CN111967655A - Short-term load prediction method and system - Google Patents
Short-term load prediction method and system Download PDFInfo
- Publication number
- CN111967655A CN111967655A CN202010738511.9A CN202010738511A CN111967655A CN 111967655 A CN111967655 A CN 111967655A CN 202010738511 A CN202010738511 A CN 202010738511A CN 111967655 A CN111967655 A CN 111967655A
- Authority
- CN
- China
- Prior art keywords
- meteorological data
- historical
- correlation coefficient
- data
- power grid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000010586 diagram Methods 0.000 claims description 10
- 238000010219 correlation analysis Methods 0.000 claims description 9
- 238000010220 Pearson correlation analysis Methods 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 6
- 238000012417 linear regression Methods 0.000 claims description 6
- 230000008030 elimination Effects 0.000 claims description 5
- 238000003379 elimination reaction Methods 0.000 claims description 5
- 230000001131 transforming effect Effects 0.000 claims description 4
- 238000010998 test method Methods 0.000 claims description 3
- 238000013277 forecasting method Methods 0.000 claims 1
- 238000010801 machine learning Methods 0.000 abstract description 5
- 238000012360 testing method Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
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—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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a short-term load prediction method and a system, wherein the method comprises the following steps: calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid; calculating the reliability of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data; calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid; calculating the reliability of a partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as key meteorological data of a target industry; and according to the key meteorological data of the target industry at the prediction time, performing load prediction on the target power grid. The embodiment of the invention effectively improves the machine learning effectiveness, thereby improving the short-term load prediction precision of the target power grid.
Description
Technical Field
The invention relates to the technical field of electronics, in particular to a short-term load prediction method and a short-term load prediction system.
Background
The short-term load prediction is an important component of load prediction, has important significance on optimal combination of a unit, economic dispatching, optimal power flow, electric power market transaction and the like, and is more beneficial to improving the utilization rate of power generation equipment and enhancing the effectiveness of economic dispatching when the load prediction precision is higher.
Machine learning algorithms such as BP neural network and the like are used as common short-term load prediction methods, have strong self-learning capability and complex nonlinear function fitting capability, have self-adaption capability to a large number of non-structural and non-precise rules, and have the characteristics of information memory, autonomous learning, knowledge reasoning and optimized calculation. However, the selection of the input samples plays an important role in whether the training result can reflect the operation rule of the power load, the accuracy of the model is affected if the input quantity is too small, and the convergence speed and the convergence of the training are affected if the input quantity is too large.
In addition, one of the methods commonly used in the current short-term load prediction methods is to improve the prediction accuracy by using meteorological factors. However, the influence of meteorological factors on load prediction is complex: firstly, various meteorological factors such as maximum temperature, minimum temperature, average humidity, average rainfall, average air pressure and the like are used, and all the meteorological factors are used as input data to influence the convergence speed and convergence of training; and secondly, the influence of various meteorological factors on the load prediction effect is different in different seasons and different industries.
Therefore, a method for extracting key meteorological factors is needed to improve the short-term load prediction accuracy.
Disclosure of Invention
In order to solve the above problems, embodiments of the present invention provide a short-term load prediction method and system.
In a first aspect, an embodiment of the present invention provides a short-term load prediction method, including:
acquiring historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
calculating the reliability of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data;
calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
calculating the reliability of a partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the reliability is greater than the second preset threshold as the key meteorological data of the target industry;
and performing load prediction on the target power grid according to the key meteorological data of the target industry at the prediction time.
Preferably, before the obtaining the historical load data and the several historical meteorological data of the target power grid in the target industry, the method further includes:
acquiring initial historical load data and a plurality of initial historical meteorological data of the target power grid in the target industry;
performing missing value completion, abnormal value correction or elimination on the initial historical load data of the target power grid and a plurality of initial historical meteorological data;
and taking the initial historical load data after pretreatment as the historical load data of the target power grid, and taking the initial historical meteorological data after pretreatment as the historical meteorological data.
Preferably, the calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and obtaining the correlation coefficient corresponding to each historical meteorological data specifically includes:
and calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid through a Pearson correlation analysis method, and obtaining the correlation coefficient corresponding to each historical meteorological data.
Preferably, the reliability of the correlation coefficient corresponding to each historical meteorological data and the reliability of the partial correlation coefficient corresponding to each alternative meteorological data are both obtained by a P-value test method.
Preferably, the calculating a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid, and obtaining the partial correlation coefficient corresponding to each candidate meteorological data specifically includes:
and calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid by adopting a partial correlation analysis method, and obtaining the partial correlation coefficient corresponding to each alternative meteorological data.
Preferably, the calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid by using a Pearson correlation analysis method specifically includes:
for any historical meteorological data, calculating a correlation coefficient between the historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to the historical meteorological data;
if the correlation coefficient corresponding to any historical meteorological data is smaller than the first preset threshold, acquiring a target nonlinear function according to a scatter diagram of sample data of the correlation coefficient corresponding to any historical meteorological data and the historical load data of the target power grid;
and transforming the sample data by using the target nonlinear function, and calculating a Pearson correlation coefficient of the transformed sample data.
Preferably, the calculating a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid by using a partial correlation analysis method specifically includes:
and calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid through a linear regression method or a correlation matrix inversion method.
In a second aspect, an embodiment of the present invention provides a short-term load prediction system, including:
the data module is used for acquiring historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
the correlation coefficient module is used for calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
the alternative module is used for calculating the credibility of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the credibility is greater than a second preset threshold as alternative meteorological data;
the partial correlation coefficient module is used for calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
the key module is used for calculating the credibility of the partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the credibility is greater than the second preset threshold as the key meteorological data of the target industry;
and the prediction module is used for predicting the load of the target power grid according to the key meteorological data of the target industry at the prediction time.
In a third aspect, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the short-term load prediction method according to the first aspect of the present invention.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the short-term load prediction method provided in the first aspect of the present invention.
According to the short-term load prediction method and system provided by the embodiment of the invention, the main factors influencing the target power grid load are effectively selected by calculating the correlation coefficient between each type of historical meteorological data and historical load data, so that the problem of overlarge data size caused by a large number of data dimensions is effectively reduced; by the aid of the partial correlation coefficient, key influence factors of the target power grid load are accurately obtained, machine learning effectiveness is effectively improved, and therefore short-term load prediction accuracy of the target power grid is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a flowchart of a short-term load prediction method according to an embodiment of the present invention;
FIG. 2 is a block diagram of a short term load prediction method according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of a short term load prediction system according to an embodiment of the present invention;
fig. 4 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of a short-term load prediction method according to an embodiment of the present invention, and as shown in fig. 1, the embodiment of the present invention provides a short-term load prediction method, including:
short-term load prediction is a load prediction method frequently used in a power grid, and the short term generally refers to 1 day or several hours, and also includes ultra-short term load prediction and long-term load prediction.
S1, acquiring historical load data and a plurality of historical meteorological data of a target power grid in the target industry;
when short-term load prediction is carried out on a target power grid in a target industry, firstly, key meteorological data influencing the target industry need to be selected, and the key meteorological data corresponding to different industries are different.
The method comprises the steps of firstly, selecting historical load data and a plurality of historical meteorological data of a target power grid in a target industry, wherein the historical load data of the target power grid can be expressed as Y ═ Y1,y2,…,ypThe historical meteorological data comprises one or more of meteorological data such as highest temperature, lowest temperature, average humidity, average rainfall, average air pressure and the like of historical days, and can be expressed as X ═ X1,x2,…,xq}。
S2, calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
and then calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, wherein the correlation coefficient can represent the correlation degree between each historical meteorological data and the historical load data of the target power grid, the larger the correlation coefficient is, the larger the relationship degree between the historical meteorological data and the target power grid is, the smaller the correlation coefficient is, the smaller the relationship degree between the historical meteorological data and the target power grid is, and obtaining the correlation coefficient corresponding to each historical meteorological data.
S3, calculating the credibility of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the credibility is greater than a second preset threshold as alternative meteorological data;
and then calculating the reliability of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation number is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data.
Generally, reliability calculation and correlation calculation are usually used in one block, and the obtained candidate meteorological data is data obtained by primary screening, and is denoted as B ═ B1,b2,…,bs}。
S4, calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
and then calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid to obtain a partial correlation coefficient corresponding to each alternative meteorological data.
Because the key meteorological data is screened through the correlation coefficient, secondary screening is carried out by combining the partial correlation coefficient, and the screening reliability is further ensured through the combination of the two methods.
S5, calculating the credibility of the partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the credibility is greater than the second preset threshold as the key meteorological data of the target industry;
and calculating the reliability of the partial correlation coefficient corresponding to each alternative meteorological data again, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the reliability is greater than the second preset threshold as the key meteorological data of the target industry, wherein the key meteorological data is the key influence factor influencing the target power grid load in the target industry.
And S6, performing load prediction on the target power grid according to the key meteorological data of the target industry at the prediction time.
And finally, according to key meteorological data of the target industry at the prediction time, a solid foundation is laid for short-term load prediction of the target power grid by using the BP neural network, and the prediction precision is favorably improved.
According to the short-term load prediction method provided by the embodiment of the invention, the main factors influencing the target power grid load are effectively selected by calculating the correlation coefficient between each type of historical meteorological data and historical load data, so that the problem of overlarge data size caused by a large number of data dimensions is effectively reduced; by the aid of the partial correlation coefficient, key influence factors of the target power grid load are accurately obtained, machine learning effectiveness is effectively improved, and therefore short-term load prediction accuracy of the target power grid is improved.
On the basis of the foregoing embodiment, preferably, before the acquiring the historical load data and the several historical meteorological data of the target power grid in the target industry, the method further includes:
acquiring initial historical load data and a plurality of initial historical meteorological data of the target power grid in the target industry;
performing missing value completion, abnormal value correction or elimination on the initial historical load data of the target power grid and a plurality of initial historical meteorological data;
and taking the initial historical load data after pretreatment as the historical load data of the target power grid, and taking the initial historical meteorological data after pretreatment as the historical meteorological data.
Specifically, the method further comprises the following steps before calculating the historical load data and the plurality of historical meteorological data of the target power grid in the target industry:
the method comprises the steps of firstly, acquiring initial historical load data and initial historical meteorological data of a target power grid in a target industry, and carrying out data preprocessing such as missing value completion, abnormal value correction or elimination on the collected initial historical load data, wherein the specific preprocessing comprises the following two aspects:
(1) for the case of single or continuous multiple missing of load data: the single data loss is replaced by the average value of the previous moment and the next moment; if the number of the continuous missing values does not exceed the limited range, linear interpolation completion is adopted, otherwise, the load mean values of the same time of the same type of days before and after are adopted for substitution.
(2) And (3) carrying out abnormal value test on the meteorological data: and setting a reasonable boundary value for each meteorological factor based on multi-year meteorological data experience, and if an abnormal value exceeding the boundary value occurs, adopting an adjacent date mean value or deleting the date sample point.
On the basis of the foregoing embodiment, preferably, the calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and obtaining the correlation coefficient corresponding to each historical meteorological data specifically includes:
and calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid through a Pearson correlation analysis method, and obtaining the correlation coefficient corresponding to each historical meteorological data.
Specifically, the correlation coefficient between each historical meteorological data and the historical load data of the target power grid is calculated by a Pearson correlation analysis method.
On the basis of the foregoing embodiment, preferably, the calculating, by using a Pearson correlation analysis method, a correlation coefficient between each historical meteorological data and the historical load data of the target power grid specifically includes:
for any historical meteorological data, calculating a correlation coefficient between the historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to the historical meteorological data;
if the correlation coefficient corresponding to any historical meteorological data is smaller than the first preset threshold, acquiring a target nonlinear function according to a scatter diagram of sample data of the correlation coefficient corresponding to any historical meteorological data and the historical load data of the target power grid;
and transforming the sample data by using the target nonlinear function, and calculating a Pearson correlation coefficient of the transformed sample data.
Specifically, the Pearson correlation analysis method specifically includes: recording a certain historical meteorological data as variable xjThe historical load data of the target industry is yhThe Pearson correlation coefficient is used to describe the degree of linear correlation between two variables:
wherein r is a variable xj、yhThe correlation coefficient of (a); n is the number of sample points contained in the variable;
are respectively a variable xj、yhThe sample mean of (2);
are respectively a variable xj、yhThe standard deviation of (a) can be specifically expressed as:
wherein r > 0 represents the variable xj、yhIs in positive correlation, r < 0 represents variable xj、yhAnd presents negative correlation. The larger the absolute value of the correlation coefficient is, the stronger the correlation is; the closer the correlation coefficient is to 1 or-1, the stronger the correlation, the closer the correlation coefficient is to 0, and the weaker the correlation. When r < 0.4, the variable x is indicatedj、yhAnd weak or no correlation, so that a variable with a load correlation coefficient of more than 0.4 is selected as a key influencing factor.
Since the Pearson correlation verification method can only judge the strength of the linear relationship and has no way to judge the nonlinear relationship between the variables, the embodiment of the present invention further needs to judge the strength of the nonlinear relationship between the two variables.
If the Pearson correlation coefficient between the two variables is less than 0.4, it needs to be further determined whether the nonlinear relationship exists: a sample data scatter diagram of the two variables is made, and the possible types of the nonlinear function are judged through the scatter diagram.
And transforming the sample data based on the possible functional relation, and calculating a Pearson correlation coefficient of the transformed data.
If the correlation coefficient is greater than 0.4 after the conversion of the steps, the variable x is considered to bej、yhThe correlation of (A) and (B) is strong, whereas the correlation of (A) and (B) is weak or no correlation.
On the basis of the foregoing embodiment, preferably, the calculating a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid, and obtaining the partial correlation coefficient corresponding to each candidate meteorological data specifically includes:
and calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid by adopting a partial correlation analysis method, and obtaining the partial correlation coefficient corresponding to each alternative meteorological data.
Specifically, calculating a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid is obtained through calculation of a partial correlation analysis method.
The partial correlation analysis method can calculate the partial correlation coefficient between the alternative meteorological data and the historical load data of the target power grid through a linear regression method and a correlation matrix inversion method.
Linear regression method: assuming that the correlation between X and Y needs to be calculated, Z represents all other variables, and the partial correlation coefficient of X and Y can be regarded as the residual R obtained by linear regression of X and ZxResidual R obtained by linear regression with Y, ZyThe correlation coefficient between them, i.e. Pearson correlation coefficient.
The correlation matrix inversion method firstly obtains a correlation matrix:
then, the covariance matrix is obtained for the correlation matrix, and the inverse matrix is obtained:
where X is the covariance matrix of all variables and r is the inverse of the covariance matrix.
The partial correlation matrix is calculated as:
on the basis of the above embodiment, preferably, the reliability of the correlation coefficient corresponding to each historical meteorological data and the reliability of the partial correlation coefficient corresponding to each candidate meteorological data are both obtained by a P-value test method.
The specific method for checking the P value is as follows: z represents the statistical quantity of the test, and ZC represents the statistical quantity of the test calculated according to the sample data.
(1) And (4) left side inspection: h0:μ≥μ0，H1:μ＜μ0。
The value of P is when mu equals mu0When the test statistic is less than or equal to the probability of the test statistic calculated from the actual observation sample data, that is, the value P is P (Z < ZC | μ ═ μ -0)。
(2) And (3) right-side checking: h0:μ≤μ0，H1:μ＞μ0。
The value of P is when mu equals mu0When the test statistic is greater than or equal to the probability of the test statistic calculated from the actual observation sample data, that is, the value P is P (Z < ZC | mu ═ mu)0)。
(3) And (3) double-side inspection: h0:μ＝μ0
The value of P is when mu equals mu0When the test statistic is greater than or equal to the probability of the test statistic calculated according to the actual observation sample data, namely the value P is 2P (Z is more than or equal to ZC | | | mu ═ mu ≧0)。
Fig. 2 is a flowchart of a short-term load prediction method according to another embodiment of the present invention, and as shown in fig. 2, historical load data and a plurality of historical meteorological data of a target industry are first obtained, then data preprocessing such as missing value completion, abnormal value correction or elimination is performed, a Pearson correlation coefficient between each historical meteorological data and the historical load data is calculated, the reliability of the correlation coefficient corresponding to each historical meteorological data is calculated, and the historical meteorological data of which the correlation coefficient is greater than 0.4 and the reliability is greater than 0.05 is used as candidate meteorological data.
And then calculating a partial correlation coefficient between each alternative meteorological data and the historical load data, calculating the reliability of the partial correlation coefficient corresponding to each alternative meteorological data, taking the alternative meteorological data with the partial correlation coefficient being more than 0.4 and the reliability being more than 0.05 as key meteorological data, and performing short-term load prediction on a target power grid in the target industry according to the selected key meteorological data.
In conclusion, the embodiment of the invention effectively solves the common data problems such as incomplete historical load data and the like through data preprocessing means such as missing value completion, abnormal value correction and the like; main factors influencing the power load are effectively selected through correlation analysis, and the problem of overlarge data volume caused by a large number of data dimensions is effectively reduced; by means of partial correlation analysis, key influence factors of the power load are accurately obtained, machine learning effectiveness is effectively improved, and therefore short-term load prediction accuracy is improved.
Fig. 3 is a schematic structural diagram of a short-term load prediction system according to an embodiment of the present invention, as shown in fig. 3, the short-term load prediction system includes: a data module 301, a correlation coefficient module 302, an alternative module 303, a partial correlation coefficient module 304, a key module 305, and a prediction module 306, wherein:
the data module 301 is configured to obtain historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
the correlation coefficient module 302 is configured to calculate a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and obtain a correlation coefficient corresponding to each historical meteorological data;
the alternative module 303 is configured to calculate a reliability of a correlation coefficient corresponding to each historical meteorological data, and use historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data;
the partial correlation coefficient module 304 is configured to calculate a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid, and obtain a partial correlation coefficient corresponding to each candidate meteorological data;
the key module 305 is configured to calculate a reliability of a partial correlation coefficient corresponding to each candidate meteorological data, and use the candidate meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the reliability is greater than the second preset threshold as the key meteorological data of the target industry;
the prediction module 306 is configured to perform load prediction on the target power grid according to the key meteorological data of the target industry at the prediction time.
The device embodiment provided in the embodiments of the present invention is for implementing the above method embodiments, and for details of the process and the details, reference is made to the above method embodiments, which are not described herein again.
Fig. 4 is a schematic entity structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 4, the electronic device may include: a processor (processor)401, a communication Interface (communication Interface)402, a memory (memory)403 and a bus 404, wherein the processor 401, the communication Interface 402 and the memory 403 complete communication with each other through the bus 404. The communication interface 402 may be used for information transfer of an electronic device. Processor 401 may call logic instructions in memory 403 to perform a method comprising:
acquiring historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
calculating the reliability of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data;
calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
calculating the reliability of a partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the reliability is greater than the second preset threshold as the key meteorological data of the target industry;
and performing load prediction on the target power grid according to the key meteorological data of the target industry at the prediction time.
In addition, the logic instructions in the memory 403 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the above-described method embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented to perform the transmission method provided in the foregoing embodiments when executed by a processor, and for example, the method includes:
acquiring historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
calculating the reliability of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data;
calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
calculating the reliability of a partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the reliability is greater than the second preset threshold as the key meteorological data of the target industry;
and performing load prediction on the target power grid according to the key meteorological data of the target industry at the prediction time.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for short-term load prediction, comprising:
acquiring historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
calculating the reliability of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the reliability is greater than a second preset threshold as alternative meteorological data;
calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
calculating the reliability of a partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the reliability is greater than the second preset threshold as the key meteorological data of the target industry;
and performing load prediction on the target power grid according to the key meteorological data of the target industry at the prediction time.
2. The short term load forecasting method of claim 1, wherein the obtaining historical load data and a plurality of historical meteorological data of a target power grid in a target industry further comprises:
acquiring initial historical load data and a plurality of initial historical meteorological data of the target power grid in the target industry;
performing missing value completion, abnormal value correction or elimination on the initial historical load data of the target power grid and a plurality of initial historical meteorological data;
and taking the initial historical load data after pretreatment as the historical load data of the target power grid, and taking the initial historical meteorological data after pretreatment as the historical meteorological data.
3. The short-term load prediction method according to claim 1, wherein the calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and obtaining a correlation coefficient corresponding to each historical meteorological data specifically includes:
and calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid through a Pearson correlation analysis method, and obtaining the correlation coefficient corresponding to each historical meteorological data.
4. The short-term load prediction method according to claim 1, wherein the confidence level of the correlation coefficient corresponding to each historical meteorological data and the confidence level of the partial correlation coefficient corresponding to each candidate meteorological data are obtained by a P-value test method.
5. The short-term load prediction method according to claim 1, wherein the calculating a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid, and obtaining the partial correlation coefficient corresponding to each candidate meteorological data specifically comprises:
and calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid by adopting a partial correlation analysis method, and obtaining the partial correlation coefficient corresponding to each alternative meteorological data.
6. The short-term load prediction method according to claim 3, wherein the calculating of the correlation coefficient between each historical meteorological data and the historical load data of the target power grid by the Pearson correlation analysis method specifically comprises:
for any historical meteorological data, calculating a correlation coefficient between the historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to the historical meteorological data;
if the correlation coefficient corresponding to any historical meteorological data is smaller than the first preset threshold, acquiring a target nonlinear function according to a scatter diagram of sample data of the correlation coefficient corresponding to any historical meteorological data and the historical load data of the target power grid;
and transforming the sample data by using the target nonlinear function, and calculating a Pearson correlation coefficient of the transformed sample data.
7. The short-term load prediction method according to claim 5, wherein the calculating a partial correlation coefficient between each candidate meteorological data and the historical load data of the target power grid by using a partial correlation analysis method specifically comprises:
and calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid through a linear regression method or a correlation matrix inversion method.
8. A short term load prediction system, comprising:
the data module is used for acquiring historical load data and a plurality of historical meteorological data of a target power grid in a target industry;
the correlation coefficient module is used for calculating a correlation coefficient between each historical meteorological data and the historical load data of the target power grid, and acquiring a correlation coefficient corresponding to each historical meteorological data;
the alternative module is used for calculating the credibility of the correlation coefficient corresponding to each historical meteorological data, and taking the historical meteorological data of which the correlation coefficient is greater than a first preset threshold and the credibility is greater than a second preset threshold as alternative meteorological data;
the partial correlation coefficient module is used for calculating a partial correlation coefficient between each alternative meteorological data and the historical load data of the target power grid, and acquiring a partial correlation coefficient corresponding to each alternative meteorological data;
the key module is used for calculating the credibility of the partial correlation coefficient corresponding to each alternative meteorological data, and taking the alternative meteorological data of which the partial correlation coefficient is greater than the first preset threshold and the credibility is greater than the second preset threshold as the key meteorological data of the target industry;
and the prediction module is used for predicting the load of the target power grid according to the key meteorological data of the target industry at the prediction time.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the short term load prediction method as claimed in any one of claims 1 to 7.
10. A non-transitory computer readable storage medium, having stored thereon a computer program, which, when being executed by a processor, carries out the steps of the short term load prediction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010738511.9A CN111967655A (en) | 2020-07-28 | 2020-07-28 | Short-term load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010738511.9A CN111967655A (en) | 2020-07-28 | 2020-07-28 | Short-term load prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111967655A true CN111967655A (en) | 2020-11-20 |
Family
ID=73363103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010738511.9A Pending CN111967655A (en) | 2020-07-28 | 2020-07-28 | Short-term load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111967655A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112598181A (en) * | 2020-12-23 | 2021-04-02 | 国网河北省电力有限公司经济技术研究院 | Load prediction method, device, equipment and storage medium |
| CN112614010A (en) * | 2020-12-07 | 2021-04-06 | 国网北京市电力公司 | Load prediction method and device, storage medium and electronic device |
| CN112686463A (en) * | 2021-01-07 | 2021-04-20 | 合肥阳光新能源科技有限公司 | Demand data processing method and device and electronic equipment |
| CN116933040A (en) * | 2023-09-13 | 2023-10-24 | 国网四川省电力公司广安供电公司 | Big data technology-based power supply highest load medium-short term prediction method |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001227792A (en) * | 2000-02-16 | 2001-08-24 | Daikin Ind Ltd | Method of estimating air-conditioning load, and its device |
| JP2012244897A (en) * | 2011-05-13 | 2012-12-10 | Fujitsu Ltd | Apparatus and method for predicting short-term power load |
| CN102930353A (en) * | 2012-11-06 | 2013-02-13 | 北京国电通网络技术有限公司 | Method and device for predicating electrical loads of residential area |
| US20130110756A1 (en) * | 2011-10-31 | 2013-05-02 | Siemens Corporation | Short-term Load Forecast Using Support Vector Regression and Feature Learning |
| US20160246272A1 (en) * | 2014-08-22 | 2016-08-25 | Siemens Corporation | Weather pattern based electrical demand forecasting for a building |
| CN106408223A (en) * | 2016-11-30 | 2017-02-15 | 华北电力大学（保定） | Short-term load prediction based on meteorological similar day and error correction |
| CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学（保定） | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
| KR20190023903A (en) * | 2017-08-30 | 2019-03-08 | 사단법인 한국벰스협회 | Building energy management system and method based on multi dimension building energy load prediction per unit using various factors |
| CN109948281A (en) * | 2019-03-29 | 2019-06-28 | 上海电力学院 | It is effectively identified based on weather pattern and the straight of combined prediction dissipates separated modeling method |
| CN110598937A (en) * | 2019-09-18 | 2019-12-20 | 柳州市工人医院 | CO poisoning prediction method based on meteorological data |
| CN110766291A (en) * | 2019-09-29 | 2020-02-07 | 西安建筑科技大学 | Horizontal plane daily total radiation data acquisition method based on solar radiation subareas |
-
2020
- 2020-07-28 CN CN202010738511.9A patent/CN111967655A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001227792A (en) * | 2000-02-16 | 2001-08-24 | Daikin Ind Ltd | Method of estimating air-conditioning load, and its device |
| JP2012244897A (en) * | 2011-05-13 | 2012-12-10 | Fujitsu Ltd | Apparatus and method for predicting short-term power load |
| US20130110756A1 (en) * | 2011-10-31 | 2013-05-02 | Siemens Corporation | Short-term Load Forecast Using Support Vector Regression and Feature Learning |
| CN102930353A (en) * | 2012-11-06 | 2013-02-13 | 北京国电通网络技术有限公司 | Method and device for predicating electrical loads of residential area |
| US20160246272A1 (en) * | 2014-08-22 | 2016-08-25 | Siemens Corporation | Weather pattern based electrical demand forecasting for a building |
| CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学（保定） | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
| CN106408223A (en) * | 2016-11-30 | 2017-02-15 | 华北电力大学（保定） | Short-term load prediction based on meteorological similar day and error correction |
| KR20190023903A (en) * | 2017-08-30 | 2019-03-08 | 사단법인 한국벰스협회 | Building energy management system and method based on multi dimension building energy load prediction per unit using various factors |
| CN109948281A (en) * | 2019-03-29 | 2019-06-28 | 上海电力学院 | It is effectively identified based on weather pattern and the straight of combined prediction dissipates separated modeling method |
| CN110598937A (en) * | 2019-09-18 | 2019-12-20 | 柳州市工人医院 | CO poisoning prediction method based on meteorological data |
| CN110766291A (en) * | 2019-09-29 | 2020-02-07 | 西安建筑科技大学 | Horizontal plane daily total radiation data acquisition method based on solar radiation subareas |
Non-Patent Citations (3)
| Title |
|---|
| 周江;: "南方电网中气象性电力负荷的特性研究――以浙江省金华市为例", 中国集体经济, no. 28, pages 92 - 93 * |
| 阙雷挺;苏亮;周灵江;蔡柯兴;阙亚妮;: "基于多气象指标的地区短期负荷预测研究", 云南电力技术, no. 01, pages 32 - 35 * |
| 黎洁仪;梁之彦;杨国杰;: "广州市空气污染影响因子与预报建模", 广东气象, no. 04, pages 49 - 52 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112614010A (en) * | 2020-12-07 | 2021-04-06 | 国网北京市电力公司 | Load prediction method and device, storage medium and electronic device |
| CN112598181A (en) * | 2020-12-23 | 2021-04-02 | 国网河北省电力有限公司经济技术研究院 | Load prediction method, device, equipment and storage medium |
| CN112686463A (en) * | 2021-01-07 | 2021-04-20 | 合肥阳光新能源科技有限公司 | Demand data processing method and device and electronic equipment |
| CN116933040A (en) * | 2023-09-13 | 2023-10-24 | 国网四川省电力公司广安供电公司 | Big data technology-based power supply highest load medium-short term prediction method |
| CN116933040B (en) * | 2023-09-13 | 2023-11-28 | 国网四川省电力公司广安供电公司 | Big data technology-based power supply highest load medium-short term prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111967655A (en) | Short-term load prediction method and system | |
| CN105989441A (en) | Model parameter adjustment method and device | |
| CN111324862A (en) | Method and system for monitoring behavior in loan | |
| CN111625516A (en) | Method and device for detecting data state, computer equipment and storage medium | |
| CN110942248A (en) | Training method and device for transaction wind control network and transaction risk detection method | |
| CN112561320A (en) | Training method of mechanism risk prediction model, mechanism risk prediction method and device | |
| CN113822488A (en) | Risk prediction method and device for financing lease, computer equipment and storage medium | |
| CN110020939B (en) | Device, method and storage medium for establishing default loss rate prediction model | |
| CN116362894A (en) | Multi-objective learning method, multi-objective learning device, electronic equipment and computer readable storage medium | |
| CN114066081B (en) | Enterprise risk prediction method and device based on graph attention network and electronic equipment | |
| CN115600818A (en) | Multi-dimensional scoring method and device, electronic equipment and storage medium | |
| CN114399289A (en) | Work order supervision time setting method and device | |
| CN114723239A (en) | Multi-party collaborative modeling method, device, equipment, medium and program product | |
| CN111190800B (en) | Method, system, device and storage medium for predicting batch operation duration of host | |
| CN113361625A (en) | Error data detection method with privacy protection in federated learning scene | |
| CN112529303A (en) | Risk prediction method, device, equipment and storage medium based on fuzzy decision | |
| CN111507578A (en) | Risk assessment method and related device and equipment | |
| VI et al. | Building a neural network model for diagnosing the probability of bankruptcy of innovative-active enterprises and checking its adequacy | |
| CN114581249B (en) | Financial product recommendation method and system based on investment risk bearing capacity assessment | |
| CN111815442B (en) | Link prediction method and device and electronic equipment | |
| CN117611229A (en) | Customer loss type prediction method and device, electronic equipment and storage medium | |
| CN116012634A (en) | Strip steel shape classification method, device, computer medium and equipment | |
| CN117556935A (en) | Time sequence prediction method, device, equipment and readable storage medium | |
| CN115659599A (en) | Data processing method and device | |
| CN113837815A (en) | Quotation prediction model construction method, prediction method, system, device and medium |
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 |