the extensible degree of the power grid current data set comprises the following steps:
in the method, in the process of the invention,representing the degree of scalability of the grid current dataset; />Representing the average value of all high current data in the power grid current data set; />Representing the average of all low current data in the grid current dataset; />Representing a preset hyper-parameter, preset +.>For preventing denominator from being 0; />Representing the number of all current data in the grid current dataset; />Representing the number of all grid current datasets; />Indicate->The number of all current data in the individual grid current data sets; />The representation takes absolute value; />Representing a difference between high current data and low current data in the grid current dataset;representing the depth at which the grid current dataset builds an orphan tree. And if the extensible degree of the power grid current data set is larger, the difference among all current data in the power grid current data set is larger, and the correlation among all current data in the power grid current data set after initial judgment is reflected to be stronger. And acquiring the expansibility of all the power grid current data sets, carrying out linear normalization processing on all the expansibility, and recording each normalized expansibility as a set expansibility.
Further, a set of threshold values of scalability T2 is preset, where the present embodiment is described by taking t2=0.7 as an example, and the present embodiment is not specifically limited, where T2 may be determined according to the specific implementation situation; recording a power grid current data set with the set expansion degree larger than T2 as an expandable power grid current data set; recording a grid current data set with the set extensibility less than or equal to T2 as a non-extensible grid current data set; all scalable grid current datasets and all non-scalable grid current datasets are acquired.
So far, all the expandable power grid current data sets and all the unexpanded power grid current data sets are obtained through the method.
The adding feasibility obtaining module 103 obtains the current difference quantity and the voltage difference quantity of each current data by the difference between adjacent current data and the difference between corresponding voltage data in the power grid current data sequence; obtaining the current voltage affinity of each current data according to the comparison difference between the current difference and the voltage difference; screening a plurality of current data to be audited from the current data on two sides of the current data set of the expandable power grid; obtaining the final adding necessity of each current data to be audited according to the difference of the current voltage affinities between the current data to be audited and the difference between the current data in the current data set of the scalable power grid; and obtaining the adding feasibility of each piece of pending nuclear power stream data according to the final adding necessity and the set expandability.
It should be noted that, in the process of constructing the isolated tree according to the data set, the data set is randomly acquired according to time, and for any one data set, there may be current data with greater relevance to all current data in the data set; in order to ensure the accuracy of the abnormal monitoring result, the current data with larger relevance are required to be added into the data set; meanwhile, as the voltage data collected by the voltage sensor also has zero drift phenomenon to a certain extent, each current data has irrelevance to a certain extent with the voltage data; for any current data outside the data set, if the correlation between the current data and the corresponding voltage data is larger, the current data is closer to the real current data; meanwhile, if the numerical value difference between the current data and all the current data in the data set is larger, the current data is related to all the current data in the data set; therefore, the embodiment can obtain the current voltage affinity of the current data and the corresponding voltage data according to the relation between the current data and the corresponding voltage data, obtain the adding necessity of the current data according to the current voltage affinity and the relation between the current data and all the current data in the data set, and obtain the adding feasibility of the current data according to the adding necessity of the current data and the set extensibility of the data set so as to facilitate the subsequent analysis and processing.
Specifically, taking any one current data in the power grid current data sequence as an example, and recording the previous current data of the current data in the power grid current data sequence as the front current data of the current data; recording the absolute value of the difference value between the current data and the front current data of the current data as the current difference value of the current data; recording the absolute value of the difference value between the voltage data corresponding to the current data and the voltage data corresponding to the front current data as the voltage difference value of the current data; and obtaining the current voltage affinity of the current data according to the current difference and the voltage difference of the current data. In addition, if the current data is the first current data in the grid current data sequence, the front current data of the current data defaults to 0, and the voltage data corresponding to the front current data of the current data defaults to 0. The method for calculating the current voltage affinity of the current data comprises the following steps:
in the method, in the process of the invention,a current-voltage affinity representing the current data; />A voltage difference amount representing the current data; />Representing the sameCurrent difference amount of the current data; />Representing a preset hyper-parameter, preset +.>For preventing denominator from being 0; />Representing the average value of the voltage difference amounts of all the current data; />The average value of the current difference amounts of all the current data is represented. The greater the current-voltage affinity of the current data, the closer the current data is related to the voltage data corresponding to the current data, reflecting that the current data is closer to the real current data. The current-voltage affinity of all current data is obtained.
Further, a current data amount T3 is preset, where the embodiment is described by taking t3=10 as an example, and the embodiment is not specifically limited, where T3 may be determined according to specific implementation situations; taking any one expandable power grid current data set as an example, in a power grid current data sequence, the front T3 current data of the expandable power grid current data set and the rear T3 current data of the expandable power grid current data set are recorded as current data to be checked of the expandable power grid current data set. Wherein each scalable grid current dataset corresponds to a plurality of current data to be audited. It should be further noted that, if the current data quantity actually existing before and after the expandable power grid current data set does not meet the preset T3, the current data to be audited of the expandable power grid current data set is obtained based on the current data quantity actually existing before and after the expandable power grid current data set.
Further, sequentially taking the maximum value, the minimum value, the median and the mode of all current data in the expandable power grid current data set as reference characteristic current data in the expandable power grid current data set, and obtaining the adding necessity of the to-be-inspected nuclear power flow data according to the reference characteristic current data of the expandable power grid current data set and the current voltage affinities of all to-be-inspected current data. The calculation method for the adding necessity of the current data to be audited comprises the following steps:
in the method, in the process of the invention,indicating the necessity of adding the current data to be audited; />Representing the number of all reference characteristic current data within the scalable grid current dataset; />The current voltage affinity of the current data to be audited is represented; />Representing +.>Current voltage affinities of the individual reference characteristic current data; />Representing +.>Reference characteristic current data; />Representing the mean value of all reference characteristic current data in the scalable grid current dataset; />Representing a preset hyper-parameter, preset +.>For preventing denominator from being 0; />Representing the number of all current data within the scalable grid current dataset; />Representing the current data to be audited; />Representing +.>-individual current data; />The representation takes absolute value. And if the necessity of adding the current data to be audited is larger, the current data to be audited can be added into the expandable power grid current data set. Obtaining the adding necessity of all current data to be audited of the expandable power grid current data set, carrying out linear normalization on all the adding necessity, and recording the adding necessity after normalization as the final adding necessity.
Further, taking any one current data to be audited of the expandable power grid current data set as an example, according to the final adding necessity of the current data to be audited and the set expansion degree of the expandable power grid current data set, the adding feasibility of the current data to be audited is obtained. The calculation method of the adding feasibility of the nuclear power flow data to be checked comprises the following steps:
in the method, in the process of the invention,indicating the adding feasibility of the current data to be audited; />Representing a collective scalability of the scalable grid current dataset; />Indicating the final added necessity of the current data to be audited. If the adding feasibility of the current data to be audited is larger, the exception monitoring result obtained by constructing an isolated tree of the expandable power grid current data set is more accurate after the current data to be audited is added into the expandable power grid current data set. Acquiring the adding feasibility of all current data to be audited of the expandable power grid current data set; and obtaining the adding feasibility of all current data to be audited of all the expandable power grid current data sets.
So far, the adding feasibility of all current data to be audited of all expandable power grid current data sets is obtained through the method.
The abnormal current data monitoring module 104 adds the to-be-inspected nuclear current data into the expandable power grid current data set according to the adding feasibility and the inextensible power grid current data set to obtain a plurality of monitorable power grid current data sets, and performs abnormal monitoring on the monitorable power grid current data sets.
Specifically, an addition feasibility threshold T4 is preset, where the embodiment is described by taking t4=0.65 as an example, and the embodiment is not specifically limited, where T4 may be determined according to the specific implementation situation; taking any one expandable power grid current data set as an example, in all current data to be audited of the expandable power grid current data set, recording the current data to be audited with the adding feasibility larger than T4 as current data to be combined, adding all current data to be combined of the expandable power grid current data set into the expandable power grid current data set, recording the expandable power grid current data set as an initial monitorable power grid current data set, and acquiring all initial monitorable power grid current data sets.
Further, all the initial monitorable grid current data sets and the unexpanded grid current data sets are recorded as one monitorable grid current data set; constructing an isolated tree for each monitorable power grid current data set, and acquiring an abnormal score of each current data through the isolated tree; presetting an abnormality score threshold T5, wherein the present embodiment is described by taking t5=0.7 as an example, and the present embodiment is not particularly limited, wherein T5 may be determined according to specific implementation conditions; recording current data with abnormal scores greater than T5 as abnormal current data; and acquiring all abnormal current data. The process of obtaining the abnormal score according to the isolated tree is known content of the isolated forest algorithm, and the embodiment is not repeated. If there are a plurality of anomaly scores in one piece of current data, the average value of all anomaly scores in the current data is used as the final anomaly score of the current data.
This embodiment is completed.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (8)
1. The power grid operation abnormal state monitoring system based on data driving is characterized by comprising the following modules:
the system comprises a current data and voltage data acquisition module, a power supply module and a power supply module, wherein the current data and voltage data acquisition module is used for acquiring a plurality of current data of a power grid and voltage data corresponding to each current data;
the system comprises an expandable power grid current data set and a non-expandable power grid current data set acquisition module, wherein the expandable power grid current data set and the non-expandable power grid current data set acquisition module are used for recording a sequence of all current data sequenced according to the sequence from the early to the late of the acquisition time as a power grid current data sequence; dividing a power grid current data sequence into a plurality of power grid current data sets; obtaining the set extensibility of each power grid current data set according to the duty ratio difference between the current data in the power grid current data sets; screening a plurality of expandable power grid current data sets and a plurality of unexpanded power grid current data sets from the power grid current data sets according to the degree of expansion of the set;
the adding feasibility obtaining module is used for obtaining the current difference quantity and the voltage difference quantity of each current data according to the difference between adjacent current data and the difference between corresponding voltage data in the power grid current data sequence; obtaining the current voltage affinity of each current data according to the comparison difference between the current difference and the voltage difference; screening a plurality of current data to be audited from the current data on two sides of the current data set of the expandable power grid; obtaining the final adding necessity of each current data to be audited according to the difference of the current voltage affinities between the current data to be audited and the difference between the current data in the current data set of the scalable power grid; obtaining the adding feasibility of each piece of to-be-examined nuclear power flow data according to the final adding necessity and the set expandability, wherein the adding feasibility is used for describing the probability that the current data belong to abnormal current data;
the abnormal current data monitoring module is used for adding the to-be-inspected nuclear power flow data into the expandable power grid current data set according to the adding feasibility and the inextensible power grid current data set to obtain a plurality of monitorable power grid current data sets, and performing abnormal monitoring on the monitorable power grid current data sets;
the method for obtaining the set extensibility degree of each power grid current data set by the duty ratio difference between the current data in the power grid current data sets comprises the following specific steps:
for any one power grid current data set, the median of all current data in the power grid current data set is recorded as a first median, the current data which are larger than the first median in the power grid current data set are recorded as high current data, and the current data which are smaller than the first median in the power grid current data set are recorded as low current data; obtaining the expansibility of the power grid current data set according to all high current data and all low current data in the power grid current data set; the method for calculating the scalability degree of the power grid current data set comprises the following steps:
in the method, in the process of the invention,representing the degree of scalability of the grid current dataset; />Representing the average value of all high-current data in the power grid current data set; />Representing the average value of all low current data in the power grid current data set; />Representing preset super parameters; />Representing the number of all current data in the grid current data set; />Representing the number of all grid current datasets; />Indicate->The number of all current data in the individual grid current data sets; />The representation takes absolute value;
acquiring the expansibility of all the power grid current data sets, carrying out linear normalization processing on all the expansibility, and recording each normalized expansibility as a set expansibility;
the final adding necessity of each current data to be audited is obtained according to the difference of the current voltage affinities among the current data to be audited and the difference of the current data in the current data set of the scalable power grid, and the specific method comprises the following steps:
for any one expandable power grid current data set, sequentially taking the maximum value, the minimum value, the median and the mode of all current data in the expandable power grid current data set as reference characteristic current data in the expandable power grid current data set; for any one auditing current data, obtaining the adding necessity of the current data to be audited according to the reference characteristic current data of the expandable power grid current data set and the current voltage affinities of all the current data to be audited; the calculation method for the adding necessity of the current data to be audited comprises the following steps:
in the method, in the process of the invention,indicating the necessity of adding current data to be audited; />Representing the number of all reference characteristic current data within the scalable grid current dataset; />The current voltage affinity of the current data to be audited is represented; />Representing +.>Current voltage affinities of the individual reference characteristic current data; />Representing +.>Reference characteristic current data; />Representing the mean value of all reference characteristic current data in the expandable grid current data set; />Representing preset super parameters; />Representing the number of all current data within the scalable grid current dataset; />Representing current data to be audited; />Representing +.>-individual current data; />The representation takes absolute value;
obtaining the adding necessity of all current data to be audited of the expandable power grid current data set, carrying out linear normalization on all the adding necessity, and recording the adding necessity after normalization as the final adding necessity.
2. The system for monitoring abnormal operation of a power grid based on data driving according to claim 1, wherein the dividing the power grid current data sequence into a plurality of power grid current data sets comprises the following specific steps:
presetting a current data quantity T1; and (3) carrying out simple random sampling on the power grid current data sequence to obtain T1 current data, taking the T1 current data as demarcation points, and recording a set formed by all current data between two adjacent demarcation points as a power grid current data set.
3. The system for monitoring abnormal operation of a power grid based on data driving according to claim 1, wherein the specific method for screening a plurality of expandable power grid current data sets and a plurality of non-expandable power grid current data sets from the power grid current data sets according to the degree of scalability of the aggregate comprises the following steps:
presetting a set extensible degree threshold T2, and recording a power grid current data set with the set extensible degree larger than T2 as an extensible power grid current data set; the grid current data set with the set extensibility less than or equal to T2 is noted as a non-extensible grid current data set.
4. The system for monitoring abnormal operation of a power grid based on data driving according to claim 1, wherein the method for obtaining the current difference and the voltage difference of each current data from the difference between adjacent current data and the difference between corresponding voltage data in the power grid current data sequence comprises the following specific steps:
for any one current data in the power grid current data sequence, recording the previous current data of the current data in the power grid current data sequence as the front current data of the current data; recording the absolute value of the difference value between the current data and the front current data of the current data as the current difference value of the current data; and recording the absolute value of the difference value between the voltage data corresponding to the current data and the voltage data corresponding to the front current data as the voltage difference value of the current data.
5. The system for monitoring abnormal operation of a power grid based on data driving according to claim 1, wherein the method for obtaining the current-voltage affinity of each current data according to the comparison difference between the current difference and the voltage difference comprises the following specific steps:
in the method, in the process of the invention,the current voltage affinity of any one current data is represented; />A voltage difference amount representing current data; />A current difference amount representing current data; />Representing preset super parameters; />Representing the average value of the voltage difference amounts of all the current data; />The average value of the current difference amounts of all the current data is represented.
6. The system for monitoring abnormal operation of a power grid based on data driving according to claim 1, wherein the method for screening a plurality of current data to be audited from the current data on two sides of the current data set of the expandable power grid comprises the following specific steps:
and presetting a current data quantity T3 for any one expandable power grid current data set, and recording the front T3 current data of the expandable power grid current data set and the rear T3 current data of the expandable power grid current data set in a power grid current data sequence as current data to be audited of the expandable power grid current data set.
7. The system for monitoring abnormal operation state of a power grid based on data driving according to claim 1, wherein the method for obtaining the adding feasibility of each piece of pending nuclear power flow data according to the final adding necessity and the set expandability comprises the following specific steps:
in the method, in the process of the invention,representing the addition feasibility of any one audit current data of any one extensible power grid current data set;representing a set scalability of the scalable grid current dataset; />Indicating the final added necessity of the current data to be audited.
8. The system for monitoring abnormal operation of a power grid based on data driving according to claim 1, wherein the method for adding the pending nuclear power flow data to the expandable power grid current data set to obtain a plurality of monitorable power grid current data sets and performing abnormal monitoring on the monitorable power grid current data sets according to the adding feasibility and the inextensible power grid current data sets comprises the following specific steps:
for any one expandable power grid current data set, presetting an adding feasibility threshold T4, recording current data to be audited with adding feasibility larger than T4 as current data to be combined in all current data to be audited of the expandable power grid current data set, recording the expandable power grid current data set as an initial monitorable power grid current data set after adding all current data to be combined of the expandable power grid current data set into the expandable power grid current data set, and acquiring all initial monitorable power grid current data sets;
all the initial monitorable grid current data sets and the unexpanded grid current data sets are recorded as one monitorable grid current data set; constructing an isolated tree for each monitorable power grid current data set, and acquiring an abnormal score of each current data through the isolated tree; if a plurality of abnormal scores exist in one piece of current data, taking the average value of all abnormal scores of the current data as the final abnormal score of the current data;
an abnormal scoring threshold T5 is preset, and current data with abnormal scores greater than T5 are recorded as abnormal current data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311733726.1A CN117420345B (en) | 2023-12-18 | 2023-12-18 | Power grid operation abnormal state monitoring system based on data driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311733726.1A CN117420345B (en) | 2023-12-18 | 2023-12-18 | Power grid operation abnormal state monitoring system based on data driving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117420345A CN117420345A (en) | 2024-01-19 |
| CN117420345B true CN117420345B (en) | 2024-03-08 |
Family
ID=89528668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311733726.1A Active CN117420345B (en) | 2023-12-18 | 2023-12-18 | Power grid operation abnormal state monitoring system based on data driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117420345B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995510A (en) * | 2014-05-14 | 2014-08-20 | 中国传媒大学 | Light amplifier monitoring technology based on SNMP4J |
| JP7240691B1 (en) * | 2021-09-08 | 2023-03-16 | 山東大学 | Data drive active power distribution network abnormal state detection method and system |
| CN116974268A (en) * | 2023-09-20 | 2023-10-31 | 青岛朗兹环保科技有限公司 | Intelligent monitoring and early warning method for control system circuit |
-
2023
- 2023-12-18 CN CN202311733726.1A patent/CN117420345B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995510A (en) * | 2014-05-14 | 2014-08-20 | 中国传媒大学 | Light amplifier monitoring technology based on SNMP4J |
| JP7240691B1 (en) * | 2021-09-08 | 2023-03-16 | 山東大学 | Data drive active power distribution network abnormal state detection method and system |
| CN116974268A (en) * | 2023-09-20 | 2023-10-31 | 青岛朗兹环保科技有限公司 | Intelligent monitoring and early warning method for control system circuit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117420345A (en) | 2024-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115982602B (en) | Photovoltaic transformer electrical fault detection method | |
| CN114282579A (en) | Aviation bearing fault diagnosis method based on variational modal decomposition and residual error network | |
| CN102435910B (en) | Power electronic circuit health monitoring method based on support vector classification | |
| CN110008898B (en) | Industrial equipment data edge processing method based on symbol and convolutional neural network | |
| CN108333468B (en) | The recognition methods of bad data and device under a kind of active power distribution network | |
| CN112257767B (en) | Product key part state classification method aiming at class imbalance data | |
| CN109389325B (en) | Method for evaluating state of electronic transformer of transformer substation based on wavelet neural network | |
| CN111626360B (en) | Method, apparatus, device and storage medium for detecting boiler fault type | |
| CN117473351B (en) | Power supply information remote transmission system based on Internet of things | |
| CN111881594A (en) | Non-stationary signal state monitoring method and system for nuclear power equipment | |
| CN109726770A (en) | A kind of analog circuit fault testing and diagnosing method | |
| CN117972618A (en) | Method and system for detecting secondary power failure of hybrid integrated circuit | |
| CN116127354A (en) | High-voltage cable partial discharge fault positioning method and system based on deep learning | |
| CN115600928A (en) | OAO-RVM-based GIS basin-type insulator state evaluation method | |
| CN117420345B (en) | Power grid operation abnormal state monitoring system based on data driving | |
| CN117172601A (en) | Non-invasive load monitoring method based on residual total convolution neural network | |
| CN112067298A (en) | Rolling bearing fault diagnosis method based on hierarchical global fuzzy entropy | |
| CN116561692A (en) | Dynamic update real-time measurement data detection method | |
| CN109765878A (en) | The aided analysis method of new-energy automobile CAN bus network failure | |
| CN115345192A (en) | Fault diagnosis method for rolling bearing under variable working conditions and complex noise environment | |
| Jiang et al. | Recurrence plot quantitative analysis-based fault recognition method of rolling bearing | |
| CN116224950A (en) | Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line | |
| CN114239759A (en) | Fault diagnosis method and device for electric vehicle, electronic equipment and storage medium | |
| CN111476392A (en) | Metering device clock error trend prediction method based on social perception | |
| CN115169434B (en) | Host working condition characteristic value extraction method and system based on K-means clustering algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |