CN110084448A - A kind of analysis method of electric system, device and its equipment - Google Patents
A kind of analysis method of electric system, device and its equipment Download PDFInfo
- Publication number
- CN110084448A CN110084448A CN201810067602.7A CN201810067602A CN110084448A CN 110084448 A CN110084448 A CN 110084448A CN 201810067602 A CN201810067602 A CN 201810067602A CN 110084448 A CN110084448 A CN 110084448A
- Authority
- CN
- China
- Prior art keywords
- power
- layout
- grid
- grid nodes
- project
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 103
- 238000000034 method Methods 0.000 claims abstract description 108
- 238000012549 training Methods 0.000 claims abstract description 66
- 230000005611 electricity Effects 0.000 claims description 74
- 230000008569 process Effects 0.000 claims description 61
- 230000002159 abnormal effect Effects 0.000 claims description 43
- 238000010801 machine learning Methods 0.000 claims description 40
- 238000004422 calculation algorithm Methods 0.000 claims description 24
- 238000010248 power generation Methods 0.000 claims description 24
- 238000004040 coloring Methods 0.000 claims description 18
- 238000003860 storage Methods 0.000 claims description 12
- 239000004744 fabric Substances 0.000 claims description 11
- 235000013399 edible fruits Nutrition 0.000 claims description 2
- 238000007726 management method Methods 0.000 description 38
- 238000010586 diagram Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 230000003068 static effect Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 239000003086 colorant Substances 0.000 description 3
- 230000015654 memory Effects 0.000 description 3
- 241001282153 Scopelogadus mizolepis Species 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 210000004209 hair Anatomy 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 208000009119 Giant Axonal Neuropathy Diseases 0.000 description 1
- 230000002547 anomalous effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 201000003382 giant axonal neuropathy 1 Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- 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
- G06Q10/20—Administration of product repair or maintenance
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application provides analysis method, device and its equipment of a kind of electric system, this method comprises: obtaining the first power project；The first grid nodes layout is obtained according to first power project；According to the label value of first grid nodes layout and the first grid nodes layout, the corresponding relationship of training power grid feature and characteristic value；Wherein, the corresponding relationship is used to analyze the state of electric system.By the technical solution of the application, quick Contingency Analysis of Power Systems can be carried out, supports power planning and scheduling.The time that safety analysis can be saved saves the computing resource of safety analysis.
Description
Technical field
This application involves power domain, especially a kind of analysis method of electric system, device and its equipment.
Background technique
In the power system, it usually needs static security analysis is carried out, to evaluate the safety and stabilization of electric system
Property.N-1 safety criterion is the important means of static security analysis, and realization principle is: in N number of node of electric system, having and appoints
When one node is disconnected because of failure, electric system being capable of stable operation and normal power supply.
In order to realize static security analysis using N-1 safety criterion, electricity can be successively disconnected in such a way that violence is enumerated
Each node of Force system, and safety analysis is carried out to electric system, whole process needs take a substantial amount of time and calculate money
Source.For example, when electric system is there are each node for when 3000 nodes, needing successively to disconnect in this 3000 nodes, and it is right
Electric system carries out safety analysis, that is, needs at least 3000 analytic processes, takes a substantial amount of time and computing resource, such as may be used
It can need to expend at least 70 minutes.
Summary of the invention
The application provides a kind of analysis method of electric system, which comprises
Obtain the first power project；
The first grid nodes layout is obtained according to first power project；
According to the label value of first grid nodes layout and the first grid nodes layout, training power grid feature and feature
The corresponding relationship of value；Wherein, the corresponding relationship is used to analyze the state of electric system.
The application provides a kind of analysis method of electric system, which comprises
Obtain the second power project for needing to analyze state；
Third grid nodes layout is obtained according to second power project；
The power grid characteristic query corresponding relationship being laid out by the third grid nodes, obtains corresponding with the power grid feature
Characteristic value；Wherein, the corresponding relationship is the corresponding relationship of the power grid feature and characteristic value according to the first power project training；
The state of the second power project according to obtained Eigenvalues analysis.
The application provides a kind of analysis method of electric system, which comprises
Obtain the third power project for needing to analyze state；
Inquiring in normal sample library whether there is and the matched normal electricity plan of the third power project；
If so, the state for analyzing the third power project is normal；
If it is not, the state for then analyzing the third power project is abnormal.
The application provides a kind of analytical equipment of electric system, and described device includes:
Module is obtained, for obtaining the first power project；The first grid nodes cloth is obtained according to first power project
Office；
Training module, for the label value according to first grid nodes layout and the first grid nodes layout, training
The corresponding relationship of power grid feature and characteristic value；The corresponding relationship is used to analyze the state of electric system.
The application provides a kind of analytical equipment of electric system, and described device includes:
Module is obtained, needs to analyze the second power project of state for obtaining；It is obtained according to second power project
Third grid nodes layout；The power grid characteristic query corresponding relationship being laid out by the third grid nodes, obtains and the electricity
The corresponding characteristic value of net feature；Wherein, the corresponding relationship is the power grid feature and characteristic value according to the first power project training
Corresponding relationship；
Analysis module, the state for the second power project according to obtained Eigenvalues analysis.
The application provides a kind of analytical equipment of electric system, and described device includes:
Module is obtained, needs to analyze the third power project of state for obtaining；
Enquiry module whether there is and the matched normal electricity of third power project for inquiring in normal sample library
Plan；
Analysis module, for when query result, which is, is, then it to be normal for analyzing the state of the third power project；When looking into
When inquiry result is no, then it is abnormal for analyzing the state of the third power project.
The application provides a kind of power management apparatus, and the power management apparatus includes:
Processor, for obtaining the first power project；The first grid nodes layout is obtained according to first power project；
According to the label value of first grid nodes layout and first grid nodes layout, training power grid feature and characteristic value
Corresponding relationship；Wherein, the corresponding relationship is used to analyze the state of electric system.
The application provides a kind of power management apparatus, and the power management apparatus includes: processor, needs to divide for obtaining
Second power project of analysis state；Third grid nodes layout is obtained according to second power project；Pass through the third electricity
The power grid characteristic query corresponding relationship for netting node layout, obtains characteristic value corresponding with the power grid feature；Wherein, the correspondence
Relationship is the corresponding relationship of the power grid feature and characteristic value according to the first power project training；According to obtained Eigenvalues analysis institute
State the state of the second power project.
The application provides a kind of power management apparatus, and the power management apparatus includes:
Processor needs to analyze the third power project of state for obtaining；Inquire normal sample library in the presence or absence of with
The matched normal electricity plan of third power project；If so, the state for analyzing the third power project is normal；If
No, then it is abnormal for analyzing the state of the third power project.
Based on the above-mentioned technical proposal, in the embodiment of the present application, the corresponding relationship of power grid feature and characteristic value can be trained,
And the state of electric system is analyzed according to the corresponding relationship, and Electric Power Network Planning, traffic control and inspection are instructed based on analysis result
Power failure plan is repaired, quick Contingency Analysis of Power Systems can be carried out, supports power planning and scheduling.It does not need using sudden and violent
The mode that power is enumerated successively disconnects each node of electric system, can save the time of safety analysis, save safety analysis
Computing resource.Safety analysis can be carried out in a short time, operation power personnel is made quickly to make scheduling decision, avoided big
Scale power failure.
Detailed description of the invention
It, below will be to the application in order to clearly illustrate the embodiment of the present application or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing is only some embodiments as described in this application, for those of ordinary skill in the art, can also be according to this Shen
Please these attached drawings of embodiment obtain other attached drawings.
Figure 1A and Figure 1B is the application scenarios schematic diagram in a kind of embodiment of the application；
Fig. 2A is the flow chart of the analysis method of the electric system in a kind of embodiment of the application；
Fig. 2 B is a schematic diagram of the topological structure of electric in a kind of embodiment of the application；
Fig. 2 C is a schematic diagram of the first intermediate node layout in a kind of embodiment of the application；
Fig. 2 D is the flow chart of the analysis method of the electric system in a kind of embodiment of the application；
Fig. 2 E is the flow chart of the analysis method of the electric system in the application another embodiment；
Fig. 3 is the flow chart of the analysis method of the electric system in the application another embodiment；
Fig. 4 is the structure chart of the analytical equipment of the electric system in a kind of embodiment of the application；
Fig. 5 is the structure chart of the analytical equipment of the electric system in the application another embodiment；
Fig. 6 is the structure chart of the analytical equipment of the electric system in the application another embodiment.
Specific embodiment
In the term that the embodiment of the present application uses merely for the sake of for the purpose of describing particular embodiments, rather than limit this Shen
Please.The "an" of singular used in the application and claims, " described " and "the" are also intended to including most shapes
Formula, unless context clearly shows that other meanings.It is also understood that term "and/or" used herein refers to comprising one
A or multiple associated any or all of project listed may combine.
