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
- point
- power system
- data points
- outliers
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 44
- 238000001514 detection method Methods 0.000 claims abstract description 12
- 238000005259 measurement Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 4
- 238000009529 body temperature measurement Methods 0.000 claims 3
- 230000005856 abnormality Effects 0.000 claims 1
- 238000005183 dynamical system Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000003381 stabilizer Substances 0.000 description 2
- 238000007418 data mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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 the technical field of anomaly detection, and particularly relates to a method for detecting an abnormal operating point of a power system.
Background
There may be some abnormal data points in the operating data of the powertrain system that deviate to varying degrees from the normal range of operating parameters, as evidenced by the sudden appearance of "peaks" in the plot of smooth operating parameters.
Most data mining methods deal with such outliers by treating them as noise and discarding them. However, since an abnormal point may be caused by abnormal operation of the power plant during operation of the power system, an abnormal operation state of the power plant may be identified by detecting it.
Therefore, detecting abnormal operating points of the power system is a problem to be solved in the industry.
Disclosure of Invention
The embodiment of the invention provides a method for detecting an abnormal operating point of a power system, which is used for meeting the urgent need in the industry at present.
According to a first aspect of the embodiments of the present invention, there is provided a method for detecting an abnormal operating point of a power system, including:
selecting outliers from a running data point set of the power system based on a DBSCAN algorithm;
calculating the outlier degree of each outlier based on an LOF algorithm;
and corresponding to any one of the outliers, determining whether the outlier is an abnormal point according to the outlier degree of the outlier.
According to a second aspect of the embodiments of the present invention, there is also provided an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor calls the program instructions to perform the method for detecting an abnormal operating point of a power system provided in any one of the various possible implementations of the first aspect.
According to a third aspect of the embodiments of the present invention, there is also provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method for detecting an abnormal point of operation of a power system according to any one of the various possible implementation manners of the first aspect.
The embodiment of the invention provides a method for detecting abnormal operating points of a power system, which comprises the steps of firstly selecting outliers, namely suspected abnormal data points, from an operating data point set of the power system by a base DBSCAN algorithm, then calculating the outlier degree of each outlier based on an LOF algorithm, and further determining whether each outlier is an abnormal point according to the outlier degree, so that the abnormal operating points of the power system are detected, the calculating speed is high, and the detection accuracy is high.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic overall flow chart of a method for detecting an abnormal operating point of a power system according to an embodiment of the present invention;
fig. 2 is a schematic diagram illustrating a direct density reachable defined in a dbs can algorithm in the method for detecting an abnormal operating point of a power system according to an embodiment of the present invention;
fig. 3 is a schematic diagram illustrating that the defined density in the DBSCAN algorithm in the method for detecting an abnormal operating point of a power system according to the embodiment of the present invention is reachable;
fig. 4 is a schematic diagram of density connection defined in a DBSCAN algorithm in the method for detecting an abnormal operating point of a power system according to the embodiment of the present invention;
FIG. 5 is a schematic diagram of a DBSCAN algorithm flow in the method for detecting an abnormal operating point of a power system according to the embodiment of the present invention;
FIG. 6 is a schematic flow chart of an LOF algorithm in the method for detecting the operating abnormal point of the power system according to the embodiment of the invention;
fig. 7 is a schematic view of an overall structure of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
In an embodiment of the present invention, a method for detecting an abnormal operating point of a power system is provided, and fig. 1 is a schematic overall flow chart of the method for detecting an abnormal operating point of a power system according to the embodiment of the present invention, where the method includes: s101, selecting outliers from a running data point set of the power system based on a DBSCAN algorithm;
among them, DBSCAN (Density-Based Clustering of Applications with Noise) is a Density-Based Clustering algorithm that defines clusters as the maximum set of Density-connected points, can divide areas with sufficiently high Density into clusters, and can find clusters of arbitrary shapes in a Spatial database of Noise. The present embodiment is not limited to the kind of power system. The operation data points of the power system are data generated at each moment when the power system operates, and the operation data point set is a set of operation data points generated at all the moments. An outlier, which is a suspected outlier, is selected from a set of operational data points of the power system using the DBSCAN algorithm.
S102, calculating the outlier degree of each outlier based on an LOF algorithm;
the Local Outlier Factor (LOF) algorithm is an unsupervised discrete detection method, and calculates a Local Outlier Factor (LOF) for each point in a data set, and determines whether the LOF is an Outlier by determining whether the LOF is close to 1. The present embodiment measures the degree of outlier of each discrete point by using the local anomaly factor, and represents the degree of outlier of each discrete point by calculating the ratio of the density of each discrete point field to the density of itself. The larger this relative density ratio, the higher the degree of outlier of the discrete point.
S103, corresponding to any outlier, determining whether the outlier is an abnormal point according to the outlier degree of the outlier.
If the degree of outlier of the discrete point is larger than a certain set threshold, if the threshold is 1, the outlier is determined to be an abnormal point, otherwise, the outlier is determined to be a normal point. There are many causes of abnormal points, and further analysis of the causes of abnormal points is required by combining the characteristics of the equipment, the operating environment, and other factors.
In the embodiment, the DBSCAN algorithm is used for selecting outliers, namely suspected abnormal data points, from the running data point set of the power system, then the outlier degree of each outlier is calculated based on the LOF algorithm, and whether each outlier is an abnormal point is further determined according to the outlier degree, so that the running abnormal points of the power system are detected, the calculating speed is high, and the detection accuracy is high.
On the basis of the foregoing embodiment, the step of selecting an outlier from the operating data points of the power system based on the DBSCAN algorithm in this embodiment specifically includes: selecting any one of the set of operational data points that is not clustered and is not labeled as an outlier; judging whether the selected operation data point is a core object; if not, judging whether the operation data point is an edge point or an outlier, and marking the operation data point as the edge point or the outlier according to a judgment result; if so, establishing a new cluster based on the selected operation data points, and adding the operation data points with the density which can be connected with the selected density of the operation data points into the new cluster until all the operation data points are classified into the cluster or marked as outliers.
Specifically, the set of operation data points is set to S ═ x (x)1,x2,…,xm) The operational data points are clustered based on the following definitions:
1) ε -neighborhood: any of the operational data points xjIs within a preset radius epsilon, epsilon is an input parameter, let | Nε(x) And | is the total number of run data points within the ε -neighborhood of any run data point x.
2) Core object: | Nε(x) The running data point x | ≧ MinPts. MinPts is a preset threshold value, which is an input parameter.
3) The direct density can reach: as shown in FIG. 2, point q is within ε -neighborhood of point p, and | Nε(xp) | is equal to or greater than MinPts, which means that q is directly density-reachable by object p.
4) The density can reach: as shown in fig. 3. Point pt+1At point ptAnd (t ═ 1,2) in the neighborhood, and the intermediate point p satisfies | Nε(xp) | is not less than MinPts, called p3From p1The density can be reached.
5) Density connection: as shown in FIG. 4, point o satisfies | Nε(xo) And | is more than or equal to MinPts, the point p can be reached by the density o, the point q can be reached by the density o, and the density p and the density q are connected.
The present embodiment classifies the operation data points in the operation data point set into three categories:
core point: namely a core object, and at least MinPts sample objects are contained in the epsilon-neighborhood;
edge points: the requirements of the core object are not met, but the density of the core object can reach;
outliers: i.e. suspected outliers which are neither core nor edge points.
The flow of the DBSCAN clustering algorithm is shown in fig. 5, and the processing steps of the DBSCAN clustering algorithm in this embodiment are as follows:
(1) inputting a running data point set S, neighborhood parameters (epsilon, MinPts) and a distance measurement mode;
(2) randomly selecting an operation data point p which is not accessed yet from S, and if | N is satisfiedε(xp) If | ≧ MinPts, turning to step (3), otherwise turning to step (2) after processing according to outliers or edge points;
(3) building a cluster based on the operating data point p, and adding an object which can be connected with the density by the density p into the cluster;
(4) and (4) continuously repeating the steps (2) and (3) until no unvisited operation data point exists in the operation data point set.
On the basis of the foregoing embodiment, the step of determining whether the selected operation data point is a core object in this embodiment specifically includes: calculating a direct distance between the selected operating data point and each of the other operating data points according to the value of each characteristic parameter of the selected operating data point and the values of each characteristic parameter of the other operating data points in the operating data point set except the selected operating data point; and acquiring the total number of other operation data points of which the direct distance is smaller than a preset radius, and if the total number is larger than a preset threshold, taking the selected operation data point as a core object.
Specifically, when the power system is a ship power system, a plurality of operation records of a voltage stabilizer in the ship power system are selected for abnormal point detection. The selected relevant characteristic parameters include pressure, water level, temperature, YKG temperature and PFG temperature. For convenience of expression, each characteristic parameter is replaced by a corresponding parameter serial number, and each parameter serial number corresponds to one measuring device, as shown in table 1.
TABLE 1 parameter number table
The characteristic parameter values of selected and other operational data points also need to be pre-processed before calculating the direct distances between them. Because the characteristic parameters of the voltage stabilizer have multiple redundancy measurements, the redundancy measurement values are preprocessed firstly, the influence of a fault measurement channel is eliminated firstly, and then the average value of the redundancy measurements is selected as the value of the characteristic parameters after the redundancy measurements are combined. For example, 24 and 25 are redundant measurements, 41 and 42 are redundant measurements, and 43, 44 and 45 are redundant measurements. The pretreatment is carried out on 8 characteristic parameters in total.
To eliminate the influence of the parameter dimension, normalization is performed according to the following formula.
Wherein,is the i-th feature x after normalizationiNormalized kth value, xikIs the ith feature xiHas a k value of, Max being the characteristic xiMin is a characteristic xiIs measured.
The distance between the running data points i and j is measured by the following euclidean distance:
the two running data points i, j are both n-8 dimensional, and the euclidean distance is the square root of the sum of the squares of the differences between the same attributes of the two running data points.
On the basis of the above embodiment, the present embodiment further includes: setting the preset radius and the preset threshold value for multiple times respectively; calculating discrete points corresponding to the preset radius and the preset threshold which are set each time; taking the operation data point determined as a discrete point each time as a final discrete point; correspondingly, the step of calculating the outlier degree of each outlier based on the LOF algorithm specifically includes: calculating the degree of outlier of each final outlier based on LOF algorithm.
In particular, although DBSCAN clustering is an unsupervised learning, the accuracy of the algorithm cannot be measured in false positives and false negatives. However, the running data of the power system always keeps stable in most of time, fluctuates only in a small range, the scale of abnormal data is small, and if the whole data is directly analyzed, the detection of most normal data is meaningless. Therefore, most of the data which cannot be abnormal can be eliminated by adjusting the parameters epsilon and MinPts of the DBSCAN algorithm, so that the scale of the abnormal data set to be analyzed is reduced, and the subsequent further abnormal point detection is facilitated.
The two input parameters ε and MinPts of the DBSCAN algorithm directly affect the number of operational data points that are eventually labeled as outliers. When epsilon decreases or MinPts increases, more outliers can be detected, and when the proportion of the outliers is high, false alarm may exist, namely, normal data is marked as the outliers; when the outlier percentage is low, there may be false positives, i.e., marking some outliers as normal. In order to reduce the missing report and the false report, algorithm parameters epsilon and MinPts are respectively set for a plurality of times. And classifying the operation data points which are divided into outliers each time as final discrete points into the suspected abnormal point data set.
Setting parameters epsilon and MinPts for multiple times, and performing experimentsThe obtained cases of the number of clusters and the outlier ratio are shown in tables 2 and 3, and it can be seen that in the case of the unchanged MinPts, due to the increase of ε, in step (2) of the DBSCAN algorithm, | N is satisfiedε(xp) The object p for the ≧ MinPts condition is increased so that more objects are added to the existing cluster. And, | N is satisfied when ε is not changed and MinPts is increasedε(xp) The object p for | ≧ MinPts conditions decreases, and phasing increases the number of outliers, such that the outlier ratio increases.
TABLE 2 percentage of outliers (%)
TABLE 3 number of clusters under different parameters
The number of clusters is 1 from multiple experiments, the operation data points which are divided into outliers in each experiment are the final discrete points, the operation data points are added into a suspected abnormal data set, and the size of the suspected abnormal data set is 143. Based on the principle of the DBSCAN algorithm, most of the different data points which are suspected to be abnormal are considered to be normal, and the subsequent LOF algorithm is utilized to detect the data points which are suspected to be abnormal.
On the basis of the foregoing embodiments, in this embodiment, the step of calculating the outlier of each of the outliers based on the LOF algorithm specifically includes: for any outlier, acquiring k running data points closest to the outlier in the running data point set; wherein k is a positive integer; obtaining the local reachable density of the outlier according to the reachable distance between each operating data point of the k operating data points and the outlier, and calculating the average value of the local reachable densities of the k operating data points; and taking the ratio of the average value of the local reachable densities of the k running data points to the local reachable density of the outlier as a local abnormal factor of the outlier, and taking the local abnormal factor as the outlier degree of the outlier.
The core concept of the local outlier LOF is as follows:
(1) the kth distance dk(p): indicating the distance to the kth point closest to point p.
(2) Reachable distance reach-dist (p, o): given parameter k, the reachable distance from data point p to data point o is the maximum of the kth distance from data point o and the direct distance between data points p and o, i.e. reach-dist (p, o) ═ max { d }k(o),d(p,o)}。
(3) Local achievable density lrdk(p) the calculation formula is as follows:
in the formula | Nk(p) | denotes the set of k points nearest to point p.
(4) Local anomaly factor LOFk(p) the calculation formula is as follows:
the flow chart of the LOF algorithm is shown in fig. 6.
On the basis of the above embodiment, the value range of the k value in this embodiment is:
k∈[klb,kub]＝[max{10,|0.01*|S||},|r*|S||]
wherein k islbIs the minimum value of k, kubIs the maximum value of k, | S | is the total number of the outliers, r is the ratio of the outliers in the set of operational data points; correspondingly, the step of calculating the outlier degree of each outlier based on the LOF algorithm specifically includes taking a plurality of values for k within the value range; for any discrete point, calculating the outlier degree of the outlier under each k value; and taking the maximum value of the outliers corresponding to all the k values of the outliers as the final outlier of the outliers.
This embodiment provides an electronic device, and fig. 7 is a schematic diagram of an overall structure of the electronic device according to the embodiment of the present invention, where the electronic device includes: at least one processor 701, at least one memory 702, and a bus 703; wherein,
the processor 701 and the memory 702 communicate with each other via a bus 703;
the memory 702 stores program instructions executable by the processor 701, and the processor calls the program instructions to perform the methods provided by the method embodiments, for example, the methods include: selecting outliers from a running data point set of the power system based on a DBSCAN algorithm; calculating the outlier degree of each outlier based on an LOF algorithm; and corresponding to any one of the outliers, determining whether the outlier is an abnormal point according to the outlier degree of the outlier.
The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the methods provided by the above method embodiments, for example, including: selecting outliers from a running data point set of the power system based on a DBSCAN algorithm; calculating the outlier degree of each outlier based on an LOF algorithm; and corresponding to any one of the outliers, determining whether the outlier is an abnormal point according to the outlier degree of the outlier.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The above-described system embodiments are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for detecting an abnormal operating point of a power system is characterized by comprising the following steps:
selecting outliers from a running data point set of the power system based on a DBSCAN algorithm;
calculating the outlier degree of each outlier based on an LOF algorithm;
and corresponding to any one of the outliers, determining whether the outlier is an abnormal point according to the outlier degree of the outlier.
2. The method for detecting the abnormal operating point of the power system according to claim 1, wherein the step of selecting the outlier from the operating data points of the power system based on the DBSCAN algorithm specifically comprises the following steps:
selecting any one of the set of operational data points that is not clustered and is not labeled as an outlier;
judging whether the selected operation data point is a core object;
if not, judging whether the operation data point is an edge point or an outlier, and marking the operation data point as the edge point or the outlier according to a judgment result;
if so, establishing a new cluster based on the selected operation data points, and adding the operation data points with the density which can be connected with the selected density of the operation data points into the new cluster until all the operation data points are classified into the cluster or marked as outliers.
3. The method for detecting the abnormal operating point of the power system according to claim 2, wherein the step of judging whether the selected operating data point is a core object specifically comprises the steps of:
calculating a direct distance between the selected operating data point and each of the other operating data points according to the value of each characteristic parameter of the selected operating data point and the values of each characteristic parameter of the other operating data points in the operating data point set except the selected operating data point;
and acquiring the total number of other operation data points of which the direct distance is smaller than a preset radius, and if the total number is larger than a preset threshold, taking the selected operation data point as a core object.
4. The power system operation abnormal point detection method according to claim 3, wherein when the power system is a ship power system, the characteristic parameters are characteristic parameters of a pressurizer in the ship power system, and include the pressure measured by each pressure test device, the water level measured by each water level measuring device, the temperature measured by each first temperature measuring device, the YLG temperature measured by each second temperature measuring device, and the PFG temperature measured by each third temperature measuring device.
5. The method of detecting a power system operation anomaly point according to claim 4, further comprising, prior to calculating a distance between the selected operational data point and each of the other operational data points:
acquiring redundant equipment in the pressure test equipment, the water level measurement equipment, the first temperature measurement equipment, the second temperature measurement equipment and the third temperature measurement equipment;
taking the average value of the values measured by the same redundant equipment as the value of one characteristic parameter;
the values of all the characteristic parameters are normalized.
6. The power system operation abnormality point detection method according to claim 3, characterized by further comprising:
setting the preset radius and the preset threshold value for multiple times respectively;
calculating discrete points corresponding to the preset radius and the preset threshold which are set each time;
taking the operation data point determined as a discrete point each time as a final discrete point;
correspondingly, the step of calculating the outlier degree of each outlier based on the LOF algorithm specifically includes:
calculating the degree of outlier of each final outlier based on LOF algorithm.
7. The power system operation anomaly point detection method according to any one of claims 1-6, wherein the step of calculating the degree of outlier of each of the outliers based on the LOF algorithm specifically comprises:
for any outlier, acquiring k running data points closest to the outlier in the running data point set; wherein k is a positive integer;
obtaining local reachable density of the outlier according to reachable distance between each operating data point of the k operating data points and the outlier, and calculating an average value of the local reachable density of the k operating data points;
and taking the ratio of the average value of the local reachable densities of the k running data points to the local reachable density of the outlier as a local abnormal factor of the outlier, and taking the local abnormal factor as the outlier degree of the outlier.
8. The method for detecting the operating abnormal point of the power system according to claim 7, wherein the value range of the k value is as follows:
k∈[klb,kub]＝[max{10,|0.01*|S||},|r*|S||]
wherein k islbIs the minimum value of k, kubIs the maximum value of k, | S | is the total number of the outliers, r is the ratio of the outliers in the set of operational data points;
correspondingly, the step of calculating the degree of outlier of each of the outliers based on the LOF algorithm specifically comprises
Taking a plurality of values for k within the value range;
for any one outlier, calculating the outlier degree of the outlier under each k value;
and taking the maximum value of the outliers corresponding to all the k values of the outliers as the final outlier of the outliers.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for detecting an abnormal operating point of a power system according to any one of claims 1 to 8 when executing the program.
10. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method for detecting an operating abnormal point of a power system according to any one of claims 1 to 8.
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 | |
| Liu et al. | Short-term prognostics of PEM fuel cells: A comparative and improvement study | |
| CN110083507B (en) | Key performance index classification method and device | |
| US7716152B2 (en) | Use of sequential nearest neighbor clustering for instance selection in machine condition monitoring | |
| WO2012147078A1 (en) | A system and a method for detecting abnormal occurrences | |
| CN111860698B (en) | Method and device for determining stability of learning model | |
| CN116663747B (en) | Intelligent early warning method and system based on data center infrastructure | |
| CN117193088B (en) | Industrial equipment monitoring method and device and server | |
| CN109711450A (en) | A kind of power grid forecast failure collection prediction technique, device, electronic equipment and storage medium | |
| CN113987027A (en) | Equipment state determination method, device and medium based on fault prediction | |
| CN110348005B (en) | Distribution network equipment state data processing method and device, computer equipment and medium | |
| CN113191432A (en) | Outlier factor-based virtual machine cluster anomaly detection method, device and medium | |
| CN117454324A (en) | Anemoclinograph fault early warning method, system, electronic equipment and storage medium | |
| CN116032016B (en) | Intelligent wind power plant running state on-line monitoring and early warning system and early warning method | |
| CN116365519A (en) | Power load prediction method, system, storage medium and equipment | |
| CN116245505A (en) | Line loss main factor analysis method and related equipment thereof | |
| CN115310516A (en) | Method and system for judging state stability of automation equipment and readable storage medium | |
| Hai-Jun et al. | Fuzzy entropy clustering using possibilistic approach | |
| CN114491416A (en) | Characteristic information processing method and device, electronic equipment and storage medium | |
| CN109436980A (en) | The condition detection method and system of elevator components | |
| CN116541252B (en) | Computer room fault log data processing method and device | |
| CN112651564A (en) | Machine learning-based dynamic baseline warning system and method for power internet of things | |
| CN117687996A (en) | Big data quality alarm detection method and system | |
| CN117171520A (en) | Data processing method, device and medium | |
| Li | An adaptive wavelet networks algorithm for prediction of gas delay outburst |
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 |