CN114021861A - Power load prediction method, device, terminal and storage medium - Google Patents

Power load prediction method, device, terminal and storage medium Download PDF

Info

Publication number
CN114021861A
CN114021861A CN202111535698.3A CN202111535698A CN114021861A CN 114021861 A CN114021861 A CN 114021861A CN 202111535698 A CN202111535698 A CN 202111535698A CN 114021861 A CN114021861 A CN 114021861A
Authority
CN
China
Prior art keywords
historical data
data set
real
time data
prediction
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
Application number
CN202111535698.3A
Other languages
Chinese (zh)
Inventor
尹艺霏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University of Science and Technology
Original Assignee
Shandong University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong University of Science and Technology filed Critical Shandong University of Science and Technology
Priority to CN202111535698.3A priority Critical patent/CN114021861A/en
Publication of CN114021861A publication Critical patent/CN114021861A/en
Pending legal-status Critical Current

Links

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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
    • 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

Abstract

The invention discloses a power load prediction method, a device, a terminal and a storage medium.K groups of real-time data are selected in each prediction period and added into the last historical data set to form a new historical data set for storage; training a new historical data set based on a neural network algorithm to obtain and store a new prediction model; before power load prediction is carried out in a current prediction period, P groups of real-time data of the current prediction period are obtained, the P groups of real-time data are respectively compared with each stored historical data set, and a historical data set with the highest similarity to the P groups of real-time data is selected and recorded as a target historical data set; and in the current prediction period, performing power load prediction based on a prediction model corresponding to the target historical data set. The method and the device enable prediction to consider recent data and an optimal model, effectively improve prediction precision, and provide reliable basis for planning and operation of the power system.

Description

Power load prediction method, device, terminal and storage medium
Technical Field
The invention relates to the field of power load prediction, in particular to a power load prediction method, a power load prediction device, a power load prediction terminal and a storage medium.
Background
The important components of the power system planning during power load prediction are used for predicting the future power demand, power consumption and the like and providing decision basis for power system planning and operation. Many factors affect the predicted value of the power load to varying degrees, such as meteorological factors. The existing power load prediction is generally carried out by training by adopting a neural network algorithm to obtain a prediction model, and prediction is carried out based on the prediction model. However, after the prediction model is initially generated, the prediction model is used for prediction for a long time, so that the subsequent prediction accuracy is reduced, the prediction result is influenced, and the planning and operation of the power system are further influenced.
Disclosure of Invention
In order to solve the problems, the invention provides a power load prediction method, a power load prediction device, a power load prediction terminal and a power load prediction storage medium, a prediction period is defined, an optimal prediction model is trained and selected in each prediction period, and the optimal prediction model is used for prediction, so that the prediction accuracy is effectively improved, and a reliable basis is provided for planning and running of a power system.
In a first aspect, an embodiment of the present invention provides a power load prediction method, including:
selecting K groups of real-time data in each prediction period, adding the real-time data into the previous historical data set to form a new historical data set, and storing the new historical data set; wherein K is more than or equal to 1;
training a new historical data set based on a neural network algorithm to obtain and store a new prediction model;
before power load prediction is carried out in a current prediction period, P groups of real-time data of the current prediction period are obtained, the P groups of real-time data are respectively compared with each stored historical data set, and a historical data set with the highest similarity to the P groups of real-time data is selected and recorded as a target historical data set; wherein P is more than or equal to 1;
and in the current prediction period, performing power load prediction based on a prediction model corresponding to the target historical data set.
Further, when the K groups of real-time data are added into the last historical data set, the K groups of historical data with the longest time in the last historical data set are deleted.
Further, let a set of real-time data be X ═ X (X)1，x2，…，xn) WhereinxiThe ith element of the real-time data is i ═ 1, 2, …, n;
recording the historical data set as
Figure BDA0003413080320000021
Wherein m represents m samples in the historical data set;
the similarity between a set of real-time data and a historical data set is:
Figure BDA0003413080320000022
a group of real-time data corresponds to a historical data set with the highest similarity;
for the P groups of real-time data, if the number of real-time data groups with the highest similarity corresponding to a certain historical data set is the largest, the historical data set is the historical data set with the highest similarity of the P groups of real-time data.
Further, when the number of the stored historical data sets reaches a preset threshold value and a new historical data set is stored, the historical data set with the longest time and the corresponding prediction model are deleted.
In a second aspect, an electrical load prediction apparatus according to an aspect of the present invention includes,
the first real-time data selecting module: selecting K groups of real-time data in each prediction period; wherein K is more than or equal to 1;
a historical data set composition module: adding K groups of real-time data selected by the first real-time data selecting module into the last historical data set to form a new historical data set and storing the new historical data set;
a model training module: training a historical data set based on a neural network algorithm to obtain and store a prediction model;
a second real-time data selection module: acquiring P groups of real-time data of a current prediction period before power load prediction is carried out in the current prediction period; wherein P is more than or equal to 1;
a target historical data set determination module: comparing the P groups of real-time data acquired by the second real-time data selection module with each stored historical data set respectively, selecting the historical data set with the highest similarity to the P groups of real-time data, and recording the historical data set as a target historical data set;
a power load prediction module: and in the current prediction period, performing power load prediction based on a prediction model corresponding to the target historical data set.
Further, when the historical data set composing module adds the K groups of real-time data selected by the first real-time data selecting module into the previous historical data set, the K groups of historical data with the longest time in the previous historical data set are deleted.
Further, let a set of real-time data be X ═ X (X)1，x2，…，xn) Wherein x isiThe ith element of the real-time data is i ═ 1, 2, …, n;
recording the historical data set as
Figure BDA0003413080320000031
Wherein m represents m samples in the historical data set;
the target historical data set determining module calculates the similarity between a group of real-time data and a historical data set as follows:
Figure BDA0003413080320000032
a group of real-time data corresponds to a historical data set with the highest similarity;
the target historical data set determined by the target historical data set determination module specifically includes:
for the P groups of real-time data, if the number of real-time data groups with the highest similarity corresponding to a certain historical data set is the largest, the historical data set is the historical data set with the highest similarity of the P groups of real-time data.
Further, when a new historical data set is formed by the historical data set forming module, whether the number of the stored historical data sets reaches a preset threshold value or not is detected, and if the number of the stored historical data sets reaches the preset threshold value, the historical data set with the longest time and the corresponding prediction model are deleted when the new historical data set is stored.
In a third aspect, a technical solution of the present invention provides a terminal, including:
a processor;
a memory for storing instructions for execution by the processor;
wherein the processor is configured to perform any of the methods described above.
In a fourth aspect, the invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the method of any one of the above.
Compared with the prior art, the power load prediction method, the power load prediction device, the terminal and the storage medium provided by the invention have the following beneficial effects: defining a prediction period, extracting new real-time data in each prediction period to combine with historical data to form a new training set, and training based on the new training set to obtain a prediction model; and when predicting in the new prediction cycle, firstly, selecting the optimal prediction model according to the real-time data of the prediction cycle, and predicting according to the prediction model. The method extracts real-time data in time to update the prediction model, and selects the optimal prediction model according to the real-time data to be predicted, so that the latest data and the optimal model are considered in prediction, the prediction precision is effectively improved, and a reliable basis is provided for planning and operating the power system.
Drawings
For a clearer explanation of the embodiments or technical solutions of the prior art of the present application, the drawings needed for the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a power load prediction method according to an embodiment of the present invention;
fig. 2 is a schematic block diagram of a power load prediction apparatus according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a terminal according to a third embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the disclosure, the following detailed description will be given with reference to the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. 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 application.
Example one
The embodiment provides a power load prediction method, which performs model training based on the latest data in time, selects an optimal prediction model before prediction, and improves prediction accuracy.
As shown in fig. 1, the power load prediction method of the present embodiment includes the following steps.
S1, selecting K groups of real-time data in each prediction period, adding the real-time data into the previous historical data set to form a new historical data set and storing the new historical data set; wherein K is more than or equal to 1.
The prediction period is set in advance, for example, one day, two days, one week, one month, or the like, and the prediction period can be adjusted as necessary or by changing seasons or the like.
And in the first prediction period, performing prediction based on a prediction model of the preset historical data set, and in the later prediction period, extracting corresponding K groups of real-time data and adding the corresponding K groups of real-time data into the previous historical data set to form a new historical data set. Accordingly, the historical data set corresponding to the first prediction period is the initial preset historical data set.
To improve accuracy, the K sets of real-time data are selected near the end of a prediction period, and the K sets can be selected on average in the early, middle and late stages of the whole prediction period, so that the selected K sets of real-time data are more representative.
In addition, when the K groups of real-time data are added into the last historical data set, the K groups of historical data with the longest time in the last historical data set are deleted, namely the number of samples in the historical data set is always kept constant, so that the training efficiency is improved.
And meanwhile, monitoring the number of the stored historical data sets in real time, and deleting the historical data set with the longest time and the corresponding prediction model when a new historical data set is stored when the number of the stored historical data sets reaches a preset threshold value. Even if the number of the stored historical data sets is up to the preset threshold value, the historical data sets with small influence are deleted, the memory occupation is reduced, and the prediction efficiency is improved.
And S2, training the new historical data set based on the neural network algorithm, obtaining a new prediction model and storing the new prediction model.
After each acquisition of the historical data set, the calculation is trained to obtain a predictive model. The historical data sets and the corresponding prediction models are stored in time for subsequent selection of the target historical data set and the corresponding optimal prediction model.
The neural network algorithm may be implemented by using an existing neural network algorithm, and is not described again.
In addition, in order to improve the calculation efficiency and reduce the influence on other performances, the training can be performed at night. Accordingly, when the prediction period is divided, the time at night is used as a starting point.
S3, before power load prediction is carried out in the current prediction period, P groups of real-time data of the current prediction period are obtained, the P groups of real-time data are respectively compared with the stored historical data sets, and the historical data set with the highest similarity to the P groups of real-time data is selected and recorded as a target historical data set; wherein P is more than or equal to 1.
At S4, at the current prediction cycle, the power load is predicted based on the prediction model corresponding to the target historical data set.
Before the prediction is carried out in the current prediction period, a prediction model needs to be determined. In this embodiment, first, P groups of real-time data (data before prediction) of the current prediction period are selected, a history data set most similar to the selected real-time data is selected according to the P groups of real-time data, an optimal prediction model is selected accordingly, and prediction accuracy is provided by using the optimal prediction model for prediction.
The principle of selecting the target history data set in step S3 is as follows.
Let a set of real-time data be X ═ X1，x2，…，xn) Wherein x isiI is the ith element of the real-time data, i is 1, 2, …, n. That is, a set of real-time data includes n elements, such as temperature, humidity, air volume, etc., and is set as required.
Recording the historical data set as
Figure BDA0003413080320000071
Where m indicates that there are m samples in the historical dataset.
The similarity between a set of real-time data and a historical data set is:
Figure BDA0003413080320000072
a group of real-time data corresponds to a historical data set with the highest similarity.
And for the P groups of real-time data, each group of real-time data carries out similarity calculation on each historical data set, and a historical data set with the highest similarity is selected. And if the number of the real-time data groups with the highest similarity corresponding to a certain historical data set is the largest, the historical data set is the historical data set with the highest similarity of the P groups of real-time data, namely the target historical data set.
For example, if there are 5 groups of real-time data and there are 10 historical data sets, where the historical data set with the highest similarity corresponding to 3 groups of real-time data is the 8 th historical data set, then the 8 th historical data set is the target historical data set.
It should be noted that, if each group of real-time data corresponds to different historical data sets, the target historical data set is selected according to the P maximum similarities.
The power load prediction method provided by the embodiment defines a prediction period, extracts new real-time data in each prediction period to combine with historical data to form a new training set, and trains based on the new training set to obtain a prediction model; and when predicting in the new prediction cycle, firstly, selecting the optimal prediction model according to the real-time data of the prediction cycle, and predicting according to the prediction model. The method extracts real-time data in time to update the prediction model, and selects the optimal prediction model according to the real-time data to be predicted, so that the latest data and the optimal model are considered in prediction, the prediction precision is effectively improved, and a reliable basis is provided for planning and operating the power system.
Example two
As shown in fig. 2, the present embodiment provides a power load prediction apparatus for implementing the aforementioned power load prediction method, which includes the following functional modules.
The first real-time data selection module 101: selecting K groups of real-time data in each prediction period; wherein K is more than or equal to 1;
historical data set composition module 102: adding K groups of real-time data selected by the first real-time data selecting module 101 into the last historical data set to form a new historical data set and storing the new historical data set;
the model training module 103: training a historical data set based on a neural network algorithm to obtain and store a prediction model;
the second real-time data selection module 104: acquiring P groups of real-time data of a current prediction period before power load prediction is carried out in the current prediction period; wherein P is more than or equal to 1;
target historical data set determination module 105: comparing the P groups of real-time data acquired by the second real-time data selection module 104 with the stored historical data sets respectively, selecting the historical data set with the highest similarity to the P groups of real-time data, and recording the historical data set as a target historical data set;
power load prediction module 106: and in the current prediction period, performing power load prediction based on a prediction model corresponding to the target historical data set.
When the historical data set composing module 102 adds the K sets of real-time data selected by the first real-time data selecting module 101 to the previous historical data set, the K sets of historical data with the longest time in the previous historical data set are deleted. When the historical data set composing module 102 composes a new historical data set, it also detects whether the number of the stored historical data sets reaches a preset threshold, and if the number of the stored historical data sets reaches the preset threshold, deletes the historical data set with the longest time and the corresponding prediction model when storing the new historical data set.
Let a set of real-time data be X ═ X1，x2，…，xn) Wherein x isiI is the ith element of the real-time data, i is 1, 2, …, n.
Recording the historical data set as
Figure BDA0003413080320000091
Where m indicates that there are m samples in the historical dataset.
The target historical data set determination module 105 calculates the similarity between a set of real-time data and a historical data set as:
Figure BDA0003413080320000092
a group of real-time data corresponds to a historical data set with the highest similarity.
The target historical data set determined by the target historical data set determining module 105 specifically includes:
for the P groups of real-time data, if the number of real-time data groups with the highest similarity corresponding to a certain historical data set is the largest, the historical data set is the historical data set with the highest similarity of the P groups of real-time data.
The power load prediction device of the present embodiment is used to implement the foregoing power load prediction method, and therefore, the specific implementation of the device can be seen from the above example portion of the power load prediction method, and therefore, the specific implementation thereof can refer to the description of the corresponding partial embodiment, and will not be further described herein.
In addition, since the power load prediction apparatus of the present embodiment is used to implement the foregoing power load prediction method, the function corresponds to that of the foregoing method, and details are not described here.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a terminal device 300 according to an embodiment of the present invention, where the terminal device 300 may be used to execute the power load prediction method according to the embodiment of the present invention.
Among them, the terminal apparatus 300 may include: a processor 310, a memory 320, and a communication unit 330. The components communicate via one or more buses, and those skilled in the art will appreciate that the architecture of the servers shown in the figures is not intended to be limiting, and may be a bus architecture, a star architecture, a combination of more or less components than those shown, or a different arrangement of components.
The memory 320 may be used for storing instructions executed by the processor 310, and the memory 320 may be implemented by any type of volatile or non-volatile storage terminal or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The executable instructions in memory 320, when executed by processor 310, enable terminal 300 to perform some or all of the steps in the method embodiments described below.
The processor 310 is a control center of the storage terminal, connects various parts of the entire electronic terminal using various interfaces and lines, and performs various functions of the electronic terminal and/or processes data by operating or executing software programs and/or modules stored in the memory 320 and calling data stored in the memory. The processor may be composed of an Integrated Circuit (IC), for example, a single packaged IC, or a plurality of packaged ICs connected with the same or different functions. For example, the processor 310 may include only a Central Processing Unit (CPU). In the embodiment of the present invention, the CPU may be a single operation core, or may include multiple operation cores.
A communication unit 330, configured to establish a communication channel so that the storage terminal can communicate with other terminals. And receiving user data sent by other terminals or sending the user data to other terminals.
Example four
The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, and the program may include some or all of the steps in the embodiments provided by the present invention when executed. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a Random Access Memory (RAM).
Those skilled in the art will readily appreciate that the techniques of the embodiments of the present invention may be implemented as software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied in the form of a software product, where the computer software product is stored in a storage medium, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like, and the storage medium can store program codes, and includes instructions for enabling a computer terminal (which may be a personal computer, a server, or a second terminal, a network terminal, and the like) to perform all or part of the steps of the method in the embodiments of the present invention.
The same and similar parts in the various embodiments in this specification may be referred to each other. Especially, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant points can be referred to the description in the method embodiment.
In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, 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 above disclosure is only for the preferred embodiments of the present invention, but the present invention is not limited thereto, and any non-inventive changes that can be made by those skilled in the art and several modifications and amendments made without departing from the principle of the present invention shall fall within the protection scope of the present invention.

Claims (10)

1. A method for predicting an electrical load, comprising the steps of:
selecting K groups of real-time data in each prediction period, adding the real-time data into the previous historical data set to form a new historical data set, and storing the new historical data set; wherein K is more than or equal to 1;
training a new historical data set based on a neural network algorithm to obtain and store a new prediction model; before power load prediction is carried out in a current prediction period, P groups of real-time data of the current prediction period are obtained, the P groups of real-time data are respectively compared with each stored historical data set, and a historical data set with the highest similarity to the P groups of real-time data is selected and recorded as a target historical data set; wherein P is more than or equal to 1;
and in the current prediction period, performing power load prediction based on a prediction model corresponding to the target historical data set.
2. The method of claim 1, wherein when K sets of real-time data are added to a previous historical data set, the K sets of historical data that have been in the previous historical data set with the longest time are deleted.
3. The power load prediction method according to claim 2,
let a set of real-time data be X ═ X1，x2，…，xn) Wherein x isiIs the ith element of real-time data, i ═ 1, 2, ·, n;
recording the historical data set as
Figure FDA0003413080310000011
Wherein m represents m samples in the historical data set;
the similarity between a set of real-time data and a historical data set is:
Figure FDA0003413080310000012
a group of real-time data corresponds to a historical data set with the highest similarity;
for the P groups of real-time data, if the number of real-time data groups with the highest similarity corresponding to a certain historical data set is the largest, the historical data set is the historical data set with the highest similarity of the P groups of real-time data.
4. A power load prediction method as claimed in claim 1, 2 or 3, characterized in that when a new historical data set is stored when the number of stored historical data sets reaches a preset threshold, the historical data set with the longest time and the corresponding prediction model are deleted.
5. An electric load prediction apparatus, comprising,
the first real-time data selecting module: selecting K groups of real-time data in each prediction period; wherein K is more than or equal to 1;
a historical data set composition module: adding K groups of real-time data selected by the first real-time data selecting module into the last historical data set to form a new historical data set and storing the new historical data set;
a model training module: training a historical data set based on a neural network algorithm to obtain and store a prediction model;
a second real-time data selection module: acquiring P groups of real-time data of a current prediction period before power load prediction is carried out in the current prediction period; wherein P is more than or equal to 1;
a target historical data set determination module: comparing the P groups of real-time data acquired by the second real-time data selection module with each stored historical data set respectively, selecting the historical data set with the highest similarity to the P groups of real-time data, and recording the historical data set as a target historical data set;
a power load prediction module: and in the current prediction period, performing power load prediction based on a prediction model corresponding to the target historical data set.
6. The power load prediction device according to claim 5, wherein the historical data set composing module deletes the K sets of historical data that have been collected most recently from the previous historical data set when adding the K sets of real-time data collected by the first real-time data collecting module to the previous historical data set.
7. The power load prediction device according to claim 6, wherein the set of real-time data is represented by X ═ X (X ═ X)1，x2，…，xn) Wherein x isiIs the ith element of real-time data, i ═ 1, 2, ·, n; recording the historical data set as
Figure FDA0003413080310000021
Wherein m represents m samples in the historical data set;
the target historical data set determining module calculates the similarity between a group of real-time data and a historical data set as follows:
Figure FDA0003413080310000031
a group of real-time data corresponds to a historical data set with the highest similarity;
the target historical data set determined by the target historical data set determination module specifically includes:
for the P groups of real-time data, if the number of real-time data groups with the highest similarity corresponding to a certain historical data set is the largest, the historical data set is the historical data set with the highest similarity of the P groups of real-time data.
8. The power load prediction device according to claim 5, 6 or 7, wherein when the historical data set composition module composes a new historical data set, it further detects whether the number of the stored historical data sets reaches a preset threshold, and if the number of the stored historical data sets reaches the preset threshold, the oldest historical data set and the corresponding prediction model thereof are deleted when the new historical data set is stored.
9. A terminal, comprising:
a processor;
a memory for storing instructions for execution by the processor;
wherein the processor is configured to perform the method of any one of claims 1-4.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-4.
CN202111535698.3A 2021-12-15 2021-12-15 Power load prediction method, device, terminal and storage medium Pending CN114021861A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111535698.3A CN114021861A (en) 2021-12-15 2021-12-15 Power load prediction method, device, terminal and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111535698.3A CN114021861A (en) 2021-12-15 2021-12-15 Power load prediction method, device, terminal and storage medium

Publications (1)

Publication Number Publication Date
CN114021861A true CN114021861A (en) 2022-02-08

Family

ID=80068843

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111535698.3A Pending CN114021861A (en) 2021-12-15 2021-12-15 Power load prediction method, device, terminal and storage medium

Country Status (1)

Country Link
CN (1) CN114021861A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117113137A (en) * 2023-08-07 2023-11-24 国网冀北电力有限公司信息通信分公司 Power model matching method and device, storage medium and electronic equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117113137A (en) * 2023-08-07 2023-11-24 国网冀北电力有限公司信息通信分公司 Power model matching method and device, storage medium and electronic equipment

Similar Documents

Publication Publication Date Title
CN109165983A (en) Insurance products recommended method, device, computer equipment and storage medium
CN111064614A (en) Fault root cause positioning method, device, equipment and storage medium
CN109918498B (en) Problem warehousing method and device
CN111143178B (en) User behavior analysis method, device and equipment
CN111198799A (en) Machine room power consumption early warning method, system, terminal and storage medium based on LSTM
CN110414569B (en) Clustering implementation method and device
CN112016190A (en) Power battery temperature correction method and device and computer readable storage medium
CN109685805B (en) Image segmentation method and device
CN113268403A (en) Time series analysis and prediction method, device, equipment and storage medium
CN111563077A (en) Power grid voltage data missing filling method, system, terminal and storage medium
CN114021861A (en) Power load prediction method, device, terminal and storage medium
CN111159481B (en) Edge prediction method and device for graph data and terminal equipment
CN108667877B (en) Method and device for determining recommendation information, computer equipment and storage medium
CN113743650B (en) Power load prediction method, device, equipment and storage medium
CN113205230A (en) Data prediction method, device and equipment based on model set and storage medium
CN112381272A (en) Power grid load prediction method, system, terminal and storage medium
CN112767056B (en) Service data prediction method, device, computer equipment and storage medium
CN112784165A (en) Training method of incidence relation estimation model and method for estimating file popularity
CN114782780A (en) Data set construction method and device and electronic equipment
CN114118411A (en) Training method of image recognition network, image recognition method and device
CN114239983A (en) Target area population flow prediction method and related device
CN112836827A (en) Model training method and device and computer equipment
CN110415006B (en) Advertisement click rate estimation method and device
CN113516275A (en) Power distribution network ultra-short term load prediction method and device and terminal equipment
CN111368887B (en) Training method of thunderstorm weather prediction model and thunderstorm weather prediction method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination