CN109949178A - One kind is based on the judgement of support vector machines middle voltage distribution networks power-off event and complementing method - Google Patents
One kind is based on the judgement of support vector machines middle voltage distribution networks power-off event and complementing method Download PDFInfo
- Publication number
- CN109949178A CN109949178A CN201910131792.9A CN201910131792A CN109949178A CN 109949178 A CN109949178 A CN 109949178A CN 201910131792 A CN201910131792 A CN 201910131792A CN 109949178 A CN109949178 A CN 109949178A
- Authority
- CN
- China
- Prior art keywords
- power
- node
- event
- measuring point
- electric current
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Abstract
The invention discloses one kind based on the judgement of support vector machines middle voltage distribution networks power-off event and complementing method, belongs to power distribution network running technology field.This method includes 4 links altogether, is distribution network model building, vector machine building, vector machine solution analysis and the judgement and merging of fault type respectively.The present invention is based on support vector machines middle voltage distribution networks power-off event judgment methods, determine the final power failure node and type of power-off event, and the power-off event progress completion that medium voltage distribution network leakage is adopted, is accidentally adopted, it solves the problems, such as failing to report, reporting by mistake, lying about in mesolow distribution power-off event statistics, promotes the statistical accuracy of distribution network reliability.
Description
Technical field
The invention belongs to power distribution network running technology fields, and in particular to one kind is based on support vector machines middle voltage distribution networks power failure thing
Part judgement and complementing method.
Background technique
Current power distribution network scale is big, broad covered area, and equipment operating environment is more severe, and acquisition device quality is irregular,
There is the case where failing to report to power-off event in part terminal, the customer interrupted data integrity rate for pushing to power quality system is insufficient
70%.Some electrical power company even will appear the phenomenon that lying about for examination pressure.But by centering press power-off event fail to report with
Track analysis, discovery device function problem itself is the most prominent, accounts for 80%, if terminal specification upgrades not in time, adverse circumstances
Middle operation leads to the defects of aging of battery antenna, SIM card damage or failure, leads to not acquisition upload and stops powering on event.This portion
Although separating device, which can not upload, stops powering on event, most case points and proximal event point can collect voltage, electricity
The load operations data-signals such as stream, capable of helping to screen the practical power failure of user, a situation arises, can be used as the breakthrough of event completion
Point.
In view of the above-mentioned problems, solution is in the prior art, with power information acquisition system, marketing system, 95598 systems
Based on the systems such as system, PMS system and distribution O&M control platform, carries out operation system aggregation of data and integrate, pressed in exploitation
Power-off event intelligence completion module (abbreviation completion module), is uploaded to again after carrying out analysis completion to medium voltage network power-off event
State's network electric energy quality on-line monitoring system.The first stage of construction module is mainly endeavoured each operation system data carrying out integrated go forward side by side
The analysis of row power-off event and completion, but power-off event completion rule is based only upon simple logic rules and engineering experience, such as route
Upstream power failure downstream centainly has a power failure, and 70% user of certain route, which has a power failure, thinks completely power failure etc., and experience completion method is to power network topology
Archive information dependence is high, and the accuracy of completion and intelligence still need to further be promoted.
The present invention is directed to this problem, has studied a kind of based on support vector machines middle voltage distribution networks power-off event complementing method.
Summary of the invention
It is an object of the invention to overcome deficiency in the prior art, provide a kind of based on support vector machines middle voltage distribution networks
Power-off event judgement and complementing method, determine final power failure node, avoid low and medium voltage distribution network power-off event count fail to report,
It reports by mistake, lie about, promote the statistical accuracy of power supply reliability.
Judged in order to solve the above technical problems, the present invention provides one kind based on support vector machines middle voltage distribution networks power-off event
Method, characterized in that including following procedure:
S1 obtains the electric network model structure of middle voltage distribution networks, divides distribution node type according to equipment at distribution node；
S2, establishes vector machine corresponding with node type at each distribution node, and the input of each vector machine is each section
Measuring point flow data at point, the output of vector machine are whether power-off event to occur at each node；
S3 is judged using the trained vector machine at each distribution node according to the measuring point flow data at each node
Whether power-off event occurs at each node out；
S4, by all distribution nodes whether power failure judging result, merged upwards step by step according to electric network model structure, determine
Final power failure node.