It will be appreciated that though various letters may be described using term first, second, third, etc. in the embodiment of the present application
Breath, but these information should not necessarily be limited by these terms.These terms are only used to for same type of information being distinguished from each other out.For example,
In the case where not departing from the application range, the first information can also be referred to as the second information, and similarly, the second information can also be with
The referred to as first information.Depending on context, in addition, used word " if " can be construed to " ... when " or
" when ... " or " in response to determination ".
The embodiment of the present application proposes that a kind of analysis method of electric system, this method can be applied to power management apparatus,
The power management apparatus is used to analyze the state (for example, safety) of electric system, such as carries out static security analysis.Wherein, electric
Power management equipment can be PC (PersonalComputer, personal computer), laptop, mobile terminal, server etc.,
The type of this power management apparatus is not done and is limited in detail.
In one example, the corresponding relationship of power grid feature and characteristic value can be trained, machine learning network is such as passed through
(such as CNN (Convolutional Neural Networks, convolutional neural networks), LSTM (Long Short-Term
Memory, length remember recursive neural network) etc. neural networks) train the corresponding relationship of power grid feature and characteristic value.Then,
The state of corresponding relationship analysis electric system can be used, to realize the state analysis of electric system, do not need using sudden and violent
The mode that power is enumerated successively disconnects each node of electric system, can save the time of safety analysis, save safety analysis
Computing resource.
In order to realize the state analysis of electric system, training stage and analysis phase can be related to, it, can in the training stage
To train the corresponding relationship of power grid feature and characteristic value.In the analysis phase, corresponding to for being obtained based on the training stage is closed
System, analyzes the state of electric system.Treatment process and analysis below in conjunction with specific application scenarios, to the training stage
The treatment process in stage is illustrated.
Referring to shown in Figure 1A, process shown in dotted line is the treatment process of training stage, and process shown in solid is analysis
The treatment process in stage.In order to distinguish conveniently, the power project of training stage is known as the first power project, by the training stage
Grid nodes layout is known as the first grid nodes layout.In addition, the power project of analysis phase is known as the second power project, it will
The grid nodes layout of analysis phase is known as third grid nodes layout.In practical applications, grid nodes layout may include
But it is not limited to power grid thermodynamic chart.
Referring to shown in Figure 1A, in the training stage, available first power project obtains first according to the first power project
Grid nodes layout, and give the layout output of the first grid nodes to machine learning network, by machine learning network according to the first electricity
Net node layout trains the corresponding relationship of power grid feature and characteristic value.
In the analysis phase, available second power project obtains third grid nodes layout according to the second power project,
And give the layout output of third grid nodes to machine learning network, it is laid out by the machine learning network according to the third grid nodes
With the corresponding relationship, a characteristic value is obtained, and exports obtained characteristic value.
It is laid out in machine learning network according to grid nodes, when training the corresponding relationship of power grid feature and characteristic value, power grid
The quantity of node layout determines the accuracy of training result, it is, therefore, possible to provide a large amount of grid nodes layouts.But in reality
In the application of border, the first abnormal power project quantity may be fewer, causes the first abnormal grid nodes to be laid out quantity and compares
It is few, and hence it is also possible to using process flow shown in Figure 1B.
As shown in Figure 1B, process shown in dotted line is the treatment process of training stage, and process shown in solid is analysis rank
The treatment process of section.It is convenient to distinguish, the power project of training stage is known as the first power project, the first kilowatt meter will be based on
It draws obtained grid nodes layout and is known as the first grid nodes layout, the grid nodes that will be laid out based on the first grid nodes
Layout is known as the second grid nodes layout.The power project of analysis phase is known as the second power project, by the electricity of analysis phase
Net node layout is known as third grid nodes layout.
It is available to the first power project in the training stage referring to shown in Figure 1B, the is obtained according to the first power project
One grid nodes layout, and give the layout output of the first grid nodes to machine learning network.Further, it is also possible to by the first power grid section
Generation confrontation network is given in the output of point layout, is laid out the second grid nodes cloth of generation according to the first grid nodes by generating confrontation network
Office's (such as a large amount of second grid nodes layouts), and give the layout output of the second grid nodes to machine learning network.By above-mentioned place
Reason, machine learning network can receive to the first grid nodes layout and the second grid nodes layout, and according to the received
One grid nodes layout and the second grid nodes layout train the corresponding relationship of power grid feature and characteristic value.
In the analysis phase, available second power project obtains third grid nodes layout according to the second power project,
And give the layout output of third grid nodes to machine learning network, it is laid out by the machine learning network according to the third grid nodes
With the corresponding relationship, a characteristic value is obtained, and exports obtained characteristic value.
In one example, machine learning network can be used in the corresponding relationship of training power grid feature and characteristic value
There are the machine learning algorithm (such as deep learning algorithm) or unsupervised machine learning algorithm of supervision, training power grid feature and spy
The corresponding relationship of value indicative.Wherein, the machine learning algorithm of supervision refers to: in the training process, for the first power grid of input
Node layout, the second grid nodes layout, having corresponding labeled data, (labeled data is for indicating the first grid nodes cloth
Office, the second grid nodes layout are normal grid nodes layouts, or abnormal grid nodes are laid out, in the follow-up process,
The labeled data is known as to the label value of grid nodes layout).Unsupervised machine learning algorithm refers to: in the training process,
The first grid nodes layout, the second grid nodes layout for input, without corresponding labeled data.For the convenience of description,
It is subsequent by taking the machine learning algorithm for having supervision as an example, illustrate the training process of machine learning network.
It is the flow chart of the analysis method of electric system referring to fig. 2 shown in A, Fig. 2A is the flow chart of training stage, and is schemed
2A is the flow chart based on application scenarios shown in Figure 1A, and this method may include:
Step 211, the first power project is obtained.
Step 212, the first grid nodes layout is obtained according to first power project.
Step 213, according to the label value of first grid nodes layout and the first grid nodes layout, training power grid feature
With the corresponding relationship of characteristic value；Wherein, which is used to analyze the state of electric system.
In one example, above-mentioned execution sequence is intended merely to facilitate description to provide example, in practical applications,
Sequence is executed between can also changing the step, with no restrictions to this execution sequence.Moreover, in other embodiments, and it is different
The fixed sequence for showing and describing according to this specification is come the step of executing correlation method, step included by method can be than this
It is more or less described in specification.In addition, single step described in this specification, it in other embodiments may quilt
Multiple steps are decomposed into be described；Multiple steps described in this specification may also be merged into other embodiments
Single step is described.
For step 211, in one example, power management apparatus can be with maintenance history database, the historical data base
It is known as the first power project for storing power project, and by the power project stored in historical data base.Based on this, power pipe
Reason equipment can get the first power project from the historical data base.
It in one example, can be by new power project when power management apparatus obtains new power project every time
It stores in historical data base, in this way, the power project in historical data base is constantly updated, to adapt to the change of external environment
Change.On this basis, power management apparatus can execute above-mentioned steps when the power project in historical data base is changed
211- step 213；Alternatively, above-mentioned steps 211- step 213 can also be periodically executed, such as every 24 hours according to predetermined period
It executes once, with no restrictions to this process.
In one example, the label value of the first power project, the first power project can also be recorded in historical data base
Label value can be first identifier (such as 1) or second identifier (such as 0).First identifier indicates that the first power project is normal electricity
Power plan, second identifier indicate that the first power project is abnormal power project.
For example, including the first power project 1, the first power project 2, the first power project 3, the first electricity in historical data base
The label value of power plan 1 and the first power project 2 is first identifier, and the label value of the first power project 3 is second identifier.In reality
In the application of border, the first power project quantity in historical data base is far longer than 3, does not limit this first power project quantity
System.
For step 212, in one example, the first grid nodes layout be can include but is not limited to: in electric system
Each node power information.For obtaining the mode of the first grid nodes layout according to the first power project, do not limit herein
System, as long as the first grid nodes layout may include the power information of each node in electric system.Below in conjunction with specific reality
Mode is applied to be illustrated the process for obtaining the first grid nodes layout.
Specifically, may include but not for the process of " obtaining the first grid nodes layout according to the first power project "
Be limited to such as under type: from topological structure of electric is obtained in first power project, in the topological structure of electric each node electric power
Information；Wherein, which may include the topological structure of each node in electric system；According to the power network topology knot
Power information generation the first grid nodes layout of structure, each node.
Further, for " according to the power information of the topological structure of electric, each node the first grid nodes cloth of generation
The process of office " may include: to be generated in first according to the topological structure (i.e. the topological structure of electric) of node each in electric system
Intermediate node layout；Wherein, it in order to distinguish conveniently, by the training stage, is laid out and is claimed based on the intermediate node that topological structure of electric generates
For the first intermediate node layout.Then, the power information based on each node, it is corresponding in each node of first intermediate node layout
Position, mark the corresponding colouring information of power information of the node, thus by the first intermediate node layout be converted to first electricity
Net node layout.
Node can include but is not limited to: route, generating set, transformer, power unit etc., it is, of course, also possible to there is it
The node of its type, it is without limitation.Power information can include but is not limited to: charge value, the electricity generated such as generating set
Magnitude, the charge value of line transmission, the charge value etc. of power unit consumption.When counting the power information of node, section here
Point can be individual node, be also possible to multiple node compositions.For example, there are multiple generating sets, this multiple hairs by power plant A
Motor group corresponds to a node, and the sum of the charge value of this multiple generating sets generation is the charge value of the node, the i.e. node
Power information.In another example there are multiple power units, the corresponding node of this multiple power unit, and this multiple use for vendor A
The sum for the charge value that electric unit uses is the charge value of the node, i.e. the power information of the node.
Below in conjunction with the first power project 1, the above process is illustrated, for the first power project 2, the first kilowatt meter
Other power projects such as 3 are drawn, similar with the processing of the first power project 1, details are not described herein.
Wherein, it can include but is not limited to topological structure of electric, each in the topological structure of electric in first power project 1
The power information of node.It is an example of the topological structure of electric in the first power project 1, in Fig. 2 B referring to fig. 2 shown in B
Node can be for route, generating set, transformer, power unit etc..In addition, the first power project 1 can also include each section
The power information (charge value of i.e. each node) of point.
Topological structure of electric may include 11 nodes, and generating set A is node 1, and generating set B is node 2, route 1-
Route 6 is respectively node 3- node 8, and power unit C- power unit E is respectively node 9- node 11.If generating set A is generated
Charge value be 1000, the charge value that generating set B is generated is 500, and the power information of node 1 is 1000, the electric power of node 2 letter
Breath is 500.If the charge value that route 1- route 6 transmits is respectively 400,100,100,100,300,500, node 3- node 8
Power information be 400,100,100,100,300,500 respectively.If the charge value difference of power unit C- power unit E consumption
Be 500,200,800, the power information of node 9- node 11 be 500,200,800 respectively.
Under above-mentioned application scenarios, topological structure of electric shown in Fig. 2 B can be parsed from the first power project 1, from
The power information of the node 1- node 11 in topological structure of electric is parsed in first power project 1.By power grid shown in Fig. 2 B
Topological structure is converted into the first intermediate node layout shown in fig. 2 C, no longer superfluous to the conversion process of this first intermediate node layout
It states, as long as including each node in topological structure of electric in first intermediate node layout, is illustrated by taking Fig. 2 C as an example.
In fig. 2 c, the region where generating set A is the corresponding position of node 1, i.e. position in 1 place frame of node；Generating set B
The region at place is the corresponding position of node 2, i.e. position in 2 place frame of node；Region where route 1 is corresponding for node 3
Position, i.e. position in 3 place frame of node；And so on, other nodes are repeated no more.
Power information 1000 based on node 1, in the corresponding position of node 1 (i.e. 1 institute of node of the first intermediate node layout
In frame) the mark corresponding colouring information of power information 1000.Power information 500 based on node 2 is laid out in the first intermediate node
The corresponding position of node 2 mark the corresponding colouring information of power information 500.And so on, it is laid out to the first intermediate node
The corresponding colouring information of 11 node label power informations after, so that it may obtain the first grid nodes layout, to this first electricity
Net the structure of node layout with no restrictions.
Wherein, in order to mark the corresponding colouring information of power information, pair of power information and colouring information can also be configured
It should be related to, in this way, after obtaining the power information of node, so that it may by inquiring the corresponding relationship, obtain the power information
Corresponding colouring information, and mark the corresponding colouring information of power information.
For example, the pass corresponding with the corresponding relationship of color 1, power information 11-20 and color 2 configuration power information 1-10
System, and so on, when power information is bigger, then color is deeper.In practical applications, different colors can be different
RGB (Red Green Blue, RGB) color, it is without limitation.
In one example, when configuring the corresponding relationship of power information and colouring information, it can also distinguish between cool colour and warm up
Color, warm colour can be the colors such as red, orange, and cool colour can be the colors such as green, blue.Specifically, if node be generate charge value or
The node of charge value is transmitted, then the corresponding colouring information of the power information of the node is warm colour, and when power information is bigger, is warmed up
Color is deeper.If node is the node for consuming charge value, the corresponding colouring information of the power information of the node is cool colour, and when electricity
When force information is bigger, cool colour is deeper.
In conclusion for the corresponding relationship of power information and colouring information, with no restrictions, as long as can based on power information
To find corresponding colouring information, and the colouring information is marked in the corresponding position of node.
For step 213, in one example, after obtaining the first grid nodes layout, the first power grid can also be determined
The label value of node layout, the label value are first identifier or second identifier, and first identifier indicates that the first grid nodes layout is
Normal grid nodes layout, second identifier indicate that the first grid nodes layout is abnormal grid nodes layout.For example, can be with
Get the label value of the first power project from historical data base, and the label value of first power project, that is, this
The label value of the corresponding first grid nodes layout of one power project, which can be first identifier (such as 1) or second
It identifies (such as 0).
Wherein, which indicates that first grid nodes layout is normal grid nodes layout, that is to say, that should
First grid nodes layout is a normal node layout sample, i.e. normal sample；In addition, the second identifier indicate this first
Grid nodes layout is abnormal grid nodes layout, that is to say, that first grid nodes layout is the node of an exception
It is laid out sample, i.e. exceptional sample.
Obtain the first grid nodes layout and the first grid nodes layout label value after, so that it may according to this first
The label value of grid nodes layout and first grid nodes layout, trains the corresponding relationship of power grid feature and characteristic value, and
In the analysis phase, the state of electric system can be analyzed according to the corresponding relationship.
In one example, for " according to the label value of the first grid nodes layout and first grid nodes layout, instruction
The process of the corresponding relationship of white silk power grid feature and characteristic value ", can include but is not limited to such as under type: can be by first power grid
Machine learning network (such as CNN, LSTM nerve net is given in the label value of node layout and first grid nodes layout, output
Network), it is laid out by the machine learning network according to first grid nodes and the label value of first grid nodes layout, training is electric
The corresponding relationship of net feature and characteristic value.
Due to can store a large amount of first power projects in historical data base, it is available to arrive a large amount of first power grids
The label value of node layout and these first grid nodes layout.It is then possible to a large amount of first grid nodes are laid out, with
And the label value of these first grid nodes layout exports and gives machine learning network.Mass data based on input, machine learning
Network can use machine learning algorithm (such as deep learning algorithm), train the corresponding relationship of power grid feature and characteristic value,
This training process is repeated no more.Finally, the corresponding relationship of power grid feature and characteristic value can be output to machine learning network
Classifier.
Wherein, power grid feature can be the characteristics of image of the first grid nodes layout, such as color characteristic, textural characteristics, shape
Shape feature, spatial relation characteristics etc., with no restrictions to this power grid feature.
Wherein, characteristic value can be the probability value of normal electricity plan or the probability value of abnormal power plan.For example,
If machine learning network is used to train the corresponding relationship of the probability value of power grid feature and normal electricity plan, characteristic value be can be
The probability value of normal electricity plan, such as 95%, the 95% expression power project is that the probability value of normal electricity plan is
95%.If machine learning network is used to train the corresponding relationship of the probability value of power grid feature and abnormal power plan, characteristic value
It can be the probability value of abnormal power plan, such as 95%, the 95% expression power project is the probability of abnormal power plan
Value is 95%.
Machine learning network is laid out according to grid nodes, when training the corresponding relationship of power grid feature and characteristic value, power grid section
Point layout quantity determines the accuracy of training result, therefore can provide a large amount of grid nodes layouts, such as a large amount of normal samples
(normal grid nodes layout) and a large amount of exceptional samples (abnormal grid nodes layout).
In view of the first power project quantity is possible fewer (the first power project quantity as abnormal can be less), thus
Cause the first grid nodes to be laid out small number, then it can also be using process flow shown in Fig. 2 D.Referring to fig. 2 shown in D, it is
The flow chart of the analysis method of electric system, Fig. 2 B is the flow chart of training stage, and Fig. 2 D is based on applied field shown in Figure 1B
The flow chart of scape, this method may include:
Step 221, the first power project is obtained.
Step 222, the first grid nodes layout is obtained according to first power project.
Step 223, first grid nodes layout is handled using generation confrontation network algorithm, obtains multiple second
The label value of grid nodes layout, the second grid nodes layout is identical as the label value that the first grid nodes are laid out.
Step 224, according to the label value of the first grid nodes layout and the first grid nodes layout, the second grid nodes cloth
The label value of office and the second grid nodes layout, the corresponding relationship of training power grid feature and characteristic value.
Wherein, in the analysis phase, the state of corresponding relationship analysis electric system can be used.
Wherein, step 221, step 222 are similar with step 211, step 212, and details are not described herein.
For step 223, if the label value of the first grid nodes layout is that first identifier (indicates normal grid nodes cloth
Innings), then when being handled using generation confrontation network algorithm the first grid nodes layout, the second obtained grid nodes are laid out
Label value be first identifier.If the label value of the first grid nodes layout is that second identifier (indicates abnormal grid nodes cloth
Innings), then when being handled using generation confrontation network algorithm the first grid nodes layout, the second obtained grid nodes are laid out
Label value be second identifier.
Specifically, choosing label from these first grid nodes layout after obtaining a large amount of first grid nodes layouts
Value is laid out for the first grid nodes of first identifier, gives the first grid nodes layout output of selection to generation confrontation network.It is raw
It is laid out at confrontation network based on these first grid nodes, obtains a large amount of second grid nodes cloth using confrontation network algorithm is generated
Office, the label value of these second grid nodes layout are first identifier.In addition, after obtaining a large amount of first grid nodes layouts,
It chooses the first grid nodes that label value is second identifier from these first grid nodes layout to be laid out, by the first electricity of selection
Net node layout, which exports, fights network to generation.It generates confrontation network to be laid out based on these first grid nodes, using generation pair
Anti- network algorithm obtains a large amount of second grid nodes layouts, and the label value of these second grid nodes layout is second identifier.
Wherein, generating confrontation network (GANs, Generative Adversarial Networks) is deep learning field
Extremely innovative method, main thought is: by the study to existing sample, generating more new samples, new sample
Originally the inherent attribute of original sample can be embodied, but original sample is effectively supplemented, new sample has the more of information
Sample, so that the reliability of sample is higher, accuracy of identification is higher.
Wherein, generating confrontation network can consist of two parts, and first part is generator, for generating new sample,
Second part is arbiter, generates new sample for going to differentiate.By way of generating and differentiating and fight, foot can lacked
Under enough, to have supervision labeled data scenes, sufficiently excavate existing unlabeled data, find and distinguish it is possible normal or
Anomalous variation, to obtain significantly more efficient electrical nature.
In conclusion the first grid nodes can be laid out to output to confrontation network is generated, so that generating confrontation network base
It is laid out in the first grid nodes, generates more second grid nodes layouts (namely new sample), not to this generating process
It is limited, these second grid nodes layout is that precision is higher, image quality better image.
In the above-described embodiments, the quantity being laid out for generating the second grid nodes that confrontation network generates, can basis
Experience is configured, or is generated according to actual needs by generating confrontation network, without limitation.
For step 224, in one example, the mark of the first grid nodes layout and the first grid nodes layout is being obtained
After the label value of label value, the second grid nodes layout and the second grid nodes layout, so that it may according to the first grid nodes cloth
The label value of office and the first grid nodes layout, the label value of the second grid nodes layout and the second grid nodes layout, training
The corresponding relationship of power grid feature and characteristic value out.It further, can be corresponding with characteristic value according to power grid feature in the analysis phase
Relationship analyzes the state of electric system.
In one example, for " according to the first grid nodes layout and the first grid nodes layout label value, second
Grid nodes layout and the second grid nodes layout label value, train the corresponding relationship of power grid feature and characteristic value " mistake
Journey, may include: can be by the label value that first grid nodes are laid out and first grid nodes are laid out, the second power grid section
Machine learning network (such as CNN, LSTM neural network) is given in the label value of point layout and second grid nodes layout, output,
By the machine learning network according to the label value of first grid nodes layout and first grid nodes layout, second power grid
The label value of node layout and second grid nodes layout, trains the corresponding relationship of power grid feature and characteristic value.
Getting a large amount of grid nodes layouts (such as the first grid nodes layout and the second grid nodes layout), these electricity
After the label value for netting node layout, a large amount of grid nodes can be laid out, the label value of these grid nodes layout is exported to machine
Device learning network.Mass data based on input, machine learning network can use machine learning algorithm (such as deep learning algorithm
Deng), the corresponding relationship of power grid feature and characteristic value is trained, this training process is repeated no more.It finally, can be by power grid feature
The classifier of machine learning network is output to the corresponding relationship of characteristic value.Wherein, power grid feature can be grid nodes layout
Characteristics of image, such as color characteristic, textural characteristics, shape feature, spatial relation characteristics, with no restrictions to this power grid feature.
In addition, characteristic value can be the probability value of normal electricity plan or the probability value of abnormal power plan.
For training stage (Fig. 2A and Fig. 2 D), can with the corresponding relationship of regular exercise power grid feature and characteristic value, thus
The corresponding relationship of power grid feature and characteristic value is kept, there is higher reliability.
It is the flow chart of the analysis method of electric system referring to fig. 2 shown in E, Fig. 2 E is the flow chart of analysis phase, is fitted
It may refer to shown in Figure 1A and Figure 1B with scene, this method may include:
Step 231, the second power project for needing to analyze state is obtained.
Step 232, third grid nodes layout is obtained according to second power project.
Step 233, the power grid characteristic query corresponding relationship being laid out by the third grid nodes, obtains and the power grid feature
Corresponding characteristic value.Wherein, the corresponding relationship is corresponding with characteristic value according to the power grid feature of the first power project training
Relationship.
Step 234, according to the state of obtained Eigenvalues analysis second power project.
For step 231, in one example, for the process of " the second power project that acquisition needs to analyze state ",
It can include but is not limited to: obtaining power generation plan, electric power demand forecasting, topological structure of electric, and according to power generation meter
It draws, electric power demand forecasting, topological structure of electric obtain the second power project.
Wherein, above-mentioned power generation plan can be the charge value of generating set generation, and power management apparatus can be from hair
Power plant gets this power generation plan, with no restrictions to the acquisition modes of this power generation plan.
Wherein, above-mentioned electric power demand forecasting can be the charge value of the power unit consumption predicted, power management apparatus
The charge value that power unit consumption can be predicted according to historical data, does not limit the acquisition modes of this electric power demand forecasting
System.For example, if being based on historical data when knowing that the average charge value consumed daily of power unit is 500, then use can be predicted
The charge value of electric unit consumption is 500；Alternatively, can then be predicted when the charge value consumed on the day before knowing power unit is 520
The charge value of power unit consumption is 520 out.
Wherein, above-mentioned topological structure of electric can be the topological structure of each node in electric system, and the node can be with
Including but not limited to: route, generating set, transformer, power unit etc..Power management apparatus can be according to the position of generating set
It sets, the position of transformer, the position of power unit, and, generating set, transformer, all routes between power unit etc.,
Topological structure of electric is generated, with no restrictions to the acquisition modes of this topological structure of electric.For example, power management apparatus can be generated
Topological structure of electric shown in Fig. 2 B.
After obtaining power generation plan, electric power demand forecasting, topological structure of electric, so that it may according to the power generation meter
It draws, the electric power demand forecasting, the topological structure of electric obtain the second power project.Wherein, second power project may include
The power information of each node in above-mentioned topological structure of electric, topological structure of electric.
For the topological structure of electric shown in Fig. 2 B, it is based on power generation plan, if the charge value that generating set A is generated
It is 1000, the charge value that generating set B is generated is 500, and the power information of the corresponding node 1 of generating set A is 1000, generator
The power information of the corresponding node 2 of group B is 500.Based on electric power demand forecasting, disappear if predicting power unit C- power unit E
The charge value of consumption is respectively 500,200,800, the power information point of the corresponding node 9- node 11 of power unit C- power unit E
It is not 500,200,800.
Based on power generation plan and electric power demand forecasting, the charge value that route 1- route 6 transmits, such as route can be known
The sum of 1 and the charge value of the transmission of route 2 are 500, and the sum of the charge value that route 3 and route 4 transmit is 200, route 5 and route 6
The sum of charge value of transmission is 800, and the sum of charge value that route 1- route 5 transmits is 1000, and the charge value that route 6 transmits is
500.In order to reach above-mentioned target, then the example of the charge value that route 1- route 6 transmits can for 400,100,100,100,
300,500, i.e. the power information of the corresponding node 3- node 8 of route 1- route 6 be 400 respectively, 100,100,100,300,
500。
Certainly, the above process has been merely given as " according to the power generation plan, the electric power demand forecasting, the power network topology
One example of the second power project of structure acquisition ", with no restrictions to this acquisition modes.
For step 232, in one example, third grid nodes layout be can include but is not limited to: in electric system
Each node power information.For obtaining the mode of third grid nodes layout according to the second power project, do not limit herein
System, as long as third grid nodes layout may include the power information of each node in electric system.Below in conjunction with specific reality
Mode is applied to be illustrated the process for obtaining third grid nodes layout.
Specifically, may include but not for the process of " obtaining third grid nodes layout according to the second power project "
Be limited to such as under type: from topological structure of electric is obtained in second power project, in the topological structure of electric each node electric power
Information；Wherein, which may include the topological structure of each node in electric system；According to the power network topology knot
The power information generation third grid nodes layout of structure, each node.
Further, for " according to the power information of the topological structure of electric, each node generation third grid nodes cloth
The process of office " may include: to be generated in second according to the topological structure (i.e. the topological structure of electric) of node each in electric system
Intermediate node layout；Wherein, it in order to distinguish conveniently, by the analysis phase, is laid out and is claimed based on the intermediate node that topological structure of electric generates
For the second intermediate node layout.Then, the power information based on each node, it is corresponding in each node of second intermediate node layout
Position, mark the corresponding colouring information of power information of the node, thus by the second intermediate node layout be converted to third electricity
Net node layout.
Wherein, the process for " third grid nodes layout being obtained according to the second power project ", and in step 212, " according to the
The process of one power project the first grid nodes of acquisition layout " is similar, and details are not described herein.
The electricity of third grid nodes layout can also be determined after obtaining third grid nodes layout for step 233
Net feature, and by the above-mentioned corresponding relationship of power grid characteristic query (i.e. in the training stage, the power grid feature and feature that train
The corresponding relationship of value), to obtain characteristic value corresponding with the power grid feature.
For " the power grid characteristic query corresponding relationship being laid out by third grid nodes obtains corresponding with the power grid feature
Characteristic value " process, may include: by the third grid nodes layout output give machine learning network, by machine learning net
Network determines that the third grid nodes are laid out corresponding power grid feature, and the correspondence generated by the power grid characteristic query training stage
Relationship, to obtain characteristic value corresponding with the power grid feature.
Wherein, the power grid feature can be third grid nodes layout characteristics of image, as color characteristic, textural characteristics,
Shape feature, spatial relation characteristics etc., with no restrictions to this power grid feature.In addition, features described above value can be normal electricity meter
The probability value of the probability value or abnormal power plan drawn.
For step 234, in one example, for " according to the shape of obtained Eigenvalues analysis second power project
The process of state " can include but is not limited to such as under type: if this feature value is the probability value of normal power project, work as this feature
When value is greater than preset first threshold value, it is determined that the state of above-mentioned second power project is normal；Otherwise, it determines above-mentioned second electric power
The state of plan is abnormal.Alternatively, if this feature value is the probability value of abnormal power plan, when this feature value is greater than default second
When threshold value, it is determined that the state of above-mentioned second power project is abnormal；Otherwise, it determines the state of above-mentioned second power project is just
Often.
Wherein, above-mentioned preset first threshold value can be configured rule of thumb, and above-mentioned default second threshold can be according to warp
It tests and is configured, preset first threshold value and default second threshold may be the same or different.For example, preset first threshold value can
Think 90%, 85% etc., default second threshold can be 90%, 95% etc..
In one example, after determining that the state of the second power project is normal, the second power project can also be deposited
Store up historical data base, and record the label value of the second power project in historical data base, the label value be first identifier (such as
1) normal electricity plan, is indicated.After determining that the state of the second power project is exception, the second power project can also be stored
To historical data base, and record in historical data base the label value of the second power project, the label value be second identifier (such as
0) abnormal power plan, is indicated.
In one example, after determining that the state of the second power project is exception, can also according to power generation plan,
Electric power demand forecasting, topological structure of electric obtain the second new power project and execute above-mentioned step based on the second new power project
Rapid 232- step 234 continues if being still abnormal power plan according to power generation plan, electric power demand forecasting, power network topology
Structure obtains the second new power project, and so on.
In addition, exception information can also be notified to electrical management after determining that the state of the second power project is exception
Personnel are adjusted power generation plan, electric power demand forecasting, topological structure of electric by electrical management personnel, to reach
Instruct the purpose of Electric Power Network Planning, traffic control and maintenance power failure plan.
Based on the above-mentioned technical proposal, in the embodiment of the present application, the corresponding relationship of power grid feature and characteristic value can be trained,
And the state of electric system is analyzed according to the corresponding relationship, and Electric Power Network Planning, traffic control and inspection are instructed based on analysis result
Power failure plan is repaired, quick Contingency Analysis of Power Systems can be carried out, supports power planning and scheduling.It does not need using sudden and violent
The mode that power is enumerated successively disconnects each node of electric system, can save the time of safety analysis, save safety analysis
Computing resource.Safety analysis can be carried out in a short time, operation power personnel is made quickly to make scheduling decision, avoided big
Scale power failure.
Based on similarly applying conceiving with the above method, the embodiment of the present application also proposes a kind of analysis side of electric system
Method, this method can be applied to power management apparatus, which is used to analyze the state of electric system, such as carry out quiet
State safety analysis.Wherein, which can be PC, laptop, mobile terminal, server etc., to this electric power
The type of management equipment is not done to be limited in detail.
It is shown in Figure 3, it is the flow chart of the analysis method of electric system, this method may include:
Step 301, the third power project for needing to analyze state is obtained.
Step 302, inquiring in normal sample library whether there is and the matched normal electricity plan of the third power project.Such as
Fruit is can then to execute step 303；If it is not, then step 304 can be executed.
Step 303, it is normal for analyzing the state of the third power project.
Step 304, it is abnormal for analyzing the state of the third power project.
For step 301, in one example, for the process of " the third power project that acquisition needs to analyze state ",
It can include but is not limited to: obtaining power generation plan, electric power demand forecasting, topological structure of electric, and according to power generation meter
It draws, electric power demand forecasting, topological structure of electric obtain third power project.
Wherein, the treatment process of step 301 and the treatment process of step 231 are similar, and it is convenient to be intended merely to distinguish, and will need
The power project of analysis state is known as third power project, and it is no longer repeated herein.
For step 302, in one example, power management apparatus can safeguard normal sample library, the normal sample library
For recording a large amount of normal electricity plan.Based on this, after obtaining third power project, can inquire in normal sample library is
It is no to exist and the matched normal electricity plan of third power project.If so, i.e. normal sample inventory is in third power project, then
Illustrate that third power project is normal electricity plan.If not, third power project is not present i.e. in normal sample library, then illustrate
Third power project is abnormal power plan.
It can also include: to be obtained from historical data base to record a large amount of normal electricity plan in normal sample library
It takes normal electricity plan (can be multiple normal electricity plans), and fights network algorithm to the normal electricity meter using generating
It draws and is handled, obtain multiple new normal electricity plans；Then, by the normal electricity plan obtained from historical data base,
The multiple normal electricity plans obtained using confrontation network algorithm is generated, store into normal sample library, so that normal sample library
In record a large amount of normal electricity plan.
Wherein, power management apparatus can be used to store power project with maintenance history database, the historical data base, these
Power project may include normal electricity plan and abnormal power plan.For example, each electric power can also be recorded in historical data base
The label value of plan, the label value can be first identifier (such as 1) or second identifier (such as 0), and the first identifier is for indicating
Power project is normal electricity plan, and the second identifier is for indicating that power project is abnormal power plan.Based on this, power pipe
The normal electricity plan that reason equipment can get normal electricity plan from the historical data base, and will acquire is stored to just
In normal sample database.
Power management apparatus can also fight network using generating after getting normal electricity plan in historical data base
Algorithm handles normal electricity plan, obtains multiple new normal electricity plans, and by these new normal electricity plans
Store normal sample library.For example, normal electricity plan can be exported to confrontation network is generated, generates confrontation network and be based on this
A little normal electricity plans can obtain a large amount of new normal electricity plans using confrontation network algorithm is generated, and these are new
Normal electricity plan is stored to normal sample library.
Wherein, the content for generating confrontation network has been introduced above, and details are not described herein.In short, by normal electricity
After plan output is to confrontation network is generated, normal electricity plan can be based on by generating confrontation network, generate more normal electricities
Plan (namely new sample), with no restrictions to this generating process.
In one example, after the state of analysis third power project is normal, third power project can also be deposited
It stores up in normal sample library, and third power project storage is recorded into third electric power to historical data base in historical data base
The label value of plan, i.e. first identifier (such as 1) indicate normal electricity plan.
It in one example, can also be according to power generation meter after the state of analysis third power project is exception
Draw, electric power demand forecasting, topological structure of electric obtain new third power project, based on new third power project, hold again
Row above-mentioned steps 302 continue if being still abnormal power plan according to power generation plan, electric power demand forecasting, power network topology
Structure obtains new third power project, and so on.
In addition, exception information can also be notified to electrical management after determining that the state of third power project is exception
Personnel are adjusted power generation plan, electric power demand forecasting, topological structure of electric by electrical management personnel, to reach
Instruct the purpose of Electric Power Network Planning, traffic control and maintenance power failure plan.
It based on the above-mentioned technical proposal, can be based on the normal electricity plan in normal sample library point in the embodiment of the present application
The state of electric system is analysed, the result based on safety analysis instructs Electric Power Network Planning, traffic control and maintenance power failure plan, Ke Yijin
The quick Contingency Analysis of Power Systems of row, supports power planning and scheduling.Aforesaid way does not need to enumerate using violence
Mode successively disconnects each node of electric system, can save the time of safety analysis, saves the computing resource of safety analysis.
In emergency circumstances, safety analysis can be carried out in a short time, so that operation power personnel is quickly made scheduling decision, avoided
Large-scale power failure.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of analysis dress of electric system
It sets, can be applied to power management apparatus, as shown in figure 4, being the structure chart of described device.
Module 401 is obtained, for obtaining the first power project；The first grid nodes are obtained according to first power project
Layout；Training module 402, for the label value according to first grid nodes layout and the first grid nodes layout, training
The corresponding relationship of power grid feature and characteristic value；The corresponding relationship is used to analyze the state of electric system.
The acquisition module 401, specifically for obtaining the first grid nodes layout according to first power project
In the process, from topological structure of electric is obtained in first power project, in the topological structure of electric each node electric power letter
Breath；Wherein, the topological structure of electric includes the topological structure of each node in electric system；According to the topological structure of electric,
The power information of each node generates the first grid nodes layout；
The training module 402 is specifically used for being laid out according to first grid nodes layout and the first grid nodes
Label value, during the corresponding relationship of training power grid feature and characteristic value, using generating confrontation network algorithm to described the
One grid nodes layout is handled, and multiple second grid nodes layouts, the label value of the second grid nodes layout are obtained
It is identical as the label value of first grid nodes layout；According to first grid nodes layout, first grid nodes
The label value of the label value of layout, the multiple second grid nodes layout, each second grid nodes layout, training power grid feature
With the corresponding relationship of characteristic value.
The acquisition module 401 is also used to obtain the second power project for needing to analyze state；According to the second power project
Obtain third grid nodes layout；Corresponding relationship described in the power grid characteristic query being laid out by the third grid nodes, obtains
Characteristic value corresponding with the power grid feature of third grid nodes layout；
Described device further includes (not shown in FIG.):
Analysis module, the state for the second power project according to obtained Eigenvalues analysis.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of power management apparatus, described
Power management apparatus may include processor；Wherein: the processor, for obtaining the first power project；According to described first
Power project obtains the first grid nodes layout；It is laid out according to first grid nodes layout and first grid nodes
Label value, the corresponding relationship of training power grid feature and characteristic value；Wherein, the corresponding relationship is used to analyze the shape of electric system
State.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of machine readable storage medium,
The machine readable storage medium can be applied to power management apparatus, and several meters are stored on the machine readable storage medium
The instruction of calculation machine, the computer instruction, which is performed, to be handled as follows: obtaining the first power project；According to first electric power
Plan obtains the first grid nodes layout；According to the label of first grid nodes layout and first grid nodes layout
Value, the corresponding relationship of training power grid feature and characteristic value；Wherein, the corresponding relationship is used to analyze the state of electric system.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of analysis dress of electric system
It sets, can be applied to power management apparatus, as shown in figure 5, being the structure chart of described device.
Module 501 is obtained, needs to analyze the second power project of state for obtaining；It is obtained according to second power project
Third grid nodes are taken to be laid out；By the third grid nodes be laid out power grid characteristic query corresponding relationship, obtain with it is described
The corresponding characteristic value of power grid feature；Wherein, the corresponding relationship is the power grid feature and feature according to the first power project training
The corresponding relationship of value；
Analysis module 502, the state for the second power project according to obtained Eigenvalues analysis.
The acquisition module 501, specifically for obtaining third grid nodes layout according to second power project
In the process, from topological structure of electric is obtained in the second power project, in the topological structure of electric each node power information；Its
In, the topological structure of electric includes the topological structure of each node in electric system；According to the topological structure of electric, described each
The power information of node generates third grid nodes layout；
The analysis module 502, if being the probability value of normal power project specifically for the characteristic value, when the feature
When value is greater than preset first threshold value, it is determined that the state of second power project is normal；Alternatively, if the characteristic value is different
The probability value of normal power project, when the characteristic value is greater than default second threshold, it is determined that the shape of second power project
State is abnormal.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of power management apparatus, described
Power management apparatus includes: processor, obtains the second power project for needing to analyze state；It is obtained according to second power project
Third grid nodes are taken to be laid out；By the third grid nodes be laid out power grid characteristic query corresponding relationship, obtain with it is described
The corresponding characteristic value of power grid feature；Wherein, the corresponding relationship is the power grid feature and feature according to the first power project training
The corresponding relationship of value；The state of the second power project according to obtained Eigenvalues analysis.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of machine readable storage medium,
It can be applied to power management apparatus, several computer instructions, the computer be stored on the machine readable storage medium
Instruction, which is performed, to be handled as follows: obtaining the second power project for needing to analyze state；According to second power project
Obtain third grid nodes layout；The power grid characteristic query corresponding relationship being laid out by the third grid nodes, obtains and institute
State the corresponding characteristic value of power grid feature；Wherein, the corresponding relationship is the power grid feature according to the first power project training and spy
The corresponding relationship of value indicative；The state of the second power project according to obtained Eigenvalues analysis.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of analysis dress of electric system
It sets, can be applied to power management apparatus, as shown in fig. 6, being the structure chart of described device.
Module 601 is obtained, needs to analyze the third power project of state for obtaining；
Enquiry module 602, for inquiring in normal sample library with the presence or absence of matched normal with the third power project
Power project；
Analysis module 603, for when query result, which is, is, then it to be normal for analyzing the state of the third power project；
When query result is no, then it is abnormal for analyzing the state of the third power project.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of power management apparatus, described
Power management apparatus may include processor；Wherein: the processor needs to analyze the third kilowatt meter of state for obtaining
It draws；Inquiring in normal sample library whether there is and the matched normal electricity plan of the third power project；If so, analysis institute
It is normal for stating the state of third power project；If it is not, the state for then analyzing the third power project is abnormal.
Based on similarly applying conceiving with the above method, the embodiment of the present application also provides a kind of machine readable storage medium,
The machine readable storage medium can be applied to power management apparatus, and several meters are stored on the machine readable storage medium
The instruction of calculation machine, the computer instruction, which is performed, to be handled as follows: obtaining the third power project for needing to analyze state；It looks into
Asking in normal sample library whether there is and the matched normal electricity plan of the third power project；If so, analysis described the
The state of three power projects is normal；If it is not, the state for then analyzing the third power project is abnormal.
System, device, module or the unit that above-described embodiment illustrates can specifically realize by computer chip or entity,
Or it is realized by the product with certain function.A kind of typically to realize that equipment is computer, the concrete form of computer can
To be personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media play
In device, navigation equipment, E-mail receiver/send equipment, game console, tablet computer, wearable device or these equipment
The combination of any several equipment.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when application.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes computer usable program code that the embodiment of the present application, which can be used in one or more,
The computer implemented in computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of program product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It is generally understood that being realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computer journeys
Sequence instruct to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices processor with
A machine is generated, so that the instruction generation executed by computer or the processor of other programmable data processing devices is used for
Realize the dress for the function of specifying in one or more flows of the flowchart and/or one or more blocks of the block diagram
It sets.
Moreover, these computer program instructions also can store be able to guide computer or other programmable datas processing set
In standby computer-readable memory operate in a specific manner, so that instruction stored in the computer readable memory generates
Manufacture including command device, the command device are realized in one process of flow chart or multiple processes and/or block diagram one
The function of being specified in a box or multiple boxes.
These computer program instructions can also be loaded into computer or other programmable data processing devices, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer
Or the instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram
The step of function of being specified in one box or multiple boxes.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (30)
1. a kind of analysis method of electric system, which is characterized in that the described method includes:
Obtain the first power project；
The first grid nodes layout is obtained according to first power project；
According to the label value of first grid nodes layout and the first grid nodes layout, training power grid feature and characteristic value
Corresponding relationship；Wherein, the corresponding relationship is used to analyze the state of electric system.
2. the method according to claim 1, wherein
The first grid nodes layout includes the power information of each node in electric system；The label value be first identifier or
Second identifier；The first identifier indicates that the first grid nodes layout is normal grid nodes layout, the second identifier table
Show that the first grid nodes layout is abnormal grid nodes layout.
3. the method according to claim 1, wherein
The process that the first grid nodes layout is obtained according to first power project, specifically includes:
From topological structure of electric is obtained in the first power project, in the topological structure of electric each node power information；Wherein,
The topological structure of electric includes the topological structure of each node in electric system；
The first grid nodes layout is generated according to the power information of the topological structure of electric, each node.
4. according to the method described in claim 3, it is characterized in that, described according to the topological structure of electric, each node
Power information generate the first grid nodes layout process, specifically include:
The first intermediate node layout is generated according to the topological structure of node each in the electric system；
Based on the power information of each node, in the corresponding position of each node of first intermediate node layout, mark should
The corresponding colouring information of the power information of node obtains the first grid nodes layout.
5. the method according to claim 1, wherein according to first grid nodes layout and the first power grid section
The label value of point layout, the corresponding relationship of training power grid feature and characteristic value, comprising:
First grid nodes layout is handled using confrontation network algorithm is generated, obtains multiple second grid nodes cloth
The label value of office, the second grid nodes layout is identical as the label value that first grid nodes are laid out；
According to the first grid nodes layout, the label value of the first grid nodes layout, multiple second grid nodes layout, each second
The label value of grid nodes layout, the corresponding relationship of training power grid feature and characteristic value.
6. method according to claim 1 or 5, which is characterized in that according to first grid nodes layout and the first electricity
Net the label value of node layout, the corresponding relationship of training power grid feature and characteristic value, comprising:
First grid nodes layout and the label value are exported and give machine learning network, by machine learning network according to the
One grid nodes layout and label value, the corresponding relationship of training power grid feature and characteristic value.
7. the method according to claim 1, wherein
The label value according to first grid nodes layout and the first grid nodes layout, training power grid feature and feature
After the corresponding relationship of value, the method also includes:
Obtain the second power project for needing to analyze state；
Third grid nodes layout is obtained according to second power project；
Corresponding relationship described in the power grid characteristic query being laid out by the third grid nodes, obtains and the third grid nodes
The corresponding characteristic value of power grid feature of layout；
The state of the second power project according to obtained Eigenvalues analysis.
8. the method according to the description of claim 7 is characterized in that
It is described to obtain the process for needing to analyze the second power project of state, it specifically includes:
Power generation plan, electric power demand forecasting, topological structure of electric are obtained, and according to the power generation plan, the electricity
Power requirement forecasting, the topological structure of electric obtain the second power project.
9. the method according to the description of claim 7 is characterized in that
The process that third grid nodes layout is obtained according to second power project, specifically includes:
From topological structure of electric is obtained in the second power project, in the topological structure of electric each node power information；Wherein,
The topological structure of electric includes the topological structure of each node in electric system；
Third grid nodes layout is generated according to the power information of the topological structure of electric, each node.
10. according to the method described in claim 9, it is characterized in that, described according to the topological structure of electric, each node
Power information generate third grid nodes layout process, specifically include:
The second intermediate node layout is generated according to the topological structure of node each in the electric system；
Based on the power information of each node, in the corresponding position of each node of second intermediate node layout, mark should
The corresponding colouring information of the power information of node obtains the third grid nodes layout.
11. the method according to the description of claim 7 is characterized in that
Corresponding relationship described in the power grid characteristic query being laid out by the third grid nodes, obtains and the third power grid
The process of the corresponding characteristic value of power grid feature of node layout, specifically includes:
It gives third grid nodes layout output to machine learning network, the third electricity is determined by the machine learning network
The corresponding power grid feature of node layout is netted, and by corresponding relationship described in the power grid characteristic query, obtained special with the power grid
Levy corresponding characteristic value.
12. the method according to the description of claim 7 is characterized in that
The characteristic value is specially the probability value of normal electricity plan or the probability value of abnormal power plan；
The process of the state of second power project according to obtained Eigenvalues analysis, specifically includes:
If the characteristic value is the probability value of normal power project, when the characteristic value is greater than preset first threshold value, it is determined that
The state of second power project is normal；Alternatively,
If the characteristic value is the probability value of abnormal power plan, when the characteristic value is greater than default second threshold, it is determined that
The state of second power project is abnormal.
13. a kind of analysis method of electric system, which is characterized in that the described method includes:
Obtain the second power project for needing to analyze state；
Third grid nodes layout is obtained according to second power project；
The power grid characteristic query corresponding relationship being laid out by the third grid nodes, obtains spy corresponding with the power grid feature
Value indicative；Wherein, the corresponding relationship is the corresponding relationship of the power grid feature and characteristic value according to the first power project training；
The state of the second power project according to obtained Eigenvalues analysis.
14. according to the method for claim 13, which is characterized in that
It is described to obtain the process for needing to analyze the second power project of state, it specifically includes:
Power generation plan, electric power demand forecasting, topological structure of electric are obtained, and according to the power generation plan, the electricity
Power requirement forecasting, the topological structure of electric obtain the second power project.
15. according to the method for claim 13, which is characterized in that
The process that third grid nodes layout is obtained according to second power project, specifically includes:
From topological structure of electric is obtained in the second power project, in the topological structure of electric each node power information；Wherein,
The topological structure of electric includes the topological structure of each node in electric system；
Third grid nodes layout is generated according to the power information of the topological structure of electric, each node.
16. according to the method for claim 15, which is characterized in that described according to the topological structure of electric, each section
The power information of point generates the process of third grid nodes layout, specifically includes:
The second intermediate node layout is generated according to the topological structure of node each in the electric system；
Based on the power information of each node, in the corresponding position of each node of second intermediate node layout, mark should
The corresponding colouring information of the power information of node obtains the third grid nodes layout.
17. according to the method for claim 13, which is characterized in that
The power grid characteristic query corresponding relationship being laid out by the third grid nodes, obtains corresponding with the power grid feature
Characteristic value process, specifically include:
It gives third grid nodes layout output to machine learning network, the third electricity is determined by the machine learning network
The corresponding power grid feature of node layout is netted, and by corresponding relationship described in the power grid characteristic query, obtained special with the power grid
Levy corresponding characteristic value.
18. according to the method for claim 13, which is characterized in that
The characteristic value is specially the probability value of normal electricity plan or the probability value of abnormal power plan；
The process of the state of second power project according to obtained Eigenvalues analysis, specifically includes:
If the characteristic value is the probability value of normal power project, when the characteristic value is greater than preset first threshold value, it is determined that
The state of second power project is normal；Alternatively,
If the characteristic value is the probability value of abnormal power plan, when the characteristic value is greater than default second threshold, it is determined that
The state of second power project is abnormal.
19. a kind of analysis method of electric system, which is characterized in that the described method includes:
Obtain the third power project for needing to analyze state；
Inquiring in normal sample library whether there is and the matched normal electricity plan of the third power project；
If so, the state for analyzing the third power project is normal；
If it is not, the state for then analyzing the third power project is abnormal.
20. according to the method for claim 19, which is characterized in that
It is described to obtain the process for needing to analyze the third power project of state, it specifically includes:
Power generation plan, electric power demand forecasting, topological structure of electric are obtained, and according to the power generation plan, the electricity
Power requirement forecasting, the topological structure of electric obtain third power project.
21. according to the method for claim 19, which is characterized in that in the inquiry normal sample library whether there is with it is described
Before the matched normal electricity plan of third power project, the method also includes:
Normal electricity plan is obtained, and the normal electricity plan is handled using confrontation network algorithm is generated, is obtained more
A new normal electricity plan；
Institute is arrived in the normal electricity plan that will acquire, the multiple normal electricity plans obtained using generation confrontation network algorithm, storage
It states in normal sample library.
22. a kind of analytical equipment of electric system, which is characterized in that described device includes:
Module is obtained, for obtaining the first power project；The first grid nodes layout is obtained according to first power project；
Training module, for the label value according to first grid nodes layout and the first grid nodes layout, training power grid
The corresponding relationship of feature and characteristic value；The corresponding relationship is used to analyze the state of electric system.
23. device according to claim 22, which is characterized in that
The acquisition module is specifically used for during obtaining the first grid nodes layout according to first power project,
From topological structure of electric is obtained in first power project, in the topological structure of electric each node power information；Wherein,
The topological structure of electric includes the topological structure of each node in electric system；According to the topological structure of electric, each section
The power information of point generates the first grid nodes layout；
The training module, specifically in the label according to first grid nodes layout and the first grid nodes layout
During the corresponding relationship of value, training power grid feature and characteristic value, network algorithm is fought to first power grid using generating
Node layout is handled, and obtains multiple second grid nodes layout, the label value of the second grid nodes layout with it is described
The label value of first grid nodes layout is identical；It is laid out according to first grid nodes layout, first grid nodes
The label value of label value, the multiple second grid nodes layout, each second grid nodes layout, training power grid feature and feature
The corresponding relationship of value.
24. device according to claim 22, which is characterized in that
The acquisition module is also used to obtain the second power project for needing to analyze state；The is obtained according to the second power project
Three grid nodes layout；By the third grid nodes be laid out power grid characteristic query described in corresponding relationship, obtain with it is described
The corresponding characteristic value of power grid feature of third grid nodes layout；
Described device further include:
Analysis module, the state for the second power project according to obtained Eigenvalues analysis.
25. a kind of analytical equipment of electric system, which is characterized in that described device includes:
Module is obtained, needs to analyze the second power project of state for obtaining；Third is obtained according to second power project
Grid nodes layout；The power grid characteristic query corresponding relationship being laid out by the third grid nodes obtains special with the power grid
Levy corresponding characteristic value；Wherein, the corresponding relationship is pair of the power grid feature and characteristic value according to the first power project training
It should be related to；
Analysis module, the state for the second power project according to obtained Eigenvalues analysis.
26. device according to claim 25, which is characterized in that
The acquisition module is specifically used for during obtaining third grid nodes layout according to second power project,
From topological structure of electric is obtained in the second power project, in the topological structure of electric each node power information；Wherein, described
Topological structure of electric includes the topological structure of each node in electric system；According to the topological structure of electric, each node
Power information generates third grid nodes layout；
The analysis module, if being the probability value of normal power project specifically for the characteristic value, when the characteristic value is greater than
When preset first threshold value, it is determined that the state of second power project is normal；Alternatively, if the characteristic value is abnormal power
The probability value of plan, when the characteristic value is greater than default second threshold, it is determined that the state of second power project is different
Often.
27. a kind of analytical equipment of electric system, which is characterized in that described device includes:
Module is obtained, needs to analyze the third power project of state for obtaining；
Enquiry module whether there is based on the matched normal electricity of third power project by inquiring in normal sample library
It draws；
Analysis module, for when query result, which is, is, then it to be normal for analyzing the state of the third power project；When inquiry is tied
When fruit is no, then it is abnormal for analyzing the state of the third power project.
28. a kind of power management apparatus, which is characterized in that the power management apparatus includes:
Processor, for obtaining the first power project；The first grid nodes layout is obtained according to first power project；According to
The label value of the first grid nodes layout and first grid nodes layout, training power grid feature are corresponding with characteristic value
Relationship；Wherein, the corresponding relationship is used to analyze the state of electric system.
29. a kind of power management apparatus, which is characterized in that the power management apparatus includes: processor, needs to divide for obtaining
Second power project of analysis state；Third grid nodes layout is obtained according to second power project；Pass through the third electricity
The power grid characteristic query corresponding relationship for netting node layout, obtains characteristic value corresponding with the power grid feature；Wherein, the correspondence
Relationship is the corresponding relationship of the power grid feature and characteristic value according to the first power project training；According to obtained Eigenvalues analysis institute
State the state of the second power project.
30. a kind of power management apparatus, which is characterized in that the power management apparatus includes:
Processor needs to analyze the third power project of state for obtaining；Inquire normal sample library in the presence or absence of with it is described
The matched normal electricity plan of third power project；If so, the state for analyzing the third power project is normal；If it is not,
It is abnormal for then analyzing the state of the third power project.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810067602.7A CN110084448A (en) | 2018-01-24 | 2018-01-24 | A kind of analysis method of electric system, device and its equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810067602.7A CN110084448A (en) | 2018-01-24 | 2018-01-24 | A kind of analysis method of electric system, device and its equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110084448A true CN110084448A (en) | 2019-08-02 |
Family
ID=67412134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810067602.7A Pending CN110084448A (en) | 2018-01-24 | 2018-01-24 | A kind of analysis method of electric system, device and its equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084448A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112560249A (en) * | 2020-12-09 | 2021-03-26 | 中国电力科学研究院有限公司 | Power grid static security analysis method and system based on dual-discriminator generated countermeasure network |
| CN112926725A (en) * | 2020-12-01 | 2021-06-08 | 杭州电力设备制造有限公司 | Data analysis method and system based on thermal image |
| CN112926725B (en) * | 2020-12-01 | 2024-05-14 | 杭州电力设备制造有限公司 | Data analysis method and system based on thermal image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778841A (en) * | 2016-11-30 | 2017-05-31 | 国网上海市电力公司 | The method for building up of abnormal electricity consumption detection model |
| CN106786533A (en) * | 2016-12-26 | 2017-05-31 | 国网山东省电力公司泰安供电公司 | The acquisition methods and device of electric dispatching plan data |
| CN106992517A (en) * | 2017-04-28 | 2017-07-28 | 国网山东省电力公司莱西市供电公司 | Electric power line loss analysis method, device and system |
| CN107016410A (en) * | 2017-03-27 | 2017-08-04 | 国网江苏省电力公司电力科学研究院 | Power information acquisition system method for diagnosing faults and trouble-shooter |
| CN107085164A (en) * | 2017-03-22 | 2017-08-22 | 清华大学 | A kind of electric network fault type determines method and device |
-
2018
- 2018-01-24 CN CN201810067602.7A patent/CN110084448A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778841A (en) * | 2016-11-30 | 2017-05-31 | 国网上海市电力公司 | The method for building up of abnormal electricity consumption detection model |
| CN106786533A (en) * | 2016-12-26 | 2017-05-31 | 国网山东省电力公司泰安供电公司 | The acquisition methods and device of electric dispatching plan data |
| CN107085164A (en) * | 2017-03-22 | 2017-08-22 | 清华大学 | A kind of electric network fault type determines method and device |
| CN107016410A (en) * | 2017-03-27 | 2017-08-04 | 国网江苏省电力公司电力科学研究院 | Power information acquisition system method for diagnosing faults and trouble-shooter |
| CN106992517A (en) * | 2017-04-28 | 2017-07-28 | 国网山东省电力公司莱西市供电公司 | Electric power line loss analysis method, device and system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926725A (en) * | 2020-12-01 | 2021-06-08 | 杭州电力设备制造有限公司 | Data analysis method and system based on thermal image |
| CN112926725B (en) * | 2020-12-01 | 2024-05-14 | 杭州电力设备制造有限公司 | Data analysis method and system based on thermal image |
| CN112560249A (en) * | 2020-12-09 | 2021-03-26 | 中国电力科学研究院有限公司 | Power grid static security analysis method and system based on dual-discriminator generated countermeasure network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | Intelligent edge: Leveraging deep imitation learning for mobile edge computation offloading | |
| Cai et al. | Simultaneous optimization of airspace congestion and flight delay in air traffic network flow management | |
| Kremers | Modelling and simulation of electrical energy systems through a complex systems approach using agent-based models | |
| CN104580284B (en) | Traffic assignments device and method for distributing business | |
| CN108681667A (en) | A kind of unit type recognition methods, device and processing equipment | |
| Fan et al. | Federated generative adversarial learning | |
| Wang et al. | Metaheuristics for robust graph coloring | |
| CN109344162A (en) | Business data processing method and device | |
| Fateen et al. | Intelligent firefly algorithm for global optimization | |
| CN110188210A (en) | One kind is based on figure regularization and the independent cross-module state data retrieval method of mode and system | |
| Erdelyi et al. | Approximate dynamic programming for dynamic capacity allocation with multiple priority levels | |
| US11455544B2 (en) | Prediction model generation device, prediction model generation method, and recording medium | |
| Sun et al. | Edge learning with timeliness constraints: Challenges and solutions | |
| CN111494964A (en) | Virtual article recommendation method, model training method, device and storage medium | |
| Tarutani et al. | Temperature distribution prediction in data centers for decreasing power consumption by machine learning | |
| CN116700277A (en) | Method and device for generating inspection path, electronic equipment and storage medium | |
| Reece et al. | Determining future capacity for an ambulatory surgical center with discrete event simulation | |
| CN110084448A (en) | A kind of analysis method of electric system, device and its equipment | |
| CN114418093A (en) | Method and device for training path characterization model and outputting information | |
| CN109857829A (en) | A kind of geographic information data fusion system | |
| Chen | Intelligent scheduling approaches for a wafer fabrication factory | |
| Chen et al. | Castor: Contextual IoT time series data and model management at scale | |
| Hu et al. | Query-policy misalignment in preference-based reinforcement learning | |
| WO2021229626A1 (en) | Learning device, learning method, and learning program | |
| CN114297041A (en) | Network heterogeneous computing platform testing method and device and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40010990 Country of ref document: HK |
|
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190802 |
|
| RJ01 | Rejection of invention patent application after publication |