CN110033140A - A kind of distribute-electricity transformer district tripping prediction technique, system and device - Google Patents
A kind of distribute-electricity transformer district tripping prediction technique, system and device Download PDFInfo
- Publication number
- CN110033140A CN110033140A CN201910324619.0A CN201910324619A CN110033140A CN 110033140 A CN110033140 A CN 110033140A CN 201910324619 A CN201910324619 A CN 201910324619A CN 110033140 A CN110033140 A CN 110033140A
- Authority
- CN
- China
- Prior art keywords
- tripping
- data
- distribute
- electricity transformer
- transformer district
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000007613 environmental effect Effects 0.000 claims abstract description 21
- 238000009826 distribution Methods 0.000 claims abstract description 17
- 230000001131 transforming effect Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 40
- 238000012549 training Methods 0.000 claims description 40
- 238000013528 artificial neural network Methods 0.000 claims description 24
- 239000002245 particle Substances 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 6
- 241000283153 Cetacea Species 0.000 claims description 4
- 230000002068 genetic effect Effects 0.000 claims description 4
- 235000013399 edible fruits Nutrition 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 230000002265 prevention Effects 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 239000010410 layer Substances 0.000 description 11
- 239000008186 active pharmaceutical agent Substances 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012952 Resampling Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application discloses the invention discloses a kind of distribute-electricity transformer district tripping prediction techniques, comprising: obtains historical data, the tripping prediction model of distribution transforming tripping is established according to historical data；Historical data include history distribute-electricity transformer district operation data and its corresponding environmental data；Target data is obtained, target data is inputted into tripping prediction model, obtains the prediction result of target distribute-electricity transformer district；Target data include target distribute-electricity transformer district operation data and its corresponding environmental data.The present invention utilizes tripping prediction model, target data is predicted, to obtain the prediction result that whether can be tripped, according to the prediction result, network system can find distribute-electricity transformer district security risk in time, carry out accident prevention and pre-adjust power load distributing, to avoid tripping from occurring or improve the power grid processing speed after tripping, to improving, power supply quality is significant.The application further correspondingly discloses a kind of tripping forecasting system of distribute-electricity transformer district with identical beneficial effect and device.
Description
Technical field
The present invention relates to power grid distribution transforming field, in particular to a kind of distribute-electricity transformer district tripping prediction technique, system and device.
Background technique
As the most end level-one power supply unit towards low-voltage customer, the operating status of the area Qi Tai power supply unit is straight for distribute-electricity transformer district
Connect the power supply quality and reliability influenced in platform area.As economy continues to develop, industry and commerce and electricity consumption of resident are all on constantly
It rises, the especially Spring Festival and power load when summer high temperature is very big, so that distribution transforming was often in heavy service state, transformer fault
Tripping occurs often.
It is directed to distribute-electricity transformer district trip problem at present, it is subsequent that only replacement switchs, adjustment route shunts load, distribution transforming increase-volume etc.
Processing means, incident response and the speed that power grid cooperates are not fast enough, so that power supply department is often in passive state, it can not active needle
Operation of power networks state is adjusted to tripping.
Therefore, how to provide a kind of scheme of solution above-mentioned technical problem is that current those skilled in the art need to solve
Problem.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of distribute-electricity transformer district tripping prediction technique, system and devices.It has
Body scheme is as follows:
A kind of distribute-electricity transformer district tripping prediction technique, comprising:
Historical data is obtained, the tripping prediction model of distribution transforming tripping is established according to the historical data；The historical data
Operation data and its corresponding environmental data including history distribute-electricity transformer district；
Target data is obtained, the target data is inputted into the tripping prediction model, obtains the pre- of target distribute-electricity transformer district
Survey result；The target data include the target distribute-electricity transformer district operation data and its corresponding environmental data.
Preferably, the acquisition historical data establishes the tripping prediction model of distribution transforming tripping according to the historical data
Process specifically includes:
Obtain historical data；
The historical data is handled, training sample is obtained；
According to the training sample, the tripping prediction model of neural network is established.
Preferably, described that the historical data is handled, the process of training sample is obtained, is specifically included:
The historical data is cleaned, the high-quality data after being cleaned；
The high-quality data are sampled, the training sample of data balancing is obtained.
Preferably, described that the high-quality data are sampled, the process of the training sample of data balancing is obtained, it is specific to wrap
It includes:
By SMOTE or Easy Casde, the high-quality data are sampled, the training sample of data balancing is obtained.
Preferably, described according to the training sample, it establishes after the tripping prediction model of neural network, further includes:
Using hybrid optimization algorithm, optimize the tripping prediction model.
Preferably, described to utilize hybrid optimization algorithm, optimize the process of the tripping prediction model, specifically includes:
Using genetic algorithm, particle swarm algorithm, in length and breadth crossover algorithm and/or whale algorithm, optimizes the tripping and predict mould
Type.
Preferably, described that the process of the tripping prediction model of neural network is established according to the training sample, it is specific to wrap
It includes:
According to the training sample, the tripping prediction model of Elman neural network is established.
Preferably, the operation data specifically includes the capacity of distribute-electricity transformer district, the duration that puts into operation, highest load factor, more prescribes a time limit
Between and/or trip event.
Correspondingly, the invention discloses a kind of distribute-electricity transformer district tripping forecasting systems, comprising:
Model creation module is predicted for obtaining historical data according to the tripping that the historical data establishes distribution transforming tripping
Model；The historical data include history distribute-electricity transformer district operation data and its corresponding environmental data；
The target data is inputted the tripping prediction model, obtained by model running module for obtaining target data
The prediction result of target distribute-electricity transformer district；The target data include the target distribute-electricity transformer district operation data and its corresponding ring
Border data.
Correspondingly, the invention also discloses a kind of distribute-electricity transformer district tripping prediction meanss, comprising:
Memory, for storing computer program；
Processor realizes the step of tripping prediction technique in distribute-electricity transformer district as described above when for executing the computer program
Suddenly.
The invention discloses a kind of distribute-electricity transformer district tripping prediction techniques, comprising: historical data is obtained, according to the history number
According to the tripping prediction model for establishing distribution transforming tripping；The historical data includes the operation data of history distribute-electricity transformer district and its corresponding
Environmental data；Target data is obtained, the target data is inputted into the tripping prediction model, obtains the pre- of target distribute-electricity transformer district
Survey result；The target data include the target distribute-electricity transformer district operation data and its corresponding environmental data.Benefit of the invention
With tripping prediction model, target data is predicted, so that the prediction result that whether can be tripped is obtained, according to the prediction
As a result, network system can find distribute-electricity transformer district security risk in time, carry out accident prevention and pre-adjust power load distributing, thus
Tripping is avoided to occur or improve the power grid processing speed after tripping, power supply quality is significant to improving.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of step flow chart of distribute-electricity transformer district tripping prediction technique in the embodiment of the present invention；
Fig. 2 is a kind of step flow chart of specific distribute-electricity transformer district tripping prediction technique in the embodiment of the present invention；
Fig. 3 is the ROC curve figure of a variety of prediction algorithms in the embodiment of the present invention；
Fig. 4 is a kind of structure distribution figure of distribute-electricity transformer district tripping forecasting system in the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The prior art is directed to distribute-electricity transformer district trip problem, only post-processing means, so that power supply department is often in passive
State actively can not adjust operation of power networks state for tripping.The embodiment of the present invention is made whether using tripping prediction model
The prediction that can be tripped improves power supply quality to actively avoid the power grid processing speed after tripping that trips or improve.
It is shown in Figure 1 the embodiment of the invention discloses a kind of distribute-electricity transformer district tripping prediction technique, comprising:
S11: obtaining historical data, and the tripping prediction model of distribution transforming tripping is established according to the historical data；The history
Data include history distribute-electricity transformer district operation data and its corresponding environmental data；
Wherein, the operation data specifically includes the capacity of distribute-electricity transformer district, the duration that puts into operation, highest load factor, out-of-limit event
And/or trip event；Environmental data includes the temperature and/or humidity of the operation data of corresponding synchronization, in addition to accurately corresponding to
Outside the environmental data at a certain moment, it is also an option that the maximum value, minimum value, median of a period of time are as more in environmental data
Environmental data in the period of front and back occurs for representational data, corresponding a certain event.
S12: obtaining target data, and the target data is inputted the tripping prediction model, obtains target distribute-electricity transformer district
Prediction result；The target data include the target distribute-electricity transformer district operation data and its corresponding environmental data.
It is understood that target data is identical as the format of historical data, tripping prediction model is built by historical data
It is vertical, training sample of the historical data as tripping prediction model；Target data is input in tripping prediction model, it is available
Corresponding prediction result.Wherein, the sample size of training sample is bigger, accuracy and the degree of correlation are higher, and the prediction model that trips is to mesh
The prediction result for marking data is more accurate.
Specifically, prediction result includes the prediction to movement or event in following distribute-electricity transformer district.
The embodiment of the invention discloses a kind of distribute-electricity transformer district tripping prediction techniques, comprising: historical data is obtained, according to described
Historical data establishes the tripping prediction model of distribution transforming tripping；The historical data include history distribute-electricity transformer district operation data and its
Corresponding environmental data；Target data is obtained, the target data is inputted into the tripping prediction model, obtains target distribution transforming platform
The prediction result in area；The target data include the target distribute-electricity transformer district operation data and its corresponding environmental data.This
Inventive embodiments are predicted target data, using tripping prediction model to obtain the prediction knot that whether can be tripped
Fruit, according to the prediction result, network system can find distribute-electricity transformer district security risk in time, carry out accident prevention and pre-adjust
Power load distributing has raising power supply quality important so that tripping be avoided to occur or improve the power grid processing speed after tripping
Meaning.
The embodiment of the invention discloses a kind of specific distribute-electricity transformer district tripping prediction techniques, relative to a upper embodiment, originally
Embodiment has made further instruction and optimization to technical solution.Specifically, shown in Figure 2:
S21: historical data is obtained；The historical data include history distribute-electricity transformer district operation data and its corresponding environment
Data；
S22: the history data are handled, training sample is obtained；
Wherein, described that the historical data is handled, the process of training sample is obtained, is specifically included:
S221: cleaning the historical data, the high-quality data after being cleaned；
It is understood that instrument fault leads to historical data inaccuracy, the note of test when may test in historical data
There is situations such as mistake in record, therefore cleans to historical data, retains the preferably high-quality data of the quality of data.
S222: the high-quality data are sampled, the training sample of data balancing is obtained.
It is understood that the problem of high-quality data equally exist data nonbalance, therefore it is sampled, it is counted
According to balanced training sample.
Specifically, being sampled to the high-quality data by SMOTE or Easy Casde, obtaining the instruction of data balancing
Practice sample.
Wherein, SMOTE i.e. Borderline-SMOTE, the step of being sampled using the algorithm include:
Calculate each sample point p in minority class sampleiWith the Euclidean distance of all sample points, sample point m neighbour is obtained；
Minority class sample is divided, it is assumed that have x most class sample (0≤x≤m) in m neighbour, think if x=m
piFor noise sample, p is thought if m/2≤x≤miFor boundary sample, p is thought if 0≤x≤m/2iFor safe sample；
The k neighbour of boundary sample and minority class sample is calculated, s k neighbour progress linear interpolation is chosen and synthesizes new minority
Class sample；
New synthesis sample merges with original sample, forms balance sample；
Wherein minority class sample, most class samples and original sample are obtained from high-quality data above, last shape
At balance sample i.e. data balancing training sample.
S23: according to the training sample, the tripping prediction model of neural network is established.
Wherein, the tripping prediction model for establishing neural network can then be selected by all kinds of neural network algorithms, the present embodiment
Elman neural network algorithm is selected, according to the training sample, establishes the tripping prediction model of Elman neural network, step tool
Body includes:
Input layer, hidden layer, output layer neuron number are set, the weight between each layer is initialized；
Training sample is inputted, input layer output is calculated；
Calculate hidden layer output；
Output layer input is calculated, is calculated with its result and accepts layer output, feedback arrives hidden layer；
Output result and actual result error are calculated, feedback updates the weight of each layer, comes back to step2, until error
Less than preset range, i.e. completion network training.
It is understood that Elman neural network algorithm is a kind of more mature algorithm, it can refer to other documents and materials
It realizes, details are not described herein again.
Further, after step 223, can also include:
S24: hybrid optimization algorithm is utilized, the tripping prediction model is optimized.
It is since traditional Elman neural network algorithm is in training it is understood why optimizing tripping prediction model
Local optimum can be fallen into the process, using hybrid optimization algorithm to the power of tripping each interlayer of prediction model of Elman neural network
Value and threshold value optimize, and obtain optimal weight and threshold value, can effectively improve precision of prediction.
Specifically, optimizing the tripping using genetic algorithm, particle swarm algorithm, in length and breadth crossover algorithm and/or whale algorithm
Prediction model.
It is traditional BP neural network model (BP-NN), in length and breadth crossover algorithm (CSO), excellent referring to shown in following table and Fig. 3
The Elman network model for changing BP network model (CSO-BP) and routine compares, and can obtain the corresponding tripping of algorithms of different
(Receiver Operating Characteristic, recipient operate special the accuracy rate accuracy and ROC of prediction model
Levy curve), AUC (Area under the Curve, area under the curve), compare it can be seen that in length and breadth crossover algorithm optimization
The performance of Elman neural network algorithm (CSO-Elman) is more excellent.
In actual operation, the process of crossover algorithm optimization tripping prediction model in length and breadth, actually with to establish tripping pre-
It surveys models coupling together: Elman network structure being determined according to training sample first, example is encoded；Then it uses
Crossover algorithm in length and breadth, initialization population calculate particle fitness, and execution is lateral, crossed longitudinally, calculate fitness and are simultaneously compared
Afterwards, best initial weights and threshold value are obtained；Training sample is reused to instruct the tripping prediction model of the Elman network after optimization
Practice, obtains the tripping prediction model that can be ultimately utilized in the prediction result of prediction target data.
Specific algorithm movement the following steps are included:
According to given training sample, the neuron number of Elman neural network topology structure and each layer is determined, and determine
The lateral cross probability P of crossover algorithm in length and breadthh, crossed longitudinally probability Pv, population scale M, maximum number of iterations Tmaxgen；
To the particle to be optimized coding, in the solution space of coding, and initial population X=[X is randomly generated1,X2,...,
XM]T；
Fitness evaluation is carried out to particle each in group using following formula:
Wherein, ptIndicate the reality output of Elman network,Indicate the target output of Elman network, n indicates training sample
This input total number of samples, m indicate that training sample exports number.
Lateral cross operation is carried out according to the following formula, and lateral cross is the intersection behaviour that counts done with two one-dimensional particles
Make, and two particle is randomly generated with one-dimensional.The filial generation that crossover operation obtains is stored in matrix MShcThe inside (MShcIn referred to as
Mediocre solution), then the adaptive value of all particles in the matrix is calculated, by obtained adaptive value and parent population X (i.e. DSvc, the first generation removes
It compares outside), selects the better particle of fitness and be retained in DShcIn (DShcReferred to as optimal solution).
MShc(i, d)=r1×X(i,d)+(1-r1)×X(j,d)+c1×(X(i,d)-X(j,d))；
MShc(j, d)=r2×X(j,d)+(1-r2)×X(i,d)+c2×(X(j,d)-X(i,d))；
I, j ∈ N (1, M), d ∈ N (1, D)；
In formula, r1、r2It is the random number between [0,1]；c1、c2It is the random number between [- 1,1]；M is the model of population
It encloses；D is the dimension of variable；The d that X (i, d), X (j, d) respectively indicate parent particle X (i) and X (j) is tieed up；MShc(i,d)、MShc
(j, d) respectively indicates X (i, d) and X (j, d) and ties up generation filial generation in d by lateral cross.
Crossed longitudinally operation is carried out according to the following formula, and crossed longitudinally is a kind of calculation that all particles carry out between different dimensions
Number intersects, and bidimensional is random combine, and the filial generation that crossover operation obtains is stored in matrix MSvcIn (MSvcReferred to as golden mean of the Confucian school solution),
Then the adaptive value for calculating each particle in the matrix, with its parent population X (i.e. DSvc) be compared, select more excellent grain
Son is retained in DSvcIn (DSvcReferred to as optimal solution).
MSvc(i,d1)=rX (i, d1)+(1-r)·X(i,d2),
I ∈ N (1, M), d1,d2∈ N (1, D), r ∈ [0,1],
In formula: D=MSvc(i,d1) be particle d1Peacekeeping d2Tie up the filial generation by generating after crossed longitudinally operation.
Judge current iteration number T > Tmaxgen, then terminate optimizing, then iteration ends.And by DSvcMiddle fitness it is best one
Group solution is set as weight corresponding to Elman neural network and threshold value.Otherwise, lateral cross operation iteration again is gone to.
S25: obtaining target data, and the target data is inputted the tripping prediction model, obtains target distribute-electricity transformer district
Prediction result；The target data include the target distribute-electricity transformer district operation data and its corresponding environmental data.
Borderline-SMOTE used in the present embodiment is a kind of novel sample process technology, it can be by original pole
It is converted into new balance sample for unbalanced data, reduces influence of the unbalanced data to precision of prediction；Due to using single
One Elman neural network prediction model easily falls into local optimum problem, and the present embodiment through crossover algorithm in length and breadth by optimizing
Elman neural network model efficiently solves the above problem, improves the generalization ability of neural network.Therefore the present embodiment utilizes
Borderline-SMOTE sampling techniques carries out resampling to original sample, using the Elman nerve of the optimization of crossover algorithm in length and breadth
Network model predicts that this method, which efficiently solves unbalanced data, causes prediction model degraded performance to be asked to distribution transforming tripping
Topic, while solving the problems, such as that the easy of Elman neural network falls into local optimum, accuracy is higher, and prediction effect is more preferable.
Correspondingly, the embodiment of the invention discloses a kind of distribute-electricity transformer district tripping forecasting system, it is shown in Figure 4, comprising:
Model creation module 01, for obtaining historical data, the tripping for establishing distribution transforming tripping according to the historical data is pre-
Survey model；The historical data include history distribute-electricity transformer district operation data and its corresponding environmental data；
The target data is inputted the tripping prediction model, obtained by model running module 02 for obtaining target data
To the prediction result of target distribute-electricity transformer district；The target data includes the operation data of the target distribute-electricity transformer district and its corresponding
Environmental data.
In some specific embodiments, model creation module 01 is specifically used for: obtaining historical data；To the history number
According to being handled, training sample is obtained；According to the training sample, the tripping prediction model of neural network is established.
In some specific embodiments, model creation module 01 is specifically used for: cleaning, obtains to the historical data
High-quality data after to cleaning；The high-quality data are sampled, the training sample of data balancing is obtained.
In some specific embodiments, model creation module 01 is specifically used for: right by SMOTE or Easy Casde
The high-quality data are sampled, and obtain the training sample of data balancing.
In some specific embodiments, model creation module 01 is specifically also used to: being utilized hybrid optimization algorithm, is optimized institute
State tripping prediction model.
In some specific embodiments, model creation module 01 is specifically used for: using genetic algorithm, particle swarm algorithm,
Crossover algorithm and/or whale algorithm in length and breadth optimize the tripping prediction model.
In some specific embodiments, model creation module 01 is specifically used for: according to the training sample, establishing
The tripping prediction model of Elman neural network.
In some specific embodiments, the operation data specifically includes the capacity of distribute-electricity transformer district, the duration that puts into operation, highest
Load factor, threshold crossing time and/or trip event.
Whether the embodiment of the present invention is predicted target data, using tripping prediction model to obtain to jump
The prediction result of lock, according to the prediction result, network system can find distribute-electricity transformer district security risk in time, carry out accident prevention
And power load distributing is pre-adjusted, to avoid tripping from occurring or improve the power grid processing speed after tripping, to raising electric service
Quality is significant.
Correspondingly, the embodiment of the invention also discloses a kind of distribute-electricity transformer district tripping prediction meanss, comprising:
Memory, for storing computer program；
Processor realizes the step of tripping prediction technique in distribute-electricity transformer district as described above when for executing the computer program
Suddenly.
Wherein, the particular content of related distribute-electricity transformer district tripping prediction technique is referred to associated description in foregoing embodiments,
Details are not described herein again.
Wherein, the present embodiment has beneficial effect identical with distribute-electricity transformer district tripping prediction technique in foregoing embodiments, this
Place repeats no more.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Detailed Jie has been carried out to a kind of distribute-electricity transformer district tripping prediction technique, system and device provided by the present invention above
It continues, used herein a specific example illustrates the principle and implementation of the invention, and the explanation of above embodiments is only
It is to be used to help understand method and its core concept of the invention；At the same time, for those skilled in the art, according to this hair
Bright thought, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification should not manage
Solution is limitation of the present invention.
Claims (10)
- The prediction technique 1. a kind of distribute-electricity transformer district is tripped characterized by comprisingHistorical data is obtained, the tripping prediction model of distribution transforming tripping is established according to the historical data；The historical data includes The operation data of history distribute-electricity transformer district and its corresponding environmental data；Target data is obtained, the target data is inputted into the tripping prediction model, obtains the prediction knot of target distribute-electricity transformer district Fruit；The target data include the target distribute-electricity transformer district operation data and its corresponding environmental data.
- 2. trip prediction technique for distribute-electricity transformer district according to claim 1, which is characterized in that the acquisition historical data, according to The historical data establishes the process of the tripping prediction model of distribution transforming tripping, specifically includes:Obtain historical data；The historical data is handled, training sample is obtained；According to the training sample, the tripping prediction model of neural network is established.
- 3. trip prediction technique for distribute-electricity transformer district according to claim 2, which is characterized in that described to be carried out to the historical data Processing, obtains the process of training sample, specifically includes:The historical data is cleaned, the high-quality data after being cleaned；The high-quality data are sampled, the training sample of data balancing is obtained.
- 4. trip prediction technique for distribute-electricity transformer district according to claim 3, which is characterized in that described to be carried out to the high-quality data Sampling, obtains the process of the training sample of data balancing, specifically includes:By SMOTE or Easy Casde, the high-quality data are sampled, the training sample of data balancing is obtained.
- 5. trip prediction technique for distribute-electricity transformer district according to claim 2, which is characterized in that it is described according to the training sample, It establishes after the tripping prediction model of neural network, further includes:Using hybrid optimization algorithm, optimize the tripping prediction model.
- 6. trip prediction technique for distribute-electricity transformer district according to claim 5, which is characterized in that it is described to utilize hybrid optimization algorithm, The process for optimizing the tripping prediction model, specifically includes:Using genetic algorithm, particle swarm algorithm, in length and breadth crossover algorithm and/or whale algorithm, optimize the tripping prediction model.
- 7. trip prediction technique for distribute-electricity transformer district according to claim 6, which is characterized in that it is described according to the training sample, The process for establishing the tripping prediction model of neural network, specifically includes:According to the training sample, the tripping prediction model of Elman neural network is established.
- 8. according to claim 1 to any one of 7 distribute-electricity transformer districts tripping prediction techniques, which is characterized in thatThe operation data specifically includes the capacity of distribute-electricity transformer district, the duration that puts into operation, highest load factor, threshold crossing time and/or tripping Event.
- The forecasting system 9. a kind of distribute-electricity transformer district is tripped characterized by comprisingModel creation module establishes the tripping prediction model of distribution transforming tripping according to the historical data for obtaining historical data； The historical data include history distribute-electricity transformer district operation data and its corresponding environmental data；The target data is inputted the tripping prediction model, obtains target by model running module for obtaining target data The prediction result of distribute-electricity transformer district；The target data include the target distribute-electricity transformer district operation data and its corresponding environment number According to.
- The prediction meanss 10. a kind of distribute-electricity transformer district is tripped characterized by comprisingMemory, for storing computer program；Processor realizes that distribute-electricity transformer district tripping is pre- as described in any one of claim 1 to 8 when for executing the computer program The step of survey method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910324619.0A CN110033140A (en) | 2019-04-22 | 2019-04-22 | A kind of distribute-electricity transformer district tripping prediction technique, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910324619.0A CN110033140A (en) | 2019-04-22 | 2019-04-22 | A kind of distribute-electricity transformer district tripping prediction technique, system and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110033140A true CN110033140A (en) | 2019-07-19 |
Family
ID=67239595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910324619.0A Pending CN110033140A (en) | 2019-04-22 | 2019-04-22 | A kind of distribute-electricity transformer district tripping prediction technique, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110033140A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111817265A (en) * | 2020-06-29 | 2020-10-23 | 北京智芯微电子科技有限公司 | Low-voltage transformer area power distribution protection method and system |
| CN111832664A (en) * | 2020-07-31 | 2020-10-27 | 华北电力大学（保定） | Borderline SMOTE-based power transformer fault sample equalization and fault diagnosis method |
| CN112684281A (en) * | 2020-11-12 | 2021-04-20 | 国网河北省电力有限公司电力科学研究院 | Power distribution network single-phase earth fault section positioning method and device and terminal equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570582A (en) * | 2016-10-26 | 2017-04-19 | 中国电力科学研究院 | Method and system for building transmission line dancing tripping risk prediction network model |
| CN109410089A (en) * | 2018-12-29 | 2019-03-01 | 广州供电局有限公司 | Low-voltage tripping and customer complaint prediction technique, device and storage medium |
-
2019
- 2019-04-22 CN CN201910324619.0A patent/CN110033140A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570582A (en) * | 2016-10-26 | 2017-04-19 | 中国电力科学研究院 | Method and system for building transmission line dancing tripping risk prediction network model |
| CN109410089A (en) * | 2018-12-29 | 2019-03-01 | 广州供电局有限公司 | Low-voltage tripping and customer complaint prediction technique, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 谢晨等: "数据驱动的燃气涡轮机跳闸预警方法的研究", 《计算机科学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111817265A (en) * | 2020-06-29 | 2020-10-23 | 北京智芯微电子科技有限公司 | Low-voltage transformer area power distribution protection method and system |
| CN111832664A (en) * | 2020-07-31 | 2020-10-27 | 华北电力大学（保定） | Borderline SMOTE-based power transformer fault sample equalization and fault diagnosis method |
| CN112684281A (en) * | 2020-11-12 | 2021-04-20 | 国网河北省电力有限公司电力科学研究院 | Power distribution network single-phase earth fault section positioning method and device and terminal equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Soliman et al. | Hybrid ANFIS‐GA‐based control scheme for performance enhancement of a grid‐connected wind generator | |
| CN110033140A (en) | A kind of distribute-electricity transformer district tripping prediction technique, system and device | |
| Zhao et al. | Optimisation of electrical system for offshore wind farms via genetic algorithm | |
| Radu et al. | A multi-objective genetic algorithm approach to optimal allocation of multi-type FACTS devices for power systems security | |
| Huang et al. | Three-phase balancing of distribution feeders using immune algorithm | |
| Hajforoosh et al. | Coordinated aggregated-based particle swarm optimisation algorithm for congestion management in restructured power market by placement and sizing of unified power flow controller | |
| CN107516170A (en) | A kind of difference self-healing control method based on probability of equipment failure and power networks risk | |
| Saonerkar et al. | Optimized DG placement in radial distribution system with reconfiguration and capacitor placement using genetic algorithm | |
| CN107316113B (en) | Power transmission network planning method and system | |
| CN103703643B (en) | System voltage stabilizing device and stabilization method | |
| EP2362978A1 (en) | Reactive power optimization | |
| Gao et al. | Multi-objective dynamic reconfiguration for urban distribution network considering multi-level switching modes | |
| Jordehi et al. | Heuristic methods for solution of FACTS optimization problem in power systems | |
| CN101739025A (en) | Immunity genetic algorithm and DSP failure diagnostic system based thereon | |
| CN105203869A (en) | Microgrid island detection method based on extreme learning machine | |
| CN105429134A (en) | Grid voltage stability prediction method based on power big data | |
| CN107681655A (en) | A kind of tidal current energy generating field coordinated planning method | |
| CN117117876B (en) | Power grid full-element resource coordination control method and system | |
| Kazemi et al. | On the use of harmony search algorithm in optimal placement of FACTS devices to improve power system security | |
| Sun et al. | Edge computing terminal equipment planning method for real-time online monitoring service of power grid | |
| Reddy et al. | Placement of distributed generator, capacitor and DG and capacitor in distribution system for loss reduction and reliability improvement | |
| JPH11289661A (en) | Power flow-calculating device for power distribution system control system | |
| CN201666923U (en) | Device for processing DSP failure diagnosis of electric power system by using immune genetic algorithm | |
| Kumar et al. | Comparative analysis of particle swarm optimization variants on distributed generation allocation for network loss minimization | |
| Wang et al. | Pareto optimization of power system reconstruction using NSGA-II algorithm |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190719 |