Further, distribution node type includes: transformer node, outgoing line cabinet node, ring network cabinet node, specially switch political loyalty point and
Public affairs five types of traitorous point.
Further, five kinds of distribution node types are described in detail below:
1) transformer node at the root node of 110/10kV Circuit Fault on Secondary Transformer outlet, is defined as transformer node, is denoted as
Node type A；
2) outgoing line cabinet node is defined as outgoing line cabinet node in power distribution room outgoing line cabinet at the root node of every outlet, it is denoted as
Node type B；
3) ring network cabinet node, several ring network cabinet positions in every outlet of medium voltage distribution network, is defined as ring network cabinet node, remembers
For node type C；
4) specially switch political loyalty point, at 10/380V user's dedicated transformer, be defined as point of specially switching political loyalty, be denoted as node type D；
5) public traitorous point at 10/380V common transformer, is defined as public traitorous point, is denoted as node type E.
Further, each node type establishes a corresponding vector machine, the selection of all kinds of vector machine measuring points
And the flow data type difference of measuring point acquisition is as follows:
1) vector machine A: measuring point are as follows: branch's outlet is respectively randomly selected under the power transformation device, in each outgoing line cabinet, each
The head end of outlet respectively chooses a special or public variable measuring point；
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；It is flat
Equal voltage, electric current and power；Minimum voltage, electric current and power；
2) vector machine B: measuring point: for the outlet first, last, in respectively choose a special or public variable measuring point, in addition with this
In 2 adjacent outlets of outlet, a special or public variable measuring point is respectively selected；
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；It is flat
Equal voltage, electric current and power；Minimum voltage, electric current and power；
3) vector machine C: a special or public variable measuring point respectively measuring point: is selected for judgement point front and back；
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；It is flat
Equal voltage, electric current and power；Minimum voltage, electric current and power；
4) vector machine D: measuring point: being measuring point specially to become all special change node box changes of user, in addition in the special change
Node it is neighbouring, reselection is adjacent to 1 special or public variable measuring point；
The flow data of initial input: case height specially becomes 96 metric data of the same day of the special point of switching political loyalty of user itself: maximum
Voltage, electric current and power；Average voltage, electric current and power；Minimum voltage, electric current and power；Specially become neighbouring measure of user to click
Take average voltage, electric current, power；
5) vector machine E: measuring point are as follows: the single public measuring point become, neighbouring selection 1 specially becomes or public variable measuring point；
The flow data of initial input: 96 metric data of the same day of case height；Maximum voltage, electric current and power；It is average
Voltage, electric current and power；Minimum voltage, electric current and power；Neighbouring measuring point chooses average voltage, electric current, power.
Further, original input data carries out data normalization to original input data before input vector machine respectively
After processing and data dimension-reduction treatment.
Further, the corresponding power failure type of distribution node is respectively as follows: transformer node and corresponds to power failure type A, outgoing line cabinet
Node corresponds to power failure type B, ring network cabinet node corresponds to power failure Type C, the corresponding power failure type D of point of specially switching political loyalty, and public point of switching political loyalty corresponds to
Power failure type E.
Further, the inclusion relation of event type is power failure typeOr
Correspondingly, the present invention provides one kind to be based on support vector machines middle voltage distribution networks power-off event complementing method, feature
It is, including following procedure:
Acquire the power-off event of fault point in middle voltage distribution networks；
Using above-mentioned power-off event judgment method, final power failure node is determined；
The fault point of acquisition is compared with the power failure node determined, according to comparison result completion power-off event.
Compared with prior art, the beneficial effects obtained by the present invention are as follows being: the present invention is based on support vector machines middle voltage distribution networks
Power-off event judgment method is analyzed using node trend operation data, determines final power failure node, solves medium-voltage distribution
The leakage of net power-off event adopts, accidentally adopts problem, avoids failing to report, report by mistake, lying about for power-off event statistics, promotes the system of power supply reliability
Count accuracy.
Detailed description of the invention
Fig. 1 is the middle voltage distribution networks power-off event completion flow chart the present invention is based on support vector machines；
Fig. 2 is the business framework figure between completion module of the present invention and related system；
Fig. 3 is node of the present invention and power failure type schematic diagram；
Fig. 4 is the dimensionality reduction result figure of E class vector machine in example of the present invention；
Fig. 5 is the parameter optimization procedure chart of E class vector machine in example of the present invention.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings.Following embodiment is only used for clearly illustrating the present invention
Technical solution, and not intended to limit the protection scope of the present invention.
One kind of the invention is based on support vector machines middle voltage distribution networks power-off event judgment method, and this method includes 4 rings altogether
Section is distribution network model building, vector machine building, vector machine solution analysis and the judgement and merging of fault type respectively.Referring to Fig. 1
It is shown, comprising the following steps:
Step S1 obtains the electric network model structure of middle voltage distribution networks, divides distribution node type according to equipment at distribution node.
Distribution topological data, the public special change account data become in account data and power information acquisition system in integrated PMS,
To construct the electric network model structure of middle voltage distribution networks.
The building of electric network model, business framework figure such as Fig. 2 institute between the data interaction with other systems, system
Show, wherein the Role delineation of each system is as follows:
1) route, the area Zhuan Gongbiantai files on each of customers power information acquisition system: are provided；Terminal is original to stop powering on event, uses
The information on load such as family voltage, electric current, power and Business Process System procedure information.
2) PMS (engineering production management system): providing route, public change Equipment Foundations account, and power grid topology model provides
There is switching-state information, feeder current in 10kV feeder line.
3) 95598: the information such as plan, fault outage and the breakdown repair processing result issued outward are provided.
4) fiber channel monitors: providing the concentrator power down information of fiber optic communication, the area Ji Tai power-off event.
5) power quality system: specially becoming basic machine account information from marketing, and public affairs become basic machine account information and come from PMS system,
Power quality is realized corresponding with marketing, PMS archives account；Middle Voltage power-off event receives power information acquisition system and pushes away automatically
The power-off event sent manually carries out power failure property and reason judgement and selection；Count power supply reliability index.
The abstract circuit model of medium voltage distribution network has several as shown in figure 3, starting point is 110/10kV transformer in network
Outlet, every outlet are divided into several segments, then several public changes of section branch line connection or specially change.Outlet is connected in branch below
The equipment such as cabinet, ring network cabinet, common transformer, dedicated transformer.The present invention is different according to influence face of the distinct device in failure
Sample has divided different node types to them.The fault coverage of different node types is different, extract equipment key position
It is abstracted into circuit node.It is transformer node, outgoing line cabinet node, ring network cabinet node, specially switch political loyalty point and public affairs by distribution node division
Five types of traitorous point；It is described in detail below:
1) transformer node at the root node of 110/10kV Circuit Fault on Secondary Transformer outlet, is defined as transformer node, is denoted as
Node type A；
2) outgoing line cabinet node is defined as outgoing line cabinet node in power distribution room outgoing line cabinet at the root node of every outlet, it is denoted as
Node type B；
3) ring network cabinet node, several ring network cabinet positions in every outlet of medium voltage distribution network, is defined as ring network cabinet node, remembers
For node type C；
4) specially switch political loyalty point, at 10/380V user's dedicated transformer, be defined as point of specially switching political loyalty, be denoted as node type D；
5) public traitorous point at 10/380V common transformer, is defined as public traitorous point, is denoted as node type E.
Step S2, establishes vector machine corresponding with node type at each distribution node, and the input of each vector machine is
Measuring point flow data at each node, the output of vector machine are whether power-off event to occur at each node.
According to failure, corresponding 5 kinds of power failure types, corresponding relationship are described below at 5 kinds of node types in distribution:
1) power failure type A is corresponded to and is had a power failure at power transformation tiny node --- since transformer equipment failure or higher level have a power failure, causes
Pressure transformer radiation area full cut-off electricity in entire；
2) power failure type B, corresponding outgoing line cabinet node have a power failure --- and route head end outgoing line cabinet switch disconnects, and leads to whole route
Power loss；
3) power failure Type C, corresponding ring network cabinet node have a power failure --- and certain looped network cabinet switch tripping among route leads to route
One section of power failure；
4) power failure type D, corresponding specially traitorous point have a power failure --- and separate unit specially becomes user and specially becomes power failure, and more specially become user and own
Specially become and has a power failure；
5) power failure type E, corresponding public point of switching political loyalty have a power failure --- and the area Gong Biantai has a power failure.
Establish five class vector machines respectively at corresponding each node according to five kinds of power failure types, as shown in figure 1 corresponding A, B,
C, D, E class vector machine, each node type establish a corresponding vector machine.The node type of the same node has a power failure
Type and vector machine type are all corresponding, so being indicated with the same letter.Vector machine is a machine learning method, input
Some data volumes, are analyzed and determined, the result is that 0 or 1, that is, judging the node on earth is to have a power failure or do not have a power failure for output.
The input data of the vector machine of each type is each different, the selection of all kinds of vector machine measuring points and measuring point acquisition
Data type difference it is as follows:
1) vector machine A: measuring point are as follows: branch's outlet is respectively randomly selected under the power transformation device, in each outgoing line cabinet, each
The head end of outlet respectively chooses a special or public variable measuring point.
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；It is flat
Equal voltage, electric current and power；Minimum voltage, electric current and power；
2) vector machine B: measuring point: for the outlet first, last, in respectively choose a special or public variable measuring point, in addition with this
In 2 adjacent outlets of outlet, a special or public variable measuring point is respectively selected.
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；It is flat
Equal voltage, electric current and power；Minimum voltage, electric current and power；
3) vector machine C: a special or public variable measuring point respectively measuring point: is selected for judgement point front and back.
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；It is flat
Equal voltage, electric current and power；Minimum voltage, electric current and power；
4) vector machine D: measuring point: specially to become all special change node box changes of user, (it may include 1 that one, which specially becomes user,
Specially become, it is also possible to include multiple special changes) it is measuring point, in addition in the neighbouring of point of specially switching political loyalty, reselection is specially or public adjacent to 1
Variable measuring point.
The flow data of initial input: case height specially becomes 96 metric data of the same day of the special point of switching political loyalty of user itself: maximum
Voltage, electric current and power；Average voltage, electric current and power；Minimum voltage, electric current and power；Specially become neighbouring measure of user to click
Take average voltage, electric current, power.
5) vector machine E: measuring point are as follows: the single public measuring point become, neighbouring selection 1 specially becomes or public variable measuring point.
The flow data of initial input: 96 metric data of the same day of case height；Maximum voltage, electric current and power；It is average
Voltage, electric current and power；Minimum voltage, electric current and power；Neighbouring measuring point chooses average voltage, electric current, power.
Step S3, using the trained vector machine at each distribution node, according to the measuring point flow data at each node
Judge whether power-off event occurs at each node.
Vector machine solves analysis, including three data normalization processing, Data Dimensionality Reduction and SVM operating analysis processes, wherein
Vector machine parameter training and vector machine test judgement are required to the processing of these three links.
Vector machine parameter training process, including following treatment process:
1) all types of vector machines are directed to, the original input data of vector machine is obtained；
2) data normalization is handled, and sample current in electric system, voltage, the dimension of power each dimension are different and poor
Away from very greatly, [0,1] normalization is carried out to every dimension, and screen to normalization result according to data fluctuations, eliminated individually
Outlier is on normalized influence；
3) Data Dimensionality Reduction is analyzed, and is principle to cover 95% or more prime information amount, using being locally linear embedding into LLE
(Locally linear embedding) Method of Data with Adding Windows carries out Data Dimensionality Reduction, and the low-dimensional data group after dimensionality reduction is as vector
The training input of machine；As shown in figure 4, histogram represents preceding 4 dimension ingredient after dimensionality reduction in figure, every dimension ingredient covers former total information
Percentage, broken line represent the accumulative percentage for covering former total information, and 4 tie up cumulative percentages up to 99.2%, therefore to E class vector
Machine, 13 dimension datas can be near 4 dimensions.
4) vector machine parameter training finds optimal penalty factor c and core parameter using the method for mesh parameter optimizing
γ, and substitute into vector machine model.It is illustrated in figure 5 the training process of E class vector machine penalty factor and core parameter, shows E in figure
Type vector machine best penalty factor c=0.125, best core parameter g=0.5.
Vector machine tests deterministic process, including following treatment process:
1) all types of vector machines are directed to, obtain vector machine original input data from extraction system；
2) data normalization is handled, and sample current in electric system, voltage, the dimension of power each dimension are different and poor
Away from very greatly, [0,1] normalization is carried out to every dimension, and screen to normalization result according to data fluctuations, eliminated individually
Outlier is on normalized influence；
3) Data Dimensionality Reduction is analyzed, and is that principle carries out Data Dimensionality Reduction to cover 95% or more prime information amount, low after dimensionality reduction
Training input of the dimension data group as vector machine；
4) data after dimensionality reduction are substituted into vector machine model, carries out test judgement, determines whether distribution node stops
Electric fault, being is 1, and no is -1.
Step S4, by all distribution nodes whether power failure judging result, merged upwards step by step according to electric network model structure,
Determine final power failure node.
1 time power-off event fault point only has 1, but will lead to multiple nodes in power grid and all have a power failure, before be each
Judge whether the power failure of a node, below a step be further to be analyzed according to the case where above-mentioned each node, to look for
Final fault point out determines final power-off event.
According to the inclusion relation in event type influence face, it is respectivelyOrD, E thing
Part is end power-off event arranged side by side, is merged to power-off fault event.Specific combination principle is that prime event merges rear class
Event, even A event occur, then B, C, D, E event all occur certainly, finally exporting the result is that A；If C occurs, portion
Divide D or E event that can occur, then result output is still C event.
According to the power-off event type of final output, final power failure node is determined.
Power-off event is acquired with extraction system, but mistake may be adopted, it is also possible to which leakage is adopted.Therefore, the present invention also provides one kind
Based on support vector machines middle voltage distribution networks power-off event complementing method, wrong correction will be adopted after exactly passing through judgement, leaks the supplement adopted,
Detailed process are as follows:
Obtain power-off event record sheet original in power information acquisition system, comprising:
1) power-off event sent on each acquisition terminal node, i.e., all D classes and E class node location；
2) it according to the power-off event sent in each terminal, is combined in conjunction with power network topology, comprehensive descision position of failure point, in formation
Final power-off event type is netted in press-fitting；
Completion is carried out to the power-off event record sheet in extraction system, comprising:
1) it using E type and the judging result of D class vector machine, is tied with the power-off event sent on each acquisition terminal node
Fruit compares, and carries out completion；
2) final power failure node is determined using vector machine is comprehensive, determine power-off event type, and with original in extraction system
Power-off event type carry out result comparison, carry out completion.
The fault point of acquisition is compared with the power failure node determined, if occurring adopting mistake or leaking adopts, is entangled to mistake is adopted
Just, the supplement adopted is leaked.For example, collecting power-off event is B type, and judge that power-off event is B type or C using vector machine
Type then judges that system adopts wrong power-off event, resurveys correct power-off event A；If collecting power-off event is B type,
And judge that power-off event is type-A using vector machine, then judge that power-off event, supplement acquisition power-off event A are adopted in system leakage.
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 the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of 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 should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, 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 one
The step of function of being specified in a box or multiple boxes.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, several improvement and deformations can also be made, these improvement and deformations
Also it should be regarded as protection scope of the present invention.
Claims (8)
1. one kind is based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that including following procedure:
S1 obtains the electric network model structure of middle voltage distribution networks, divides distribution node type according to device type at distribution node；
S2, establishes the vector machine of type corresponding with node type at each distribution node, and the input of each vector machine is everywhere
The flow data of the correspondence measuring point of node, the output of vector machine are whether power-off event to occur at each node；
S3 is judged using the trained vector machine at each distribution node according to the correspondence measuring point flow data at each node
Whether power-off event occurs at each node out；
S4, by all distribution nodes whether power failure judging result, merged upwards step by step according to electric network model structure, determined final
Power failure node.
2. according to claim 1 a kind of based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that
Distribution node type includes: transformer node, outgoing line cabinet node, ring network cabinet node, specially switch political loyalty point and public five types of point of switching political loyalty.
3. according to claim 2 a kind of based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that
Five node types are described in detail below:
1) transformer node at the root node of 110/10kV Circuit Fault on Secondary Transformer outlet, is defined as transformer node, is denoted as node
Type A；
2) outgoing line cabinet node is defined as outgoing line cabinet node in power distribution room outgoing line cabinet at the root node of every outlet, it is denoted as node
Type B；
3) ring network cabinet node, several ring network cabinet positions in every outlet of medium voltage distribution network, is defined as ring network cabinet node, is denoted as section
Vertex type C；
4) specially traitorous point, 10/380V user specially at change, are defined as point of specially switching political loyalty, are denoted as node type D；
5) public traitorous point at 10/380V common transformer, is defined as public traitorous point, is denoted as node type E.
