CN109961160A - A kind of power grid future operation trend predictor method and system based on trend parameter - Google Patents
A kind of power grid future operation trend predictor method and system based on trend parameter Download PDFInfo
- Publication number
- CN109961160A CN109961160A CN201711335875.7A CN201711335875A CN109961160A CN 109961160 A CN109961160 A CN 109961160A CN 201711335875 A CN201711335875 A CN 201711335875A CN 109961160 A CN109961160 A CN 109961160A
- Authority
- CN
- China
- Prior art keywords
- data
- electric network
- key feature
- network swim
- history
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000005611 electricity Effects 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims description 59
- 238000012549 training Methods 0.000 claims description 35
- 230000010354 integration Effects 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 241001269238 Data Species 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 3
- 235000013399 edible fruits Nutrition 0.000 claims description 2
- 230000008859 change Effects 0.000 abstract description 4
- 238000011217 control strategy Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000011160 research Methods 0.000 abstract description 3
- 238000004458 analytical method Methods 0.000 abstract description 2
- 238000004590 computer program Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000006872 improvement Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000005283 ground state Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
Abstract
The present invention relates to a kind of power grid future operation trend predictor methods and system based on trend parameter, acquire current electric grid flow data and on the corresponding data of the influential factor of current electric grid tool;Data after acquisition are input in the relational model between the history electric network swim constructed in advance and influence factor, power grid future operation trend is estimated;Relational model between the history electric network swim and influence factor includes history electric network swim data weighting shared by key feature.The present invention analyzes influence of all kinds of uncertain factors to security and stability analysis conclusion in grid operation situation change procedure, extract the key feature of the uncertain informations such as generation of electricity by new energy fluctuation, bus load prediction error, operation plan exception, research considers probabilistic power grid future operation trend trend parameter prediction method, and comprehensive study considers that the operation of power networks trend data of operation situation uncertainty and control strategy forms technology online.
Description
Technical field
The present invention relates to power grid future trend trend and computerized algorithm field, and in particular to a kind of based on trend parameter
Power grid future operation trend predictor method and system.
Background technique
Transmission cross-section, also referred to as trend section.In practical power systems, system coordinator is often according only to geographical position
It sets, if the main line for getting in touch with power center and load center is selected as a transmission cross-section.Compare specification to be defined as follows: at certain
Under one ground state trend, effective power flow direction is identical and electrical distance similar in the collection of one group of transmission line of electricity be collectively referred to as transmission cross-section.
There are many factor for influencing power grid section tidal current, are to have rule between these usual factors and the practical trend numerical value of power grid
Rule and rule, but we to be not aware that these factors have the actual influence of electric network swim much, cause to power grid future
Trend Trend judgement inaccuracy, causes trend section accident.
Summary of the invention
To solve above-mentioned deficiency of the prior art, the object of the present invention is to provide a kind of power grids based on trend parameter not
Coming operation trend predictor method and system, research considers probabilistic power grid future operation trend trend parameter prediction method,
Comprehensive study considers that the operation of power networks trend data of operation situation uncertainty and control strategy forms technology online.
The purpose of the present invention is adopt the following technical solutions realization:
The present invention provides a kind of power grid future operation trend predictor method based on trend parameter, thes improvement is that:
Acquire current electric grid flow data and on the corresponding data of the influential factor of current electric grid tool;
Data after acquisition are input in the relational model between the history electric network swim constructed in advance and influence factor,
Estimate power grid future operation trend;
Relational model between the history electric network swim and influence factor includes the tide of history power grid shared by key feature
Flow data weight.
Further: the model of relationship between the history electric network swim constructed in advance and influence factor, comprising:
Acquire the history electric network swim data of each section of power grid;
Data corresponding to each section corresponding key feature acquisition key feature;The key feature includes: new energy
Source power generation fluctuation, bus load prediction error or operation plan are abnormal;
History electric network swim data weighting shared by key feature is calculated according to history electric network swim data and key feature;
History electric network swim data weighting shared by key feature is trained, the parameters weighting training after being optimized
As a result;Based on history electric network swim data weighting structure shared by the history electric network swim data, key feature, key feature
Build the model of relationship between electric network swim and influence factor.
Further: calculating history power grid tide shared by key feature according to history electric network swim data and key feature
Before flow data weight, further includes: stored to history electric network swim data and key feature;
It is described to history electric network swim data and key feature carry out storage include:
The history electric network swim data and key feature of acquisition are subjected to Data Integration, it, will be crucial special as unit of section
Input file of history of the seeking peace electric network swim Data Integration in a data file, as parameter training;
By after integration history electric network swim data and key feature store into history data file.
Further: it is described that history electric network swim data weighting shared by key feature is trained, after obtaining optimization
Parameters weighting training result, comprising:
History electric network swim data and key feature are obtained from history data file;
History electric network swim data and key feature are trained, history electric network swim number shared by key feature is calculated
It is optimized according to weight, and to the weight.
Further, described that the weight is optimized, comprising: the weighted value so that minimum risk function is calculated,
Include:
Building includes the weight fitting function of key feature;
Risk function, and the residual error in calculation risk function between fitting function and objective function are constructed according to fitting function
The maximum weighted value comprising key feature.
Further, the vector representation of the weight fitting function are as follows:
The risk function indicates are as follows:
Minimizing risk function indicates are as follows:
To each weight parameter θ minimized in risk function J (θ)jIts partial derivative is sought, each weight parameter θ is obtainedjIt is right
The gradient answered:
According to each weight parameter θjNegative gradient direction update each weight parameter θj, indicate are as follows:
In formula: hθ(xi) be the weight fitting function comprising key feature vector representation, θT=[θ0,θ1,...,
θn], n is weight number of components, and m indicates key feature sample data, xiIndicate i-th of key feature sample point,It indicates i-th
Key feature sample point xiJth weight component;yiFor preset practical correctness objective function;α is weight parameter variance
Step-length.
It further, further include being verified to weight, including partial data is used to verify above-mentioned weight, training of judgement knot
Whether the history electric network swim data of fruit are consistent with actual electric network flow data, obtain history electric network swim data and key is special
Success rate after sign training, until success rate meets given threshold.
Further: the model pre-estimating power grid future operation based on relationship between the electric network swim and influence factor becomes
Gesture, comprising:
To current generation of electricity by new energy fluctuation, bus are inputted between the electric network swim and influence factor in the model of relationship
The key feature of load prediction error, operation plan exception meets the history electric network swim data of given threshold according to success rate
Power grid future trend trend is estimated with key feature.
The present invention provides a kind of power grid future operation trend Prediction System based on trend parameter, thes improvement is that:
Include:
Acquisition module, for acquiring current electric grid flow data and being corresponded on the influential factor of current electric grid tool
Data;
Module is estimated, for the data after acquisition to be input between the history electric network swim constructed in advance and influence factor
Relational model in, estimate power grid future operation trend;
Relational model between the history electric network swim and influence factor includes the tide of history power grid shared by key feature
Flow data weight.
Further: further including building module, for constructing the relational model between history electric network swim and influence factor.
Further: the building module includes:
Submodule is acquired, for acquiring the history electric network swim data of each section of power grid;Pass corresponding to each section
The corresponding data of key feature described in key collection apparatus;The key feature includes: that generation of electricity by new energy fluctuation, bus load are pre-
It surveys error and operation plan is abnormal;
Computational submodule, for calculating the electricity of history shared by key feature according to history electric network swim data and key feature
Net flow data weight;
Training submodule is optimized for being trained to history electric network swim data weighting shared by key feature
Parameters weighting training result afterwards;
Setting up submodule, for based on the electricity of history shared by the history electric network swim data, key feature, key feature
Net flow data weight constructs the model of relationship between electric network swim and influence factor.
Further: further including memory module, for calculating satisfaction according to history electric network swim data and key feature
Before weight shared by each feature under all flow datas, history electric network swim data and key feature are stored;
The memory module includes:
Submodule is integrated, the history electric network swim data and key feature for that will acquire carry out Data Integration, with section
For unit, by key feature and history electric network swim Data Integration in a data file, the input as parameter training is literary
Part;
Save submodule, for after integrate history electric network swim data and key feature store to history data file
In.
Further: the computational submodule is also used to calculate so that minimizing the weighted value of risk function, comprising:
First construction unit, for constructing the weight fitting function comprising key feature;
Second construction unit constructs risk function according to fitting function, and fitting function and target letter in calculation risk function
The maximum weighted value comprising key feature of residual error between number.
Further: further including authentication unit, for using partial data to verify above-mentioned weight, training of judgement result is gone through
Whether history electric network swim data are consistent with actual electric network flow data, obtain history electric network swim data and key feature training
Success rate afterwards, until success rate meets given threshold.
Further: it is described to estimate module, it is also used between the electric network swim and influence factor in the model of relationship
The key feature for inputting current generation of electricity by new energy fluctuation, bus load prediction error, operation plan exception, according to success rate
The history electric network swim data and key feature for meeting given threshold estimate power grid future trend trend.
Compared with the immediate prior art, technical solution provided by the invention is had the beneficial effect that
The present invention acquires the corresponding numerical value of preset influence factor first;Acquire current electric grid flow data and right
The current electric grid has the corresponding data of influential factor;Data after acquisition are input to the history power grid constructed in advance
In relational model between trend and influence factor, power grid future operation trend is estimated, the influence factor of selection is accurate, improves
Accuracy that power grid section tidal current situation is estimated.
Electric network swim data and influence factor are accurate: system can by the real-time data of current electric grid, carry out integration and
Processing, is stored according to chronomere, forms the set of effective training data, and can pass through a variety of uncertain influences
Factor feature is trained the weight of these features.
Accurate to the training of weight: the data of system are real-time updates, can be according to current time to the training of factor feature
The actual conditions optimization of section, guarantees that the estimation results of trend more level off to actual numerical value, and can extract partial data and join to weight
Number is verified, and more ensure that the preparatory of estimation results.
Detailed description of the invention
Fig. 1 is a kind of detailed process of power grid future operation trend predictor method based on trend parameter provided by the invention
Figure;
Fig. 2 is a kind of structure chart of power grid future operation trend Prediction System based on trend parameter provided by the invention;
Fig. 3 is a kind of brief process of power grid future operation trend predictor method based on trend parameter provided by the invention
Figure.
Specific embodiment
Specific embodiments of the present invention will be described in further detail with reference to the accompanying drawing.
The following description and drawings fully show specific embodiments of the present invention, to enable those skilled in the art to
Practice them.Other embodiments may include structure, logic, it is electrical, process and other change.Implement
Example only represents possible variation.Unless explicitly requested, otherwise individual component and function are sequences that is optional, and operating
It can change.The part of some embodiments and feature can be included in or replace part and the feature of other embodiments.
The range of embodiment of the present invention includes obtained by the entire scope of claims and all of claims
Equivalent.Herein, these embodiments of the invention individually or can be indicated generally with term " invention ", this
Just for the sake of convenient, and if in fact disclosing the invention more than one, it is not meant to automatically limit the model of the application
It encloses for any single invention or inventive concept.
The present invention constructs the model of a relationship between electric network swim and various factors, passes through largely going through for acquisition
History data are trained this model, and are optimized by the data of real-time update to model and parameter, to reach
The numerical value of each influence factor at a given time point can estimate the power grid section tidal current situation of this time.This hair
A kind of bright power grid future operation trend predictor method and system based on trend parameter analyzes grid operation situation change procedure
In influence of all kinds of uncertain factors to security and stability analysis conclusion, extract generation of electricity by new energy fluctuation, bus load prediction miss
The key feature of the uncertain informations such as difference, operation plan exception, research consider probabilistic power grid future operation trend trend
The operation of power networks trend data of parameter prediction method, comprehensive study consideration operation situation uncertainty and control strategy is linear
At technology.It include: the acquisition and storage to history electric network swim data and key feature;According to history electric network swim data and
Key feature is trained parameters weighting shared by key feature;By the parameters weighting of historical data training as a result, right
Power grid future operation trend is estimated.
Embodiment one,
The present invention provides a kind of power grid future operation trend predictor method based on trend parameter, flow chart such as Fig. 1 and 3
It is shown, include the following steps:
1) acquisition and storage of the data acquisition module to history electric network swim data and key feature;
2) parameter training module is according to history electric network swim data and key feature, to parameters weighting shared by key feature
It is trained;
3) Trend Prediction module is by the parameters weighting of historical data training as a result, carrying out to power grid future operation trend pre-
Estimate.
Further, the step 1) includes the following steps:
Step 101: the trend of the data acquisition module acquirement each section of power grid;
Step 102: the corresponding generation of electricity by new energy fluctuation of each section being obtained by external input, bus load prediction misses
The key feature of the uncertain informations such as difference, operation plan exception;
Step 103: the data of acquisition are stored into history data file.
Further, the step 2) includes the following steps:
Step 201: a large amount of flow data and key feature are obtained from history data file.
Step 202: above-mentioned big data being trained, power shared by each feature met under all flow datas is calculated
Weight, and constantly these weights are optimized;Above-mentioned weight is substituted into using partial data (20% of such as overall amount of data), is sentenced
Whether the flow data of disconnected training result is consistent with actual flow data, the success rate after obtaining data training, until verifying
Success rate meet the requirements (such as success rate reaches 90%).
Step 203: verifying above-mentioned weight using partial data, the success rate after judging data training.
Step 204: step 202 and step 203 are executed repeatedly, until the success rate of verifying is met the requirements.
Algorithm citing:
Following algorithm is the gradient descent algorithm of machine learning linear regression, it is assumed that the value of feature is x, and the weight of feature is
θ, below this h (θ) be our fitting function:
It can also be indicated with the form of vector:
hθ(x)=θTX
Lower surface function is the risk function that we are optimized, each of these hθ(xi)-yiAll indicate
Residual error on some training sets between our fitting function and y calculates the risk that its quadratic loss function is constructed as us
Function:
The data of m expression sample.
Our target seeks to minimize risk function, enables our fitting function to the greatest extent to target
Function y is fitted, it may be assumed that
According to traditional thought, it would be desirable to each θ in above-mentioned risk functionjIts partial derivative is sought, each θ is obtainedjIt is right
The gradient answered:
HereIndicate i-th of sample point xiJth component, i.e. θ in h (θ)jxj。
Next since we will minimize risk function, therefore according to each parameter θjNegative gradient direction it is each to update
It is a
Further, the step 3) includes the following steps:
Step 301: the corresponding generation of electricity by new energy fluctuation of input section, bus load prediction error, operation plan are abnormal etc.
The key feature data of uncertain information.
Step 302: obtaining above-mentioned training result, calculate the following operation of power networks trend after being estimated.
Embodiment two,
Based on same inventive concept, the present invention also provides a kind of, and the power grid future operation trend based on trend parameter is estimated
System, structure chart are as shown in Figure 2, comprising:
Acquisition module, for acquiring current electric grid flow data and being corresponded on the influential factor of current electric grid tool
Data;
Module is estimated, for the data after acquisition to be input between the history electric network swim constructed in advance and influence factor
Relational model in, estimate power grid future operation trend;
Relational model between the history electric network swim and influence factor includes the tide of history power grid shared by key feature
Flow data weight.
Further: further including building module, for constructing the relational model between history electric network swim and influence factor.
Further: the building module includes:
Submodule is acquired, for acquiring the history electric network swim data of each section of power grid;Pass corresponding to each section
The corresponding data of key feature described in key collection apparatus;The key feature includes: that generation of electricity by new energy fluctuation, bus load are pre-
It surveys error and operation plan is abnormal;
Computational submodule, for calculating the electricity of history shared by key feature according to history electric network swim data and key feature
Net flow data weight;
Training submodule is optimized for being trained to history electric network swim data weighting shared by key feature
Parameters weighting training result afterwards;
Setting up submodule, for based on the electricity of history shared by the history electric network swim data, key feature, key feature
Net flow data weight constructs the model of relationship between electric network swim and influence factor.
Further: further including memory module, for calculating satisfaction according to history electric network swim data and key feature
Before weight shared by each feature under all flow datas, history electric network swim data and key feature are stored;
The memory module includes:
Submodule is integrated, the history electric network swim data and key feature for that will acquire carry out Data Integration, with section
For unit, by key feature and history electric network swim Data Integration in a data file, the input as parameter training is literary
Part;
Save submodule, for after integrate history electric network swim data and key feature store to history data file
In.
Further: the computational submodule is also used to calculate so that minimizing the weighted value of risk function, comprising:
First construction unit, for constructing the weight fitting function comprising key feature;
Second construction unit constructs risk function according to fitting function, and fitting function and target letter in calculation risk function
The maximum weighted value comprising key feature of residual error between number.
Further: the computational submodule further includes authentication unit, for using partial data to verify above-mentioned weight, is sentenced
Whether the history electric network swim data of disconnected training result are consistent with actual electric network flow data, obtain history electric network swim data
Success rate after being trained with key feature, until success rate meets given threshold.
Further: it is described to estimate module, it is also used between the electric network swim and influence factor in the model of relationship
The key feature for inputting current generation of electricity by new energy fluctuation, bus load prediction error, operation plan exception, according to success rate
The history electric network swim data and key feature for meeting given threshold estimate power grid future trend trend.
The present invention acquires the corresponding numerical value of preset influence factor first;Numerical value based on the acquisition with set in advance
The model of relationship between fixed electric network swim and various factors, estimates the power grid section tidal current situation.Wherein
The model of relationship includes that history electric network swim data and key are special between preset electric network swim and various factors
Sign, the influence factor of selection is accurate, improves the accuracy estimated to power grid section tidal current situation.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute
For realizing the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram
Device.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that instruction stored in the computer readable memory generation includes
The manufacture of command device, the command device are realized in one box of one or more flows of the flowchart and/or block diagram
Or the function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer
Or the instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or box
The step of function of being specified in figure one box or multiple boxes.
The above embodiments are merely illustrative of the technical scheme of the present invention and are not intended to be limiting thereof, although referring to above-described embodiment pair
The present invention is described in detail, and those of ordinary skill in the art still can be to a specific embodiment of the invention
It is modified or replaced equivalently, these are without departing from any modification of spirit and scope of the invention or equivalent replacement, in Shen
Within claims of the invention that please be pending.
Claims (15)
1. a kind of power grid future operation trend predictor method based on trend parameter, it is characterised in that:
Acquire current electric grid flow data and on the corresponding data of the influential factor of current electric grid tool;
Data after acquisition are input in the relational model between the history electric network swim constructed in advance and influence factor, are estimated
Power grid future operation trend;
Relational model between the history electric network swim and influence factor includes history electric network swim number shared by key feature
According to weight.
2. power grid future operation trend predictor method as described in claim 1, it is characterised in that: the history constructed in advance
The model of relationship between electric network swim and influence factor, comprising:
Acquire the history electric network swim data of each section of power grid;
Data corresponding to each section corresponding key feature acquisition key feature;The key feature includes: new energy hair
Electro-mechanical wave, bus load prediction error or operation plan are abnormal;
History electric network swim data weighting shared by key feature is calculated according to history electric network swim data and key feature;
History electric network swim data weighting shared by key feature is trained, the parameters weighting training knot after being optimized
Fruit;Electricity is constructed based on history electric network swim data weighting shared by the history electric network swim data, key feature, key feature
The model of relationship between net trend and influence factor.
3. power grid future operation trend predictor method as claimed in claim 2, it is characterised in that: according to history electric network swim
Data and key feature calculate before history electric network swim data weighting shared by key feature, further includes: to history power grid tide
Flow data and key feature are stored;
It is described to history electric network swim data and key feature carry out storage include:
The history electric network swim data and key feature of acquisition are subjected to Data Integration, as unit of section, by key feature and
Input file of the history electric network swim Data Integration in a data file, as parameter training;
By after integration history electric network swim data and key feature store into history data file.
4. power grid future operation trend predictor method as claimed in claim 2, it is characterised in that: described to shared by key feature
History electric network swim data weighting be trained, the parameters weighting training result after being optimized, comprising:
History electric network swim data and key feature are obtained from history data file;
History electric network swim data and key feature are trained, the power of history electric network swim data shared by key feature is calculated
Weight, and the weight is optimized.
5. power grid future operation trend predictor method as claimed in claim 4, which is characterized in that described to be carried out to the weight
Optimization, comprising: calculate so that minimizing the weighted value of risk function, comprising:
Building includes the weight fitting function of key feature;
Risk function is constructed according to fitting function, and the residual error in calculation risk function between fitting function and objective function is most
The big weighted value comprising key feature.
6. power grid future operation trend predictor method as claimed in claim 5, which is characterized in that the weight fitting function
Vector representation are as follows:
The risk function indicates are as follows:
Minimizing risk function indicates are as follows:
To each weight parameter θ minimized in risk function J (θ)jIts partial derivative is sought, each weight parameter θ is obtainedjIt is corresponding
Gradient:
According to each weight parameter θjNegative gradient direction update each weight parameter θj, indicate are as follows:
In formula: hθ(xi) be the weight fitting function comprising key feature vector representation, θT=[θ0,θ1,...,θn], n
For weight number of components, m indicates key feature sample data, xiIndicate i-th of key feature sample point,Indicate i-th of key
Feature samples point xiJth weight component;yiFor preset practical correctness objective function;α is the step-length of weight parameter variance.
7. power grid future operation trend predictor method as claimed in claim 5, which is characterized in that further include testing weight
Card, including partial data is used to verify above-mentioned weight, the history electric network swim data and actual electric network trend of training of judgement result
Whether data are consistent, and the success rate after obtaining history electric network swim data and key feature training is set until success rate meets
Threshold value.
8. power grid future operation trend predictor method as described in claim 1, it is characterised in that: based on the electric network swim and
The model pre-estimating power grid future operation trend of relationship between influence factor, comprising:
To current generation of electricity by new energy fluctuation, bus load are inputted between the electric network swim and influence factor in the model of relationship
The key feature for predicting error, operation plan exception, history electric network swim data and the pass of given threshold are met according to success rate
Key feature estimates power grid future trend trend.
9. a kind of power grid future operation trend Prediction System based on trend parameter, it is characterised in that: include:
Acquisition module, for acquiring current electric grid flow data and on the corresponding number of the influential factor of current electric grid tool
According to;
Module is estimated, for the data after acquisition to be input to the pass between the history electric network swim constructed in advance and influence factor
It is to estimate power grid future operation trend in model;
Relational model between the history electric network swim and influence factor includes history electric network swim number shared by key feature
According to weight.
10. power grid future operation trend Prediction System as claimed in claim 9, it is characterised in that: further include building module, use
Relational model between building history electric network swim and influence factor.
11. power grid future operation trend Prediction System as claimed in claim 10, it is characterised in that: the building module packet
It includes:
Submodule is acquired, for acquiring the history electric network swim data of each section of power grid;It is corresponding to each section crucial special
Sign acquires the corresponding data of the key feature;The key feature includes: generation of electricity by new energy fluctuation, bus load prediction error
With operation plan exception;
Computational submodule, for calculating history power grid tide shared by key feature according to history electric network swim data and key feature
Flow data weight;
Training submodule, for being trained to history electric network swim data weighting shared by key feature, after being optimized
Parameters weighting training result;
Setting up submodule, for based on the tide of history power grid shared by the history electric network swim data, key feature, key feature
Flow data weight constructs the model of relationship between electric network swim and influence factor.
12. power grid future operation trend Prediction System as claimed in claim 10, it is characterised in that: it further include memory module,
For calculating power shared by each feature met under all flow datas according to history electric network swim data and key feature
Before weight, history electric network swim data and key feature are stored;The memory module includes:
Submodule is integrated, the history electric network swim data and key feature for that will acquire carry out Data Integration, are single with section
Position, by the input file of key feature and history electric network swim Data Integration in a data file, as parameter training;
Save submodule, for after integrate history electric network swim data and key feature storage into history data file.
13. power grid future operation trend Prediction System as claimed in claim 11, it is characterised in that: the computational submodule,
It is also used to calculate so that minimizing the weighted value of risk function, comprising:
First construction unit, for constructing the weight fitting function comprising key feature;
Second construction unit constructs risk function according to fitting function, and in calculation risk function fitting function and objective function it
Between residual error the maximum weighted value comprising key feature.
14. power grid future operation trend Prediction System as claimed in claim 13, it is characterised in that: it further include authentication unit,
For using partial data to verify above-mentioned weight, the history electric network swim data and actual electric network flow data of training of judgement result
Whether consistent, after obtaining history electric network swim data and key feature training success rate, until success rate meets given threshold.
15. power grid future operation trend Prediction System as claimed in claim 9, it is characterised in that: it is described to estimate module, also use
In pre- to current generation of electricity by new energy fluctuation, bus load is inputted between the electric network swim and influence factor in the model of relationship
The key feature for surveying error, operation plan exception, the history electric network swim data and key of given threshold are met according to success rate
Feature estimates power grid future trend trend.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711335875.7A CN109961160B (en) | 2017-12-14 | 2017-12-14 | Power grid future operation trend estimation method and system based on tide parameters |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711335875.7A CN109961160B (en) | 2017-12-14 | 2017-12-14 | Power grid future operation trend estimation method and system based on tide parameters |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109961160A true CN109961160A (en) | 2019-07-02 |
CN109961160B CN109961160B (en) | 2024-01-19 |
Family
ID=67017709
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711335875.7A Active CN109961160B (en) | 2017-12-14 | 2017-12-14 | Power grid future operation trend estimation method and system based on tide parameters |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109961160B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111131331A (en) * | 2020-01-15 | 2020-05-08 | 国网陕西省电力公司电力科学研究院 | Network vulnerability guided information attack-oriented moving target defense deployment optimization method |
CN111952959A (en) * | 2020-07-14 | 2020-11-17 | 北京科东电力控制系统有限责任公司 | Method and device for compressing power grid process simulation time and storage medium |
CN115344830A (en) * | 2022-08-02 | 2022-11-15 | 无锡致为数字科技有限公司 | Event probability estimation method based on big data |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104008512A (en) * | 2014-06-12 | 2014-08-27 | 国家电网公司 | Online stability evaluation index system of electric power system |
US20150081129A1 (en) * | 2011-12-28 | 2015-03-19 | State Grid Electric Power Institute | Equipment overload successive approximation adaptive control method based on centralized real-time decision |
CN104600695A (en) * | 2014-12-29 | 2015-05-06 | 国家电网公司 | Trend load flow calculating method based on online status estimation and real-time scheduling plans |
CN105071385A (en) * | 2015-08-10 | 2015-11-18 | 国家电网公司 | Power grid operating data real-time analysis system |
CN106920123A (en) * | 2017-01-18 | 2017-07-04 | 广州大学 | A kind of dining room data analysis system based on machine learning and Internet of Things |
-
2017
- 2017-12-14 CN CN201711335875.7A patent/CN109961160B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150081129A1 (en) * | 2011-12-28 | 2015-03-19 | State Grid Electric Power Institute | Equipment overload successive approximation adaptive control method based on centralized real-time decision |
CN104008512A (en) * | 2014-06-12 | 2014-08-27 | 国家电网公司 | Online stability evaluation index system of electric power system |
CN104600695A (en) * | 2014-12-29 | 2015-05-06 | 国家电网公司 | Trend load flow calculating method based on online status estimation and real-time scheduling plans |
CN105071385A (en) * | 2015-08-10 | 2015-11-18 | 国家电网公司 | Power grid operating data real-time analysis system |
CN106920123A (en) * | 2017-01-18 | 2017-07-04 | 广州大学 | A kind of dining room data analysis system based on machine learning and Internet of Things |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111131331A (en) * | 2020-01-15 | 2020-05-08 | 国网陕西省电力公司电力科学研究院 | Network vulnerability guided information attack-oriented moving target defense deployment optimization method |
CN111131331B (en) * | 2020-01-15 | 2022-02-22 | 国网陕西省电力公司电力科学研究院 | Network vulnerability guided information attack-oriented moving target defense deployment optimization method |
CN111952959A (en) * | 2020-07-14 | 2020-11-17 | 北京科东电力控制系统有限责任公司 | Method and device for compressing power grid process simulation time and storage medium |
CN111952959B (en) * | 2020-07-14 | 2023-10-13 | 北京科东电力控制系统有限责任公司 | Method, device and storage medium for compressing power grid process simulation time |
CN115344830A (en) * | 2022-08-02 | 2022-11-15 | 无锡致为数字科技有限公司 | Event probability estimation method based on big data |
Also Published As
Publication number | Publication date |
---|---|
CN109961160B (en) | 2024-01-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109214605A (en) | Power-system short-term Load Probability prediction technique, apparatus and system | |
CN108121215B (en) | Process control loops method of evaluating performance and device based on full loop reconstruct emulation | |
CN105631483A (en) | Method and device for predicting short-term power load | |
CN105205570A (en) | Power grid power sale quantity prediction method based on season time sequence analysis | |
US20210133376A1 (en) | Systems and methods of parameter calibration for dynamic models of electric power systems | |
CN104636985A (en) | Method for predicting radio disturbance of electric transmission line by using improved BP (back propagation) neural network | |
KR101660102B1 (en) | Apparatus for water demand forecasting | |
CN104899135A (en) | Software defect prediction method and system | |
CN109961160A (en) | A kind of power grid future operation trend predictor method and system based on trend parameter | |
Yao et al. | Improved support vector machine regression in multi-step-ahead prediction for rock displacement surrounding a tunnel | |
Maruta et al. | Synthesis of fixed-structure robust controllers using a constrained particle swarm optimizer with cyclic neighborhood topology | |
CN110334879A (en) | Power grid bus reactive load forecasting method and device | |
CN103885867B (en) | Online evaluation method of performance of analog circuit | |
CN109829627A (en) | A kind of safe confidence appraisal procedure of Electrical Power System Dynamic based on integrated study scheme | |
CN103514488B (en) | Power-system short-term load forecasting device and method based on combination forecasting | |
CN108022004A (en) | A kind of adaptive weighting training method of multi-model weighted array Forecasting Power System Load | |
CN104993966B (en) | A kind of electric integrated service network method for predicting | |
Wang et al. | A DRL-aided multi-layer stability model calibration platform considering multiple events | |
CN110378358A (en) | A kind of power distribution network isomeric data integration method and system | |
US10635078B2 (en) | Simulation system, simulation method, and simulation program | |
JP6894724B2 (en) | Power equipment planning method, power equipment manufacturing method, power equipment planning equipment and programs | |
Jain et al. | A novel hybrid method for short term load forecasting using fuzzy logic and particle swarm optimization | |
CN111080037A (en) | Short-term power load prediction method and device based on deep neural network | |
Coelho et al. | A heuristic fuzzy algorithm bio-inspired by evolution strategies for energy forecasting problems | |
CN115619028A (en) | Clustering algorithm fusion-based power load accurate prediction method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |