CN114912678A - Online automatic detection and early warning method and system for abnormal operation of power grid regulation and control - Google Patents
Online automatic detection and early warning method and system for abnormal operation of power grid regulation and control Download PDFInfo
- Publication number
- CN114912678A CN114912678A CN202210504901.9A CN202210504901A CN114912678A CN 114912678 A CN114912678 A CN 114912678A CN 202210504901 A CN202210504901 A CN 202210504901A CN 114912678 A CN114912678 A CN 114912678A
- Authority
- CN
- China
- Prior art keywords
- power grid
- abnormal operation
- grid regulation
- early warning
- log data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 129
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001514 detection method Methods 0.000 title claims abstract description 25
- 230000006399 behavior Effects 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 9
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 230000009471 action Effects 0.000 claims description 31
- 238000010276 construction Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 150000001875 compounds Chemical class 0.000 claims description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 abstract 2
- 238000013135 deep learning Methods 0.000 abstract 1
- 238000011156 evaluation Methods 0.000 description 9
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000007619 statistical method Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/909—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
-
- 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
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- 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)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Library & Information Science (AREA)
- Development Economics (AREA)
- Power Engineering (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an online automatic detection and early warning method and system for abnormal operation of power grid regulation and control. Firstly, a power grid regulation and control abnormal operation defining system is constructed, and abnormal operation behaviors of the power grid regulation and control system are classified and defined on the basis of log data of the power grid regulation and control system; secondly, extracting abnormal operation behavior characteristics by constructing a characteristic extraction model based on a deep learning method; and finally, establishing an online automatic advanced screening early warning model for the abnormal operation of the power grid regulation and control according to the characteristics of the abnormal operation behaviors, training and optimizing model parameters according to historical log data, evaluating the prediction performance of the model, and realizing accurate online advanced screening and early warning for the abnormal operation behaviors in the power grid regulation and control system. The invention monitors abnormal operation in real time, accurately and early discriminates and warns the abnormal operation which possibly occurs, helps a dispatcher to correct error operation in time, reduces the rate of man-made error operation, and improves the operation safety and stability of a power grid.
Description
Technical Field
The invention belongs to the field of power system scheduling control, and particularly relates to an online automatic detection and early warning method and system for abnormal operation of power grid regulation and control.
Background
With the continuous state of the power grid dispatching control operation informatization level, the core service operation of the intelligent power grid increasingly depends on the reliable operation of a regulation and control system, and the inestimable risk and loss can be brought by the incorrect operation of an operator. However, the traditional power grid regulation and control system has no discrimination and control capability on abnormal control behaviors, and the system cannot recognize and timely prevent the control behaviors when the abnormal control behaviors occur. The current power grid network security situation is severe, and the safety precaution measures of the system in the aspect of remote equipment operation control are urgently needed to be improved, so that major power grid accidents caused by external environment interference and personnel misoperation are avoided.
In the prior art, statistical analysis is carried out on the power grid regulation and control operation data, and the research on personnel operation data is lacked.
In the prior art 1(CN 109919448A) "method for intelligent statistical analysis and application of power grid regulation and control operation data", a statistical analysis time period is set; acquiring one of the power grid regulation and control operation data in a statistical analysis time period; the occurrence frequency and the numerical range of the power grid regulation and control operation data are statistically analyzed; comparing the occurrence frequency and the numerical range with the limit warning value of the power grid regulation and control operation data to obtain a comparison result; and obtaining an operation state statistical analysis table of the power grid regulation and control operation data according to the comparison result. The method and the device can monitor the power grid regulation and control operation data and realize the abnormal early warning of the power grid regulation and control operation data. The defect of the prior art document 1 is that the method only performs statistical analysis on the occurrence frequency and the range of a certain item of data in the power grid regulation and control operation process, does not consider important factors such as the time and the place of the regulation and control action, and is not beneficial to finding abnormal risks caused by the wrong operation of a dispatcher or the illegal operation of non-professional personnel.
In the prior art 2(CN113807690A) "online evaluation and early warning method and system for the operation state of a regional power grid regulation and control system", the operation data of the regional power grid regulation and control system is collected and stored in a standardized manner; carrying out data preprocessing, and carrying out data priority marking and data correctness marking; determining the weight of the evaluation index, and determining the evaluation level of the system running state; constructing an online evaluation early warning model of the operation state of the power grid regulation and control system; and realizing the operation state evaluation early warning model training of the power grid regulation and control system according to a priority evaluation principle, and carrying out online evaluation early warning on the operation state of the power grid regulation and control system by adopting the trained evaluation early warning model. The defect of the prior art document 2 is that the technology mainly performs early warning evaluation on abnormal data in the operation state of the power grid regulation and control system, and does not relate to the operation action scope of a dispatcher. Manual operations can create varying degrees of operational risk due to dispatcher personnel factors and job task maturity. The early warning of such risks does not fall within the scope of prior art document 2.
In summary, at present, research on the problem of abnormal operation of power grid regulation and control is still lacking, and a method capable of accurately early warning abnormal operation of a power grid regulation and control system is urgently needed.
Disclosure of Invention
In order to solve the defects in the prior art, the invention aims to provide an online automatic detection and early warning method for abnormal operation of power grid regulation. Meanwhile, according to the historical data of the regulation and control operation, the early warning information of the abnormal operation is provided in time.
The invention adopts the following technical scheme. The online automatic detection and early warning method for abnormal operation of power grid regulation comprises the following steps:
step 1, collecting operation log data of a regional power grid regulation and control system, and performing expansion processing on the operation log data;
step 2, defining operation log data based on three factors of time, place and action behavior to obtain abnormal operation log data;
step 3, according to the abnormal operation log data, extracting the abnormal operation log characteristics from three aspects of dispatcher attribute characteristics, operation behavior characteristics and service characteristics;
step 4, adopting the structure of CNN (Convolutional Neural Networks) and LSTM (Long Short Term Neural Networks) to establish a power grid regulation and control abnormal operation early warning model, and training power grid regulation and control abnormal operation early warning model parameters by using historical operation log data;
and 5, adopting the trained power grid regulation and control abnormal operation early warning model, and utilizing the power grid regulation and control operation log data acquired in real time to perform online automatic detection and early warning on the abnormal operation of the power grid regulation and control system.
Preferably, in step 1, the operation log data includes: motion data, equipment data, site data, and personnel data.
In step 1, the operation log data is expanded, wherein personnel data characteristics in the operation log data are expanded, and the expanded personnel data characteristics comprise: name, job number, age, job level and operating age.
Preferably, in step 2, the operation log data is defined based on a time factor, and the obtained abnormal operation log data includes: log-in time exception, operation time exception, and operation time exception.
In step 2, the operation log data is defined based on the location factor, and the obtained abnormal operation log data further includes: the normal login place is abnormal, and the specific operation login place is abnormal.
In step 2, the operation log data is defined based on the action behavior factors, and the obtained abnormal operation log data further comprises: normal operation actions and abnormal operation actions.
Preferably, in step 3, the dispatcher attribute characteristics include: age, job level and operational age;
the dispatcher operation behavior characteristics comprise: login time, login place, operation time and operation action;
the operator service features include: service class, service type and service maturity.
Preferably, in step 4, the model comprises an input layer, a convolutional layer, a pooling layer, an LSTM layer, a full link layer and an output layer;
the convolutional layer and the pooling layer in the CNN are used for extracting space and motion characteristics, and the LSTM layer is used for extracting time characteristics hidden in a space and motion characteristic sequence;
the model used the mean absolute error as a loss function and an Adam optimizer.
Preferably, in step 5, the power grid regulation and control operation log data collected in real time is utilized, the residual error between the predicted value and the normal value is solved by the power grid regulation and control abnormal operation early warning model, and the power grid regulation and control abnormal operation early warning is completed by analyzing the difference between the residual error sequence and the normal data sample.
In step 5, calculating a residual RMSE value between the predicted value and the true value according to the following relation:
in the formula (I), the compound is shown in the specification,
m represents the total number of sample data,
t j representing the real value of the jth power grid regulation historical operation log data sample,
e j representing the predicted value of the jth power grid regulation historical operation log data sample,
and continuously correcting parameters of the power grid regulation abnormal operation early warning model, and stopping correction when residual error RMSE reaches the standard.
In step 5, when the average deviation of the differential residual sequence and the normal sample is less than 10%, the operation is considered to have no abnormal risk;
and when the average deviation of the difference residual sequence and the normal sample is more than 10%, judging that the abnormal operation risk exists, and sending an abnormal early warning signal.
Electric wire netting regulation and control abnormal operation on-line automated inspection early warning system includes: the system comprises a data acquisition module, a data classification module, a feature extraction module, a construction model module and an online early warning module, and is characterized in that:
the data acquisition module is used for acquiring operation log data of the regional power grid regulation and control system and performing expansion processing on the operation log data;
the data classification module is used for classifying and defining the operation log data based on three factors of time, place and action behavior to obtain abnormal operation log data;
the characteristic extraction module is used for extracting the characteristics of the abnormal operation log from three aspects of dispatcher attribute characteristics, operation behavior characteristics and service characteristics according to the abnormal operation log data;
the construction model module is used for constructing a power grid regulation and control abnormal operation early warning model, and training and optimizing model parameters according to historical operation log data;
the online early warning module is used for carrying out online automatic advanced discrimination early warning on abnormal operation behaviors of the power grid regulation and control system by adopting the power grid regulation and control abnormal operation early warning model.
The invention has the advantages that compared with the prior art,
according to the method, personnel operation data are used as research objects, and correctness analysis and risk early warning are carried out on scheduling operation according to log data generated in the operation process of the power grid regulation and control system.
The method classifies and defines the abnormal operation behaviors of the power grid based on three aspects of time, place and action behaviors, and avoids one-sidedness and incompleteness caused by abnormal identification only based on certain data. The operation classification and the exception definition are carried out from a plurality of aspects which can generate the exception operation, so that the exception operation in the scheduling can be more comprehensively and accurately identified.
Meanwhile, according to the characteristics of manual operation actions in the power grid regulation and control system, attribute characteristics are extracted from three aspects of a dispatcher, operation behaviors and service types, and an abnormal operation prediction model is constructed. The method has the advantages that factors accumulated by experience of dispatchers are considered, differences of manual regulation and control operations in tasks with different maturity levels are considered, internal relations between the dispatchers and the dispatching tasks are comprehensively analyzed, accurate early warning aiming at high-risk operations is made, safety of a power grid regulation and control system is guaranteed, and major power grid accidents caused by abnormal operations are avoided.
Drawings
FIG. 1 is a flow chart of an on-line automatic detection and early warning method for abnormal operation of power grid regulation and control according to the invention;
fig. 2 is a power grid regulation abnormal operation early warning model architecture diagram.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only used to illustrate the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
Example 1.
As shown in fig. 1, the online automatic detection and early warning method for abnormal operation of power grid regulation comprises the following steps:
step 1, collecting operation log data of a regional power grid regulation and control system, wherein the operation log data comprises action data, equipment data, field data and personnel data, and expanding the data. And expanding the personnel data into a job number, an age, a job level and an operation age, and carrying out standardized processing.
And 2, defining the operation log data based on three factors of time, place and action behavior to obtain abnormal operation log data.
And 2.1, defining abnormal operation of power grid regulation and control based on time factors, wherein the obtained abnormal operation log data comprises but is not limited to login time abnormity, operation time abnormity and operation time abnormity.
The normal login time of the power grid regulation and control system is (TL) in -△TL)～(TL out +. DELTA TL), wherein TL in Indicating the time of logging into the grid regulation system, TL out Indicating the time of exit from the grid regulation system and Δ TL indicating the time period during which logging into the grid regulation system is allowed.
If the log data has the time for logging in the power grid regulation and control system
Is defined as an abnormal log-in time.
The normal operation time of the service of the power grid regulation and control system is (TO) start -△TO)～(TO finish +. DELTA TO), wherein TO start Indicating the start time of operation, TO, of the grid regulation system finish And the delta TO represents the operation ending time of the power grid regulation and control system, and the business normal operation time period of the power grid regulation and control system.
If the log data has the login time
Is defined as an operational time anomaly.
Service operation time of the power grid regulation and control system is (TO) finish -TO start )。
If it appears in the log data (TO) finish -TO start )<0.5 Δ TO Or (TO) finish -TO start )>2 Δ TO is defined as an operational anomaly.
And 2.2, defining abnormal operation of power grid regulation and control based on site factors, wherein the obtained abnormal operation log data include but are not limited to normal login site abnormality and specific operation login site abnormality.
And IP description is adopted to represent site factors of abnormal operation of power grid regulation. The login IP address set of the power grid regulation and control normal operation is IP all If log data is registered with IP address
Is defined as a regular login location exception. For anSetting the login IP address set of the power grid regulation and control specific operation as IP s If log data is registered with IP address
Is defined as a specific operation login location exception.
And 2.3, defining the abnormal operation of the power grid regulation and control based on the action behavior factors, wherein the obtained abnormal operation log data comprises but is not limited to normal operation actions and abnormal operation actions.
And measuring the difference degree between the two log data samples according to the distance leaving of the log data samples of the same operation action by adopting a K-means clustering method. Two log data samples x of the same operation action are calculated in the following relation i And x j Euler distance therebetween:
in the formula (I), the compound is shown in the specification,
k represents the total number of attributes contained in each log data,
m represents the m-th attribute and is,
x i one log data sample representing the operation action x,
x j another log data sample representing an operation action x,
x im representing log data sample x i The (m) th attribute of (2),
x jm representing log data sample x j The mth attribute of (2).
And determining two log data samples with the nearest distance as a center sample, arranging the distances between the rest log data samples and the center sample from small to large, defining the log data sample close to the center sample as a normal operation action, and defining the log data sample far away from the center sample as an abnormal operation action.
And 3, extracting operation characteristics from three aspects of dispatcher attribute characteristics, operation behavior characteristics and service characteristics according to operation log data causing abnormal operation of power grid regulation and control.
And the four characteristics of the job number, the age, the job level and the operation age obtained by the personnel data expansion are used as the dispatcher attribute characteristics. Five characteristics of the dispatcher registration time, the registration place, the operation time and the operation action are taken as operation behavior characteristics. And taking the service grade, the service type and the service maturity as service characteristics.
And 4, constructing and establishing a power grid regulation and control abnormal operation early warning model by adopting the CNN and LSTM structures, and training power grid regulation and control abnormal operation early warning model parameters by using historical operation log data.
A power grid regulation and control abnormal operation early warning model is designed by adopting a structure of a Convolutional Neural Network (CNN) + a Long Short Term Neural network (LSTM), wherein the model comprises an input layer, a Convolutional layer, a pooling layer, an LSTM layer, a full connection layer and an output layer, and is shown in figure 2. Convolutional and pooling layers in CNN networks can extract spatial and motion features, and LSTM layers can be used to extract temporal features hidden in the spatial and motion feature sequences. Mean Absolute Error (MAE) was used as a loss function and Adam optimizer.
And training and optimizing the parameters of the power grid regulation abnormal operation early warning model by using the power grid regulation historical operation log data.
And 5, adopting the trained power grid regulation and control abnormal operation early warning model, and utilizing the power grid regulation and control operation log data acquired in real time to perform online automatic detection and early warning on the abnormal operation of the power grid regulation and control system.
After the power grid regulation and control operation log data collected in real time are screened and standardized, a residual error between a predicted value and a normal value is solved by a power grid regulation and control abnormal operation early warning model, and the power grid regulation and control abnormal operation early warning is completed by analyzing the difference between a difference residual error sequence and a normal data sample.
Meanwhile, in order to further improve the prediction accuracy, in the initial operation stage of the system, the residual RMSE value between the predicted value and the true value is calculated according to the following relation:
in the formula (I), the compound is shown in the specification,
m represents the total number of sample data,
t j representing the real value of the jth power grid regulation historical operation log data sample,
e j and representing the predicted value of the jth power grid regulation historical operation log data sample.
And continuously correcting parameters of the power grid regulation abnormal operation early warning model, and stopping correction when residual error RMSE reaches the standard.
And predicting the abnormal operation of the power grid regulation according to the power grid regulation operation log data acquired in real time, and if the predicted value deviates from a normal range, determining that the abnormal operation of the power grid regulation is possible, and sending an early warning signal of the abnormal operation of the power grid regulation.
Preferably, in this embodiment, when the average deviation of the differential residual sequence from the normal sample is less than 10%, the operation is considered to have no abnormal risk; and when the average deviation of the difference residual sequence and the normal sample is more than 10%, judging that the abnormal operation risk exists, and sending an abnormal early warning signal.
Example 2.
An online automatic detection early warning system for abnormal operation of power grid regulation and control. The method comprises the following steps: the system comprises a data acquisition module, a data classification module, a feature extraction module, a construction model module and an online early warning module, wherein:
the data acquisition module is used for acquiring operation log data of the regional power grid regulation and control system and performing expansion processing on the operation log data;
the data classification module is used for classifying and defining the operation log data based on three factors of time, place and action behavior to obtain abnormal operation log data;
the characteristic extraction module is used for extracting the characteristics of the abnormal operation log from three aspects of dispatcher attribute characteristics, operation behavior characteristics and service characteristics according to the abnormal operation log data;
the construction model module is used for constructing a power grid regulation and control abnormal operation early warning model, and training and optimizing model parameters according to historical operation log data;
the online early warning module is used for carrying out online automatic advanced discrimination early warning on abnormal operation behaviors of the power grid regulation and control system by adopting the power grid regulation and control abnormal operation early warning model.
Compared with the prior art, the method has the advantages that personnel operation data are used as research objects, and accuracy analysis and risk early warning are carried out on scheduling operation according to log data generated in the operation process of the power grid regulation and control system.
The method classifies and defines the abnormal operation behaviors of the power grid based on three aspects of time, place and action behaviors, and avoids one-sidedness and incompleteness caused by abnormal identification only based on certain data. The operation classification and the exception definition are carried out from a plurality of aspects which can generate the exception operation, so that the exception operation in the scheduling can be more comprehensively and accurately identified.
Meanwhile, according to the characteristics of manual operation actions in the power grid regulation and control system, attribute characteristics are extracted from three aspects of a dispatcher, operation behaviors and service types, and an abnormal operation prediction model is constructed. The method has the advantages that factors accumulated by experience of dispatchers are considered, differences of manual regulation and control operations in tasks with different maturity levels are considered, internal relations between the dispatchers and the dispatching tasks are comprehensively analyzed, accurate early warning aiming at high-risk operations is made, safety of a power grid regulation and control system is guaranteed, and major power grid accidents caused by abnormal operations are avoided.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are merely preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for limiting the scope of the present invention, and on the contrary, any improvement or modification made based on the spirit of the present invention should fall within the scope of the present invention.
Claims (12)
1. The online automatic detection and early warning method for abnormal operation of power grid regulation and control comprises the following steps:
step 1, collecting operation log data of a regional power grid regulation and control system, and performing expansion processing on the operation log data;
step 2, defining operation log data based on three factors of time, place and action behavior to obtain abnormal operation log data;
step 3, according to the abnormal operation log data, extracting the abnormal operation log characteristics from three aspects of dispatcher attribute characteristics, operation behavior characteristics and service characteristics;
step 4, establishing a power grid regulation abnormal operation early warning model by adopting the CNN and LSTM structures, and training power grid regulation abnormal operation early warning model parameters by using historical operation log data;
and 5, adopting the trained power grid regulation and control abnormal operation early warning model, and utilizing the power grid regulation and control operation log data acquired in real time to perform online automatic detection and early warning on the abnormal operation of the power grid regulation and control system.
2. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 1, wherein:
in step 1, the operation log data includes: motion data, equipment data, site data, and personnel data.
3. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 2, characterized in that:
in step 1, the operation log data is expanded, wherein personnel data characteristics in the operation log data are expanded, and the expanded personnel data characteristics comprise: name, job number, age, job level and operating age.
4. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 1, wherein:
in step 2, the operation log data is defined based on time factors, and the obtained abnormal operation log data comprises: log-in time exception, operation time exception and operation time exception.
5. The online automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 4, wherein:
in step 2, the operation log data is defined based on the location factor, and the obtained abnormal operation log data further includes: a normal login place exception and a specific operation login place exception.
6. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 5, wherein:
in step 2, the operation log data is defined based on the action behavior factors, and the obtained abnormal operation log data further comprises: normal operation actions and abnormal operation actions.
7. The online automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 6, characterized in that:
in the step 3, the step of the method is that,
dispatcher attribute characteristics include: age, job level and operational age;
the dispatcher operation behavior characteristics comprise: login time, login place, operation time and operation action;
the operator service features include: service class, service type and service maturity.
8. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 7, wherein:
in step 4, the model comprises an input layer, a convolution layer, a pooling layer, an LSTM layer, a full-link layer and an output layer;
the convolutional layer and the pooling layer in the CNN are used for extracting space and motion characteristics, and the LSTM layer is used for extracting time characteristics hidden in a space and motion characteristic sequence;
the model used the mean absolute error as a loss function and an Adam optimizer.
9. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 8, wherein:
and 5, solving a residual error between a predicted value and a normal value by using the power grid regulation and control operation log data acquired in real time through a power grid regulation and control abnormal operation early warning model, and completing power grid regulation and control abnormal operation early warning by analyzing the difference between a residual error sequence and a normal data sample.
10. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 9, wherein:
in step 5, calculating a residual RMSE value between the predicted value and the true value according to the following relation:
in the formula (I), the compound is shown in the specification,
m represents the total number of sample data,
t j representing the real value of the jth power grid regulation historical operation log data sample,
e j representing the predicted value of the jth power grid regulation historical operation log data sample,
and continuously correcting parameters of the power grid regulation abnormal operation early warning model, and stopping correction when residual error RMSE reaches the standard.
11. The on-line automatic detection and early warning method for abnormal operation of power grid regulation and control as claimed in claim 10, wherein:
in step 5, when the average deviation of the differential residual sequence and the normal sample is less than 10%, the operation is considered to have no abnormal risk;
and when the average deviation of the difference residual sequence and the normal sample is more than 10%, judging that the abnormal operation risk exists, and sending an abnormal early warning signal.
12. The online automatic detection and early warning system for the abnormal operation of power grid regulation is used for realizing the online automatic detection and early warning method for the abnormal operation of power grid regulation according to any one of claims 1 to 11, and comprises the following components: the system comprises a data acquisition module, a data classification module, a feature extraction module, a construction model module and an online early warning module, and is characterized in that:
the data acquisition module is used for acquiring operation log data of the regional power grid regulation and control system and performing expansion processing on the operation log data;
the data classification module is used for classifying and defining the operation log data based on three factors of time, place and action behavior to obtain abnormal operation log data;
the characteristic extraction module is used for extracting the characteristics of the abnormal operation log from three aspects of dispatcher attribute characteristics, operation behavior characteristics and service characteristics according to the abnormal operation log data;
the construction model module is used for constructing a power grid regulation and control abnormal operation early warning model, and training and optimizing model parameters according to historical operation log data;
the online early warning module is used for carrying out online automatic advanced discrimination early warning on abnormal operation behaviors of the power grid regulation and control system by adopting the power grid regulation and control abnormal operation early warning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210504901.9A CN114912678A (en) | 2022-05-10 | 2022-05-10 | Online automatic detection and early warning method and system for abnormal operation of power grid regulation and control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210504901.9A CN114912678A (en) | 2022-05-10 | 2022-05-10 | Online automatic detection and early warning method and system for abnormal operation of power grid regulation and control |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114912678A true CN114912678A (en) | 2022-08-16 |
Family
ID=82766923
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210504901.9A Pending CN114912678A (en) | 2022-05-10 | 2022-05-10 | Online automatic detection and early warning method and system for abnormal operation of power grid regulation and control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114912678A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117148802A (en) * | 2023-10-27 | 2023-12-01 | 广州汇通国信科技有限公司 | Unmanned workshop intelligent supervision system of power plant based on artificial intelligence |
| CN117710832A (en) * | 2024-01-04 | 2024-03-15 | 广州智寻科技有限公司 | Intelligent identification method for power grid satellite, unmanned aerial vehicle and video monitoring image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107528832A (en) * | 2017-08-04 | 2017-12-29 | 北京中晟信达科技有限公司 | Baseline structure and the unknown anomaly detection method of a kind of system-oriented daily record |
| CN110674021A (en) * | 2019-09-09 | 2020-01-10 | 深圳供电局有限公司 | Detection method and system for login log of mobile application |
| CN111639497A (en) * | 2020-05-27 | 2020-09-08 | 北京东方通科技股份有限公司 | Abnormal behavior discovery method based on big data machine learning |
| CN112581283A (en) * | 2020-12-28 | 2021-03-30 | 中国建设银行股份有限公司 | Method and device for analyzing and alarming business bank employee transaction behaviors |
| CN113285962A (en) * | 2021-07-21 | 2021-08-20 | 南方电网数字电网研究院有限公司 | Online operation monitoring method and system |
-
2022
- 2022-05-10 CN CN202210504901.9A patent/CN114912678A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107528832A (en) * | 2017-08-04 | 2017-12-29 | 北京中晟信达科技有限公司 | Baseline structure and the unknown anomaly detection method of a kind of system-oriented daily record |
| CN110674021A (en) * | 2019-09-09 | 2020-01-10 | 深圳供电局有限公司 | Detection method and system for login log of mobile application |
| CN111639497A (en) * | 2020-05-27 | 2020-09-08 | 北京东方通科技股份有限公司 | Abnormal behavior discovery method based on big data machine learning |
| CN112581283A (en) * | 2020-12-28 | 2021-03-30 | 中国建设银行股份有限公司 | Method and device for analyzing and alarming business bank employee transaction behaviors |
| CN113285962A (en) * | 2021-07-21 | 2021-08-20 | 南方电网数字电网研究院有限公司 | Online operation monitoring method and system |
Non-Patent Citations (3)
| Title |
|---|
| 谭娟: "《网络创新社区用户知识重混创新研究》", 31 July 2021, 知识产权出版社, pages: 156 * |
| 郭娜 等: ""基于用户画像的大数据环境中异常特征提取"", 《计算机仿真》, vol. 37, no. 8, pages 332 - 336 * |
| 郭娜;魏荣凯;沈焱萍;: "基于用户画像的大数据环境中异常特征提取", 计算机仿真, no. 08, pages 332 - 336 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117148802A (en) * | 2023-10-27 | 2023-12-01 | 广州汇通国信科技有限公司 | Unmanned workshop intelligent supervision system of power plant based on artificial intelligence |
| CN117148802B (en) * | 2023-10-27 | 2024-03-22 | 广州汇通国信科技有限公司 | Unmanned workshop intelligent supervision system of power plant based on artificial intelligence |
| CN117710832A (en) * | 2024-01-04 | 2024-03-15 | 广州智寻科技有限公司 | Intelligent identification method for power grid satellite, unmanned aerial vehicle and video monitoring image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110647539B (en) | Prediction method and system for vehicle faults | |
| US9384603B2 (en) | Failure cause classification apparatus | |
| CN111507376B (en) | Single-index anomaly detection method based on fusion of multiple non-supervision methods | |
| CN114912678A (en) | Online automatic detection and early warning method and system for abnormal operation of power grid regulation and control | |
| CN112884325A (en) | Method and system for application analysis and health condition evaluation of customer station equipment | |
| CN110851422A (en) | Data anomaly monitoring model construction method based on machine learning | |
| CN115358155A (en) | Power big data abnormity early warning method, device, equipment and readable storage medium | |
| CN108763966B (en) | Tail gas detection cheating supervision system and method | |
| CN110580492A (en) | Track circuit fault precursor discovery method based on small fluctuation detection | |
| CN116457802A (en) | Automatic real-time detection, prediction and prevention of rare faults in industrial systems using unlabeled sensor data | |
| CN112257914B (en) | Aviation safety causal prediction method based on random forest | |
| JP2019028834A (en) | Abnormal value diagnostic device, abnormal value diagnostic method, and program | |
| CN113962308A (en) | Aviation equipment fault prediction method | |
| CN117032165A (en) | Industrial equipment fault diagnosis method | |
| CN117151649A (en) | Construction method management and control system and method based on big data analysis | |
| CN113468022B (en) | Automatic operation and maintenance method for centralized monitoring of products | |
| KR20190098484A (en) | System and Method for Fault Isolation in Industrial Processes using CART based variable ranking | |
| CN117408519A (en) | Chemical reaction process risk early warning method and device based on deep learning algorithm | |
| CN116910662A (en) | Passenger anomaly identification method and device based on random forest algorithm | |
| CN115683504B (en) | Bridge acceleration monitoring data anomaly identification method and system based on multi-label classification | |
| Aftabi et al. | A Variational Autoencoder Framework for Robust, Physics-Informed Cyberattack Recognition in Industrial Cyber-Physical Systems | |
| CN114492636A (en) | Transformer winding state signal acquisition system | |
| CN113392576A (en) | Staying balloon main cable tension state assessment early warning method | |
| CN110569277A (en) | Method and system for automatically identifying and classifying configuration data information | |
| CN117688485B (en) | Fire disaster cause analysis method and system based on deep learning |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240205 Address after: 215004 No. 555 Labor Road, Gusu District, Suzhou City, Jiangsu Province Applicant after: STATE GRID JIANGSU ELECTRIC POWER Co.,Ltd. SUZHOU BRANCH Country or region after: China Applicant after: NR ENGINEERING Co.,Ltd. Applicant after: STATE GRID JIANGSU ELECTRIC POWER Co.,Ltd. Address before: 215004 No. 555 Labor Road, Gusu District, Suzhou City, Jiangsu Province Applicant before: STATE GRID JIANGSU ELECTRIC POWER Co.,Ltd. SUZHOU BRANCH Country or region before: China Applicant before: NARI TECHNOLOGY Co.,Ltd. Applicant before: NR ENGINEERING Co.,Ltd. Applicant before: STATE GRID JIANGSU ELECTRIC POWER Co.,Ltd. |
|
| TA01 | Transfer of patent application right |