4. according to claim 1 a kind of based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that
Each node type establishes the number of a corresponding vector machine, the selection of the measuring point of all kinds of vector machines and measuring point acquisition
Distinguish according to type as follows:
1) vector machine A: measuring point are as follows: branch's outlet, each outlet are respectively randomly selected under the power transformation device, in each outgoing line cabinet
Head end respectively choose a special or public variable measuring point；
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；Average electricity
Pressure, electric current and power；Minimum voltage, electric current and power；
2) vector machine B: measuring point: for the outlet first, last, in respectively choose a special or public variable measuring point, in addition with the outlet
In 2 adjacent outlets, a special or public variable measuring point is respectively selected；
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；Average electricity
Pressure, electric current and power；Minimum voltage, electric current and power；
3) vector machine C: a special or public variable measuring point respectively measuring point: is selected for judgement point front and back；
The flow data of initial input: 96 metric data of the same day of each measuring point: maximum voltage, electric current and power；Average electricity
Pressure, electric current and power；Minimum voltage, electric current and power；
4) vector machine D: measuring point: being measuring point specially to become all special change node box changes of user, in addition in the point of specially switching political loyalty
It is neighbouring, reselection is adjacent to 1 special or public variable measuring point；
The flow data of initial input: case height specially become user itself specially switch political loyalty point 96 metric data of the same day: maximum voltage,
Electric current and power；Average voltage, electric current and power；Minimum voltage, electric current and power；Specially become user to choose averagely adjacent to measuring point
Voltage, electric current, power；
5) vector machine E: measuring point are as follows: the single public measuring point become, neighbouring selection 1 specially becomes or public variable measuring point；
The flow data of initial input: 96 metric data of the same day of case height；Maximum voltage, electric current and power；Average voltage,
Electric current and power；Minimum voltage, electric current and power；Neighbouring measuring point chooses average voltage, electric current, power.
5. according to claim 1 a kind of based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that
Original input data carries out data normalization processing and data dimension-reduction treatment before input vector machine, to original input data respectively
Afterwards.
6. according to claim 2 a kind of based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that
The corresponding power failure type of distribution node be respectively as follows: transformer node corresponds to power failure type A, outgoing line cabinet node corresponds to power failure type B,
Ring network cabinet node corresponds to power failure Type C, specially switch political loyalty the corresponding power failure type D of point, the public corresponding power failure type E of point of switching political loyalty.
7. according to claim 6 a kind of based on support vector machines middle voltage distribution networks power-off event judgment method, characterized in that
The inclusion relation of event type is power failure typeOr
8. one kind is based on support vector machines middle voltage distribution networks power-off event complementing method, characterized in that including following procedure:
Acquire the power-off event of fault point in middle voltage distribution networks；
Using power-off event judgment method described in any one of claims 1 to 5, final power failure node is determined；
The fault point of acquisition is compared with the power failure node determined, according to comparison result completion power-off event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910131792.9A CN109949178B (en) | 2019-02-22 | 2019-02-22 | Method for judging and complementing power failure event of medium-voltage distribution network based on support vector machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910131792.9A CN109949178B (en) | 2019-02-22 | 2019-02-22 | Method for judging and complementing power failure event of medium-voltage distribution network based on support vector machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109949178A true CN109949178A (en) | 2019-06-28 |
| CN109949178B CN109949178B (en) | 2023-06-16 |
Family
ID=67007636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910131792.9A Active CN109949178B (en) | 2019-02-22 | 2019-02-22 | Method for judging and complementing power failure event of medium-voltage distribution network based on support vector machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109949178B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909811A (en) * | 2019-11-28 | 2020-03-24 | 国网湖南省电力有限公司 | OCSVM (online charging management system) -based power grid abnormal behavior detection and analysis method and system |
| CN111813825A (en) * | 2020-06-02 | 2020-10-23 | 国网江西省电力有限公司电力科学研究院 | Distribution transformer outlet power failure event missing report automatic detection method and system |
| CN113344364A (en) * | 2021-05-31 | 2021-09-03 | 广东电网有限责任公司佛山供电局 | Power failure plan risk analysis method and device, electronic equipment and storage medium |
| CN113377835A (en) * | 2021-06-09 | 2021-09-10 | 国网河南省电力公司电力科学研究院 | Distribution network line power failure identification method based on long-short term memory deep learning network |
| WO2023165348A1 (en) * | 2022-03-03 | 2023-09-07 | 广东电网有限责任公司梅州供电局 | Method and device for determining switching state of reactive compensation of distribution transformer |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100152910A1 (en) * | 2008-05-09 | 2010-06-17 | Accenture Global Services Gmbh | Power grid outage and fault condition management |
| CN102930007A (en) * | 2012-10-30 | 2013-02-13 | 广东电网公司 | User power supply recovery emergency degree classification method in large area power failure emergency processing |
| CN105974265A (en) * | 2016-04-29 | 2016-09-28 | 北京四方继保自动化股份有限公司 | SVM (support vector machine) classification technology-based power grid fault cause diagnosis method |
| CN106443436A (en) * | 2016-11-10 | 2017-02-22 | 国电南瑞科技股份有限公司 | Detection method for remote control function of switch by combining loop risk with operation and maintenance assistant decision |
| CN107292513A (en) * | 2017-06-21 | 2017-10-24 | 国网辽宁省电力有限公司 | A kind of method that power customer management is realized based on svm classifier algorithm |
| CN108169621A (en) * | 2017-12-05 | 2018-06-15 | 国电南瑞科技股份有限公司 | Taiwan area power-off event complementing method based on support vector machines |
| CN108364187A (en) * | 2017-12-20 | 2018-08-03 | 国网冀北电力有限公司承德供电公司 | A kind of power failure sensitive users based on power failure sensitivity characteristic determine method and system |
| CN108734381A (en) * | 2018-04-11 | 2018-11-02 | 国网山东省电力公司 | A kind of outage information collaborative management method, apparatus and system |
| CN109002937A (en) * | 2018-09-07 | 2018-12-14 | 深圳供电局有限公司 | Load Forecasting, device, computer equipment and storage medium |
| CN109035066A (en) * | 2018-09-30 | 2018-12-18 | 国网山西省电力公司阳泉供电公司 | The high breaking route genetic analysis of 10 kilovolts of distributions and administering method based on SVM |
-
2019
- 2019-02-22 CN CN201910131792.9A patent/CN109949178B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100152910A1 (en) * | 2008-05-09 | 2010-06-17 | Accenture Global Services Gmbh | Power grid outage and fault condition management |
| CN102930007A (en) * | 2012-10-30 | 2013-02-13 | 广东电网公司 | User power supply recovery emergency degree classification method in large area power failure emergency processing |
| CN105974265A (en) * | 2016-04-29 | 2016-09-28 | 北京四方继保自动化股份有限公司 | SVM (support vector machine) classification technology-based power grid fault cause diagnosis method |
| CN106443436A (en) * | 2016-11-10 | 2017-02-22 | 国电南瑞科技股份有限公司 | Detection method for remote control function of switch by combining loop risk with operation and maintenance assistant decision |
| CN107292513A (en) * | 2017-06-21 | 2017-10-24 | 国网辽宁省电力有限公司 | A kind of method that power customer management is realized based on svm classifier algorithm |
| CN108169621A (en) * | 2017-12-05 | 2018-06-15 | 国电南瑞科技股份有限公司 | Taiwan area power-off event complementing method based on support vector machines |
| CN108364187A (en) * | 2017-12-20 | 2018-08-03 | 国网冀北电力有限公司承德供电公司 | A kind of power failure sensitive users based on power failure sensitivity characteristic determine method and system |
| CN108734381A (en) * | 2018-04-11 | 2018-11-02 | 国网山东省电力公司 | A kind of outage information collaborative management method, apparatus and system |
| CN109002937A (en) * | 2018-09-07 | 2018-12-14 | 深圳供电局有限公司 | Load Forecasting, device, computer equipment and storage medium |
| CN109035066A (en) * | 2018-09-30 | 2018-12-18 | 国网山西省电力公司阳泉供电公司 | The high breaking route genetic analysis of 10 kilovolts of distributions and administering method based on SVM |
Non-Patent Citations (2)
| Title |
|---|
| 张波等: "SVM在中压配网停电事件补全中的应用研究", 《电力工程技术》 * |
| 王宇飞: "支持向量机在电网故障诊断中的应用研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909811A (en) * | 2019-11-28 | 2020-03-24 | 国网湖南省电力有限公司 | OCSVM (online charging management system) -based power grid abnormal behavior detection and analysis method and system |
| CN110909811B (en) * | 2019-11-28 | 2022-10-18 | 国网湖南省电力有限公司 | OCSVM (online charging management system) -based power grid abnormal behavior detection and analysis method and system |
| CN111813825A (en) * | 2020-06-02 | 2020-10-23 | 国网江西省电力有限公司电力科学研究院 | Distribution transformer outlet power failure event missing report automatic detection method and system |
| CN111813825B (en) * | 2020-06-02 | 2023-03-14 | 国网江西省电力有限公司电力科学研究院 | Distribution transformer outlet power failure event missing report automatic detection method and system |
| CN113344364A (en) * | 2021-05-31 | 2021-09-03 | 广东电网有限责任公司佛山供电局 | Power failure plan risk analysis method and device, electronic equipment and storage medium |
| CN113344364B (en) * | 2021-05-31 | 2022-12-23 | 广东电网有限责任公司佛山供电局 | Power failure plan risk analysis method and device, electronic equipment and storage medium |
| CN113377835A (en) * | 2021-06-09 | 2021-09-10 | 国网河南省电力公司电力科学研究院 | Distribution network line power failure identification method based on long-short term memory deep learning network |
| WO2023165348A1 (en) * | 2022-03-03 | 2023-09-07 | 广东电网有限责任公司梅州供电局 | Method and device for determining switching state of reactive compensation of distribution transformer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109949178B (en) | 2023-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109949178A (en) | One kind is based on the judgement of support vector machines middle voltage distribution networks power-off event and complementing method | |
| AU2011202886B2 (en) | A self-healing power grid and method thereof | |
| Ho et al. | Optimal placement of fault indicators using the immune algorithm | |
| Wang et al. | Impacts of operators’ behavior on reliability of power grids during cascading failures | |
| Abbasghorbani et al. | Reliability‐centred maintenance for circuit breakers in transmission networks | |
| CN107317394A (en) | Dispatching anti-misoperation method, device and system | |
| CN113328437B (en) | Intelligent power distribution network CPS topology construction method and fault recovery method | |
| Avritzer et al. | Survivability models for the assessment of smart grid distribution automation network designs | |
| CN110556920A (en) | Distribution automation monitoring method, system, terminal and storage medium | |
| CN115842342B (en) | Topology identification method and device for distributed power distribution network | |
| CN109586281B (en) | Power distribution network reliability assessment method, device and medium based on node optimization number | |
| CN112986746B (en) | Distribution network feeder fault self-healing rate automatic analysis method, system and equipment | |
| Li et al. | Importance Assessment of Communication Equipment in Cyber-Physical Coupled Distribution Networks Based on Dynamic Node Failure Mechanism | |
| Qiu | Risk assessment of power system catastrophic failures and hidden failure monitoring & control system | |
| Giacomoni et al. | A control and communications architecture for a secure and reconfigurable power distribution system: An analysis and case study | |
| CN104484546A (en) | Automatic power flow check file generation method for power grid planning project | |
| CN114186858A (en) | Method and system for evaluating importance of energy storage planning node of power distribution network | |
| Birchfield | Graph decomposition for constructing blackstart restoration strategies in benchmark cases | |
| Vrtal et al. | Power grid and data network simulator | |
| CN112465358A (en) | Voltage quality classification method and device based on support vector machine | |
| Muka et al. | Genetic algorithm for placement of IEDs for fault location in smart distribution grids | |
| CN112395715A (en) | Assessment method and system for wiring bus electric network maintenance plan | |
| CN106385027B (en) | Method and system are determined based on the medium voltage distribution network maintenance solution of reliability assessment | |
| CN109933523A (en) | IEC61850 model checking method, system, terminal device, computer readable storage medium | |
| CN105305613B (en) | A kind of traditional power network reorganization and expansion increases Allocation transformer intelligent control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |