CN113592192A - Short-term power load prediction method and device and terminal equipment - Google Patents
Short-term power load prediction method and device and terminal equipment Download PDFInfo
- Publication number
- CN113592192A CN113592192A CN202110944859.8A CN202110944859A CN113592192A CN 113592192 A CN113592192 A CN 113592192A CN 202110944859 A CN202110944859 A CN 202110944859A CN 113592192 A CN113592192 A CN 113592192A
- Authority
- CN
- China
- Prior art keywords
- data set
- historical
- attribute
- load
- load data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 239000011159 matrix material Substances 0.000 claims abstract description 64
- 238000012549 training Methods 0.000 claims abstract description 20
- 238000012216 screening Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 35
- 238000004590 computer program Methods 0.000 claims description 19
- 230000001186 cumulative effect Effects 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 10
- 238000012417 linear regression Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims description 4
- 230000008030 elimination Effects 0.000 claims description 4
- 238000003379 elimination reaction Methods 0.000 claims description 4
- 238000013277 forecasting method Methods 0.000 claims 4
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000611 regression analysis 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention is suitable for the technical field of power system load prediction, and provides a short-term power load prediction method, a device and terminal equipment, wherein the method comprises the following steps: the method comprises the steps of obtaining a historical load data set and a corresponding historical attribute data set, calculating the correlation degree between the load data and each attribute data in the historical attribute data set aiming at each load data in the historical load data set, obtaining a correlation matrix, calculating a first weighted rank comparison estimation value corresponding to each attribute according to the correlation matrix, determining the attribute with the first weighted rank comparison estimation value being larger than a preset threshold value as a key attribute, screening the historical attribute data set by using the key attribute to obtain a key attribute data set, carrying out load prediction model training according to the historical load data set and the corresponding key attribute data set, and carrying out short-term power load prediction on a power system according to the trained load prediction model. The complexity and the prediction precision of the short-term power load prediction can be considered.
Description
Technical Field
The invention belongs to the technical field of power system load prediction, and particularly relates to a short-term power load prediction method, a short-term power load prediction device and terminal equipment.
Background
The short-term power load prediction mainly refers to prediction of power load for hours, days or more in the future. The power load prediction is an important component of an energy management system, and the short-term load prediction not only provides guarantee for the safe and economic operation of the power system, but also is a basis for arranging a scheduling plan, a power supply plan and a trading plan in a market environment. The magnitude of the short-term load prediction effect mainly depends on the prediction accuracy, so how to improve the prediction accuracy is the key point of researching short-term load prediction theories and methods at present.
The current common short-term load prediction methods comprise a regression analysis method, a similar day method, an intelligent prediction method and the like. However, since the short-term load prediction is greatly influenced by factors such as weather, equipment conditions, and significant social activities, if all of these factors are used as input characteristics, the complexity of the load prediction will increase. If only factors of a certain aspect are considered, the prediction accuracy of short-term load prediction based on a data-driven method is greatly reduced due to the fact that the basic data scene of each factor is diverse and the regularity is weak.
Disclosure of Invention
In view of this, embodiments of the present invention provide a short-term power load prediction method, an apparatus, and a terminal device, and aim to solve the problem that complexity and prediction accuracy of a short-term power load prediction method in the prior art cannot be considered at the same time.
To achieve the above object, a first aspect of an embodiment of the present invention provides a short-term power load prediction method, including:
acquiring a historical load data set and a corresponding historical attribute data set;
calculating the correlation degree of each load data in the historical load data set and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation degree matrix;
calculating a first weighted rank combination ratio estimation value corresponding to each attribute according to the correlation matrix;
determining the attribute of which the first weighted rank combination ratio estimated value is greater than a preset threshold value as a key attribute;
screening the historical attribute data set by using the key attribute to obtain a key attribute data set;
and carrying out load prediction model training according to the historical load data set and the corresponding key attribute data set, and carrying out short-term power load prediction on the power system according to the trained load prediction model.
A second aspect of an embodiment of the present invention provides a short-term power load prediction apparatus, including:
the acquisition module is used for acquiring a historical load data set and a corresponding historical attribute data set;
the first processing module is used for calculating the correlation degree of each load data in the historical load data set and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation degree matrix;
the second processing module is used for calculating a first weighted rank comparison estimation value corresponding to each attribute according to the correlation matrix;
the third processing module is used for determining the attribute of which the first weighted rank-to-total ratio estimation value is greater than a preset threshold as a key attribute;
the screening module is used for screening the historical attribute data set by using the key attribute to obtain a key attribute data set;
and the training prediction module is used for carrying out load prediction model training according to the historical load data set and the corresponding key attribute data set and carrying out short-term power load prediction on the power system according to the trained load prediction model.
A third aspect of an embodiment of the present invention provides a terminal device, including: a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the short term power load prediction method as described in the first aspect above when executing the computer program.
A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, including: the computer readable storage medium stores a computer program which, when executed by a processor, implements the steps of the short term power load prediction method as described in the above first aspect.
Compared with the prior art, the embodiment of the invention has the following beneficial effects: compared with the prior art, the method comprises the steps of obtaining a historical load data set and a corresponding historical attribute data set, calculating the correlation between the load data and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation matrix, calculating a first weighted rank comparison estimation value corresponding to each attribute according to the correlation matrix, determining the attribute with the first weighted rank comparison estimation value being larger than a preset threshold value as a key attribute, screening the historical attribute data set by using the key attribute to obtain a key attribute data set, carrying out load prediction model training according to the historical load data set and the corresponding key attribute data set, and carrying out short-term power load prediction on a power system according to the trained load prediction model. The method can reduce the input characteristics of load prediction, and simultaneously reserve the key attribute with higher value of load prediction, thereby taking the complexity and prediction precision of short-term power load prediction into account.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a schematic flow chart illustrating an implementation of a short-term power load prediction method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart illustrating an implementation of a short-term power load prediction method according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of a short-term power load forecasting arrangement provided by an embodiment of the present invention;
fig. 4 is a schematic diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
In order to explain the technical means of the present invention, the following description will be given by way of specific examples.
Fig. 1 is a schematic flow chart of an implementation of the short-term power load prediction method according to the embodiment of the present invention, which is detailed as follows:
For the time length of the short-term power load prediction required, past historical load data and corresponding historical attribute data can be obtained at intervals of hours or days, and then a historical load data set and a corresponding historical attribute data set are obtained.
Alternatively, in order to improve the prediction accuracy of the short-term power load prediction, past historical load data and corresponding historical attribute data may be obtained at hourly intervals, and historical load data for one or several days may be used as one load data, and historical attribute data for one or several days may be used as one historical attribute data.
The attributes included in the historical attribute data set may be weather information, season information, date information, policy information, equipment information, emergency information, and the like. The embodiment of the invention does not limit the types of the attributes contained in the historical attribute data set, and specifically can select the regions where the short-term power load is predicted according to the needs.
Optionally, before acquiring the historical load data set and the corresponding historical attribute data set, the method may further include: acquiring an initial historical load data set and a corresponding initial historical attribute data set; performing missing value completion, abnormal value correction or elimination processing on the initial historical load data set and the initial historical attribute data set; and taking the processed initial historical load data set as a historical load data set, and taking the processed initial historical attribute data set as a historical attribute data set.
When missing value completion processing is performed on the initial historical load data set, if a single data missing occurs in the initial historical load data set, the average value of the previous time or the next time can be used for substitution. If a plurality of continuous data are missing in the initial historical load data set, if the number of continuous missing values does not exceed a limited range, linear difference value completion can be adopted; if the number of the continuous missing values exceeds the limited range, the load mean values of the same time of the same type of days before and after can be adopted for substitution. When the abnormal value correction processing is performed on the initial historical load data set, a method similar to the missing value completion processing may be employed.
For the initial historical attribute data set, a reasonable boundary value can be set according to experience, and if an abnormal value is detected, the average value of adjacent dates can be adopted or the sample point of the date can be directly removed.
For example, for the weather information data in the initial historical attribute data set, a reasonable boundary value may be set for each weather factor based on the experience of the weather information data, and if an abnormal value exceeding the boundary value or not reaching the boundary value occurs, the adjacent date mean value may be used or the date sample point may be directly rejected.
In the embodiment, missing value completion, abnormal value correction or elimination processing is performed on the initial historical load data set and the initial historical attribute data set; and taking the processed initial historical load data set as a historical load data set, and taking the processed initial historical attribute data set as a historical attribute data set. The accuracy of the data in the historical load data set and the historical attribute data set can be improved.
And 102, calculating the correlation between the load data and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation matrix.
Optionally, calculating the correlation between the load data and each attribute data in the historical attribute data set may include:
according to
The relevance of the load data to each attribute data in the historical attribute data set is calculated.
Wherein, I (x)i,xo) As load data xoRelevance to ith attribute data in the historical attribute data set, p (x)ik,xok) As a function of probability density, p (x)ik) For the marginal distribution of the kth value in the ith attribute data in the historical attribute data set, p (x)ok) As load data xoThe marginal distribution of the kth value, where k is 1,2, … n.
Wherein, since the historical load data of one or several days can be used as a load data, and the historical attribute data corresponding to one or several days can be used as a historical attribute data, a load data can be represented as xo＝[xo1,…,xok,…,xon]A historical attribute data may be represented as xi＝[xi1,…,xik,…,xin]。
The embodiment takes mutual information of a certain load data and the ith attribute data in the historical attribute data set as the correlation degree of the load data and the ith attribute data in the historical attribute data set. Since the mutual information represents the reduction of the uncertainty of the original random variable given another random variable, and the mutual information is used as the degree of correlation, the influence of certain attribute data on the load data can be more accurately expressed.
And 103, calculating a first weighted rank combination ratio estimation value corresponding to each attribute according to the correlation matrix.
Optionally, referring to fig. 2, calculating a first weighted rank-to-total ratio estimation value corresponding to each attribute according to the correlation matrix may include:
Optionally, before determining the weight corresponding to each load data according to the distribution of the correlation matrix, normalization processing may be performed on the correlation matrix, and all the correlation data are mapped between [0,1] so as to facilitate subsequent processing.
The formula for obtaining the normalized data corresponding to each correlation data may be:
wherein, I'ijNormalized data for the correlation of the jth load data with the ith attribute data, IijIs the correlation of the jth load data and the ith attribute data, Ij-minIs the minimum value of the correlation of the jth load data in the correlation matrix and each attribute, Ij-maxThe maximum value of the correlation degree of the jth load data and each attribute in the correlation degree matrix is shown, wherein i is 1,2, … m, j is 1,2, … q, m is the number of the types of the attributes in the historical attribute data set, and q is the number of the load data in the historical load data set.
Wherein, after normalization processing is performed on the correlation matrix, the correlation matrix can be obtained according to
Calculating a weight of a degree of correlation of various attributes with each load data, wherein yijIs the weight of the correlation degree of the ith attribute and the jth load data.
Based on a weight of a degree of correlation of various attributes with each load data
The information utility value of each load data can be calculated, wherein djInformation utility value of jth load data, ejThe information entropy of the jth payload data is K1/lnm.
According to the information utility value of each load data, based on
The weight w corresponding to the jth load data can be determinedj。
In this embodiment, on the basis that the weight corresponding to each load data is determined in step 201, the correlation matrix is weighted according to the weight corresponding to each load data to obtain a weighted correlation matrix, and subsequent processing is performed based on the weighted correlation matrix, which is beneficial to improving the accuracy of the subsequent processing.
When ranking each weighted correlation degree, the positive indicators can be ranked from small to large, the negative indicators can be ranked from large to small, and the weighted correlation degrees with the same value can be averaged, so that a rank matrix R (R) is obtainedij)m×qWherein R isijThe rank of the ith attribute and the jth load data in the rank matrix.
And 204, calculating a first weighted rank comparison estimation value corresponding to each attribute according to the weight and the rank matrix corresponding to each load data.
Optionally, calculating a first weighted rank-to-total ratio estimation value corresponding to each attribute according to the weight corresponding to each load data and the rank matrix, may include: calculating the weighted rank sum ratio of each attribute according to the weight and rank matrix corresponding to each load data; calculating the cumulative frequency of each attribute according to each weighted rank sum ratio and converting the cumulative frequency into a probability unit; establishing a linear regression equation by taking the probability unit as an independent variable and the weighted rank sum ratio of each attribute as a dependent variable; and calculating a first weighted rank combination ratio estimation value corresponding to each attribute according to a linear regression equation.
After the rank matrix is obtained, the weighted rank sum ratio of each attribute can be calculated according to the rank matrix and the weight corresponding to each load data:
wherein, WRSRiIs a weighted rank sum ratio of the ith attribute.
After the weighted rank sum ratio of each attribute is calculated, a WRSR frequency distribution table is compiled: will WRSRiArranged from small to large (the values are the same as one group), and the frequency f of each group is listediAnd each group of accumulated frequency sigma fi(ii) a Calculating cumulative frequency for each attribute
Then, the cumulative frequency p of each attribute is determinediConversion to ProbitiIn which p is taken as a standard normal distributioniQuantile + 5.
Calculating to obtain the cumulative frequency of each attribute, converting the cumulative frequency into a probability unit, and then using the probability unit Probit corresponding to the cumulative frequencyiAs independent variable, WRSRiFor dependent variables, a linear regression equation was established:
WRSRi＝a1+b1*Probiti。
wherein, a1And b1Are the coefficients of the linear regression equation to be determined. And calculating to obtain a first weighted rank-sum ratio estimation value corresponding to each attribute according to a linear regression equation.
In this embodiment, the value of each attribute to load prediction can be determined according to the first weighted rank-sum ratio estimation value corresponding to each attribute, an attribute having a higher value to load prediction can be determined based on the value of each attribute to load prediction, load prediction is performed according to the attribute having the higher value to load prediction, and reduction of complexity of load prediction is facilitated.
And 104, determining the attribute of which the first weighted rank comparison estimation value is greater than a preset threshold value as a key attribute.
In the embodiment, the preset threshold is used as a judgment condition, and the attribute corresponding to the first weighted rank comparison estimation value larger than the preset threshold is an attribute with a higher load prediction value, that is, a key attribute, and the load prediction is performed according to the key attribute, so that the complexity of the load prediction is favorably reduced.
And 105, screening the historical attribute data set by using the key attribute to obtain a key attribute data set.
And 106, performing load prediction model training according to the historical load data set and the corresponding key attribute data set, and performing short-term power load prediction on the power system according to the trained load prediction model.
Optionally, after calculating, for each load data in the historical load data set, a correlation between the load data and each attribute data in the historical attribute data set, and obtaining a correlation matrix, the method may further include: calculating a second weighted rank combination ratio estimation value corresponding to each load data according to the correlation matrix; and determining the load data with the maximum second weighted rank-sum-ratio estimation value as the key load data.
In this embodiment, referring to the method for calculating the first weighted rank-sum-ratio estimation value corresponding to each attribute in steps 201 to 204, according to the correlation matrix, the second weighted rank-sum-ratio estimation value corresponding to each load data may be calculated, the value of each load data to load prediction may be determined according to the second weighted rank-sum-ratio estimation value corresponding to each load data, and the load data with the largest second weighted rank-sum-ratio estimation value is determined as the key load data, that is, the load data with the largest load prediction value may be determined.
Optionally, the training of the load prediction model according to the historical load data set and the corresponding key attribute data set may include: determining an initial load prediction model according to the key load data and key attribute data corresponding to the key load data; and updating the initial load prediction model according to the load data except the key load data in the historical load data set and the corresponding key attribute data until an update termination condition is reached.
In this embodiment, since the key load data is the load data with the largest load prediction value, the key attribute data corresponding to the key load data is the attribute data with the largest load prediction value, and the initial load prediction model is determined according to the key load data and the key attribute data corresponding to the key load data, the problem that the randomly determined initial load prediction model is not accurate enough can be avoided, and thus, when the initial load prediction model is used for updating training, the speed of training the load prediction model is increased.
The short-term power load prediction method includes the steps of obtaining a historical load data set and a corresponding historical attribute data set, calculating the correlation between the load data and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation matrix, calculating a first weighted rank comparison estimation value corresponding to each attribute according to the correlation matrix, determining the attribute with the first weighted rank comparison estimation value being larger than a preset threshold value as a key attribute, screening the historical attribute data set by using the key attribute to obtain a key attribute data set, conducting load prediction model training according to the historical load data set and the corresponding key attribute data set, and conducting short-term power load prediction on a power system according to the trained load prediction model. The method can reduce the input characteristics of load prediction, and simultaneously reserve the key attribute with higher value of load prediction, thereby taking the complexity and prediction precision of short-term power load prediction into account.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
The following are embodiments of the apparatus of the invention, reference being made to the corresponding method embodiments described above for details which are not described in detail therein.
Fig. 3 shows an exemplary diagram of a short-term power load prediction apparatus provided by an embodiment of the present invention, corresponding to the short-term power load prediction method described in the above embodiments. As shown in fig. 3, the apparatus may include: an acquisition module 31, a first processing module 32, a second processing module 33, a third processing module 34, a screening module 35, and a training prediction module 36.
An obtaining module 31, configured to obtain a historical load data set and a corresponding historical attribute data set;
a first processing module 32, configured to calculate, for each piece of load data in the historical load data set, a correlation between the piece of load data and each piece of attribute data in the historical attribute data set, so as to obtain a correlation matrix;
the second processing module 33 is configured to calculate a first weighted rank-to-rank ratio estimation value corresponding to each attribute according to the correlation matrix;
a third processing module 34, configured to determine, as a key attribute, an attribute that the first weighted rank-to-rank ratio estimate is greater than a preset threshold;
the screening module 35 is configured to screen the historical attribute data set by using the key attribute to obtain a key attribute data set;
and the training prediction module 36 is configured to perform load prediction model training according to the historical load data set and the corresponding key attribute data set, and perform short-term power load prediction on the power system according to the trained load prediction model.
Optionally, the obtaining module 31 is further configured to obtain an initial historical load data set and a corresponding initial historical attribute data set;
performing missing value completion, abnormal value correction or elimination processing on the initial historical load data set and the initial historical attribute data set;
and taking the processed initial historical load data set as the historical load data set, and taking the processed initial historical attribute data set as the historical attribute data set.
Optionally, a first processing module 32 for processing the data according to
Calculating the load data and the historical attribute numberAccording to the relevancy of each attribute data in the set;
wherein, I (x)i,xo) As load data xoA degree of correlation, p (x), with the ith attribute data in the historical attribute data setik,xok) As a function of probability density, p (x)ik) A marginal distribution, p (x), for a kth value in an ith attribute data in the historical attribute data setok) As load data xoThe marginal distribution of the kth value, where k is 1,2, … n.
Optionally, the second processing module 33 is further configured to calculate a second weighted rank-to-total ratio estimation value corresponding to each load data according to the correlation matrix;
and determining the load data with the maximum second weighted rank-to-average ratio estimation value as key load data.
Optionally, the training prediction module 36 is configured to determine an initial load prediction model according to the key load data and key attribute data corresponding to the key load data;
and updating the initial load prediction model according to the load data except the key load data in the historical load data set and the corresponding key attribute data until an update termination condition is reached.
Optionally, the second processing module 33 is configured to determine a weight corresponding to each load data according to a distribution condition of the correlation matrix;
calculating to obtain a weighted correlation matrix according to the weight corresponding to each load data and the correlation matrix;
ranking each weighted correlation in the weighted correlation matrix to obtain a rank matrix;
and calculating a first weighted rank-to-total ratio estimation value corresponding to each attribute according to the weight corresponding to each load data and the rank matrix.
Optionally, the second processing module 33 is configured to calculate a weighted rank-sum ratio of each attribute according to the weight corresponding to each load data and the rank matrix;
calculating the cumulative frequency of each attribute according to each weighted rank sum ratio and converting the cumulative frequency into a probability unit;
establishing a linear regression equation by taking the probability unit as an independent variable and taking the weighted rank sum ratio of each attribute as a dependent variable;
and calculating a first weighted rank combination ratio estimation value corresponding to each attribute according to the linear regression equation.
The short-term power load prediction device calculates the correlation between the load data and each attribute data in the historical attribute data set aiming at each load data in the historical load data set by acquiring the historical load data set and the corresponding historical attribute data set to obtain a correlation matrix, calculates a first weighted rank comparison estimation value corresponding to each attribute according to the correlation matrix, determines the attribute with the first weighted rank comparison estimation value being larger than a preset threshold value as a key attribute, screens the historical attribute data set by using the key attribute to obtain a key attribute data set, performs load prediction model training according to the historical load data set and the corresponding key attribute data set, and performs short-term power load prediction on a power system according to the trained load prediction model. The method can reduce the input characteristics of load prediction, and simultaneously reserve the key attribute with higher value of load prediction, thereby taking the complexity and prediction precision of short-term power load prediction into account.
Fig. 4 is a schematic diagram of a terminal device according to an embodiment of the present invention. As shown in fig. 4, the terminal device 400 of this embodiment includes: a processor 401, a memory 402 and a computer program 403, such as a short term power load forecasting program, stored in said memory 402 and executable on said processor 401. The processor 401 implements the steps in the above-mentioned short-term power load prediction method embodiment, such as steps 101 to 106 shown in fig. 1, or steps 201 to 204 shown in fig. 2, when executing the computer program 403, and the processor 401 implements the functions of the modules in the above-mentioned device embodiments, such as the functions of the modules 31 to 36 shown in fig. 3, when executing the computer program 403.
Illustratively, the computer program 403 may be divided into one or more program modules that are stored in the memory 402 and executed by the processor 401 to implement the present invention. The one or more program modules may be a series of computer program instruction segments capable of performing specific functions for describing the execution process of the computer program 403 in the short-term power load prediction apparatus or terminal device 400. For example, the computer program 403 may be divided into an obtaining module 31, a first processing module 32, a second processing module 33, a third processing module 34, a screening module 35, and a training prediction module 36, and specific functions of each module are shown in fig. 3, which is not described in detail herein.
The terminal device 400 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The terminal device may include, but is not limited to, a processor 401, a memory 402. Those skilled in the art will appreciate that fig. 4 is merely an example of a terminal device 400 and does not constitute a limitation of terminal device 400 and may include more or fewer components than shown, or some components may be combined, or different components, e.g., the terminal device may also include input-output devices, network access devices, buses, etc.
The Processor 401 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 402 may be an internal storage unit of the terminal device 400, such as a hard disk or a memory of the terminal device 400. The memory 402 may also be an external storage device of the terminal device 400, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the terminal device 400. Further, the memory 402 may also include both an internal storage unit and an external storage device of the terminal device 400. The memory 402 is used for storing the computer programs and other programs and data required by the terminal device 400. The memory 402 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
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 units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; 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; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method for short-term power load prediction, comprising:
acquiring a historical load data set and a corresponding historical attribute data set;
calculating the correlation degree of each load data in the historical load data set and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation degree matrix;
calculating a first weighted rank combination ratio estimation value corresponding to each attribute according to the correlation matrix;
determining the attribute of which the first weighted rank combination ratio estimated value is greater than a preset threshold value as a key attribute;
screening the historical attribute data set by using the key attribute to obtain a key attribute data set;
and carrying out load prediction model training according to the historical load data set and the corresponding key attribute data set, and carrying out short-term power load prediction on the power system according to the trained load prediction model.
2. The short term power load forecasting method of claim 1, further comprising, prior to obtaining the historical load data set and the corresponding historical attribute data set:
acquiring an initial historical load data set and a corresponding initial historical attribute data set;
performing missing value completion, abnormal value correction or elimination processing on the initial historical load data set and the initial historical attribute data set;
and taking the processed initial historical load data set as the historical load data set, and taking the processed initial historical attribute data set as the historical attribute data set.
3. The short term power load forecasting method of claim 1 or 2, wherein said calculating a correlation of the load data to each attribute data in the historical attribute data set comprises:
according to
Calculating the relevance of the load data and each attribute data in the historical attribute data set;
wherein, I (x)i,xo) As load data xoA degree of correlation, p (x), with the ith attribute data in the historical attribute data setik,xok) As a function of probability density, p (x)ik) A marginal distribution, p (x), for a kth value in an ith attribute data in the historical attribute data setok) As load data xoThe marginal distribution of the kth value, where k is 1,2, … n.
4. The short-term power load forecasting method of claim 3, further comprising, after calculating, for each load data in the historical set of load data, a correlation of the load data with each attribute data in the historical set of attribute data and obtaining a correlation matrix:
calculating a second weighted rank combination ratio estimation value corresponding to each load data according to the correlation matrix;
and determining the load data with the maximum second weighted rank-to-average ratio estimation value as key load data.
5. The short term power load forecasting method of claim 4, wherein the load forecasting model training based on the historical load data sets and the corresponding key attribute data sets comprises:
determining an initial load prediction model according to the key load data and key attribute data corresponding to the key load data;
and updating the initial load prediction model according to the load data except the key load data in the historical load data set and the corresponding key attribute data until an update termination condition is reached.
6. The short-term power load prediction method according to claim 3, wherein the calculating a first weighted rank-to-total ratio estimation value for each attribute according to the correlation matrix comprises:
determining the weight corresponding to each load data according to the distribution condition of the correlation matrix;
calculating to obtain a weighted correlation matrix according to the weight corresponding to each load data and the correlation matrix;
ranking each weighted correlation in the weighted correlation matrix to obtain a rank matrix;
and calculating a first weighted rank-to-total ratio estimation value corresponding to each attribute according to the weight corresponding to each load data and the rank matrix.
7. The short-term power load prediction method according to claim 6, wherein the calculating a first weighted rank-to-average ratio estimation value for each attribute according to the weight corresponding to each load data and the rank matrix comprises:
calculating the weighted rank sum ratio of each attribute according to the weight corresponding to each load data and the rank matrix;
calculating the cumulative frequency of each attribute according to each weighted rank sum ratio and converting the cumulative frequency into a probability unit;
establishing a linear regression equation by taking the probability unit as an independent variable and taking the weighted rank sum ratio of each attribute as a dependent variable;
and calculating a first weighted rank combination ratio estimation value corresponding to each attribute according to the linear regression equation.
8. A short-term power load prediction apparatus, comprising:
the acquisition module is used for acquiring a historical load data set and a corresponding historical attribute data set;
the first processing module is used for calculating the correlation degree of each load data in the historical load data set and each attribute data in the historical attribute data set aiming at each load data in the historical load data set to obtain a correlation degree matrix;
the second processing module is used for calculating a first weighted rank comparison estimation value corresponding to each attribute according to the correlation matrix;
the third processing module is used for determining the attribute of which the first weighted rank-to-total ratio estimation value is greater than a preset threshold as a key attribute;
the screening module is used for screening the historical attribute data set by using the key attribute to obtain a key attribute data set;
and the training prediction module is used for carrying out load prediction model training according to the historical load data set and the corresponding key attribute data set and carrying out short-term power load prediction on the power system according to the trained load prediction model.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944859.8A CN113592192A (en) | 2021-08-17 | 2021-08-17 | Short-term power load prediction method and device and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944859.8A CN113592192A (en) | 2021-08-17 | 2021-08-17 | Short-term power load prediction method and device and terminal equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113592192A true CN113592192A (en) | 2021-11-02 |
Family
ID=78258387
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110944859.8A Withdrawn CN113592192A (en) | 2021-08-17 | 2021-08-17 | Short-term power load prediction method and device and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113592192A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116186548A (en) * | 2023-05-04 | 2023-05-30 | 广州三晶电气股份有限公司 | Power load prediction model training method and power load prediction method |
| CN116205355A (en) * | 2023-02-22 | 2023-06-02 | 正泰电气股份有限公司 | Power load prediction method, device and storage medium |
-
2021
- 2021-08-17 CN CN202110944859.8A patent/CN113592192A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116205355A (en) * | 2023-02-22 | 2023-06-02 | 正泰电气股份有限公司 | Power load prediction method, device and storage medium |
| CN116205355B (en) * | 2023-02-22 | 2023-12-01 | 正泰电气股份有限公司 | Power load prediction method, device and storage medium |
| CN116186548A (en) * | 2023-05-04 | 2023-05-30 | 广州三晶电气股份有限公司 | Power load prediction model training method and power load prediction method |
| CN116186548B (en) * | 2023-05-04 | 2023-08-01 | 广州三晶电气股份有限公司 | Power load prediction model training method and power load prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113592192A (en) | Short-term power load prediction method and device and terminal equipment | |
| CN112651563B (en) | Load prediction method and device, computer readable storage medium and electronic equipment | |
| CN113743673B (en) | Power load prediction method during typhoon | |
| CN110263818A (en) | Method, apparatus, terminal and the computer readable storage medium of resume selection | |
| CN115936180A (en) | Photovoltaic power generation power prediction method and device and computer equipment | |
| CN113112099A (en) | Power grid daily electric quantity prediction model training method and power grid daily electric quantity prediction method | |
| CN112002949A (en) | Monitoring management method and system for active equalization of power battery | |
| CN113177366A (en) | Comprehensive energy system planning method and device and terminal equipment | |
| CN115905927A (en) | Method and device for identifying abnormal electricity consumption user, electronic equipment and storage medium | |
| CN109325530A (en) | Compression method based on the depth convolutional neural networks on a small quantity without label data | |
| CN114461618A (en) | Metering missing data completion method and device and terminal equipment | |
| CN113139698A (en) | Load prediction method, device and equipment | |
| CN113591900A (en) | Identification method and device for high-demand response potential user and terminal equipment | |
| CN109242321B (en) | User power load online analysis method and terminal equipment | |
| CN108171417B (en) | Planting task adjusting method, electronic device and storage medium | |
| CN114971053A (en) | Training method and device for online prediction model of network line loss rate of low-voltage transformer area | |
| CN113393099A (en) | Power distribution network project group value index evaluation method and device and terminal equipment | |
| CN109582921B (en) | Distribution transformer load state evaluation method and terminal equipment | |
| CN114092276A (en) | Prediction method of short-term load of transformer area | |
| CN113570105A (en) | Power load prediction method and device and terminal | |
| CN113822464A (en) | User information processing method and device, electronic equipment and storage medium | |
| CN114580255A (en) | Method for constructing residual life prediction model of equipment and terminal equipment | |
| CN112734245B (en) | Low-voltage power distribution loop monitoring method, device and equipment | |
| CN113205259A (en) | Power grid scheduling decision evaluation method and device and terminal equipment | |
| CN113705872A (en) | Short-term load prediction method and device |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211102 |
|
| WW01 | Invention patent application withdrawn after publication |