CN110045236A - Transformer state parametric data prediction technique and system based on core pivot element analysis optimization - Google Patents
Transformer state parametric data prediction technique and system based on core pivot element analysis optimization Download PDFInfo
- Publication number
- CN110045236A CN110045236A CN201910276195.5A CN201910276195A CN110045236A CN 110045236 A CN110045236 A CN 110045236A CN 201910276195 A CN201910276195 A CN 201910276195A CN 110045236 A CN110045236 A CN 110045236A
- Authority
- CN
- China
- Prior art keywords
- transformer
- state
- transformer state
- parametric data
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
- G01R31/1281—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation of liquids or gases
Abstract
The invention discloses a kind of transformer state parametric data prediction techniques based on core pivot element analysis optimization, it is the following steps are included: S100: obtaining the transformer state amount data in a period of time, and it is converted into the transformer state moment matrix of matrix form, the transformer state amount includes the related data of transformer state parameter;S200: screening transformer state moment matrix X based on core pivot element analysis algorithm, retains main state variables, obtains transformer major state moment matrix;S300: building transformer state parametric data prediction model is trained the prediction model based on the transformer major state moment matrix;S400: transformer state parametric data is predicted based on the transformer state parametric data prediction model through step S300 training.This method can reduce model training complexity, to guarantee higher transformer state parametric data predictablity rate.In addition, the invention also discloses corresponding forecasting systems.
Description
Technical field
The present invention relates in electric system power transmission and transformation equipment operation maintenance area transformer state parameter prediction method and
System more particularly to a kind of transformer state parameter prediction method and system based on core pivot element analysis optimization.
Background technique
Essential core equipment of the power transformer as power transmission and transformation system guarantees that it can be run healthy and stablely with weight
Want meaning.Usually pass through monitoring running state of transformer and predict its variation tendency, to effectively prevent transformer fault, carry out
Prediction scheme guarantees the stable operation of transformer.In order to effectively monitor running state of transformer and predict its variation tendency, it usually needs
Each transformer state parameter of monitoring reflection running state of transformer simultaneously predicts it.
The state parameter information dimension of transformer is very high, and data volume is huge, and different state parameter data are from different perspectives
The operating status of transformer is reflected to a certain extent.The state parameter data of transformer be mainly included in line monitoring data,
Test experiment data, operation inspection data and transformer self-technique supplemental characteristic etc., these data can be in all its bearings
The operating status for reflecting transformer, carrying out prediction to running state of transformer using these data has important research significance.
Transformer online monitoring state parameter data are a complicated data sequences, not to transformer online monitoring state parameter data
The variation tendency come is predicted, the operating status variation of transformer can be better anticipated.
Traditional transformer state Parameter Prediction Models only only account for unitary variant or the data of a few variable
The variation tendency in sequence future is judged.And when being fitted to historical data, more remote go through can not be retained
Effect of the history information to current time, to cannot achieve the prediction to the following long period scale.
Elman neural network is a kind of novel single hidden layer feedforward neural network.Neural network is non-processing complexity
Good performance is shown in linear process.In various types of neural networks, feedforward network is popularized in recent years, because
The complex nonlinear mapping for being input to output sample can be generated, and directly for it to be difficult to handle using classical parametric technique
The phenomenon that model is provided.Prediction model based on neural network building can be used for the prediction of transformer state parametric data, still
There are model training complexity is higher, the problem of influencing predictablity rate.
Summary of the invention
An object of the present invention is to provide a kind of transformer state parametric data prediction based on core pivot element analysis optimization
Method, this method can reduce model training complexity, to guarantee higher transformer state parametric data predictablity rate.
According to foregoing invention purpose, the invention proposes a kind of transformer state parameter numbers based on core pivot element analysis optimization
It is predicted that method, predicts transformer state parametric data, the described method comprises the following steps:
S100: the transformer state amount data in a period of time are obtained, and are converted into the transformer shape of matrix form
State moment matrix, the transformer state amount include the related data of transformer state parameter;
S200: screening transformer state moment matrix based on core pivot element analysis algorithm, retains main state variables, obtains
Transformer major state moment matrix;
S300: building transformer state parametric data prediction model, based on the transformer major state moment matrix to described
Prediction model is trained;
S400: transformer state parameter is predicted based on the transformer state parametric data prediction model through step S300 training
Data.
Transformer state parametric data prediction technique proposed by the present invention based on core pivot element analysis optimization, uses core master
Meta analysis algorithm optimization prediction model inputs parameter, to reduce the training complexity of prediction model, to guarantee higher transformation
Device state parameter data predictablity rate.Core pivot element analysis (kernel principal component analysis, KPCA)
The basic principle of algorithm be nonlinear data is first mapped to the linear character space of higher-dimension by initial data by kernel function, then
Feature extraction is carried out using the method for pivot analysis (principal component analysis, PCA).The essence of KPCA is
Pivot analysis is carried out to the data being mapped in feature space, therefore can be by that there will be the transformer of nonlinear characteristic
State moment matrix is mapped to feature space and carries out pivot analysis screening, then main in the corresponding transformer state moment matrix retained
Quantity of state, that is, transformer major state moment matrix is considered covering most information, thus while reducing model training complexity
It ensure that the especially non-linear effective information loss of less information loss, therefore based on transformer major state moment matrix training
Prediction model can guarantee higher transformer state parametric data predictablity rate.
The usual prediction model is constructed based on neural network.
Further, the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization
In, transformer state parametric data prediction model described in step S300 is constructed based on Elman neural network.
In above scheme, Elman neural network is a kind of feed forward type network, which increases one in hidden layer and hold
Layer is connect, can achieve the purpose of memory, so that system has the ability for adapting to time-varying characteristics, can directly dynamically reflect dynamic
The feature of procedures system.Elman neural network depends on data sample itself during study, being capable of depth excavation
Hiding information characteristics, largely reduce artificial Subjective Factors, are therefore particularly suitable for having non-inside data
The prediction of the transformer state parametric data of linear character.In addition, Elman neural network can retain history letter more remote
The effect to current time is ceased, so as to realize the prediction to the following long period scale.
Further, the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization
In, the prediction model is trained using error backpropagation algorithm in step S300.
In above scheme, the structural parameters of the prediction model generally include the hidden layer number of plies of Elman neural network, mind
Through first number of nodes parameter, the error backpropagation algorithm is determined for the structural parameters of the prediction model.
Further, the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization
In, the related data of transformer state parameter described in step S100 includes the ratio between the data and/or parameter of parameter itself
Data.
In above scheme, in order to further ensure prediction effect, it is often desirable that Donna enters the transformer state of some dimensions
Amount, i.e. quantity of state range of choice are not limited to state parameter, can also include the related datas such as the ratio between state parameter.
Further, the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization
In, transformer state parameter described in step S100 includes transformer body parameter and substation parameter.
In above scheme, since running state of transformer is not only influenced by transformer body parameter, also by substation's ring
The influence of border parameter, therefore transformer body parameter and substation parameter are included in the selection model of transformer state parameter
It encloses.
Further, described in the above-mentioned transformer state parametric data prediction technique based on core pivot element analysis optimization
Transformer body parameter includes the gas content and/or temperature of oil in transformer dissolved in transformer oil.
In above scheme, the gas content and temperature of oil in transformer dissolved in transformer oil can reflect change to a certain extent
The degree of depressor insulation ag(e)ing or failure, therefore the gas content dissolved in transformer oil and temperature of oil in transformer are included in transformer
The range of choice of ontology parameter.
Further, described in the above-mentioned transformer state parametric data prediction technique based on core pivot element analysis optimization
Substation parameter includes one of temperature, surface humidity, relative humidity, mean wind speed or a variety of.
In above scheme, the factors such as temperature, surface humidity, relative humidity, mean wind speed can cause the performance of transformer
It influences, it is considered to be the correlative factor of transformer state, therefore by factors such as temperature, surface humidity, relative humidity, mean wind speeds
It is included in the range of choice of transformer body parameter.
Further, described in the above-mentioned transformer state parametric data prediction technique based on core pivot element analysis optimization
The gas dissolved in transformer oil includes H2、CO、CH4、C2H2、C2H4、C2H6One of or it is a variety of.
In above scheme, associated arguments can be obtained based on dissolved gas analysis (DGA) corresponding chromatographic data
Initial data.It includes hydrogen (H that above scheme, which picks up the gaseous species selected,2), carbon monoxide (CO), methane (CH4), acetylene
(C2H2), ethylene (C2H4) and ethane (C2H6)。
Further, it in the above-mentioned transformer state parametric data prediction technique based on core pivot element analysis optimization, is based on
The gas dissolved in the transformer oil, the ratio data between the parameter includes CH4/H2、C2H2/C2H4、C2H4/C2H6、C2H6/
CH4、C2H2/CH4、C2H6/C2H2One of or it is a variety of.
In above scheme, in order to expand the dimension of input vector, uses for reference and utilized dissolved gas analysis (DGA) ratio
To the thinking that transformer fault is diagnosed, by common International Electrotechnical Commission (IEC) ratio, Rogers ratio and
Dornenburg ratio is also included in the range of choice of input vector, i.e., by CH4/H2、C2H2/C2H4、C2H4/C2H6、C2H6/CH4、
C2H2/CH4、C2H6/C2H26 groups of DGA ratios are also included in the ratio data range of choice between the parameter.
It is a further object of the present invention to provide a kind of transformer state parametric data predictions based on core pivot element analysis optimization
System, the system can reduce model training complexity, to guarantee higher transformer state parametric data predictablity rate.
According to foregoing invention purpose, the invention proposes a kind of transformer state parameter numbers based on core pivot element analysis optimization
It is predicted that system, which includes the data acquisition module and data processing module of data connection, using above-mentioned transformer state
Any one method in parametric data prediction technique predicts transformer state parametric data.
Transformer state parametric data forecasting system proposed by the present invention based on core pivot element analysis optimization, by using
Any of the above-described transformer state parametric data prediction technique predicts transformer state parametric data, therefore, according to aforementioned
Principle, the system can reduce model training complexity, to guarantee higher transformer state parametric data predictablity rate.
Transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization can reduce model
Training complexity, to guarantee higher transformer state parametric data predictablity rate.The method of the present invention uses training vector
The combination of screening and prediction model has considerable flexibility and expansion, can be with the suitable prediction model of unrestricted choice.This hair
Bright method is experiments verify that have preferable capability of fitting and predictive ability.
It is of the present invention based on core pivot element analysis optimization transformer state parametric data forecasting system equally have with
Upper advantages and beneficial effects.
Detailed description of the invention
Fig. 1 is the process of the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization
Schematic diagram.
Fig. 2 is the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization in one kind
Flow diagram under embodiment.
Fig. 3 is concentration of methane gas prediction result schematic diagram in verifying example.
Fig. 4 is concentration of methane gas prediction result percentage error schematic diagram in verifying example.
Specific embodiment
Below in conjunction with Figure of description and specific embodiment to of the present invention based on core pivot element analysis optimization
Transformer state parametric data prediction technique and system are described in further detail.
Fig. 1 illustrates the process of the transformer state parametric data prediction technique optimized based on core pivot element analysis.
As shown in Figure 1, the stream of the transformer state parametric data prediction technique of the invention based on core pivot element analysis optimization
Journey includes:
S100: the transformer state amount data in a period of time are obtained, and are converted into the transformer shape of matrix form
State moment matrix, transformer state amount include the related data of transformer state parameter;
S200: screening transformer state moment matrix based on core pivot element analysis algorithm, retains main state variables, obtains
Transformer major state moment matrix;
S300: building transformer state parametric data prediction model, based on transformer major state moment matrix to prediction model
It is trained;
S400: transformer state parameter is predicted based on the transformer state parametric data prediction model through step S300 training
Data.
In some embodiments, transformer state parametric data prediction model is based on Elman nerve net in step S300
Network building.
In some embodiments, prediction model is trained using error backpropagation algorithm in step S300.
In some embodiments, in step S100 the related data of transformer state parameter include parameter itself data
And/or the ratio data between parameter.
In some embodiments, transformer state parameter includes transformer body parameter and substation's ring in step S100
Border parameter.
Wherein, in some embodiments, transformer body parameter include the gas content dissolved in transformer oil and/or
Temperature of oil in transformer;In some embodiments, substation parameter includes temperature, surface humidity, relative humidity, mean wind speed
One of or it is a variety of.
Wherein, in some embodiments, the gas dissolved in transformer oil includes H2、CO、CH4、C2H2、C2H4、C2H6In
It is one or more.
Wherein, in some embodiments, based on the gas dissolved in transformer oil, the ratio data between parameter includes
CH4/H2、C2H2/C2H4、C2H4/C2H6、C2H6/CH4、C2H2/CH4、C2H6/C2H2One of or it is a variety of.
The transformer state parametric data prediction technique that Fig. 2 illustrates to optimize based on core pivot element analysis is in a kind of embodiment
Under process.
As shown in Fig. 2, the transformer state parametric data prediction technique of the present invention based on core pivot element analysis optimization
Process in one embodiment includes the following steps 1- step 6:
Step 1: the transformer online monitoring state quantity data in a period of time is obtained by data acquisition module.
The step transformer online monitoring state quantity data interior for a period of time by data collecting module collected, wherein
Transformer online monitoring quantity of state includes ratio between transformer state parameter and parameter.Wherein, transformer state parameter specifically wraps
Include the isallobaric device ontology parameter of the gas content dissolved in transformer oil, temperature of oil in transformer and temperature, surface temperature, opposite
The substations parameter such as humidity and mean wind speed.Wherein, the gas dissolved in transformer oil includes hydrogen (H2), carbon monoxide
(CO), methane (CH4), acetylene (C2H2), ethylene (C2H4) and ethane (C2H6), ratio is using in 6 groups of transformer oil between parameter
The gas content ratio of dissolution indicates respectively CH in the form of the ratio between molecular formula4/H2, C2H2/C2H4, C2H4/C2H6, C2H6/
CH4, C2H2/CH4, C2H6/C2H2。
Step 2: transformer online monitoring state quantity data is converted to by transformer online monitoring by data processing module
State moment matrix X.
The step by data processing module using deviation standardized method to transformer online monitoring state quantity data into
Row normalized obtains transformer online monitoring state moment matrix X:
Wherein, X1、X2And XrIndicate that each transformer online monitoring quantity of state, subscript 1,2 ... n indicate time series.
Step 3: core pivot element analysis algorithm being based on by data processing module and screens to obtain transformer online monitoring major state
Moment matrix X '.
In the step, the transformer online monitoring quantity of state in transformer online monitoring state moment matrix X is randomly selected
50% data calculate the characteristic value of core variance matrix K using core pivot element analysis algorithm, and are ranked up to characteristic value, calculate tired
Contribution rate is counted, retains the big feature of contribution rate to get transformer online monitoring major state moment matrix X ' is arrived.
Core pivot element analysis algorithm is specific as follows:
If initial data is x={ xij}m*n, dimension m, the sample data of every dimension is n.xijIndicate i-th dimension number
According to j-th of sample.X is mapped on feature space F by Nonlinear Mapping φ, then initial data xiPicture on feature space
For φ (xi), pivot extraction then is carried out to it with the method for PCA again.
If the data φ (x after mappingi) mean value be zero, then the sample covariance matrix on high-dimensional feature space F is
If v is feature vector corresponding to the eigenvalue λ of Cov in formula (1), then have:
Covv=λ v (2)
By formula (2) both sides simultaneously with φ (xi) do inner product
φ(xi) Covv=λ (φ (xi)·v) (3)
Because feature vector v can be by data set φ (xi) Linearly Representation, if αiFor coefficient, then have
Formula (2) and formula (4) are brought into formula (3), obtained:
Define the core variance matrix K of m*m dimension, expression formula are as follows:
K=k (xi,xj)=φ (xi)·φ(xj) (6)
Then have
M λ α=K α (7)
In formula, α=[α1,α1,…αm]TFor the feature phase vector of core variance matrix K, the characteristic value of K known to formula (7) is m
λ, corresponding feature vector are α.
In order to guarantee the data φ (x of inputi) meet zero-mean condition, it needs to be modified K, revised kernel function
Are as follows:
Wherein ImIt is the value of m rank is all 1 matrix.Calculate the eigenvalue λ of K 'iWith feature vector pi, then to characteristic value into
Row sequence, corresponding feature vector will also do corresponding adjustment.Then the contribution rate of accumulative total T of characteristic value is calculatedi.According to setting
Threshold epsilon, if Tk>=ε, then choosing k sequence is pivot.Wherein the formula of contribution rate of accumulative total is
Step 4: transformer state parametric data prediction model is constructed by data processing module.
In the step, transformer state parametric data prediction model is constructed based on Elman neural network, and by repeatedly trying
The effect of comparison model is tested to determine the structure of prediction model.Wherein, the structural parameters of prediction model include the hidden of Elman network
Hide number, neuron node number parameter layer by layer.
Elman neural network includes 4 layers: input layer, accepts layer and output layer at hidden layer.The unit of input layer plays
Transmit the effect of signal.The unit of output layer plays the role of linear weighted function.The transmission function of implicit layer unit can be used linearly
Or nonlinear function.Layer is accepted to be used to remember the output valve of implicit layer unit previous moment and return to the input of network, it can
To be calculated as a step delay.
The characteristics of Elman neural network is the output of hidden layer by the delay and storage of undertaking layer, is connected himself to hidden
In input containing layer.This connection type makes it have sensibility to the data of historic state.The addition energy of internal feedback network
Enough abilities of enhancing network itself processing multidate information, to achieve the purpose that dynamic modeling.In addition to this, Elman nerve net
Network can not consider external noise to the concrete form of systematic influence.It is correctly output and input if can provide, just
Elman neural network can use to carry out modeling analysis to system.
The non-linear state space expression of Elman neural network are as follows:
Y (k)=g (ω3x(k)) (10)
X (k)=f (ω1xc(k)+ω2(u(k-1))) (11)
xc(k)=x (k-1) (12)
In above formula, y is that m ties up output node vector;X is that n ties up middle layer node unit vector;U is r dimensional input vector;xc
Feedback state vector, ω are tieed up for n3For middle layer to output layer connection weight;ω2For input layer to middle layer connection weight;ω1
For the connection weight of undertaking layer to middle layer;G (*) is the transmission function of output neuron, is the linear combination of middle layer output;
F (*) is the transmission function of middle layer neuron, generally uses Sigmoid function.
Elman neural network can also carry out the amendment of weight using BP algorithm, and study target function can use error
Sum of squares function.
In above formula,For the input vector of target.
Step 5: transformer online monitoring major state moment matrix X ' being based on by data processing module, prediction model is carried out
Training.
In the step, by 80% transformer online monitoring quantity of state in transformer online monitoring major state moment matrix X '
As input, prediction model is trained using error backpropagation algorithm, determines the structural parameters of the prediction model.
Step 6: housebroken transformer state parametric data prediction model being based on by data processing module and predicts transformation
Device state parameter data.
In the step, by another part transformer online monitoring quantity of state in transformer online monitoring state moment matrix X
Trained prediction model is inputted as input data, inputs to Feedforward Neural Networks network layers, neural net layer output prediction knot
Fruit.
Above-mentioned data acquisition module and the mutual data connection of data processing module constitute dividing based on core pivot for the present embodiment
Analyse the transformer state parametric data forecasting system of optimization.The system uses above-mentioned transformer state parametric data prediction technique pair
Transformer state parametric data is predicted.
Test verifying is carried out to above-described embodiment below.
Fig. 3 illustrates concentration of methane gas prediction result in this verifying example, and Fig. 4 illustrates methane gas in this verifying example
Bulk concentration prediction result percentage error.
This verifying example is using the transformer state parametric data prediction technique and system of above-described embodiment to transformer shape
State property data are predicted.Existed based on certain 220kV transformer oil chromatographic on-Line Monitor Device oil colours Data acquisition transformer
Line monitoring state amount data, the sampling interval of data are 1 day.Using 400 groups of monitoring data as training sample, by 30 groups of monitoring numbers
According to as test sample.
In order to evaluate the accuracy and validity of combination forecasting proposed by the present invention, using following interpretational criteria
To be analyzed:
The true value of test set and the root-mean-square error value R of predicted valuemse, expression formula are as follows:
The average percentage error of true value and predicted value, expression formula are as follows:
Maximum percentage error, expression formula are as follows:
In formula, N is the number of test set data, xiIt is true value,It is predicted value.
This verifying example is with methane CH4Illustrate the process of entire combined prediction for forecasting of Gas Concentration result.Model is defeated
The transformer online monitoring quantity of state vector entered is successively are as follows: CH4(only history concentration), H2、CO、C2H2、C2H4、C2H6, it is total hydrocarbon, total
Flammable gas concentration, C2H2/C2H4、C2H4/C2H6, environment temperature and oil temperature totally 11 groups of monitoring data.These vectors are standardized
Processing, is mapped between [0,1], transfer function are as follows:
In formula, xminFor the minimum value of vector sample data, xmaxFor the maximum value of vector sample data.
Principal component analysis is carried out to these vectors using KPCA method, it is as follows to obtain characteristic value:
λ1=1.563, λ2=0.121, λ3=0.008, λ4=3.737, λ5=3.5*10-4, λ6=4.585, λ7=9.4*
10-5, λ8=2.7*10-9, λ9=2.093, λ10=1.6*10-8, λ11=3.5*10-4
The contribution rate of each ingredient is calculated, wherein the 1st (H2)、4(C2H4), 6 (total hydrocarbons), 9 (C2H4/C2H6) this 4 it is main at
The contribution rate of accumulative total for dividing vector is 98.93%, to choose these vectors to train the change based on the building of Elman neural network
Depressor state parameter data prediction model determines the hidden layer number of plies, the neuron node number of model.
Concentration of methane gas is predicted based on housebroken transformer state parametric data prediction model, prediction result
As shown in figure 3, prediction error is as shown in Figure 4.As shown in figure 3, abscissa is time series, unit is day, and ordinate is methane
Gas concentration, unit are μ L/L, and legend A is desired output, and legend B is prediction output, as can be seen from the figure this case verification
Transformer state parametric data prediction technique and system proposed by the present invention based on core pivot element analysis optimization has preferable
Capability of fitting and predictive ability.The percentage error of true value and predicted value as shown in figure 4, abscissa be time series,
Unit is day, and ordinate is percentage error, unit %, and average percentage error is 0.55% in figure, maximum opposite
Percent error 1.32%.
It is equally based on oil chromatography sample data, other several gas concentrations are predicted using the above scheme, and utilizes
Prediction model predicts oil temperature and ambient temperature data.Training set number of samples is 400, and test set number of samples is 30.
Predict that error is as shown in table 1.
The other transformer online monitoring state parameters of table 1. predict error
Predict gas | Mean error % | Worst error % |
Hydrogen | 1.32 | 1.78 |
Carbon monoxide | 1.84 | 2.13 |
Ethylene | 2.07 | 2.76 |
Acetylene | 1.73 | 3.05 |
Oil temperature | 1.95 | 3.72 |
Environment temperature | 1.14 | 2.48 |
It can also be seen that this case verification transformer proposed by the present invention based on core pivot element analysis optimization from the table 1
State parameter data predication method and system have preferable capability of fitting and predictive ability.
It should be noted that the above list is only specific embodiments of the present invention, it is clear that the present invention is not limited to implement above
Example has many similar variations therewith.If those skilled in the art directly exports or joins from present disclosure
All deformations expected, are within the scope of protection of the invention.
Claims (10)
1. a kind of transformer state parametric data prediction technique based on core pivot element analysis optimization, to transformer state parameter number
According to being predicted, it is characterised in that the described method comprises the following steps:
S100: the transformer state amount data in a period of time are obtained, and are converted into the transformer state amount of matrix form
Matrix, the transformer state amount include the related data of transformer state parameter;
S200: screening transformer state moment matrix based on core pivot element analysis algorithm, retains main state variables, obtains transformation
Device major state moment matrix;
S300: building transformer state parametric data prediction model, based on the transformer major state moment matrix to the prediction
Model is trained;
S400: transformer state parameter number is predicted based on the transformer state parametric data prediction model through step S300 training
According to.
2. the transformer state parametric data prediction technique as described in claim 1 based on core pivot element analysis optimization, feature
It is, transformer state parametric data prediction model described in step S300 is constructed based on ELman neural network.
3. the transformer state parametric data prediction technique as described in claim 1 based on core pivot element analysis optimization, feature
It is, the prediction model is trained using error backpropagation algorithm in step S300.
4. the transformer state parametric data prediction technique as described in claim 1 based on core pivot element analysis optimization, feature
It is, the related data of transformer state parameter described in step S100 includes the ratio between the data and/or parameter of parameter itself
Value Data.
5. the transformer state parametric data based on core pivot element analysis optimization as described in any claim in claim 1-4
Prediction technique, which is characterized in that transformer state parameter described in step S100 includes transformer body parameter and substation's ring
Border parameter.
6. the transformer state parametric data prediction technique as claimed in claim 5 based on core pivot element analysis optimization, feature
It is, the transformer body parameter includes the gas content and/or temperature of oil in transformer dissolved in transformer oil.
7. the transformer state parametric data prediction technique as claimed in claim 5 based on core pivot element analysis optimization, feature
It is, the substation parameter includes one of temperature, surface humidity, relative humidity, mean wind speed or a variety of.
8. the transformer state parametric data prediction technique as claimed in claim 6 based on core pivot element analysis optimization, feature
It is, the gas dissolved in the transformer oil includes H2、CO、CH4、C2H2、C2H4、C2H6One of or it is a variety of.
9. the transformer state parametric data prediction technique as claimed in claim 8 based on core pivot element analysis optimization, feature
It is, based on the gas dissolved in the transformer oil, the ratio data between the parameter includes CH4/H2、C2H2/C2H4、C2H4/
C2H6、C2H6/CH4、C2H2/CH4、C2H6/C2H2One of or it is a variety of.
10. a kind of transformer state parametric data forecasting system based on core pivot element analysis optimization, which is characterized in that the system packet
The data acquisition module and data processing module for including data connection, using the transformer state of any one of claim 1-9
Parametric data prediction technique predicts transformer state parametric data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910276195.5A CN110045236A (en) | 2019-04-08 | 2019-04-08 | Transformer state parametric data prediction technique and system based on core pivot element analysis optimization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910276195.5A CN110045236A (en) | 2019-04-08 | 2019-04-08 | Transformer state parametric data prediction technique and system based on core pivot element analysis optimization |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110045236A true CN110045236A (en) | 2019-07-23 |
Family
ID=67276437
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910276195.5A Pending CN110045236A (en) | 2019-04-08 | 2019-04-08 | Transformer state parametric data prediction technique and system based on core pivot element analysis optimization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110045236A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112327115A (en) * | 2020-10-30 | 2021-02-05 | 国网上海市电力公司 | Partial discharge pulse characteristic parameter extraction method adopting time-frequency domain waveform principal component |
CN112541634A (en) * | 2020-12-16 | 2021-03-23 | 国网江苏省电力有限公司检修分公司 | Top layer oil temperature prediction and false fire alarm discrimination method, device and storage medium |
CN112924325A (en) * | 2020-12-30 | 2021-06-08 | 广东电网有限责任公司电力科学研究院 | Gas-insulated transformer monitoring method and device based on mixed gas |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942625A (en) * | 2014-04-23 | 2014-07-23 | 天津大学 | Hydropower station plant dam structure vibration response prediction method based on FOA-GRNN |
CN107677473A (en) * | 2017-09-23 | 2018-02-09 | 哈尔滨理工大学 | A kind of GRNN rotating machinery fault Forecasting Methodologies based on FOA optimizations |
CN108390380A (en) * | 2018-05-16 | 2018-08-10 | 国网山东省电力公司青岛供电公司 | A kind of transformer state parameter trend forecasting method and system |
CN108665052A (en) * | 2017-03-30 | 2018-10-16 | 华为技术有限公司 | The method and apparatus of drosophila algorithm optimization generalized regression nerve networks |
CN109164248A (en) * | 2018-09-18 | 2019-01-08 | 山东理工大学 | A kind of predicting model for dissolved gas in transformer oil method |
CN109242205A (en) * | 2018-10-09 | 2019-01-18 | 国网上海市电力公司 | A kind of transformer state prediction technique and system |
-
2019
- 2019-04-08 CN CN201910276195.5A patent/CN110045236A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942625A (en) * | 2014-04-23 | 2014-07-23 | 天津大学 | Hydropower station plant dam structure vibration response prediction method based on FOA-GRNN |
CN108665052A (en) * | 2017-03-30 | 2018-10-16 | 华为技术有限公司 | The method and apparatus of drosophila algorithm optimization generalized regression nerve networks |
CN107677473A (en) * | 2017-09-23 | 2018-02-09 | 哈尔滨理工大学 | A kind of GRNN rotating machinery fault Forecasting Methodologies based on FOA optimizations |
CN108390380A (en) * | 2018-05-16 | 2018-08-10 | 国网山东省电力公司青岛供电公司 | A kind of transformer state parameter trend forecasting method and system |
CN109164248A (en) * | 2018-09-18 | 2019-01-08 | 山东理工大学 | A kind of predicting model for dissolved gas in transformer oil method |
CN109242205A (en) * | 2018-10-09 | 2019-01-18 | 国网上海市电力公司 | A kind of transformer state prediction technique and system |
Non-Patent Citations (1)
Title |
---|
朱沛恒: "基于果蝇算法优化的概率神经网络在变压器故障诊断中的应用", 《电力大数据》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112327115A (en) * | 2020-10-30 | 2021-02-05 | 国网上海市电力公司 | Partial discharge pulse characteristic parameter extraction method adopting time-frequency domain waveform principal component |
CN112541634A (en) * | 2020-12-16 | 2021-03-23 | 国网江苏省电力有限公司检修分公司 | Top layer oil temperature prediction and false fire alarm discrimination method, device and storage medium |
CN112541634B (en) * | 2020-12-16 | 2024-03-15 | 国网江苏省电力有限公司检修分公司 | Method and device for predicting top-layer oil temperature and discriminating false alarm and storage medium |
CN112924325A (en) * | 2020-12-30 | 2021-06-08 | 广东电网有限责任公司电力科学研究院 | Gas-insulated transformer monitoring method and device based on mixed gas |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ke et al. | Short-term electrical load forecasting method based on stacked auto-encoding and GRU neural network | |
Kalogirou | Artificial neural networks in energy applications in buildings | |
CN107274067B (en) | Distribution transformer overload risk assessment method | |
CN110045236A (en) | Transformer state parametric data prediction technique and system based on core pivot element analysis optimization | |
CN110929847A (en) | Converter transformer fault diagnosis method based on deep convolutional neural network | |
CN112163371B (en) | Transformer bushing state evaluation method | |
CN108764540A (en) | Water supply network pressure prediction method based on parallel LSTM series connection DNN | |
CN110542819A (en) | transformer fault type diagnosis method based on semi-supervised DBNC | |
CN110443417A (en) | Multiple-model integration load forecasting method based on wavelet transformation | |
CN110045237A (en) | Transformer state parametric data prediction technique and system based on drosophila algorithm optimization | |
CN107993012A (en) | A kind of adaptive electric system on-line transient stability appraisal procedure of time | |
CN111369120A (en) | Non-invasive load monitoring method based on equipment transfer learning | |
CN109583645A (en) | A kind of public building short-term load forecasting method | |
Rizwan et al. | Artificial intelligence based approach for short term load forecasting for selected feeders at madina saudi arabia | |
CN114707754A (en) | Intelligent ammeter fault prediction method and system based on BiLSTM-CNN model | |
Chukwu et al. | Analysis of some meteorological parameters using artificial neural network method for Makurdi, Nigeria | |
CN115358437A (en) | Power supply load prediction method based on convolutional neural network | |
CN114239292B (en) | Comprehensive evaluation method and system for low-carbon economic operation-oriented multifunctional demand park | |
CN112016241B (en) | Soft measurement modeling method based on mLASSO-MLP model | |
CN113379116A (en) | Cluster and convolutional neural network-based line loss prediction method for transformer area | |
Matsumoto et al. | Short-term load forecasting by artificial neural networks using individual and collective data of preceding years | |
CN115907198A (en) | Long-distance heat supply load intelligent prediction system | |
CN115759458A (en) | Load prediction method based on comprehensive energy data processing and multi-task deep learning | |
Kowalski et al. | The comparison of linear models for PM10 and PM2. 5 forecasting | |
Li et al. | Residential Photovoltaic Power Forecasting Considering Division of Weather Type Index Interval |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190723 |