CN110532119A - Power system operation abnormal point detecting method - Google Patents
Power system operation abnormal point detecting method Download PDFInfo
- Publication number
- CN110532119A CN110532119A CN201910681810.0A CN201910681810A CN110532119A CN 110532119 A CN110532119 A CN 110532119A CN 201910681810 A CN201910681810 A CN 201910681810A CN 110532119 A CN110532119 A CN 110532119A
- Authority
- CN
- China
- Prior art keywords
- outlier
- operation data
- point
- data point
- power system
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Quality & Reliability (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention provides a kind of power system operation abnormal point detecting method, and method includes: to select outlier from the operation data point set of dynamical system based on DBSCAN algorithm；The degree that peels off of each outlier is calculated based on LOF algorithm；Corresponding any outlier determines whether the outlier is abnormal point according to the degree that peels off of the outlier.The present invention realizes the detection of power system operation abnormal point, and calculating speed is fast, and Detection accuracy is high.
Description
Technical field
The invention belongs to abnormality detection technical field more particularly to a kind of power system operation abnormal point detecting methods.
Background technique
There can be some exceptional data points in the operation data of dynamical system, these abnormal points deviate from fortune to some extent
The normal range (NR) of row parameter shows as " wave crest " occur suddenly in stable operating parameter curve graph.
Most of data digging method is to be regarded as noise and abandon to the processing of this kind of abnormal point.However, due to
During power system operation abnormal point may due to power-equipment misoperation caused by, therefore by being detected to it
It can identify the abnormal operating condition of power-equipment.
Therefore, to power system operation abnormal point carry out detection be current industry it is urgently to be resolved need project.
Summary of the invention
The embodiment of the present invention provides a kind of power system operation abnormal point detecting method, urgently to be resolved to meet current industry
Needs.
According to a first aspect of the embodiments of the present invention, a kind of power system operation abnormal point detecting method is provided, comprising:
Outlier is selected from the operation data point set of dynamical system based on DBSCAN algorithm；
The degree that peels off of each outlier is calculated based on LOF algorithm；
Corresponding any outlier determines whether the outlier is abnormal point according to the degree that peels off of the outlier.
The second aspect according to an embodiment of the present invention, also offer a kind of electronic equipment, including memory, processor and deposit
The computer program that can be run on a memory and on a processor is stored up, the processor calls described program instruction to be able to carry out
Power system operation abnormal point provided by any possible implementation in the various possible implementations of first aspect
Detection method.
In terms of third according to an embodiment of the present invention, a kind of non-transient computer readable storage medium is also provided, it is described
Non-transient computer readable storage medium stores computer instruction, and the computer instruction makes the computer execute first aspect
Various possible implementations in power system operation abnormal point detecting method provided by any possible implementation.
The embodiment of the present invention provides a kind of power system operation abnormal point detecting method, and this method is calculated by elder generation base DBSCAN
Method selects outlier from the operation data point set of dynamical system, i.e., doubtful exceptional data point, is then based on the calculating of LOF algorithm
The degree that peels off of each outlier further determines whether each outlier is abnormal point according to the degree of peeling off, to realize dynamic
Force system is operating abnormally the detection of point, and calculating speed is fast, and Detection accuracy is high.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is power system operation abnormal point detecting method overall flow schematic diagram provided in an embodiment of the present invention；
Fig. 2 is that definition is straight in DBSCAN algorithm in power system operation abnormal point detecting method provided in an embodiment of the present invention
Connect the reachable schematic diagram of density；
Fig. 3 is close to define in DBSCAN algorithm in power system operation abnormal point detecting method provided in an embodiment of the present invention
Spend reachable schematic diagram；
Fig. 4 is close to define in DBSCAN algorithm in power system operation abnormal point detecting method provided in an embodiment of the present invention
The connected schematic diagram of degree；
Fig. 5 is DBSCAN algorithm flow signal in power system operation abnormal point detecting method provided in an embodiment of the present invention
Figure；
Fig. 6 is LOF algorithm flow schematic diagram in power system operation abnormal point detecting method provided in an embodiment of the present invention；
Fig. 7 is electronic equipment overall structure diagram provided in an embodiment of the present invention.
Specific embodiment
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
A kind of power system operation abnormal point detecting method is provided in one embodiment of the invention, and Fig. 1 is the present invention
The power system operation abnormal point detecting method overall flow schematic diagram that embodiment provides, this method comprises: S101, is based on
DBSCAN algorithm selects outlier from the operation data point set of dynamical system；
Wherein, DBSCAN (Density-Based Spatial Clustering of Applications with
Noise has noisy density clustering method) it is a kind of density-based algorithms, cluster is defined as density phase
The maximum set of point even can be cluster having region division highdensity enough, and can be in the spatial database of noise
It was found that the cluster of arbitrary shape.The present embodiment is not limited to the type of dynamical system.The operation data point of dynamical system is dynamical system
The system data that each moment generates at runtime, operation data point set are combined into the set for the operation data point that all moment generate.
Outlier is selected from the operation data point set of dynamical system using DBSCAN algorithm, outlier is suspicious exception object.
S102 calculates the degree that peels off of each outlier based on LOF algorithm；
Wherein, LOF algorithm (Local Outlier Factor, locally peel off factors check method) is a kind of unsupervised
Discrete detection method, which calculates a local outlier factor LOF to each point in data set, by whether judging LOF
Judge whether it is abnormal point close to 1.The present embodiment measures the degree that peels off of each discrete point using local outlier factor, passes through meter
The ratio of the density and autologous density of calculating each discrete point field indicates the degree that peels off of each discrete point.The ratio of this relative density
Value is bigger, indicates that the degree that peels off of the discrete point is higher.
S103, corresponding any outlier determine whether the outlier is abnormal according to the degree that peels off of the outlier
Point.
If the degree that peels off of discrete point is 1 greater than a certain given threshold, such as threshold value, it is determined that the outlier is abnormal
Otherwise point is normal point.The producing cause of abnormal point has very much, and to analyze these abnormal point Producing reasons and also need to combine
Device characteristics, running environment etc. is because of usually further analysis.
The present embodiment passes through first base DBSCAN algorithm and selects outlier from the operation data point set of dynamical system, that is, doubts
Like exceptional data point, it is then based on the degree that peels off that LOF algorithm calculates each outlier, is further determined according to the degree that peels off every
Whether a outlier is abnormal point, thus realize the detection of power system operation abnormal point, and calculating speed is fast, Detection accuracy
It is high.
On the basis of the above embodiments, in the present embodiment based on DBSCAN algorithm from the operation data point of dynamical system
The step of selecting outlier specifically includes: selecting any not being included into cluster and unmarked to peel off from the operation data point set
The operation data point of point；Judge the operation data point selected whether for kernel object；If it is not, then judging the operation data
Point is marginal point or outlier, and the operation data point is labeled as marginal point or outlier according to judging result；If so,
The operation data point based on selection establishes a new cluster, will be by the operation data dot density selected is reachable and density phase
The operation data point even is added in the new cluster, until all operation data points are included into cluster or labeled as outlier.
Specifically, setting operation data point set is combined into S=(x1,x2,…,xm), operation data is clicked through based on defined below
Row cluster:
1) ε-neighborhood: any operation data point xjPre-set radius ε in region, ε be input parameter, enable | Nε(x) | to appoint
The sum of operation data point in ε-neighborhood of one operation data point x.
2) kernel object: | Nε(x) | the operation data point x of >=MinPts.MinPts is preset threshold, to input parameter.
3) directly density is reachable: as shown in Fig. 2, point q is in ε-neighborhood of point p, and | Nε(xp) | >=MinPts claims q by right
As the direct density of p is reachable.
4) density is reachable: as shown in Figure 3.Point pt+1In point ptε-neighborhood in (t=1,2), and intermediate point p meet | Nε
(xp) | >=MinPts claims p3By p1Density is reachable.
5) density is connected: as shown in figure 4, point o meets | Nε(xo) | >=MinPts, point p is reachable by o density, and point q is by o density
It is reachable, claim p, q density to be connected.
Operation data point in operation data point set is divided into three classes by the present embodiment:
Core point: i.e. kernel object, ε-neighborhood is interior to include at least MinPts sample object；
Marginal point: it is unsatisfactory for the requirement of kernel object, but reachable by kernel object density；
Outlier: not being the doubtful abnormal point that core point is also not marginal point.
The process of DBSCAN clustering algorithm is as shown in figure 5, the processing step of DBSCAN clustering algorithm is as follows in the present embodiment:
(1) operation data point set S, Neighbourhood parameter (ε, MinPts) and distance metric mode are inputted；
(2) also not visited operation data point p is arbitrarily chosen from S, if meeting | Nε(xp) | >=MinPts then turns to walk
Suddenly (3), otherwise according to peel off or marginal point processing after go to step (2)；
(3) cluster is built based on operation data point p, will be added in cluster by the reachable object being connected with density of p density；
(4) continue repeat the above steps (2) and (3), until there is no the operation of not visited mistake in operation data point set
Data point.
On the basis of the above embodiments, judge the operation data point selected whether for kernel object in the present embodiment
The step of specifically include: according in the value and the operation data point set of each characteristic parameter of the operation data point of selection
The value of each characteristic parameter of other operation data points in addition to the operation data point of selection, calculates the operation of selection
Direct range between data point and other each described operation data points；Obtain the direct range be less than pre-set radius other
The sum of operation data point, if the sum is greater than preset threshold, using the operation data point selected as kernel object.
Specifically, when dynamical system is ship power system, a plurality of operation of voltage-stablizer in ship power system is chosen
Record carries out outlier detection.The relevant feature parameters of selection include pressure, water level, temperature, YKG temperature and PFG temperature.For table
It states conveniently, each characteristic parameter is replaced with into corresponding parameter serial number, the corresponding measuring device of each parameter serial number, such as 1 institute of table
Show.
1 parameter serial number table of table
Before direct range between the operation data point for calculating selection and other operation data points, it is also necessary to these
The characteristic ginseng value of operation data point is pre-processed.Since voltage-stablizer characteristic parameter has multiple redundancy measurement, so first to superfluous
Remaining measured value is pre-processed, and Measurement channel of first debugging influences, and the average value for then choosing redundant measurement is surveyed as redundancy
The value of characteristic parameter after amount merging.Such as 24 and 25 be redundant measurement, 41 and 42 be redundant measurement, 43,44 and 45 be redundancy
Measurement.Totally 8 characteristic parameters after pretreatment.
For the influence for eliminating parameter dimension, it is standardized according to the following formula.
Wherein,It is the ith feature x after standardizationiK-th of value after standardization, xikFor ith feature xiKth
A value, Max are characterized xiMaximum value, Min is characterized xiMinimum value.
Operation data point the distance between i and j is measured with following Euclidean distance:
Two operation data points i, j are n=8 dimension, and Euclidean distance is the difference between two operation data point same alike results
Square root sum square of value.
On the basis of the above embodiments, in the present embodiment further include: to the pre-set radius and preset threshold respectively into
Row repeatedly setting；Calculate the pre-set radius and the corresponding discrete point of preset threshold set every time；It will be confirmed as every time discrete
The operation data point of point is as final discrete point；Correspondingly, the degree that peels off of each outlier is calculated based on LOF algorithm
The step of specifically include: the degree that peels off of each final outlier is calculated based on LOF algorithm.
Specifically, although DBSCAN cluster is a kind of unsupervised study, measure algorithm can not be carried out to report by mistake and fail to report
Accuracy.But since power system operation data are always to maintain stabilization most of the time, only fluctuate to a very small extent, it is abnormal
Data scale is smaller, if directly analyzed overall data, the detection to most of normal data is meaningless.Therefore it can
It is unlikely to be abnormal data rejecting by most of with the parameter ε and MinPts by adjusting DBSCAN algorithm, to reduce
The scale of the abnormal data set to be analyzed provides help for subsequent further outlier detection.
Two inputs parameter ε and MinPts of DBSCAN algorithm directly affect the operation data for being finally marked as outlier
Points.More outliers are detected when ε reduction or MinPts increase Shi Douhui, when outlier accounting is high, it is understood that there may be wrong report,
Normal data is labeled as outlier；When outlier accounting is low, it is understood that there may be fail to report, i.e., outlier label in part is positive
Often.It fails to report and reports by mistake to reduce, algorithm parameter ε and MinPts are repeatedly set respectively.To be divided into every time from
The operation data point of group's point is included into doubtful abnormal point numerical collection as final discrete point.
Multiple setup parameter ε and MinPts, 3 institute of number and outlier ratio situation such as table 2 and table of experiment gained sub-clustering
Show, it can be seen that in the case where MinPts is constant, due to the increase of ε, the DBSCAN algorithm the step of in (2), meet | Nε
(xp) | the object p of >=MinPts condition increases, so that more objects are added into existing cluster.And it is constant in ε,
Under the case where MinPts increases, meet | Nε(xp) | the object p of >=MinPts condition is reduced, and in a disguised form increases the number of outlier
Amount, so that outlier ratio increases.
The ratio (%) of outlier under 2 different parameters of table
Sub-clustering number under 3 different parameters of table
The number that sub-clustering is chosen from many experiments is 1, and is all divided into the operation data of outlier in experiment every time
Point discrete point the most final, is added into doubtful abnormal data set, and the size of doubtful abnormal data set is 143.Based on DBSCAN
The principle of algorithm, it is believed that the different most of normal data points of these doubtful exceptional data points, it is doubtful containing abnormal data,
It is detected using subsequent LOF algorithm.
On the basis of the various embodiments described above, the journey that peels off of each outlier is calculated in the present embodiment based on LOF algorithm
The step of spending specifically includes: for any outlier, it is nearest to obtain distance outlier in the operation data point set
K operation data point；Wherein, the k is positive integer；According to each operation data point in the k operation data point and it is somebody's turn to do
Reach distance between outlier, obtains the local reachability density of the outlier, and calculates the part of the k operation data point
Up to the average value of density；It can by the part of the average value of the local reachability density of the k operation data point and the outlier
Local outlier factor up to the ratio between density as the outlier, using the local outlier factor as the outlier from
Group's degree.
Part peel off factor LOF key concept it is as follows:
(1) kth distance dk(p): indicate with point p recently k-th point and its at a distance from.
(2) reach distance reach-dist (p, o): when given parameters k, the reach distance of data point p to data point o is number
The maximum value of direct range between the kth distance and data point p and o of strong point o, i.e. reach-dist (p, o)=max { dk
(o),d(p,o)}。
(3) local reachability density lrdk(p), calculation formula is as follows:
In formula | Nk(p) | indicate the set with k point p nearest point.
(4) local outlier factor LOFk(p), calculation formula is as follows:
The flow chart of LOF algorithm is as shown in Figure 6.
On the basis of the above embodiments, in the present embodiment k value value range are as follows:
k∈[klb,kub]=[max 10, | 0.01* | S | | }, | r* | S | |]
Wherein, klbFor the minimum value of k, kubFor the maximum value of k, | S | it is the total number of the outlier, r is described peels off
Ratio of the point in the operation data point set；Correspondingly, the degree that peels off of each outlier is calculated based on LOF algorithm
Step is specifically included in the value range and takes multiple values for k；For any discrete point, calculating separately should under each k value
The degree that peels off of outlier；Most using the maximum value in the corresponding degree that peels off of all k values of the outlier as the outlier
Peel off degree eventually.
The present embodiment provides a kind of electronic equipment, Fig. 7 is electronic equipment overall structure provided in an embodiment of the present invention signal
Figure, which includes: at least one processor 701, at least one processor 702 and bus 703；Wherein,
Processor 701 and memory 702 pass through bus 703 and complete mutual communication；
Memory 702 is stored with the program instruction that can be executed by processor 701, and the instruction of processor caller is able to carry out
Method provided by above-mentioned each method embodiment, for example, based on DBSCAN algorithm from the operation data point set of dynamical system
Outlier is selected in conjunction；The degree that peels off of each outlier is calculated based on LOF algorithm；Corresponding any outlier, according to
The degree that peels off of the outlier determines whether the outlier is abnormal point.
The present embodiment provides a kind of non-transient computer readable storage medium, non-transient computer readable storage medium storages
Computer instruction, computer instruction make computer execute method provided by above-mentioned each method embodiment, for example, are based on
DBSCAN algorithm selects outlier from the operation data point set of dynamical system；Each outlier is calculated based on LOF algorithm
The degree that peels off；Corresponding any outlier determines whether the outlier is abnormal according to the degree that peels off of the outlier
Point.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
The various media that can store program code such as disk.
System embodiment described above is only schematical, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of power system operation abnormal point detecting method characterized by comprising
Outlier is selected from the operation data point set of dynamical system based on DBSCAN algorithm；
The degree that peels off of each outlier is calculated based on LOF algorithm；
Corresponding any outlier determines whether the outlier is abnormal point according to the degree that peels off of the outlier.
2. power system operation abnormal point detecting method according to claim 1, which is characterized in that be based on DBSCAN algorithm
The step of selecting outlier from the operation data point of dynamical system specifically includes:
It selects any not being included into cluster and the unmarked operation data point for outlier from the operation data point set；
Judge the operation data point selected whether for kernel object；
If it is not, then judging the operation data point for marginal point or outlier, according to judging result by the operation data point
Labeled as marginal point or outlier；
If so, the operation data point based on selection establishes a new cluster, by the operation data dot density by selecting
The reachable operation data point being connected with density is added in the new cluster, until all operation data points are included into cluster or mark
It is denoted as outlier.
3. power system operation abnormal point detecting method according to claim 2, which is characterized in that judge the described of selection
The step of whether operation data point is kernel object specifically includes:
According to the institute for removing selection in the value and the operation data point set of each characteristic parameter of the operation data point of selection
The value for stating each characteristic parameter of other operation data points other than operation data point calculates the operation data point of selection and each
Direct range between other described operation data points；
Sum of the direct range less than other operation data points of pre-set radius is obtained, if the sum is greater than default threshold
Value, then using the operation data point selected as kernel object.
4. power system operation abnormal point detecting method according to claim 3, which is characterized in that when the dynamical system
When for ship power system, the characteristic parameter is the characteristic parameter of voltage-stablizer in the ship power system, including each pressure
The temperature that the pressure of power test equipment measurement, the water level of each water-level measuring equipment measurement, each first temperature measurement equipment measure
The PFG temperature of degree, the YLG temperature of each second temperature measuring device measurement and the measurement of each third temperature measurement equipment.
5. power system operation abnormal point detecting method according to claim 4, which is characterized in that in the institute for calculating selection
Before stating the distance between operation data point and each other operation data points further include:
Obtain the pressure test equipment, water-level measuring equipment, the first temperature measurement equipment, second temperature measuring device and third
Redundance unit in temperature measurement equipment；
Using the average value of the value of the same redundance unit measurement as the value of a characteristic parameter；
The value of all characteristic parameters is standardized.
6. power system operation abnormal point detecting method according to claim 3, which is characterized in that further include:
The pre-set radius and preset threshold are repeatedly set respectively；
Calculate the pre-set radius and the corresponding discrete point of preset threshold set every time；
It will be confirmed as the operation data point of discrete point every time as final discrete point；
Correspondingly, based on LOF algorithm calculate each outlier peel off degree the step of specifically include:
The degree that peels off of each final outlier is calculated based on LOF algorithm.
7. -6 any power system operation abnormal point detecting method according to claim 1, which is characterized in that be based on LOF
Algorithm calculate each outlier peel off degree the step of specifically include:
For any outlier, the k operation data that the distance outlier is nearest in the operation data point set is obtained
Point；Wherein, k is positive integer；
According to each operation data point in k operation data point and the reach distance between the outlier, the outlier is obtained
Local reachability density, and calculate the average value of the local reachability density of the k operation data point；
By the ratio between the average value of local reachability density and the local reachability density of the outlier of the k operation data point
It is worth the local outlier factor as the outlier, using the local outlier factor as the degree that peels off of the outlier.
8. power system operation abnormal point detecting method according to claim 7, which is characterized in that the value range of k value
Are as follows:
k∈[klb,kub]=[max 10, | 0.01* | S | | }, | r* | S | |]
Wherein, klbFor the minimum value of k, kubFor the maximum value of k, | S | it is the total number of the outlier, r is that the outlier exists
Ratio in the operation data point set；
Correspondingly, based on LOF algorithm calculate each outlier peel off degree the step of specifically include
Multiple values are taken in the value range for k；
For any outlier, the degree that peels off of the outlier under each k value is calculated separately；
Using the maximum value in the corresponding degree that peels off of all k values of the outlier as the degree that finally peels off of the outlier.
9. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes the dynamical system as described in any one of claim 1 to 8 when executing described program
System is operating abnormally the step of point detecting method.
10. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer
The step of the power system operation abnormal point detecting method as described in any one of claim 1 to 8 is realized when program is executed by processor
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910681810.0A CN110532119B (en) | 2019-07-26 | 2019-07-26 | Method for detecting abnormal running point of power system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910681810.0A CN110532119B (en) | 2019-07-26 | 2019-07-26 | Method for detecting abnormal running point of power system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110532119A true CN110532119A (en) | 2019-12-03 |
| CN110532119B CN110532119B (en) | 2023-04-25 |
Family
ID=68661967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910681810.0A Active CN110532119B (en) | 2019-07-26 | 2019-07-26 | Method for detecting abnormal running point of power system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110532119B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541016A (en) * | 2020-11-26 | 2021-03-23 | 南方电网数字电网研究院有限公司 | Power consumption abnormality detection method, device, computer equipment and storage medium |
| CN113722384A (en) * | 2021-11-02 | 2021-11-30 | 西安热工研究院有限公司 | Detection method, system and equipment for abnormal value of measured point data based on density algorithm |
| CN115809417A (en) * | 2023-02-09 | 2023-03-17 | 新风光电子科技股份有限公司 | Production line operation signal detection method for high-voltage frequency converter control cabinet |
| CN116228603A (en) * | 2023-05-08 | 2023-06-06 | 山东杨嘉汽车制造有限公司 | Alarm system and device for barriers around trailer |
| CN116859902A (en) * | 2023-09-04 | 2023-10-10 | 西安热工研究院有限公司 | Database abnormal point detection method and system for hydropower control system |
| CN117591986A (en) * | 2024-01-18 | 2024-02-23 | 天津市职业大学 | Real-time automobile data processing method based on artificial intelligence |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109067725A (en) * | 2018-07-24 | 2018-12-21 | 成都亚信网络安全产业技术研究院有限公司 | Network flow abnormal detecting method and device |
| CN109446189A (en) * | 2018-10-31 | 2019-03-08 | 成都天衡智造科技有限公司 | A kind of technological parameter outlier detection system and method |
| CN109740175A (en) * | 2018-11-18 | 2019-05-10 | 浙江大学 | A kind of point judging method that peels off towards Wind turbines power curve data |
| CN109753991A (en) * | 2018-12-06 | 2019-05-14 | 中科恒运股份有限公司 | Abnormal deviation data examination method and device |
| WO2019095719A1 (en) * | 2017-11-14 | 2019-05-23 | 深圳中兴网信科技有限公司 | Network traffic anomaly detection method, apparatus, computer device and storage medium |
-
2019
- 2019-07-26 CN CN201910681810.0A patent/CN110532119B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019095719A1 (en) * | 2017-11-14 | 2019-05-23 | 深圳中兴网信科技有限公司 | Network traffic anomaly detection method, apparatus, computer device and storage medium |
| CN109067725A (en) * | 2018-07-24 | 2018-12-21 | 成都亚信网络安全产业技术研究院有限公司 | Network flow abnormal detecting method and device |
| CN109446189A (en) * | 2018-10-31 | 2019-03-08 | 成都天衡智造科技有限公司 | A kind of technological parameter outlier detection system and method |
| CN109740175A (en) * | 2018-11-18 | 2019-05-10 | 浙江大学 | A kind of point judging method that peels off towards Wind turbines power curve data |
| CN109753991A (en) * | 2018-12-06 | 2019-05-14 | 中科恒运股份有限公司 | Abnormal deviation data examination method and device |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541016A (en) * | 2020-11-26 | 2021-03-23 | 南方电网数字电网研究院有限公司 | Power consumption abnormality detection method, device, computer equipment and storage medium |
| CN113722384A (en) * | 2021-11-02 | 2021-11-30 | 西安热工研究院有限公司 | Detection method, system and equipment for abnormal value of measured point data based on density algorithm |
| CN115809417A (en) * | 2023-02-09 | 2023-03-17 | 新风光电子科技股份有限公司 | Production line operation signal detection method for high-voltage frequency converter control cabinet |
| CN115809417B (en) * | 2023-02-09 | 2023-05-09 | 新风光电子科技股份有限公司 | Production line operation signal detection method for high-voltage frequency converter control cabinet |
| CN116228603A (en) * | 2023-05-08 | 2023-06-06 | 山东杨嘉汽车制造有限公司 | Alarm system and device for barriers around trailer |
| CN116859902A (en) * | 2023-09-04 | 2023-10-10 | 西安热工研究院有限公司 | Database abnormal point detection method and system for hydropower control system |
| CN117591986A (en) * | 2024-01-18 | 2024-02-23 | 天津市职业大学 | Real-time automobile data processing method based on artificial intelligence |
| CN117591986B (en) * | 2024-01-18 | 2024-04-05 | 天津市职业大学 | Real-time automobile data processing method based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110532119B (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110532119A (en) | Power system operation abnormal point detecting method | |
| CN108923952B (en) | Fault diagnosis method, equipment and storage medium based on service monitoring index | |
| EP1960853A1 (en) | Evaluating anomaly for one-class classifiers in machine condition monitoring | |
| CN112148768A (en) | Index time series abnormity detection method, system and storage medium | |
| JP2008059601A (en) | Method for identifying bimodal data | |
| Bogojeska et al. | Classifying server behavior and predicting impact of modernization actions | |
| US20120203517A1 (en) | Method of determining the influence of a variable in a phenomenon | |
| CN110570544A (en) | method, device, equipment and storage medium for identifying faults of aircraft fuel system | |
| CN112416662A (en) | Multi-time series data anomaly detection method and device | |
| CN111061581A (en) | Fault detection method, device and equipment | |
| Ahsan et al. | Developing computational intelligence for smart qualification testing of electronic products | |
| Song et al. | Novel application of deep learning for adaptive testing based on long short-term memory | |
| KR101883277B1 (en) | Method and device for fault detection of manufacturing process based on dynamic time warping and exponentioal penaltu(dtwep) | |
| CN111783883A (en) | Abnormal data detection method and device | |
| JP2007164346A (en) | Decision tree changing method, abnormality determination method, and program | |
| CN113570070B (en) | Streaming data sampling and model updating method, device, system and storage medium | |
| CN111240881A (en) | Web service combination abnormity diagnosis method based on Bayesian network | |
| Chandrashekar et al. | In-vivo fault analysis and real-time fault prediction for rf generators using state-of-the-art classifiers | |
| CN113554126B (en) | Sample evaluation method, device, equipment and computer readable storage medium | |
| CN115545191B (en) | Current noise reduction network model training method and fault current limiter current noise reduction method | |
| Gandomi et al. | Deep-Learning-based Fault Detection and Location Method Applied on Isolated Dc-Dc Converter | |
| CN113447813B (en) | Fault diagnosis method and equipment for offshore wind generating set | |
| US20220188669A1 (en) | Prediction method for system errors | |
| CN117170914A (en) | Fault locating method, device, computer system and readable storage medium | |
| CN116955178A (en) | Method and device for determining server, storage medium and electronic equipment |
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 |