CN109993364A - A kind of prediction technique and device of natural gas gas consumption - Google Patents
A kind of prediction technique and device of natural gas gas consumption Download PDFInfo
- Publication number
- CN109993364A CN109993364A CN201910257250.6A CN201910257250A CN109993364A CN 109993364 A CN109993364 A CN 109993364A CN 201910257250 A CN201910257250 A CN 201910257250A CN 109993364 A CN109993364 A CN 109993364A
- Authority
- CN
- China
- Prior art keywords
- natural gas
- data
- gas consumption
- principal component
- matrix
- 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
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 392
- 239000007789 gas Substances 0.000 title claims abstract description 224
- 239000003345 natural gas Substances 0.000 title claims abstract description 198
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000012417 linear regression Methods 0.000 claims abstract description 86
- 239000011159 matrix material Substances 0.000 claims abstract description 85
- 238000000513 principal component analysis Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000007405 data analysis Methods 0.000 claims description 6
- 239000004615 ingredient Substances 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004836 empirical method Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000567 combustion gas Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application provides the prediction techniques and device of a kind of natural gas gas consumption, obtain the relevant historical data of natural gas gas consumption, and pre-process to the relevant historical data of natural gas gas consumption；Principal component analysis is carried out to pretreated data, obtains multiple principal components and main variables matrix for influencing natural gas gas consumption；Multielement linear regression model (LRM) is constructed according to multiple principal components for influencing natural gas gas consumption, and multielement linear regression model (LRM) is trained using the relevant historical data of natural gas gas consumption and main variables matrix, obtains natural gas gas consumption prediction model；The parameter value of the principal component for influencing natural gas gas consumption each in preset time period is input in the natural gas gas consumption prediction model and is calculated, the predicted value of natural gas gas consumption is obtained.The present invention uses big data technology and multielement linear regression algorithm, trains natural gas gas consumption prediction model based on mass historical data, realizes the Accurate Prediction to natural gas gas consumption.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of prediction technique and dress of natural gas gas consumption
It sets.
Background technique
In recent years, with the rapid development of our country's economy, Natural Gas Demand had obtained quick growth, natural in order to reduce
Gas carrying cost and the generation for avoiding " gas is waste " phenomenon need to carry out Accurate Prediction to natural gas gas consumption.
The general natural gas gas consumption for speculating next day according to the natural gas gas consumption on the same day using empirical method at present, due to
Artificial interference caused by subjective factors is big so that prediction natural gas gas consumption and practical natural gas gas consumption between deviation compared with
Greatly.When the natural gas gas consumption of prediction is greater than practical natural gas gas consumption, the carrying cost of natural gas is increased, and when prediction
Natural gas gas consumption when being less than practical natural gas gas consumption, cause natural gas not enough, or even the hair of " gas famine " phenomenon occur
It is raw.
Summary of the invention
In view of this, it is based on big data technology the present invention provides the prediction technique and device of a kind of natural gas gas consumption,
Multielement linear regression model (LRM) is trained using mass historical data, obtains the natural gas gas consumption prediction model of science,
Improve the accuracy of natural gas gas consumption prediction.
In order to achieve the above-mentioned object of the invention, specific technical solution provided by the invention is as follows:
A kind of prediction technique of natural gas gas consumption, comprising:
The relevant historical data of natural gas gas consumption is obtained, and the relevant historical data of natural gas gas consumption is located in advance
Reason；
Principal component analysis is carried out to pretreated data, obtains multiple principal components for influencing natural gas gas consumption and master
Component variable matrix；
Multielement linear regression model (LRM) is constructed according to multiple principal components for influencing natural gas gas consumption, and using naturally
The relevant historical data of gas gas consumption and the main variables matrix are trained the multielement linear regression model (LRM), obtain
To natural gas gas consumption prediction model；
The parameter value of each principal component for influencing natural gas gas consumption in preset time period is obtained, and by the parameter
Value is input in the natural gas gas consumption prediction model and is calculated, and obtains the predicted value of natural gas gas consumption.
Optionally, the relevant historical data to natural gas gas consumption pre-processes, comprising:
Data cleansing processing is carried out to the relevant historical data of natural gas gas consumption；
The data that there is missing after data cleaning treatment are handled using multiple interpolation, fill up the data of missing；
Data are normalized, the data of different dimensions are converted to the data with identical dimension, and will not
Data with the order of magnitude are converted to the data with same order.
Optionally, described that principal component analysis is carried out to pretreated data, obtain multiple influence natural gas gas consumptions
Principal component and principal component matrix, comprising:
Pretreated data are standardized, matrix of variables are generated, after the matrix of variables indicates pretreatment
Data in value of each variable within the corresponding time；
The related coefficient between variable is calculated, the correlation matrix between variable is generated；
The eigen vector of each principal component is calculated according to the correlation matrix；
The contribution rate of each principal component is calculated, and the principal component is ranked up by contribution rate is descending, is taken
Preceding K principal component is as the principal component for influencing natural gas gas consumption；
The feature vector of the principal component for influencing natural gas gas consumption is pressed into row generator matrix S；
Dimension-reduction treatment is carried out to the matrix of variables using the matrix S, obtains the main variables matrix Y.
Optionally, described to construct multielement linear regression mould according to multiple principal components for influencing natural gas gas consumption
Type, and using the relevant historical data and the main variables matrix of natural gas gas consumption to the multielement linear regression mould
Type is trained, and obtains natural gas gas consumption prediction model, comprising:
Multielement linear regression model (LRM) is constructed, the quantity of variable and the influence day in the multielement linear regression model (LRM)
The quantity of the principal component of right gas gas consumption is identical；
By each principal component in the main variables matrix in the variate-value of different time and the phase of natural gas gas consumption
It closes in historical data and is iterated training in the corresponding natural gas gas consumption input multielement linear regression model (LRM), obtain institute
The regression coefficient of each variable in multielement linear regression model (LRM) is stated, and then obtains natural gas gas consumption prediction model.
A kind of prediction meanss of natural gas gas consumption, comprising:
Data processing unit, for obtaining the relevant historical data of natural gas gas consumption, and to the phase of natural gas gas consumption
Historical data is closed to be pre-processed；
Data analysis unit obtains multiple influence natural gases and uses for carrying out principal component analysis to pretreated data
The principal component and main variables matrix of tolerance；
Model construction unit, for linearly being returned according to multiple principal component building multielements for influencing natural gas gas consumption
Return model, and the multielement is linearly returned using the relevant historical data of natural gas gas consumption and the main variables matrix
Return model to be trained, obtains natural gas gas consumption prediction model；
Gas consumption predicting unit, for obtaining each principal component for influencing natural gas gas consumption in preset time period
Parameter value, and the parameter value is input in the natural gas gas consumption prediction model and is calculated, obtain natural gas gas
The predicted value of amount.
Optionally, the data processing unit carries out data specifically for the relevant historical data to natural gas gas consumption
Cleaning treatment；The data that there is missing after data cleaning treatment are handled using multiple interpolation, fill up the data of missing；
Data are normalized, the data of different dimensions are converted to the data with identical dimension, and by different number grade
Data be converted to the data with same order.
Optionally, the data analysis unit generates and becomes specifically for being standardized to pretreated data
Moment matrix, the matrix of variables indicate value of each variable within the corresponding time in pretreated data；Between calculating variable
Related coefficient, generate variable between correlation matrix；The feature of each principal component is calculated according to the correlation matrix
Value and feature vector；The contribution rate of each principal component is calculated, and the principal component is arranged by contribution rate is descending
Sequence, K principal component is as the principal component for influencing natural gas gas consumption before taking；By it is described influence natural gas gas consumption it is main at
The feature vector divided presses row generator matrix S；Dimension-reduction treatment is carried out to the matrix of variables using the matrix S, obtains the master
Component variable matrix Y.
Optionally, the model construction unit, is specifically used for building multielement linear regression model (LRM), and the multielement is linear
The quantity of variable is identical as the influence quantity of principal component of natural gas gas consumption in regression model；By the main variables
Each principal component corresponding natural gas in the variate-value of different time and the relevant historical data of natural gas gas consumption in matrix
Gas consumption inputs in the multielement linear regression model (LRM) and is iterated training, obtains every in the multielement linear regression model (LRM)
The regression coefficient of a variable, and then obtain natural gas gas consumption prediction model.
Compared with the existing technology, beneficial effects of the present invention are as follows:
The prediction technique and device of natural gas gas consumption disclosed by the invention, the first phase to the natural gas gas consumption of acquisition
It closes historical data to be pre-processed, ensure that the quality of data, analyzed convenient for subsequent, then using Principal Component Analysis to pre-
Treated, and data are analyzed, and obtain multiple principal components and main variables matrix for influencing natural gas gas consumption, and benefit
Multielement linear regression model (LRM) is trained with the relevant historical data of natural gas gas consumption and main variables matrix, is obtained
Natural gas gas consumption prediction model may be implemented using the natural gas gas consumption prediction model to the accurate pre- of natural gas gas consumption
It surveys, solves and the natural gas gas consumption and practical natural gas gas consumption that lead to prediction are artificially predicted using empirical method in the prior art
Between the larger problem of deviation, improve the accuracy of natural gas gas consumption prediction.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow diagram of the prediction technique of natural gas gas consumption disclosed by the embodiments of the present invention；
Fig. 2 is a kind of flow diagram of the preprocess method of data disclosed by the embodiments of the present invention；
Fig. 3 is a kind of flow diagram of principal component analytical method disclosed by the embodiments of the present invention；
Fig. 4 is a kind of process signal of construction method of natural gas gas consumption prediction model disclosed by the embodiments of the present invention
Figure；
Fig. 5 is a kind of structural schematic diagram of the prediction meanss of natural gas gas consumption disclosed by the embodiments of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Present embodiment discloses a kind of prediction technique of natural gas gas consumption, referring to Fig. 1, this method specifically include it is following
Step:
S101: obtain natural gas gas consumption relevant historical data, and to the relevant historical data of natural gas gas consumption into
Row pretreatment；
The relevant historical data general record of natural gas gas consumption in the database, can be by way of calling database
Obtain the relevant historical data of natural gas gas consumption.
The relevant historical data of natural gas gas consumption includes area, natural gas gas consumption, weather conditions, natural gas gas class
The corresponding data of variables such as type, season, natural gas gas plan (production plan), festivals or holidays, natural gas user number.
After the relevant historical data for obtaining natural gas gas consumption, the relevant historical data of natural gas gas consumption is carried out pre-
Processing, is analyzed convenient for subsequent.
Specifically, referring to Fig. 2, the relevant historical data of natural gas gas consumption is pre-processed the following steps are included:
S201: data cleansing processing is carried out to the relevant historical data of natural gas gas consumption；
Data cleansing processing includes deleting nugatory abnormal data and nonsensical variable, wherein is not worth
Abnormal data such as business or the data that generate of system exception, nonsensical variable expression natural gas gas consumption is not influenced
Variable.
S202: the data that there is missing after data cleaning treatment are handled using multiple interpolation, fill up missing
Data；
If the relevant historical data of the natural gas gas consumption recorded in database generally comprises the data of multiple dimensions, when out
When the case where existing dimension missing, need to carry out the data of missing to fill up missing processing.
S203: being normalized data, and the data of different dimensions are converted to the data with identical dimension, and
The data of different number grade are converted to the data with same order.
Data can be normalized by linear function, to the methods of function, arc cotangent function.
S102: carrying out principal component analysis to pretreated data, obtains multiple principal components for influencing natural gas gas consumption
And main variables matrix；
Specifically, referring to Fig. 3, carrying out analysis to pretreated data using Principal Component Analysis includes following step
It is rapid:
S301: being standardized pretreated data, generates matrix of variables, and the matrix of variables indicates pre- place
Value of each variable within the corresponding time in data after reason；
It should be noted that being standardized pretreated data including by domestic gas, industry and commerce gas, good fortune
Summarized using gas for a data source A, data source A is standardized, converts the value of the various variables of data source A to
Numerical value obtains matrix of variables X.
Wherein, p indicates that the number of days of historical data, n indicate the quantity of variable.
S302: calculating the related coefficient between variable, generates the correlation matrix between variable；
The calculation formula for calculating the related coefficient between variable is as follows:
Wherein, rijIndicate variable xiWith xjBetween related coefficient, rij=rji；
I, j=1,2,3 ..., p.
The correlation matrix being calculated is as follows:
S303: the eigen vector of each principal component is calculated according to the correlation matrix；
Specifically, passing through solution characteristic equation | λ E-R |=0, calculate the characteristic value of each principal component.
Wherein, E is unit matrix, and λ is the characteristic value of principal component, and the quantity of principal component is not more than p.
Feature vector calculation method are as follows:
Characteristic value is substituted into equation λ v=R, wherein λ is characterized value, and R is correlation matrix, and ν is feature vector, and λ is
Know, R is it is known that solve ν.
S304: the contribution rate of each principal component is calculated, and the principal component is arranged by contribution rate is descending
Sequence, K principal component is as the principal component for influencing natural gas gas consumption before taking；
Specifically, the calculation method of contribution rate is as follows:
S305: the feature vector of the principal component for influencing natural gas gas consumption is pressed into row generator matrix S；
Wherein, [x11,x21,...,xm1] indicate feature vector.
S306: using the matrix S to the matrix of variables, dimension-reduction treatment is carried out, matrix Y is obtained.
Specifically, Y=SX
Wherein,
The ginseng of every a line in matrix indicates to influence some principal component of natural gas gas consumption every day in the historical data
Numerical value, each column of matrix indicate the parameter of each principal component for influencing natural gas gas consumption in some day of historical data
Value.
S103: multielement linear regression model (LRM), and benefit are constructed according to multiple principal components for influencing natural gas gas consumption
The multielement linear regression model (LRM) is carried out with the relevant historical data of natural gas gas consumption and the main variables matrix
Training, obtains natural gas gas consumption prediction model；
Specifically, referring to Fig. 4, the construction method of natural gas gas consumption prediction model is as follows:
S401: building multielement linear regression model (LRM), in the multielement linear regression model (LRM) quantity of variable with it is described
The quantity for influencing the principal component of natural gas gas consumption is identical；
Multielement linear regression model (LRM) is as follows:
Y=F1x1+F2x2+…+Fkxk
Wherein, F1,F2…FkRespectively indicate the principal component for influencing natural gas gas consumption, x1,x2…xkIt is influenced naturally to be each
The regression coefficient of the principal component of gas gas consumption, Y indicate natural gas gas consumption.
S402: by principal component each in main variables matrix in the variate-value of different time and the phase of natural gas gas consumption
It closes in historical data and is iterated training in the corresponding natural gas gas consumption input multielement linear regression model (LRM), obtain institute
The regression coefficient of each variable in multielement linear regression model (LRM) is stated, and then obtains natural gas gas consumption prediction model.
Variate-value first by principal component each in main variables matrix in different time inputs multiple linear regression mould
Type, then will be more with natural gas gas consumption input of the principal component in time corresponding historical data with main variables matrix
First linear regression model (LRM).If the principal component in main variables matrix is n-th day, then the relevant historical number of natural gas gas consumption
Natural gas gas consumption in is also n-th day.
Finally, being iterated training by the data for inputting multiple times, obtain each in multielement linear regression model (LRM)
The regression coefficient of variable, and then obtain natural gas gas consumption prediction model.
S104: the parameter value of each principal component for influencing natural gas gas consumption in preset time period is obtained, and by institute
It states parameter value and is input in the natural gas gas consumption prediction model and calculated, obtain the predicted value of natural gas gas consumption.In advance
If the period is user according to practical business demand preset a period of time, such as second day of current time, obtained day
The predicted value of right gas gas consumption is to predict the predicted value of the natural gas gas consumption of day.
It should be noted that the input data of natural gas gas consumption prediction model is each in preset time period influences naturally
The parameter value of the principal component of gas gas consumption, output data are the predicted value of natural gas gas consumption.
Wherein, the method for the parameter value of the principal component of each corresponding natural gas gas consumption in preset time period is obtained are as follows: first
Each variate-value for influencing natural gas gas consumption in preset time period is first obtained, such as weather conditions, festivals or holidays, combustion gas gas meter
It draws, gas user number etc., then further according to the principal component for influencing natural gas gas consumption and influences the line between natural gas gas consumption
Property calculated relationship, on these influence natural gas gas consumptions variate-values handle, obtain each influence natural gas gas
The parameter value of the principal component of amount.
The prediction technique of natural gas gas consumption disclosed in the present embodiment first goes through the correlation of the natural gas gas consumption of acquisition
History data are pre-processed, and ensure that the quality of data, are analyzed convenient for subsequent, then using Principal Component Analysis to pretreatment
Data afterwards are analyzed, and obtain multiple principal components and main variables matrix for influencing natural gas gas consumption, and utilize day
The relevant historical data and main variables matrix of right gas gas consumption are trained multielement linear regression model (LRM), obtain natural
The Accurate Prediction to natural gas gas consumption may be implemented using the natural gas gas consumption prediction model in gas gas consumption prediction model,
Solve in the prior art using empirical method artificially predict to cause prediction natural gas gas consumption and practical natural gas gas consumption it
Between the larger problem of deviation, improve the accuracy of natural gas gas consumption prediction.
Disclosed a kind of prediction technique of natural gas gas consumption based on the above embodiment, the present embodiment correspondence disclose one kind
The prediction meanss of natural gas gas consumption, referring to Fig. 4, the device includes:
Data processing unit 501, for obtaining the relevant historical data of natural gas gas consumption, and to natural gas gas consumption
Relevant historical data is pre-processed；
Data analysis unit 502 obtains multiple influence natural gases for carrying out principal component analysis to pretreated data
The principal component and main variables matrix of gas consumption；
Model construction unit 503, for constructing multielement line according to multiple principal components for influencing natural gas gas consumption
Property regression model, and using natural gas gas consumption relevant historical data and the main variables matrix to the multielement line
Property regression model is trained, and obtains natural gas gas consumption prediction model；
Gas consumption predicting unit 504, for obtain in preset time period it is each it is described influence natural gas gas consumption it is main at
The parameter value divided, and the parameter value is input in the natural gas gas consumption prediction model and is calculated, obtain natural gas
The predicted value of gas consumption.
Optionally, the data processing unit 501, counts specifically for the relevant historical data to natural gas gas consumption
According to cleaning treatment；The data that there is missing after data cleaning treatment are handled using multiple interpolation, fill up the number of missing
According to；Data are normalized, the data of different dimensions are converted to the data with identical dimension, and by different number
The data of grade are converted to the data with same order.
Optionally, the data analysis unit 502, it is raw specifically for being standardized to pretreated data
At matrix of variables, the matrix of variables indicates value of each variable within the corresponding time in pretreated data；It calculates and becomes
Related coefficient between amount generates the correlation matrix between variable；Each principal component is calculated according to the correlation matrix
Eigen vector；Calculate the contribution rate of each principal component, and by contribution rate it is descending to the principal component into
Row sequence, K principal component is as the principal component for influencing natural gas gas consumption before taking；By the influence natural gas gas consumption
The feature vector of principal component presses row generator matrix S；Dimension-reduction treatment is carried out to the matrix of variables using the matrix S, obtains institute
State main variables matrix Y.
Optionally, the model construction unit 503 is specifically used for building multielement linear regression model (LRM), the multielement
The quantity of variable is identical as the influence quantity of principal component of natural gas gas consumption in linear regression model (LRM)；By the principal component
Each principal component corresponding day in the variate-value of different time and the relevant historical data of natural gas gas consumption in matrix of variables
Right gas gas consumption, which inputs in the multielement linear regression model (LRM), is iterated training, obtains the multielement linear regression model (LRM)
In each variable regression coefficient, and then obtain natural gas gas consumption prediction model.
The prediction meanss of natural gas gas consumption disclosed in the present embodiment first go through the correlation of the natural gas gas consumption of acquisition
History data are pre-processed, and ensure that the quality of data, are analyzed convenient for subsequent, then using Principal Component Analysis to pretreatment
Data afterwards are analyzed, and obtain multiple principal components and main variables matrix for influencing natural gas gas consumption, and utilize day
The relevant historical data and main variables matrix of right gas gas consumption are trained multielement linear regression model (LRM), obtain natural
The Accurate Prediction to natural gas gas consumption may be implemented using the natural gas gas consumption prediction model in gas gas consumption prediction model,
Solve in the prior art using empirical method artificially predict to cause prediction natural gas gas consumption and practical natural gas gas consumption it
Between the larger problem of deviation, improve the accuracy of natural gas gas consumption prediction.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (8)
1. a kind of prediction technique of natural gas gas consumption characterized by comprising
The relevant historical data of natural gas gas consumption is obtained, and the relevant historical data of natural gas gas consumption is pre-processed；
Principal component analysis is carried out to pretreated data, obtains multiple principal components and principal component for influencing natural gas gas consumption
Matrix of variables；
Multielement linear regression model (LRM) is constructed according to multiple principal components for influencing natural gas gas consumption, and is used using natural gas
The relevant historical data of tolerance and the main variables matrix are trained the multielement linear regression model (LRM), obtain day
Right gas gas consumption prediction model；
The parameter value of each principal component for influencing natural gas gas consumption in preset time period is obtained, and the parameter value is defeated
Enter into the natural gas gas consumption prediction model and calculated, obtains the predicted value of natural gas gas consumption.
2. the method according to claim 1, wherein the relevant historical data to natural gas gas consumption carries out
Pretreatment, comprising:
Data cleansing processing is carried out to the relevant historical data of natural gas gas consumption；
The data that there is missing after data cleaning treatment are handled using multiple interpolation, fill up the data of missing；
Data are normalized, the data of different dimensions are converted to the data with identical dimension, and by different numbers
The data of magnitude are converted to the data with same order.
3. the method according to claim 1, wherein it is described to pretreated data carry out principal component analysis,
Obtain multiple principal components and principal component matrix for influencing natural gas gas consumption, comprising:
Pretreated data are standardized, matrix of variables is generated, the matrix of variables indicates pretreated number
Value of each variable within the corresponding time in；
The related coefficient between variable is calculated, the correlation matrix between variable is generated；
The eigen vector of each principal component is calculated according to the correlation matrix；
The contribution rate of each principal component is calculated, and the principal component is ranked up by contribution rate is descending, K before taking
Principal component is as the principal component for influencing natural gas gas consumption；
The feature vector of the principal component for influencing natural gas gas consumption is pressed into row generator matrix S；
Dimension-reduction treatment is carried out to the matrix of variables using the matrix S, obtains the main variables matrix Y.
4. the method according to claim 1, wherein described according to multiple masters for influencing natural gas gas consumption
Ingredient constructs multielement linear regression model (LRM), and utilizes the relevant historical data of natural gas gas consumption and the main variables square
Battle array is trained the multielement linear regression model (LRM), obtains natural gas gas consumption prediction model, comprising:
Multielement linear regression model (LRM) is constructed, the quantity of variable and the influence natural gas in the multielement linear regression model (LRM)
The quantity of the principal component of gas consumption is identical；
Variate-value by each principal component in the main variables matrix in different time related to natural gas gas consumption is gone through
Corresponding natural gas gas consumption inputs in the multielement linear regression model (LRM) and is iterated training in history data, obtains described more
The regression coefficient of each variable in element linear regression model (LRM), and then obtain natural gas gas consumption prediction model.
5. a kind of prediction meanss of natural gas gas consumption characterized by comprising
Data processing unit is gone through for obtaining the relevant historical data of natural gas gas consumption, and to the correlation of natural gas gas consumption
History data are pre-processed；
Data analysis unit obtains multiple influence natural gas gas consumptions for carrying out principal component analysis to pretreated data
Principal component and main variables matrix；
Model construction unit, for constructing multielement linear regression mould according to multiple principal components for influencing natural gas gas consumption
Type, and using the relevant historical data and the main variables matrix of natural gas gas consumption to the multielement linear regression mould
Type is trained, and obtains natural gas gas consumption prediction model；
Gas consumption predicting unit, for obtaining the parameter of each principal component for influencing natural gas gas consumption in preset time period
Value, and the parameter value is input in the natural gas gas consumption prediction model and is calculated, obtain natural gas gas consumption
Predicted value.
6. device according to claim 5, which is characterized in that the data processing unit is specifically used for using natural gas
The relevant historical data of tolerance carries out data cleansing processing；Using multiple interpolation to the number that there is missing after data cleaning treatment
According to being handled, the data of missing are filled up；Data are normalized, the data of different dimensions are converted to identical
The data of dimension, and the data of different number grade are converted to the data with same order.
7. device according to claim 5, which is characterized in that the data analysis unit, after being specifically used for pretreatment
Data be standardized, generate matrix of variables, the matrix of variables indicates each variable in pretreated data
Value within the corresponding time；The related coefficient between variable is calculated, the correlation matrix between variable is generated；According to the phase relation
Matrix number calculates the eigen vector of each principal component；The contribution rate of each principal component is calculated, and presses contribution rate
Descending to be ranked up to the principal component, K principal component is as the principal component for influencing natural gas gas consumption before taking；It will
The feature vector of the principal component for influencing natural gas gas consumption presses row generator matrix S；Using the matrix S to the variable square
Battle array carries out dimension-reduction treatment, obtains the main variables matrix Y.
8. device according to claim 5, which is characterized in that the model construction unit is specifically used for building multielement
Linear regression model (LRM), the quantity of variable and the principal component for influencing natural gas gas consumption in the multielement linear regression model (LRM)
Quantity it is identical；By each principal component in the main variables matrix different time variate-value and natural gas gas consumption
Corresponding natural gas gas consumption inputs in the multielement linear regression model (LRM) and is iterated training in relevant historical data, obtains
The regression coefficient of each variable in the multielement linear regression model (LRM), and then obtain natural gas gas consumption prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910257250.6A CN109993364A (en) | 2019-04-01 | 2019-04-01 | A kind of prediction technique and device of natural gas gas consumption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910257250.6A CN109993364A (en) | 2019-04-01 | 2019-04-01 | A kind of prediction technique and device of natural gas gas consumption |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109993364A true CN109993364A (en) | 2019-07-09 |
Family
ID=67130786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910257250.6A Pending CN109993364A (en) | 2019-04-01 | 2019-04-01 | A kind of prediction technique and device of natural gas gas consumption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109993364A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429180A (en) * | 2020-03-23 | 2020-07-17 | 上海中旖能源科技有限公司 | Liquefied natural gas demand prediction method and device |
| CN111652417A (en) * | 2020-05-25 | 2020-09-11 | 成都千嘉科技有限公司 | Automatic correction method and system for remote meter reading data |
| CN112101612A (en) * | 2020-08-03 | 2020-12-18 | 深圳市燃气集团股份有限公司 | Prediction method of town gas usage, storage medium and terminal equipment |
| CN113326654A (en) * | 2021-05-20 | 2021-08-31 | 北京市燃气集团有限责任公司 | Method and device for constructing gas load prediction model |
| CN113326983A (en) * | 2021-05-28 | 2021-08-31 | 重庆能源大数据中心有限公司 | Natural gas consumption prediction system and method |
| CN113869588A (en) * | 2021-09-29 | 2021-12-31 | 广州新科佳都科技有限公司 | Energy consumption benchmarking value prediction method, device, equipment and storage medium |
| CN114740159A (en) * | 2022-04-14 | 2022-07-12 | 成都秦川物联网科技股份有限公司 | Natural gas energy metering component acquisition method and Internet of things system |
| US11979697B2 (en) | 2021-07-26 | 2024-05-07 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods and internet of things systems for obtaining natural gas energy metering component |
| CN118128737A (en) * | 2024-05-07 | 2024-06-04 | 德耐尔能源装备有限公司 | Intelligent exhaust pressure control method and device for protecting unit |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933483A (en) * | 2015-06-17 | 2015-09-23 | 中国电力科学研究院 | Wind power forecasting method dividing based on weather process |
| WO2016144225A1 (en) * | 2015-03-12 | 2016-09-15 | Telefonaktiebolaget Lm Ericsson (Publ) | Method node and computer program for energy prediction |
| CN107730054A (en) * | 2017-11-15 | 2018-02-23 | 西南石油大学 | A kind of Gas Load combination forecasting method based on support vector regression |
-
2019
- 2019-04-01 CN CN201910257250.6A patent/CN109993364A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016144225A1 (en) * | 2015-03-12 | 2016-09-15 | Telefonaktiebolaget Lm Ericsson (Publ) | Method node and computer program for energy prediction |
| CN104933483A (en) * | 2015-06-17 | 2015-09-23 | 中国电力科学研究院 | Wind power forecasting method dividing based on weather process |
| CN107730054A (en) * | 2017-11-15 | 2018-02-23 | 西南石油大学 | A kind of Gas Load combination forecasting method based on support vector regression |
Non-Patent Citations (1)
| Title |
|---|
| 王勃等: "基于天气分型的风电功率预测方法", 《电网技术》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429180A (en) * | 2020-03-23 | 2020-07-17 | 上海中旖能源科技有限公司 | Liquefied natural gas demand prediction method and device |
| CN111652417A (en) * | 2020-05-25 | 2020-09-11 | 成都千嘉科技有限公司 | Automatic correction method and system for remote meter reading data |
| CN112101612A (en) * | 2020-08-03 | 2020-12-18 | 深圳市燃气集团股份有限公司 | Prediction method of town gas usage, storage medium and terminal equipment |
| CN113326654A (en) * | 2021-05-20 | 2021-08-31 | 北京市燃气集团有限责任公司 | Method and device for constructing gas load prediction model |
| CN113326983A (en) * | 2021-05-28 | 2021-08-31 | 重庆能源大数据中心有限公司 | Natural gas consumption prediction system and method |
| CN113326983B (en) * | 2021-05-28 | 2022-08-26 | 重庆能源大数据中心有限公司 | Data processing system and method for natural gas consumption prediction |
| US11979697B2 (en) | 2021-07-26 | 2024-05-07 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods and internet of things systems for obtaining natural gas energy metering component |
| CN113869588A (en) * | 2021-09-29 | 2021-12-31 | 广州新科佳都科技有限公司 | Energy consumption benchmarking value prediction method, device, equipment and storage medium |
| CN114740159A (en) * | 2022-04-14 | 2022-07-12 | 成都秦川物联网科技股份有限公司 | Natural gas energy metering component acquisition method and Internet of things system |
| CN114740159B (en) * | 2022-04-14 | 2023-09-19 | 成都秦川物联网科技股份有限公司 | Natural gas energy metering component acquisition method and Internet of things system |
| CN118128737A (en) * | 2024-05-07 | 2024-06-04 | 德耐尔能源装备有限公司 | Intelligent exhaust pressure control method and device for protecting unit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109993364A (en) | A kind of prediction technique and device of natural gas gas consumption | |
| Jacoby et al. | Technology and technical change in the MIT EPPA model | |
| Wang et al. | Forecasting the monthly iron ore import of China using a model combining empirical mode decomposition, non-linear autoregressive neural network, and autoregressive integrated moving average | |
| Contreras et al. | ARIMA models to predict next-day electricity prices | |
| CN111210093B (en) | Daily water consumption prediction method based on big data | |
| CN112149879B (en) | New energy medium-and-long-term electric quantity prediction method considering macroscopic volatility classification | |
| CN105930931A (en) | Electric power engineering cost management method | |
| CN104715292A (en) | City short-term water consumption prediction method based on least square support vector machine model | |
| CN107230349B (en) | A kind of online real-time short time traffic flow forecasting method | |
| CN107885951A (en) | A kind of Time series hydrological forecasting method based on built-up pattern | |
| CN105205570A (en) | Power grid power sale quantity prediction method based on season time sequence analysis | |
| CN101825622A (en) | Water quality prediction method and device | |
| CN109636007A (en) | A kind of water demands forecasting method and device based on big data | |
| CN111563662B (en) | Service quality evaluation system and method based on time-window-based deep reinforcement learning | |
| CN116703644A (en) | Attention-RNN-based short-term power load prediction method | |
| Glynn et al. | Economic impacts of future changes in the energy system—global perspectives | |
| CN115994613A (en) | Carbon emission monitoring method based on electric carbon model, terminal and storage medium | |
| CN109376937B (en) | Self-adaptive scheduling end-of-term water level prediction method based on ensemble empirical mode decomposition | |
| CN114239990A (en) | Time series data prediction method based on time series decomposition and LSTM | |
| CN101364286B (en) | Prediction method for equivalent forced outage rate of thermoelectric generating set | |
| CN101930566A (en) | Hydrological experimental simulation system and method based on parallel system | |
| CN107679651A (en) | A kind of moon electricity demand forecasting method based on grey forecasting model Yu BP artificial nerve network models | |
| CN104680400B (en) | The short-term and long-range forecast method of enterprise's electricity sales amount variation based on grey correlation | |
| Hosseini et al. | Day-ahead wind power temporal distribution forecasting with high resolution | |
| CN109389164A (en) | Regional per Unit GDP Energy Consumption prediction technique based on support vector regression model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190709 |
|
| RJ01 | Rejection of invention patent application after publication |