CN115473219A - Load prediction method, load prediction device, computer equipment and storage medium - Google Patents

Load prediction method, load prediction device, computer equipment and storage medium Download PDF

Info

Publication number
CN115473219A
CN115473219A CN202211077451.6A CN202211077451A CN115473219A CN 115473219 A CN115473219 A CN 115473219A CN 202211077451 A CN202211077451 A CN 202211077451A CN 115473219 A CN115473219 A CN 115473219A
Authority
CN
China
Prior art keywords
power distribution
load data
distribution network
model
daily load
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
CN202211077451.6A
Other languages
Chinese (zh)
Inventor
尚龙龙
罗井利
马楠
胡冉
刘洋宇
刘国伟
慈海
程卓
汪建波
刘涛
吴江龙
王冠璎
尚宇炜
周莉梅
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.)
Shenzhen Power Supply Bureau Co Ltd
Original Assignee
Shenzhen Power Supply Bureau Co Ltd
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 Shenzhen Power Supply Bureau Co Ltd filed Critical Shenzhen Power Supply Bureau Co Ltd
Priority to CN202211077451.6A priority Critical patent/CN115473219A/en
Publication of CN115473219A publication Critical patent/CN115473219A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • 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 application relates to a load prediction method, a load prediction device, computer equipment and a storage medium, and relates to the technical field of intelligent power utilization. The load prediction method comprises the following steps: acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network; constructing at least two different generalized linear models; initializing parameters in each generalized linear model by using historical daily load data of each power distribution device to obtain daily load data predicted by each generalized linear model; and according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network, selecting a target model from the generalized linear models, and predicting the future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network. By the method, the future daily load data of the power distribution network can be accurately predicted by utilizing the historical daily load data of the power distribution equipment, so that the future load demand of the power distribution system is obtained, and the planning and operation of the power distribution network are effectively supported.

Description

Load prediction method, load prediction device, computer equipment and storage medium
Technical Field
The present application relates to the field of intelligent power utilization technologies, and in particular, to a load prediction method and apparatus, a computer device, and a storage medium.
Background
Load prediction is the basic work for ensuring the planning of the power system, and has very important significance for reliable and economic good operation, so that the load of the power system needs to be accurately predicted. Currently, in the related art, a prediction model is usually constructed by using a large amount of external data such as population density, humidity and the like to predict the load.
However, since the quality of external data is poor, the acquisition efficiency is not high, and the variety of factors affecting short-term load prediction is various, the volatility and randomness are high, so that accurate prediction is difficult to realize, and improvement is urgently needed.
Disclosure of Invention
In view of the above, it is necessary to provide a load prediction method, an apparatus, a computer device, and a storage medium, which can realize accurate load prediction.
In a first aspect, the present application provides a load prediction method. The method comprises the following steps:
acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment;
initializing daily load parameters in each generalized linear model by adopting historical daily load data of each power distribution device to obtain predicted daily load data of the power distribution network predicted by each generalized linear model;
selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
In one embodiment, selecting a target model from the generalized linear models according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear models comprises:
for each generalized linear model, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network;
and selecting a target model from the generalized linear models according to the model performance index value of each generalized linear model.
In one embodiment, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model, and the total number of the historical daily load data of the power distribution network includes:
determining a residual square sum according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear model;
and determining the model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
In one embodiment, selecting the target model from the generalized linear models according to the model performance index value of the generalized linear model includes:
selecting a candidate model from each generalized linear model according to the model performance index value of each generalized linear model;
determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total number of the historical daily load data of the power distribution network; the error data comprises a root mean square error and/or an average error;
and selecting the target model from the candidate models according to the error data of the candidate models.
In one embodiment, the historical daily load data of each power distribution device in the power distribution network is the historical daily maximum load data of each power distribution device, and the historical daily load data of the power distribution network is the historical daily maximum load data of the power distribution network;
correspondingly, the method for acquiring the historical daily load data of each power distribution device in the power distribution network and the historical daily load data of the power distribution network comprises the following steps:
determining historical day maximum load data of each power distribution device according to historical day original load data of each power distribution device in the power distribution network;
and determining the historical daily maximum load data of the power distribution network according to the same-day rate and the historical daily original load data of each power distribution device.
In a second aspect, the present application further provides a load prediction apparatus. The device comprises:
the data acquisition module is used for acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
the model construction module is used for constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment;
the initialization module is used for initializing the daily load parameters in the generalized linear models by adopting the historical daily load data of the power distribution equipment to obtain the predicted daily load data of the power distribution network predicted by the generalized linear models;
the model selection module is used for selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:
acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment;
initializing daily load parameters in the generalized linear models by adopting historical daily load data of each power distribution device to obtain predicted daily load data of the power distribution network predicted by each generalized linear model;
selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
constructing at least two different generalized linear models; wherein, different generalized linear models comprise daily load parameters of different power distribution equipment;
initializing daily load parameters in the generalized linear models by adopting historical daily load data of each power distribution device to obtain predicted daily load data of the power distribution network predicted by each generalized linear model;
selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:
acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
constructing at least two different generalized linear models; wherein, different generalized linear models comprise daily load parameters of different power distribution equipment;
initializing daily load parameters in the generalized linear models by adopting historical daily load data of each power distribution device to obtain predicted daily load data of the power distribution network predicted by each generalized linear model;
selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
According to the load prediction method, the load prediction device, the computer equipment, the storage medium and the computer program product, a plurality of generalized linear models are constructed based on historical daily load data of each power distribution equipment in the power distribution network, and a target model is selected from the plurality of generalized linear models based on the historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by each generalized linear model to predict future daily load data of the power distribution network. According to the scheme, external data such as population density and humidity are not needed, a model capable of accurately predicting future daily load data of the power distribution network can be constructed by using historical daily load data of the power distribution equipment, an optional mode is provided for accurately predicting future loads of the power distribution network, and planning and operation of the power distribution network are effectively supported.
Drawings
FIG. 1 is a diagram of an exemplary load prediction method;
FIG. 2 is a flow diagram illustrating a method for load prediction according to one embodiment;
FIG. 3 is a schematic diagram of an embodiment of a power distribution network;
FIG. 4 is a schematic flow chart illustrating selection of a target model in one embodiment;
FIG. 5 is a schematic flow chart of selecting a target model in another embodiment;
FIG. 6 is a schematic diagram of a process for obtaining historical daily maximum load data in one embodiment;
FIG. 7 is a block diagram showing the structure of a load prediction apparatus according to an embodiment;
FIG. 8 is a block diagram that illustrates the structure of a model selection module in one embodiment;
FIG. 9 is a block diagram of the structure of a data acquisition module in one embodiment;
fig. 10 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It is also to be understood that the terminology used in the description of the present application is for the purpose of describing particular embodiments only, and is not intended to be limiting of the application. As used in the specification and claims of this application, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The load prediction method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process, such as historical daily load data for each distribution device in the distribution grid and historical daily loads for the distribution grid. The data storage system may be integrated on the server 104, or may be located on the cloud or other network server. The load prediction method provided by the embodiment of the application can be applied to the server 104, can also be applied to the terminal 102, and can also be realized through interaction between the terminal 102 and the server 104. For example, the server 104 may construct a plurality of generalized linear models based on historical daily load data of each power distribution device in the power distribution network, and select a target model from the plurality of generalized linear models based on the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model, so as to predict future daily load data of the power distribution network; further, the server 104 may send the predicted future daily load data of the distribution network to the terminal 102 for presentation. The terminal 102 may be, but is not limited to, various personal computers, notebook computers, and the like. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers.
In one embodiment, as shown in fig. 2, a flowchart of a load prediction method is provided, where the embodiment relates to predicting future daily load data of a power distribution network by obtaining historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network to construct a generalized linear model. This embodiment is illustrated by applying the method to the server 104 in fig. 1. In this embodiment, the method includes the steps of:
s201, historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network are obtained.
The power distribution network mentioned in S201 is any power distribution network with load prediction demand. Optionally, the distribution network may have a plurality of distribution devices for transmitting ac power, such as the distribution network shown in fig. 3 having a plurality of distribution transformers.
Further, the historical daily load data of each distribution device refers to the load data of each day of the distribution device in a period of time (such as a month, a quarter, or a year) before the current time, for example, the maximum load data of each day of the distribution device in a period of time before the current time. Accordingly, the historical daily load data of the distribution network refers to the load data of each day of the distribution system as a whole in a period of time before the current time. It can be understood that the historical daily load data of each power distribution device and the historical daily load data of the power distribution network in the embodiment are data of actual loads in the operation process of the power distribution network.
Since short-term load prediction is generally performed on an hour/day basis, the prediction results are influenced by a wide variety of factors, and have strong fluctuation and strong randomness. Therefore, in order to accurately predict future daily load data of the power distribution network, a large amount of historical daily load data (such as historical daily load data in a year) is acquired to construct a model.
Specifically, in this embodiment, historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network may be acquired from the data storage system.
S202, at least two different generalized linear models are constructed.
The generalized linear model is a method that takes a linear prediction function of an independent variable as an estimated value of a dependent variable, and can be applied to data having a nonlinear characteristic by using linear transformation.
Optionally, there are three requirements for constructing the generalized linear model, namely that the density function of the model follows a certain distribution in an exponential distribution family with η as a parameter, and has a connection function g (x), and an expectation function h (x) = E [ y | x, θ = g (x)]＝μ＝g -1 (η)＝g -1 (x T β). Common connection functions can be classified into exponential distribution families such as logarithmic functions, poisson functions, normal distribution functions and the like. Further, different generalized linear models can be constructed by selecting different connection functions.
Specifically, after the initial model after the connection function is selected is constructed, the model is gradually optimized, that is, the power distribution equipment with low correlation with the prediction is gradually deleted until the model performance index value of the generalized linear model is optimal, and a plurality of different generalized linear models can be constructed in the process. The generalized linear model in the embodiment is a model for predicting future daily load data of the power distribution network based on daily load parameters of the power distribution equipment; optionally, the different generalized linear models include daily load parameters of different power distribution devices.
It can be understood that the generalized linear model assumes the nonlinear relations in the data as linear relations for processing, so that the complex relations among the nonlinear data are blurred, and the future daily load data of the power distribution network can be predicted conveniently and rapidly in the follow-up process.
And S203, initializing daily load parameters in each generalized linear model by using the historical daily load data of each power distribution device to obtain the predicted daily load data of the power distribution network predicted by each generalized linear model.
Specifically, after the generalized linear model is constructed, for each constructed generalized linear model, historical daily load data of the power distribution equipment required by the generalized linear model can be selected from historical daily load data of each power distribution equipment; and then assigning the daily load parameters of the power distribution equipment contained in the generalized linear model by adopting the selected historical daily load data to obtain the predicted daily load data of the power distribution network predicted by the generalized linear model.
For example, a generalized linear model includes daily load parameters of the power distribution equipment a, the power distribution equipment B, and the power distribution equipment C, and the historical daily load data of the power distribution equipment a, the power distribution equipment B, and the power distribution equipment C are daily load data within one year before the current time; at this time, for each day before the current time, the daily load data of the power distribution equipment A, the power distribution equipment B and the power distribution equipment C on the day can be adopted, the daily load parameters of the power distribution equipment A, the power distribution equipment B and the power distribution equipment C in the generalized linear model are assigned, and the generalized linear model can predict the daily load data of the whole power distribution network on the day.
And S204, selecting a target model from the generalized linear models according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear models.
In this embodiment, the target model refers to a generalized linear model with the best fitting degree between the model screened out from all the established generalized linear models and the data, and is used for predicting future daily load data of the power distribution network according to current daily load data of each power distribution device in the power distribution network.
Optionally, there are many ways to select the target model from the generalized linear models, and this embodiment is not limited by contrast. For example, one possible implementation is to input the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model into a neural network trained in advance, and further select a target model from each generalized linear model based on the evaluation score of each generalized linear model output by the network.
The other implementation mode is that based on preset model selection logic, historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by each generalized linear model are further processed, and then based on processing results, a target model is determined.
Further, after the target model is determined, the current daily load data of each power distribution device in the power distribution network can be substituted into the selected target model to predict the future daily load data of the power distribution network. Specifically, the current daily load data of the power distribution equipment required by the target model is selected from the current daily load data of each power distribution equipment, and the selected current daily load data is substituted into the target model, so that the future daily load data predicted by the target model can be obtained.
In this embodiment, a plurality of generalized linear models are constructed based on historical daily load data of each power distribution device in the power distribution network, and a target model is selected from the plurality of generalized linear models based on the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model to predict future daily load data of the power distribution network. According to the scheme, external data such as population density and humidity are not needed, a model capable of accurately predicting future daily load data of the power distribution network can be built only through historical daily load data of each power distribution device inside the power distribution network, and inaccurate prediction of the future daily load data of the power distribution network caused by poor quality of the external data and low acquisition efficiency is avoided.
On the basis of the above embodiments, as shown in fig. 4, in one embodiment, a target model is selected from the generalized linear models according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model, and the steps include:
s401, aiming at each generalized linear model, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network.
In this embodiment, the model performance index value may include a score of the akabane information criterion AIC, which is used to measure the model fitting superiority, and may be specifically used to balance the complexity of the estimated model and the superiority of the model fitting data.
The number of model parameters is the number of parameters included in the generalized linear model; the total number of the historical daily load data of the power distribution network can also be called as an observation number, and specifically is the total number of days for collecting the historical daily load data of the power distribution network.
One implementation manner is that, for each generalized linear model, historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model, and the total number of the historical daily load data of the power distribution network may be input into a model evaluation network trained in advance, and the model evaluation network may output a model performance index value of the generalized linear model.
The other realization mode is that for each generalized linear model, the sum of squares of residual errors is determined according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear model; and determining the model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
The residual sum of squares is a quantity for measuring the fitting degree of the model in the generalized linear model, and a continuous curve is used for approximately describing or comparing discrete point groups on a plane so as to represent a data processing method of the functional relation between coordinates.
Alternatively, the residual sum of squares SSR can be determined by the following equation 1. Wherein y is historical daily load data of the power distribution network,
Figure BDA0003832173090000081
and the predicted daily load data of the power distribution network predicted by the generalized linear model.
Figure BDA0003832173090000082
After the residual sum of squares is determined, the model performance index value AIC may be determined using the following equation 2. And k is the number of the model parameters, and n is the total number of the historical daily load data of the power distribution network.
AIC =2k +nln (SSR/n) (formula 2)
It can be understood that in the present example, the optimal model of the plurality of generalized linear models can be more accurately selected by calculating the value of AIC, and the problem that the predicted future daily load data of the power distribution network is greatly different from the actual result due to the fact that the model is not good enough is avoided.
S402, selecting a target model from the generalized linear models according to the model performance index values of the generalized linear models.
Alternatively, increasing the number of parameters will improve the goodness of fit, and AIC encourages goodness of data fit but tries to avoid overfitting. The model to be prioritized should be the one with the smallest AIC value. For example, if a selection is made among n models, the AIC values of the n models may be calculated at a time, and the model corresponding to the minimum AIC value may be found as the target model.
In the embodiment, the optimal model can be accurately selected through the model performance index values of the generalized linear models, the inaccuracy of future daily load data prediction of the power distribution network caused by low model fitting goodness is avoided, and the prediction accuracy is improved.
In one embodiment, as shown in fig. 5, the target model is selected from the generalized linear models according to the model performance index value of each generalized linear model, and the steps of this embodiment include:
s501, selecting candidate models from the generalized linear models according to the model performance index values of the generalized linear models.
After calculating the model performance index value of each generalized linear model, for example, the score of the akabane information criterion AIC, it can be found that the AIC value differences of a plurality of generalized linear models are not obvious, and an optimal model cannot be judged, so that the generalized linear models with the part of the AIC value differences which are not obvious are called candidate models, and the next screening is performed.
And S502, determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total quantity of the historical daily load data of the power distribution network.
The error data is used for measuring the difference between the real data and the model prediction data, and particularly can be used for measuring the accuracy of the model prediction data. Optionally, the error data in this embodiment may include a root mean square error and/or an average error.
Alternatively, for each candidate model, the root mean square error RMSE for that candidate model may be determined using equation 3 below. Wherein, y i Is the historical daily load data of the distribution network,
Figure BDA0003832173090000091
the predicted daily load data of the power distribution network predicted for the candidate model, N is the total number of the historical daily load data of the power distribution network, and N is the maximum number of days, such as 365.
Figure BDA0003832173090000092
Further, for each candidate model, the mean error MAPE of the candidate model can be determined using the following equation 4.
Figure BDA0003832173090000093
S503, selecting a target model from the candidate models according to the error data of the candidate models.
Optionally, when the AIC value differences of the plurality of generalized linear models are not obvious, an optimal model cannot be judged, and the values of the root mean square error RMSE and the mean error MAPE serve as selection indexes of the final model to perform the next screening.
Optionally, a model with the smallest root mean square error RMSE and the smallest average error MAPE value is selected from the plurality of candidate models as the target model.
In the embodiment, the target model with the minimum prediction error is selected from the candidate models because the root mean square error RMSE and the average error MAPE are added as the measurement indexes of the models, so that the accuracy of model prediction is improved.
In one embodiment, as shown in fig. 6, the historical daily load data of each power distribution device in the power distribution network is historical daily maximum load data of each power distribution device, and the historical daily load data of the power distribution network is historical daily maximum load data of the power distribution network; correspondingly, the step of acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network comprises the following steps:
s601, determining historical daily maximum load data of each power distribution device according to historical daily original load data of each power distribution device in the power distribution network.
The historical daily raw load data of each power distribution device refers to the hourly load data of each day of the power distribution device in a period of time (such as one month, one quarter or one year) before the current time, for example, the maximum load data of each hour of each day of the power distribution device in a period of time before the current time. Accordingly, the historical daily maximum load data of each power distribution device refers to the maximum load data in the daily load data of the power distribution devices in a period of time before the current time.
Further, historical daily maximum load data for each distribution device may be determined by equation 5 below. Wherein X i Indicating the historical daily maximum load of the ith distribution equipment,
Figure BDA0003832173090000101
and load data of the power distribution equipment i at the time j is represented, namely power distribution data at the time with the maximum load at 24 hours in one day are selected as maximum daily negative data of the day.
Figure BDA0003832173090000102
Optionally, in the process of recording daily load data, many power distribution devices may have the problems of load data loss, large data error and the like due to overhaul, fault and the like. Therefore, according to the historical daily original load data of each power distribution device in the power distribution network, the historical daily maximum load data of each power distribution device can be determined, and missing values and/or abnormal values in the historical daily original load data of each power distribution device can be processed; and determining the historical daily maximum load data of each power distribution device according to the processed historical daily original load data.
The common processing methods include sample deletion, single variable padding, regression padding, multiple padding and the like. Taking the multi-padding method as an example, the multi-padding method includes the steps of generating a group of possible padding values from an original load data containing missing values to form a plurality of complete data sets; performing statistical analysis on the generated complete data, synthesizing the results of each filling data to generate final statistical inference, and introducing a confidence interval of a missing value; and (3) detecting abnormal values of the data, namely the load data is in a 0 load state for a long time or excessively exceeds abnormal points of a load quadridentate value, and replacing the abnormal values by using the multi-filling method.
After the historical daily raw load data of each power distribution device is processed, the maximum daily load data of each power distribution device is calculated.
And S602, determining the historical daily maximum load data of the power distribution network according to the same daily rate and the historical daily original load data of each power distribution device.
The same-day rate refers to the probability that different power distribution devices work together in the same time period, and in this embodiment, the power distribution devices working together in different time periods are different.
Furthermore, as the selected hour levels of the daily maximum daily negative data of each power distribution equipment are different, the daily maximum daily negative data of all the power distribution equipment cannot be simply summed by the historical daily maximum load data of the power distribution network, but the power distribution data of all the equipment in the power distribution network is summed according to the hour levels and multiplied by the coincidence rate, so that the load data of the power distribution network system in each hour is obtained. The hourly summation refers to summing the load data of the power distribution equipment which operates in the same time period. For example, only the power distribution equipment a and the power distribution equipment C work together in the time period of 3-00-4 of the day, the load data of the power distribution equipment a and the power distribution equipment C are added and multiplied by the concurrence rate, and the obtained data is the load data of the power distribution network system in the time period of 3-00-4 of the day.
And after load data of the power distribution network system every hour are obtained, selecting maximum load data of the power distribution network within 24 hours, wherein the selected data are historical daily maximum load data of the power distribution network system.
In the embodiment, the original load data of the historical days of each power distribution device are preprocessed, so that the accuracy of model training is improved, and the accuracy of model prediction is further improved.
In one embodiment, taking 1-year load data of each distribution device in a small distribution network composed of 15 distribution devices in a certain area as an example, the specific implementation includes:
because 15 power distribution equipment are randomly selected, the maximum daily load data of the power distribution equipment fluctuate greatly at random, and the load difference between different equipment is also large, for example, the average value of the maximum daily load of the power distribution equipment 1 and the power distribution equipment 2 is about 2500 and 1.5 respectively, and the random sampling method accords with the difference of power utilization modes of different areas in actual life. For example, the amount of electricity used in residential areas is much smaller than that used in commercial and industrial areas.
The historical daily load data of the initialized 15 pieces of power distribution equipment is put into a generalized linear model, and the generalized model can be shown as an equation 6. Wherein f (x) represents daily maximum load data of the small distribution network predicted by the generalized linear model on the next day, i represents equipment number, and x i Representing the maximum load data of the distribution equipment on the current day, the function g (x) is distributed through the data in the embodiment, a log (x) connection function is used, and the conditional probability distribution is Gaussian distribution.
f(x)＝g(x 1 +x 2 +...+x n ) (formula 6)
The initialized historical daily load data of the power distribution equipment are substituted to obtain a generalized linear model result, then the generalized linear model is further screened through the AIC value, wherein the maximum daily load data correlation of the power distribution network with the prediction of the small power distribution network in the generalized linear model is not large and the number of the power distribution equipment with multiple linear correlations is required to be reduced, and the load prediction of the generalized linear model is more accurate.
4 process models are formed in the model simplification process and are respectively marked as a model 1 to a model 4, wherein the model 1 is an initial model, and so on. And sequentially reducing the number of the power distribution equipment by using a backward stepwise regression method until a generalized linear model combined by the power distribution equipment is found, wherein the AIC value of the generalized linear model is not reduced continuously, and the generalized linear model is determined as an optimal prediction model.
The generalized linear model is further verified, the result that the basic trend of the maximum load data predicted by the generalized linear model is not greatly different from the basic trend of the actual maximum load data, and the average absolute percentage error of the maximum load data is stabilized between 5% and 6% can be obtained, and the prediction result is accurate.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the present application further provides a load prediction apparatus for implementing the above-mentioned load prediction method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so specific limitations in one or more embodiments of the load prediction device provided below can be referred to the limitations of the load prediction method in the foregoing, and details are not described here.
In one embodiment, as shown in fig. 7, there is provided a load prediction apparatus 1 including: a data acquisition module 10, a model construction module 20, an initialization module 30, and a model selection module 40, wherein:
the data acquisition module 10 is used for acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
a model construction module 20 for constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment;
the initialization module 30 is configured to initialize the daily load parameters in each generalized linear model by using the historical daily load data of each power distribution device, so as to obtain predicted daily load data of the power distribution network predicted by each generalized linear model;
the model selection module 40 is used for selecting a target model from the generalized linear models according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
In one embodiment, as shown in FIG. 8, the model selection module 40 of FIG. 7 above includes:
a model index unit 41, configured to determine, for each generalized linear model, a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model, and the total number of the historical daily load data of the power distribution network;
and a model determining unit 42, configured to select a target model from the generalized linear models according to the model performance index value of each generalized linear model.
On the basis of the previous embodiment, in one embodiment, the model index unit 41 is specifically configured to:
and determining the residual square sum according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear model.
And determining the model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
On the basis of the previous embodiment, in one embodiment, the model determining unit 42 is specifically configured to:
and selecting a candidate model from the generalized linear models according to the model performance index value of the generalized linear model.
Determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total number of the historical daily load data of the power distribution network; the error data includes a root mean square error and/or an average error.
And selecting the target model from the candidate models according to the error data of the candidate models.
In one embodiment, as shown in fig. 9, the data acquisition module 10 in fig. 7 includes:
and the data processing unit 11 is configured to determine historical daily maximum load data of each power distribution device according to historical daily original load data of each power distribution device in the power distribution network. Processing missing values and/or abnormal values in the historical daily original load data of each power distribution device; and determining the historical daily maximum load data of each power distribution device according to the processed historical daily original load data.
And the data calculation unit 12 is configured to determine historical daily maximum load data of the power distribution network according to the same-day rate and the historical daily original load data of each power distribution device.
The modules in the load prediction device can be implemented in whole or in part by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 10. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used for storing historical daily load data of each power distribution equipment in the power distribution network and historical daily load data of the power distribution network. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of load prediction. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 10 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network are obtained.
Constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment.
And initializing the daily load parameters in each generalized linear model by adopting the historical daily load data of each power distribution device to obtain the predicted daily load data of the power distribution network predicted by each generalized linear model.
Selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
On the basis of the previous embodiment, when the processor in this embodiment executes the logic of selecting the target model from each generalized linear model according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model in the computer program, the following steps are specifically implemented:
aiming at each generalized linear model, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network; and selecting a target model from the generalized linear models according to the model performance index values of the generalized linear models.
On the basis of the previous embodiment, when the processor in this embodiment executes the logic in the computer program for determining the model performance index value of the generalized linear model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model, and the total number of the historical daily load data of the power distribution network, the following steps are specifically implemented:
determining a residual sum of squares according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear model; and determining the model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
In one embodiment, when the processor executes a computer program to select a target model logic from each generalized linear model according to the model performance index value of each generalized linear model, the following steps are specifically implemented:
selecting a candidate model from each generalized linear model according to the model performance index value of each generalized linear model; determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total number of the historical daily load data of the power distribution network; the error data comprises a root mean square error and/or an average error; and selecting a target model from the candidate models according to the error data of the candidate models.
In one embodiment, when the processor executes logic for acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network in the computer program, the following steps are specifically implemented:
determining historical day maximum load data of each power distribution device according to historical day original load data of each power distribution device in the power distribution network; and determining the historical daily maximum load data of the power distribution network according to the same-day rate and the historical daily original load data of each power distribution device.
In one embodiment, when the processor executes a logic of historical daily maximum load data of each power distribution device in the computer program according to historical daily original load data of each power distribution device in the power distribution network, the following steps are specifically implemented:
processing missing values and/or abnormal values in the historical daily original load data of each power distribution device; and determining the historical daily maximum load data of each power distribution device according to the processed historical daily original load data.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network are obtained.
Constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment.
And initializing the daily load parameters in each generalized linear model by adopting the historical daily load data of each power distribution device to obtain the predicted daily load data of the power distribution network predicted by each generalized linear model.
Selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
On the basis of the previous embodiment, when the computer program is executed by the processor according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model, and a target model is selected from each generalized linear model for operation, the following steps are specifically realized:
for each generalized linear model, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network; and selecting a target model from the generalized linear models according to the model performance index values of the generalized linear models.
On the basis of the last embodiment, when the computer program is executed by the processor to determine the model performance index value of the generalized linear model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network, the following steps are specifically implemented:
determining a residual square sum according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear model; and determining the model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
In one embodiment, when the computer program is executed by the processor to select the target model operation from the generalized linear models according to the model performance index value of the generalized linear model, the following steps are specifically implemented:
selecting a candidate model from each generalized linear model according to the model performance index value of each generalized linear model; determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total number of the historical daily load data of the power distribution network; the error data includes a root mean square error and/or an average error; and selecting the target model from the candidate models according to the error data of the candidate models.
In one embodiment, when the computer program is executed by the processor to obtain the historical daily load data of each power distribution device in the power distribution network and the historical daily load data of the power distribution network, the following steps are specifically implemented:
determining historical day maximum load data of each power distribution device according to historical day original load data of each power distribution device in the power distribution network; and determining the historical daily maximum load data of the power distribution network according to the same daily rate and the historical daily original load data of each power distribution device.
In one embodiment, when the computer program is executed by the processor to obtain the historical daily load data of each power distribution device in the power distribution network and the historical daily load data of the power distribution network, the following steps are specifically implemented:
determining historical day maximum load data of each power distribution device according to historical day original load data of each power distribution device in the power distribution network; and determining the historical daily maximum load data of the power distribution network according to the same-day rate and the historical daily original load data of each power distribution device.
In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:
historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network are obtained.
Constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment.
And initializing daily load parameters in each generalized linear model by using historical daily load data of each power distribution device to obtain predicted daily load data of the power distribution network predicted by each generalized linear model.
Selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
On the basis of the previous embodiment, when the computer program is executed by the processor according to the historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by each generalized linear model, and a target model is selected from each generalized linear model for operation, the following steps are specifically realized:
for each generalized linear model, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network; and selecting a target model from the generalized linear models according to the model performance index values of the generalized linear models.
On the basis of the last embodiment, when the computer program is executed by the processor to determine the model performance index value of the generalized linear model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network, the following steps are specifically implemented:
determining a residual sum of squares according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear model; and determining the model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
In one embodiment, when the computer program is executed by the processor to select the target model operation from the generalized linear models according to the model performance index value of the generalized linear model, the following steps are specifically implemented:
selecting a candidate model from each generalized linear model according to the model performance index value of each generalized linear model; determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total number of the historical daily load data of the power distribution network; the error data comprises a root mean square error and/or an average error; and selecting the target model from the candidate models according to the error data of the candidate models.
In one embodiment, when the computer program is executed by the processor to obtain the historical daily load data of each power distribution device in the power distribution network and the historical daily load data of the power distribution network, the following steps are specifically implemented:
determining historical day maximum load data of each power distribution device according to historical day original load data of each power distribution device in the power distribution network; and determining the historical daily maximum load data of the power distribution network according to the same-day rate and the historical daily original load data of each power distribution device.
In one embodiment, the computer program when executed by the processor is further operable to determine historical daily maximum load data for each distribution device in the power distribution network based on historical daily raw load data for each distribution device, and further operable to:
processing missing values and/or abnormal values in the historical daily original load data of each power distribution device; and determining the historical daily maximum load data of each power distribution device according to the processed historical daily original load data.
It should be noted that the data related to the power distribution network (including, but not limited to, historical daily load data and current daily load data of the power distribution equipment, and historical daily load data of the power distribution network, etc.) are all data authorized by the user or fully authorized by each party.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include a Read-Only Memory (ROM), a magnetic tape, a floppy disk, a flash Memory, an optical Memory, a high-density embedded nonvolatile Memory, a resistive Random Access Memory (ReRAM), a Magnetic Random Access Memory (MRAM), a Ferroelectric Random Access Memory (FRAM), a Phase Change Memory (PCM), a graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases involved in the embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims (10)

1. A method of load prediction, the method comprising:
acquiring historical daily load data of each power distribution device in a power distribution network and historical daily load data of the power distribution network;
constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment;
initializing daily load parameters in the generalized linear models by using historical daily load data of each power distribution device to obtain predicted daily load data of the power distribution network predicted by each generalized linear model;
selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
2. The method of claim 1, wherein selecting the target model from the generalized linear models based on historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear models comprises:
for each generalized linear model, determining a model performance index value of the generalized linear model according to historical daily load data of the power distribution network, predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model and the total number of the historical daily load data of the power distribution network;
and selecting a target model from the generalized linear models according to the model performance index values of the generalized linear models.
3. The method of claim 2, wherein the determining the model performance index value of the generalized linear model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the generalized linear model, the number of model parameters of the generalized linear model, and the total number of the historical daily load data of the power distribution network comprises:
determining a residual sum of squares according to historical daily load data of the power distribution network and predicted daily load data of the power distribution network predicted by the generalized linear model;
and determining a model performance index value of the generalized linear model according to the residual sum of squares, the number of model parameters of the generalized linear model and the total number of historical daily load data of the power distribution network.
4. The method of claim 2, wherein selecting the target model from the generalized linear models based on the model performance metric values for the generalized linear models comprises:
selecting a candidate model from each generalized linear model according to the model performance index value of each generalized linear model;
determining error data of the candidate model according to the historical daily load data of the power distribution network, the predicted daily load data of the power distribution network predicted by the candidate model and the total number of the historical daily load data of the power distribution network; the error data comprises a root mean square error and/or an average error;
and selecting a target model from the candidate models according to the error data of the candidate models.
5. The method according to any one of claims 1-4, wherein the historical daily load data of each power distribution device in the power distribution network is historical daily maximum load data of each power distribution device, and the historical daily load data of the power distribution network is historical daily maximum load data of the power distribution network;
correspondingly, the acquiring historical daily load data of each power distribution device in the power distribution network and the historical daily load data of the power distribution network includes:
determining historical day maximum load data of each power distribution device according to historical day original load data of each power distribution device in the power distribution network;
and determining the historical daily maximum load data of the power distribution network according to the same daily rate and the historical daily original load data of each power distribution device.
6. The method of claim 5, wherein determining historical daily maximum load data for each distribution device in the distribution network based on historical daily raw load data for each distribution device comprises:
processing missing values and/or abnormal values in the historical daily original load data of each power distribution device;
and determining the historical daily maximum load data of each power distribution device according to the processed historical daily original load data.
7. A load prediction apparatus, characterized in that the apparatus comprises:
the data acquisition module is used for acquiring historical daily load data of each power distribution device in the power distribution network and historical daily load data of the power distribution network;
the model construction module is used for constructing at least two different generalized linear models; wherein, the different generalized linear models comprise daily load parameters of different power distribution equipment;
the initialization module is used for initializing the daily load parameters in the generalized linear models by adopting the historical daily load data of the power distribution equipment to obtain the predicted daily load data of the power distribution network predicted by the generalized linear models;
the model selection module is used for selecting a target model from the generalized linear models according to historical daily load data of the power distribution network and the predicted daily load data of the power distribution network predicted by the generalized linear models; the target model is used for predicting future daily load data of the power distribution network according to the current daily load data of each power distribution device in the power distribution network.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
CN202211077451.6A 2022-09-05 2022-09-05 Load prediction method, load prediction device, computer equipment and storage medium Pending CN115473219A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211077451.6A CN115473219A (en) 2022-09-05 2022-09-05 Load prediction method, load prediction device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211077451.6A CN115473219A (en) 2022-09-05 2022-09-05 Load prediction method, load prediction device, computer equipment and storage medium

Publications (1)

Publication Number Publication Date
CN115473219A true CN115473219A (en) 2022-12-13

Family

ID=84371043

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211077451.6A Pending CN115473219A (en) 2022-09-05 2022-09-05 Load prediction method, load prediction device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN115473219A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116979531A (en) * 2023-09-25 2023-10-31 山西京能售电有限责任公司 Novel energy data monitoring method and method for monitoring auxiliary power market

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116979531A (en) * 2023-09-25 2023-10-31 山西京能售电有限责任公司 Novel energy data monitoring method and method for monitoring auxiliary power market
CN116979531B (en) * 2023-09-25 2023-12-12 山西京能售电有限责任公司 Novel energy data monitoring method and method for monitoring auxiliary power market

Similar Documents

Publication Publication Date Title
Wisesa et al. Prediction analysis sales for corporate services telecommunications company using gradient boost algorithm
CN107506868B (en) Method and device for predicting short-time power load
CN109272142A (en) Data predication method, device, computer equipment and storage medium based on event
CN113255973A (en) Power load prediction method, power load prediction device, computer equipment and storage medium
CN115641019A (en) Index anomaly analysis method and device, computer equipment and storage medium
CN115473219A (en) Load prediction method, load prediction device, computer equipment and storage medium
CN115689018A (en) Material demand prediction method, device, equipment, storage medium and program product
CN116191398A (en) Load prediction method, load prediction device, computer equipment and storage medium
CN116933035A (en) Data anomaly detection method, device, computer equipment and storage medium
CN113837467B (en) Point-to-point energy project evaluation method and device, computer equipment and storage medium
CN116316617A (en) Multi-station intelligent fusion new energy generation power region prediction method and system
CN116468151A (en) Carbon emission calculation method, carbon emission calculation device, computer equipment and storage medium
CN115564152A (en) Carbon emission prediction method and device based on STIRPAT model
CN114784795A (en) Wind power prediction method and device, electronic equipment and storage medium
CN115481767A (en) Operation data processing method and device for power distribution network maintenance and computer equipment
CN114925919A (en) Service resource processing method and device, computer equipment and storage medium
CN112256735B (en) Power consumption monitoring method and device, computer equipment and storage medium
CN115549082A (en) Power load prediction method based on load mining and LSTM neural network
CN115511562A (en) Virtual product recommendation method and device, computer equipment and storage medium
CN114140158A (en) Power distribution network investment demand determination method, device, equipment and storage medium based on combination prediction
Chang et al. Solving quantile-based stochastic optimization problems with modified Stochastic Nelder-Mead Simplex Method
CN116382926B (en) Method and device for computing power optimization of recommendation system
CN115860273B (en) Method, apparatus, computer device and storage medium for predicting electric load
CN115204501A (en) Enterprise evaluation method and device, computer equipment and storage medium
CN117437104A (en) County carbon emission prediction method, county carbon emission prediction device, county carbon emission prediction computer device, and county carbon emission prediction storage medium

Legal Events

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