CN112365280B - Electric power demand prediction method and device - Google Patents
Electric power demand prediction method and device Download PDFInfo
- Publication number
- CN112365280B CN112365280B CN202011124185.9A CN202011124185A CN112365280B CN 112365280 B CN112365280 B CN 112365280B CN 202011124185 A CN202011124185 A CN 202011124185A CN 112365280 B CN112365280 B CN 112365280B
- Authority
- CN
- China
- Prior art keywords
- power demand
- dominant
- demand prediction
- meteorological
- support vector
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012706 support-vector machine Methods 0.000 claims abstract description 83
- 230000006870 function Effects 0.000 claims description 23
- 238000004422 calculation algorithm Methods 0.000 claims description 22
- 238000007637 random forest analysis Methods 0.000 claims description 21
- 239000013598 vector Substances 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 16
- 238000012360 testing method Methods 0.000 claims description 15
- 238000012795 verification Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 description 20
- 238000010586 diagram Methods 0.000 description 14
- 230000000875 corresponding effect Effects 0.000 description 9
- 230000008859 change Effects 0.000 description 7
- 230000005611 electricity Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000002596 correlated effect Effects 0.000 description 3
- 238000003066 decision tree Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010206 sensitivity analysis Methods 0.000 description 3
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013642 negative control Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000005612 types of electricity Effects 0.000 description 1
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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a method and a device for predicting power demand, wherein the method comprises the following steps: acquiring meteorological factors of a period to be predicted of a target area; obtaining dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor recognition model; obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established support vector machine power demand prediction model; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature samples and power load feature samples. The invention can improve the accuracy of power demand prediction.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for predicting power demand.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
The electric load is one of important indexes of planning design and operation management of an electric power system, and the research of the characteristics of the load and the change rule thereof is a primary condition for achieving safe, stable, high-quality and economic operation of a power grid.
At present, there are various solutions for analyzing and predicting the power demand, such as regression models, neural network methods, etc., but these prediction methods have the problem of low prediction accuracy.
Disclosure of Invention
The embodiment of the invention provides a power demand prediction method for improving the precision of power demand prediction, which comprises the following steps:
acquiring meteorological factors of a period to be predicted of a target area;
obtaining dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor recognition model;
obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established support vector machine power demand prediction model; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature samples and power load feature samples.
The embodiment of the invention also provides a device for predicting the power demand, which is used for improving the precision of the power demand prediction, and comprises the following steps:
The acquisition unit is used for acquiring meteorological factors of a period to be predicted of the target area;
The identification unit is used for obtaining the dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor identification model;
the prediction unit is used for obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established power demand prediction model of the support vector machine; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature samples and power load feature samples.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the method for predicting the power demand when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium storing a computer program for executing the above-described power demand prediction method.
In the embodiment of the invention, compared with the technical scheme of predicting power demand analysis by using regression models, neural networks and the like in the prior art, the prediction scheme of the power demand is characterized by comprising the following steps: acquiring meteorological factors of a period to be predicted of a target area; obtaining dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor recognition model; obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established support vector machine power demand prediction model; the support vector machine power demand prediction model is established in advance according to a plurality of dominant meteorological feature samples and power load feature samples, so that the dominant meteorological factors are accurately identified, the support vector machine is adopted to establish the power demand prediction model based on analysis of the dominant meteorological factors, further, accurate prediction of the power demand is achieved, and the prediction precision of the power demand is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 is a flow chart of a method for predicting power demand according to an embodiment of the invention;
FIG. 2 is a schematic diagram of the weather effect of summer industrial load types according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the weather exposure of summer business load types in an embodiment of the present invention;
FIG. 4 is a schematic diagram showing the degree of weather influence on the load types of residents in summer according to the embodiment of the invention;
FIG. 5 is a schematic view of weather-affected degrees of winter industrial load types in an embodiment of the present invention;
FIG. 6 is a schematic diagram of the weather exposure of the winter business load types in an embodiment of the present invention;
FIG. 7 is a schematic view of weather-affected degree of load types of winter residents in an embodiment of the present invention;
FIG. 8 is a schematic diagram of the correspondence between power load and temperature according to an embodiment of the present invention;
FIG. 9 is a schematic diagram illustrating analysis of load prediction results according to an embodiment of the present invention;
Fig. 10 is a schematic structural diagram of a power demand prediction apparatus according to an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.
The inventors found that: the analysis and prediction of current power demand has the following problems: firstly, factors influencing the power load are various, and particularly quite a large number of factors cannot be accurately and quantitatively given, and the change characteristics of the power load are represented by time variability, air variability, randomness, complexity and diversity; secondly, in recent years, the lack of electricity has become a problem of social concern, with the continuous improvement of the living standard of people and the gradual adjustment of the industrial structure, the specific gravity of resident electricity and third industry electricity is continuously increased, the two types of electricity are indistinguishable from weather conditions, and the specific gravity of the total electricity demand is further increased, so that the relationship between the change of the electric load and the weather conditions is more intimate, but the economic structures and the development levels in all places are also quite different due to the different weather conditions of different electric grids, and the relationship between the electric grid load and weather factors is also quite different; finally, the relation between the regional or urban power load and the meteorological factors is quite complex, the relation between the power load and the meteorological factors in different regions is different, and the dominant meteorological factors also change along with the regional change.
Because the inventor discovers the technical problem existing in the existing power demand prediction, the relation between the power load and the meteorological factors is necessary to be further researched, a prediction scheme based on the meteorological factors is established, and a basis is provided for analysis, prediction and operation scheduling of the power demand. Accordingly, the inventors propose a prediction scheme of power demand.
The power demand prediction scheme provided by the embodiment of the invention mainly aims to solve the problems that the existing power demand has a plurality of weather influencing factors, and main weather influencing factors are difficult to clearly identify, so that the power demand prediction precision is insufficient. According to the power demand prediction scheme provided by the embodiment of the invention, namely the power demand sensitivity analysis and prediction method based on the meteorological factor characteristics, the dominant meteorological factors can be identified, the association relation between the meteorological factors and the power demand is excavated, and the power demand prediction precision is improved.
The power demand prediction scheme provided by the embodiment of the invention is a power demand sensitivity analysis and prediction method based on meteorological factor characteristics, and has the main functions as follows:
(1) The association relation between the meteorological factors and the power requirements is rapidly analyzed, and dominant meteorological factors are identified;
(2) And combining the identified dominant factors, establishing a power demand prediction model (support vector machine power demand prediction model) based on meteorological factors, and realizing power demand sensitivity analysis and prediction.
According to the method for analyzing and predicting the power demand sensitivity based on the meteorological factor characteristics, which is provided by the embodiment of the invention, the large customer information collected by the negative control system is classified according to the classification standard (in order to obtain the dominant meteorological feature sample and the power load feature sample), then the association relation between each meteorological factor index and the power demand is analyzed by adopting a random forest algorithm, and a meteorological factor index library and a dominant factor identification system are established; analyzing the load characteristics of different types of loads, such as industrial and residential loads, extracting feature vectors, and establishing an electric quantity demand prediction model (support vector machine electric power demand prediction model) based on dominant meteorological factors by adopting a support vector machine technology to predict future electric quantity demand conditions.
Specifically, firstly, weather factors such as the highest temperature, the lowest temperature, the average temperature, the wind speed and the air pressure are identified based on a random forest algorithm, then, the influence and the load characteristics of the change of the electric quantity demand are analyzed, then, the data set is normalized and divided into a training set, a testing set and a verification set, a support vector machine electric quantity demand prediction model is obtained through training, and then, the support vector machine electric quantity demand prediction model is used for prediction, so that an accurate electric quantity demand prediction result is obtained. The prediction scheme of the power demand will be described in detail.
Fig. 1 is a flow chart of a method for predicting power demand according to an embodiment of the invention, as shown in fig. 1, the method includes the following steps:
step 101: acquiring meteorological factors of a period to be predicted of a target area;
Step 102: obtaining dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor recognition model;
Step 103: obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established support vector machine power demand prediction model; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature samples and power load feature samples.
In the embodiment of the invention, compared with the technical scheme of predicting power demand analysis by using regression models, neural networks and the like in the prior art, the prediction scheme of the power demand realizes accurate identification of dominant weather factors, and based on the analysis of the dominant weather factors, a support vector machine is adopted to establish a prediction model of the power demand, thereby realizing accurate prediction of the power demand and improving the prediction precision of the power demand.
The following describes in detail the steps involved in the embodiments of the present invention.
1. First, the step of pre-modeling is introduced.
1. Identification of dominant meteorological factors (factors)
The current numerical weather data comprises a plurality of weather indexes, wherein certain weather indexes can have great influence on the power demand, so that the embodiment of the invention adopts a correlation analysis method to identify the power load and the dominant factors of the power demand. The correlation analysis method utilizes a Pearson coefficient calculation formula to calculate a correlation coefficient, so that the correlation degree between the weather influence factor M and the power load P can be reflected, the correlation coefficient r PM has no dimension, and the value of the correlation coefficient r PM is within the range of [ -1,1 ]. When r PM =0, there is no correlation between P and M, and P and M are not correlated; when r PM >0, P increases as M increases, which is called P positively correlated with M; when r PM <0, P decreases as M increases, which is said to be inversely related to M; when r PM =1, p can be expressed exactly as a linear function of the variable M, its calculation formula is:
Wherein n represents the number of time series, and P i represents the load value at the ith moment; Mean load values are shown; m i represents the weather value at the ith moment, and generally takes a plurality of single/comprehensive weather factors (factors) such as temperature, humidity and the like; /(I) The average meteorological factor values are shown.
In specific implementation, the above formula (1) can be a dominant meteorological factor identification model.
2. Dominant meteorological factors and load feature sample extraction and classification
The embodiment of the invention adopts random forests to extract the characteristics of power load data and dominant weather index data (dominant weather factor data), forms a weather characteristic library and a load characteristic library, constructs a data characteristic sample library, stores sample data in a classified manner, and provides samples for prediction (samples for power demand prediction: dominant weather characteristic samples and power load characteristic samples).
The RF adopts a repeated sampling (Bootstrap Sampling) mode with a put back, randomly extracts a plurality of samples from the initial sample set B to generate a new sample subset, and then generates a forest set consisting of k decision trees according to each sub-sample. The essence of the method is a combined decision tree algorithm, and a model with stronger generalization learning capability is obtained by arranging and combining a plurality of decision trees. Typically, each tree in the forest has the same distribution, and the error of the fit depends on the correlation between the learning capabilities of each tree. During the sampling process, the remaining samples that are not being sampled are out-of-bag samples and are defined as a collectionWherein C and/>B and/>, respectivelyIs a subset of the subset of (c). Assuming X n+p is an n-dimensional dataset with p features and y is an n-dimensional tag vector, the RF algorithm calculates the importance of the features by rearranging the fitting errors before and after the features. There are Q out-of-bag sample sets as test sets when building Q trees. The feature importance index S is calculated as follows:
Wherein x j is a characteristic parameter (which may be dominant meteorological factor data and power load data characteristics); y i is the fitting property of the ith out-of-bag data, I is the error representation function, h k (I) is the sample fitting property parameter predicted from dataset B, The fitting attribute parameters are obtained by replacing the characteristic parameters x j.
In specific implementation, the above formula (2) can identify important features (key features described below) as samples for subsequently establishing a support vector machine power demand prediction model, and a support vector machine power demand prediction model with higher prediction accuracy can be obtained, so that a high-accuracy power demand prediction result is obtained through prediction of the high-accuracy model.
3. And establishing a Support Vector Machine (SVM) power demand prediction model.
According to the embodiment of the invention, a support vector machine is adopted, sample data in a data feature library is used as input quantity and output quantity, and a training prediction model is continuously optimized to obtain an optimal prediction model. The basic thought of using a support vector machine to classify samples is as follows: and mapping the linear inseparable disturbance signal input feature vector to a feature space with higher dimensionality, and establishing an optimal hyperplane (HYPER PLANE) in the high-dimensional feature space, so that the hyperplane is the largest in distance from the edge of a sample to be separated, and the classification accuracy is ensured to the greatest extent under the condition of smaller training set.
The support vector machine is derived from the two classification problem, for a two classification sample set (x i,yi),x∈Rd, e { -1,1}, where x i is the sample vector to be classified and y i represents the sample label, creating a classification hyperplane w x + b = 0,w is the normal vector of the hyperplane, b is a real number, and x is the d-dimensional input sample size.
Solving the problem of an optimal hyperplane can be expressed as:
s.t.y[(w×xi)+b]-1≥0,i＝1,2,...,n (4)
where y is the sample tag.
Defining a Lagrangian function:
where a is the Lagrangian coefficient (> 0) and b is the bias. Solving the optimization problem under constraint conditions by utilizing a dual principle, and further solving an optimal classification function formula as follows:
Where a * represents the lagrangian coefficient (> 0) and b * represents the threshold for sample classification. x i represents various meteorological factors and x represents sample characteristics. For the linear inseparable case, a penalty factor u is introduced, along with a relaxation variable, so the solution problem of the generalized optimal classification plane can be translated into:
in the formula, ζ represents a deviation.
The support vector machine maps the inseparable sample vector in the low-dimensional space into the high-dimensional feature space through the nonlinear mapping operation of the kernel function, converts the dot product operation (x i.x) in the optimal classification plane into the calculation of the kernel function K (x i.x) to obtain a discriminant function, the kernel function is crucial to the construction of the support vector machine, and the relation between the load and the influencing factors not only comprises a linear relation, but also has a complex nonlinear relation, so the embodiment of the invention adopts a radial basis kernel function, has excellent performance, can treat nonlinear problems more quickly and high-quality, and has the expression:
in the formula, gamma represents a coefficient corresponding to the kernel function type, and can convert low-dimensional data into high-dimensional data and clearly divide the data.
Finally, the prediction of the predicted power demand is performed according to the regression equation of the support vector machine, as shown in the formula (9):
where a is the Lagrangian coefficient, b is the threshold for sample classification, i is the time of day, As a kernel function, x is a meteorological factor of a period to be predicted of a target area, and m is the number of samples.
In specific implementation, according to the principle of the support vector machine, it can be known that a regression equation is finally generated, as shown in formula (9), the equation (9) can be a support vector machine power demand prediction model, the equation can obtain prediction of power demand according to meteorological factors x, that is, the input of the support vector machine power demand prediction model is the dominant meteorological factor of the target area to be predicted period, and the output of the model can be the power demand of the target area to be predicted period.
In summary, in one embodiment, the method for predicting power demand may further include: the power demand prediction model of the support vector machine is established according to the following method:
Acquiring sample data;
dividing sample data into a training set, a testing set and a verification set;
training the support vector machine model by using a training set to obtain a support vector machine power demand prediction model;
testing the support vector machine power demand prediction model by using a test set to obtain a tested support vector machine power demand prediction model;
And verifying the tested support vector machine power demand prediction model by using a verification set to obtain the pre-established support vector machine power demand prediction model.
In specific implementation, the above detailed implementation of obtaining the pre-established support vector machine power demand prediction model further improves the accuracy of power demand prediction.
In one embodiment, acquiring sample data may include:
According to historical meteorological factor data and historical power load data of a preset period and a pre-established dominant meteorological factor recognition model, historical dominant meteorological factor data and corresponding historical power load data are obtained;
Extracting characteristics of historical dominant meteorological factor data and corresponding historical power load data by adopting a random forest algorithm to form a dominant meteorological characteristic library and a power load characteristic library;
And obtaining sample data of power demand prediction according to the dominant meteorological feature library and the power load feature library.
In specific implementation, the historical dominant meteorological factor data are obtained through identifying the historical data, the characteristics of the historical dominant meteorological factor data and the corresponding historical power load data are extracted through adopting a random forest algorithm, important characteristics (key characteristics) can be obtained, a dominant meteorological characteristic library and a power load characteristic library are built, and finally the accuracy of power demand prediction is further improved through the detailed implementation of the sample data for power demand prediction.
In one embodiment, a random forest algorithm is adopted to extract characteristics of historical dominant meteorological factor data and corresponding historical power load data thereof to form a dominant meteorological characteristic library and a power load characteristic library, and the method can comprise the following steps:
and extracting key features of the history dominant meteorological factors and the corresponding history power load data thereof as key feature vectors of the history dominant meteorological factors and the corresponding history power load data thereof by adopting a random forest algorithm according to the following formula:
In the specific implementation, the importance index data are obtained according to the formula, so that the key feature vector is obtained, the meteorological feature library and the power load feature library are dominant, and the accuracy of power demand prediction is further improved.
2. The above step 101 is described.
In particular, the target area may be any area where a predicted power demand is required, such as Beijing changping; the period to be predicted may be any future period, for example, 3 months in the future.
3. The above step 102 is described.
In specific implementation, the specific implementation of this step 102 may refer to the implementation of identifying dominant weather factors in the modeling process, which is not described herein.
4. The above step 103 is described.
In one embodiment, obtaining the power demand of the target area to be predicted period according to the dominant meteorological factors of the target area to be predicted period and a pre-established support vector machine power demand prediction model may include:
Extracting the characteristics of the dominant meteorological factors by adopting a random forest algorithm to obtain the characteristic vectors of the dominant meteorological factors;
and inputting the feature vector of the dominant meteorological factors into a pre-established support vector machine power demand prediction model to obtain the power demand of the target area in the period to be predicted.
In specific implementation, the input of the support vector machine power demand prediction model is a meteorological factor, and the output is a prediction result of power demand.
In specific implementation, the pre-established power demand prediction model of the support vector machine can be regarded as two parts, wherein the first part is a feature extraction part, the first part firstly adopts a random forest algorithm to extract features of dominant meteorological factors to obtain feature vectors of the dominant meteorological factors to obtain important feature vectors of the dominant meteorological factors, and then the second part is a prediction part, and the feature vectors of the dominant meteorological factors are input into the pre-established power demand prediction model of the support vector machine to obtain a prediction result. This particular predictive embodiment further improves the accuracy of the prediction of power demand.
In one embodiment, a random forest algorithm is used to extract features of dominant weather factors and obtain feature vectors of the dominant weather factors, including:
And extracting key features of the dominant meteorological factors as key feature vectors of the dominant meteorological factors (feature vectors of important dominant meteorological factors) according to the following formula by adopting a random forest algorithm:
In specific implementation, the important (key) feature vector of the dominant meteorological factor is obtained through the formula, so that the subsequent power demand prediction is performed, and the precision of the power demand prediction is further improved.
5. A step of further updating the model after the above step 103 is described.
In one embodiment, the method for predicting power demand may further include:
Acquiring a current power demand prediction result;
And adjusting the power demand prediction model of the support vector machine according to the current power demand prediction result to obtain an updated power demand prediction model of the support vector machine.
When the method is specifically implemented, the current power demand prediction result is a real-time power demand prediction result, and the support vector machine power demand prediction model is adjusted according to the current power demand prediction result to obtain a model with higher prediction precision, so that power demand prediction is performed, and the precision of power demand prediction is further improved.
In the specific implementation, after the prediction model process and the prediction model are optimized, a post-evaluation system is utilized to feed back the prediction result in real time, and the model is dynamically adjusted, so that a dynamic optimization model is realized, and the prediction precision is kept. The evaluation indexes comprise root mean square error and average relative error for evaluation, and the specific formulas are as follows:
(1) Root mean square error
Is the square root of the ratio of the sum of squares of the observed value and the true value deviation to the number of observations m.
Is used to measure the deviation between the observed value and the true value.
Where P is represents an actual load value and P iy represents a predicted load value.
(2) Average relative error
The average relative error is the average of the relative errors, which is typically expressed in absolute terms, i.e., the absolute value of the average relative error.
Where n represents the number of predicted loads, P is represents the actual load value, and P iy represents the predicted load value.
In order to facilitate understanding, referring to fig. 2 to fig. 9, an example of a power demand prediction method provided by an embodiment of the present invention will be described in detail.
Examples: the analysis is performed on three types of loads, typically commercial industry, business, and residential.
1. Weather analysis of seasonal load
And (5) respectively calculating the association degree of the residential, commercial and industrial loads and the summer weather information in winter by using a random forest algorithm. As can be seen from fig. 2 to 7, the load change has the greatest correlation with temperature information, and has a certain correlation with relative humidity, so that the influence of air pressure and wind speed on various loads is small. Meanwhile, the load of residents is more sensitive to weather changes, and the highest temperature in summer and the lowest temperature in winter have great influence on the electricity consumption of the residents. Since commercial and industrial loads are more regular and are relatively less affected by weather than residential loads, where industrial loads are more susceptible to temperature, production plans are somewhat correlated to average temperatures.
The relationship of power load to temperature was further analyzed as shown in fig. 8. There is a linear relationship between temperature and electrical load, and the prevailing temperature in the region is more comfortable, with some periods of lower temperature. When the temperature is lower, the use of an air conditioner or other heating equipment increases the power load demand, and the accuracy of the random forest algorithm on the weather characteristic relevance analysis is proved through the analysis side of the relation between the air conditioner and the other heating equipment.
2. Weather analysis of seasonal load
Based on the analysis of weather characteristics, the respective predictions of load are shown in fig. 9, wherein the test data range analysis selects daily load amounts for analysis within one month. Through the experimental comparison analysis shown in fig. 9, the prediction effect of the RF-SVM model in the embodiment of the present invention is better than that of the neural network model. This is because, after correlation of meteorological features by RF analysis, features having a greater influence on future trend predictions can be selected, so that the model has good generalization ability and prediction effect.
In summary, the power demand prediction method provided by the embodiment of the invention realizes:
(1) The invention relies on classification standards and electric power big data technology to realize intelligent classification of big customer information;
(2) When a plurality of influence factors exist, the method can adopt a random forest algorithm, and adopts a mode of replacing and repeated sampling to extract the load data and the characteristic vectors of all meteorological index data, so as to identify the dominant factors and establish an influence factor index library;
(3) According to the method, based on analysis of influence factors, a prediction model of the power demand is established by adopting the support vector machine, so that accurate prediction of the power demand is realized, and the prediction precision is improved.
The embodiment of the invention also provides a device for predicting the power demand, as described in the following embodiment. Since the principle of the device for solving the problem is similar to that of the prediction method of the power demand, the implementation of the device can be referred to the implementation of the method, and the repetition is omitted.
Fig. 10 is a schematic structural diagram of a power demand prediction apparatus according to an embodiment of the present invention, as shown in fig. 10, the apparatus includes:
An obtaining unit 01, configured to obtain weather factors of a period to be predicted in a target area;
the recognition unit 02 is used for obtaining dominant weather factors of the target area to-be-predicted period according to weather factors of the target area to-be-predicted period and a pre-established dominant weather factor recognition model;
The prediction unit 03 is configured to obtain a power demand of the target area in the period to be predicted according to a dominant meteorological factor of the target area in the period to be predicted and a pre-established power demand prediction model of the support vector machine; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature samples and power load feature samples.
In one embodiment, the prediction apparatus of power demand may further include: the building unit is used for building a power demand prediction model of the support vector machine according to the following method:
Acquiring sample data;
dividing sample data into a training set, a testing set and a verification set;
training the support vector machine model by using a training set to obtain a support vector machine power demand prediction model;
testing the support vector machine power demand prediction model by using a test set to obtain a tested support vector machine power demand prediction model;
And verifying the tested support vector machine power demand prediction model by using a verification set to obtain the pre-established support vector machine power demand prediction model.
In one embodiment, acquiring sample data may include:
According to historical meteorological factor data and historical power load data of a preset period and a pre-established dominant meteorological factor recognition model, historical dominant meteorological factor data and corresponding historical power load data are obtained;
Extracting characteristics of historical dominant meteorological factor data and corresponding historical power load data by adopting a random forest algorithm to form a dominant meteorological characteristic library and a power load characteristic library;
And obtaining sample data of power demand prediction according to the dominant meteorological feature library and the power load feature library.
In one embodiment, the obtaining unit may be further configured to obtain a current power demand prediction result;
The prediction apparatus of power demand may further include: and the adjusting unit is used for adjusting the power demand prediction model of the support vector machine according to the current power demand prediction result to obtain an updated power demand prediction model of the support vector machine.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the method for predicting the power demand when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium storing a computer program for executing the above-described power demand prediction method.
In the embodiment of the invention, compared with the technical scheme of predicting power demand analysis by using regression models, neural networks and the like in the prior art, the prediction scheme of the power demand realizes accurate identification of dominant weather factors, and based on the analysis of the dominant weather factors, a support vector machine is adopted to establish a prediction model of the power demand, thereby realizing accurate prediction of the power demand and improving the prediction precision of the power demand.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (6)
1. A method of predicting power demand, comprising:
acquiring meteorological factors of a period to be predicted of a target area;
obtaining dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor recognition model;
Obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established support vector machine power demand prediction model; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature sample data and power load feature sample data, and is an RF-SVM model; wherein the process of acquiring the sample data comprises: according to historical meteorological factor data and historical power load data of a preset period and a pre-established dominant meteorological factor recognition model, historical dominant meteorological factor data and corresponding historical power load data are obtained; analyzing the load characteristics of different types of power loads by adopting a random forest algorithm, extracting historical dominant meteorological factor data and the characteristics of the corresponding historical power load data, and forming a dominant meteorological characteristic library and a power load characteristic library; obtaining sample data of power demand prediction according to the dominant meteorological feature library and the power load feature library; types of electrical loads include: industrial load type, commercial load type, resident load type for each season;
The method for extracting the characteristics of the historical dominant meteorological factor data and the corresponding historical power load data by adopting a random forest algorithm to form a dominant meteorological characteristic library and a power load characteristic library comprises the following steps: and extracting key features of the history dominant meteorological factors and the corresponding history power load data thereof as key feature vectors of the history dominant meteorological factors and the corresponding history power load data thereof by adopting a random forest algorithm according to the following formula:
Wherein x j is a characteristic parameter, which is the characteristic of dominant meteorological factor data and power load data; y i is the fitting property of the ith out-of-bag data, I is the error representation function, h k (I) is the sample fitting property parameter predicted from dataset B, Fitting attribute parameters obtained after the characteristic parameters x j are replaced;
the support vector machine power demand prediction model is as follows: wherein: a is Lagrangian coefficient, b is a threshold for sample classification, i is time,/> As a kernel function, x is a dominant meteorological factor of a period to be predicted of a target area, and m is the number of samples; the kernel function is: /(I)Wherein, gamma represents the coefficient corresponding to the kernel function type so as to clearly divide the data;
the method for predicting the power demand further comprises the following steps: acquiring a current power demand prediction result; the current power demand prediction result is a real-time power demand prediction result; according to the current power demand prediction result, adjusting a power demand prediction model of the support vector machine to obtain an updated power demand prediction model of the support vector machine; and the post-evaluation system is utilized to feed back a prediction result in real time, dynamically adjust the model and realize a dynamic optimization model so as to maintain the prediction precision, the evaluation index is evaluated by a root mean square error and an average relative error, the root mean square error is the square root of the ratio of the square sum of the observed value and the true value deviation to the observed times, the average relative error is the average value of the relative error, and the average relative error is expressed by the absolute value of the average relative error.
2. The method of predicting power demand of claim 1, further comprising: the power demand prediction model of the support vector machine is established according to the following method:
Acquiring sample data;
dividing sample data into a training set, a testing set and a verification set;
training the support vector machine model by using a training set to obtain a support vector machine power demand prediction model;
testing the support vector machine power demand prediction model by using a test set to obtain a tested support vector machine power demand prediction model;
And verifying the tested support vector machine power demand prediction model by using a verification set to obtain the pre-established support vector machine power demand prediction model.
3. A prediction apparatus of power demand, comprising:
The acquisition unit is used for acquiring meteorological factors of a period to be predicted of the target area;
The identification unit is used for obtaining the dominant weather factors of the target area to-be-predicted period according to the weather factors of the target area to-be-predicted period and a pre-established dominant weather factor identification model;
The prediction unit is used for obtaining the power demand of the target area to be predicted in the period according to the dominant meteorological factors of the target area to be predicted in the period and a pre-established power demand prediction model of the support vector machine; the support vector machine power demand prediction model is pre-established according to a plurality of dominant meteorological feature sample data and power load feature sample data, and is an RF-SVM model; wherein the process of acquiring the sample data comprises: according to historical meteorological factor data and historical power load data of a preset period and a pre-established dominant meteorological factor recognition model, historical dominant meteorological factor data and corresponding historical power load data are obtained; analyzing load characteristics of different types of loads by adopting a random forest algorithm, extracting historical dominant meteorological factor data and characteristics of corresponding historical power load data thereof, and forming a dominant meteorological characteristic library and a power load characteristic library; obtaining sample data of power demand prediction according to the dominant meteorological feature library and the power load feature library; types of electrical loads include: industrial load type, commercial load type, resident load type for each season;
The method for extracting the characteristics of the historical dominant meteorological factor data and the corresponding historical power load data by adopting a random forest algorithm to form a dominant meteorological characteristic library and a power load characteristic library comprises the following steps: and extracting key features of the history dominant meteorological factors and the corresponding history power load data thereof as key feature vectors of the history dominant meteorological factors and the corresponding history power load data thereof by adopting a random forest algorithm according to the following formula:
Wherein x j is a characteristic parameter, which is the characteristic of dominant meteorological factor data and power load data; y i is the fitting property of the ith out-of-bag data, I is the error representation function, h k (I) is the sample fitting property parameter predicted from dataset B, Fitting attribute parameters obtained after the characteristic parameters x j are replaced;
the support vector machine power demand prediction model is as follows: wherein: a is Lagrangian coefficient, b is a threshold for sample classification, i is time,/> As a kernel function, x is a dominant meteorological factor of a period to be predicted of a target area, and m is the number of samples; the kernel function is: /(I)Wherein, gamma represents the coefficient corresponding to the kernel function type so as to clearly divide the data;
The acquisition unit is also used for acquiring a current power demand prediction result; the current power demand prediction result is a real-time power demand prediction result; the power demand prediction device further includes: the adjusting unit is used for adjusting the power demand prediction model of the support vector machine according to the current power demand prediction result to obtain an updated power demand prediction model of the support vector machine; and the post-evaluation system is utilized to feed back a prediction result in real time, dynamically adjust the model and realize a dynamic optimization model so as to maintain the prediction precision, the evaluation index is evaluated by a root mean square error and an average relative error, the root mean square error is the square root of the ratio of the square sum of the observed value and the true value deviation to the observed times, the average relative error is the average value of the relative error, and the average relative error is expressed by the absolute value of the average relative error.
4. The power demand prediction apparatus according to claim 3, further comprising: the building unit is used for building a power demand prediction model of the support vector machine according to the following method:
Acquiring sample data;
dividing sample data into a training set, a testing set and a verification set;
training the support vector machine model by using a training set to obtain a support vector machine power demand prediction model;
testing the support vector machine power demand prediction model by using a test set to obtain a tested support vector machine power demand prediction model;
And verifying the tested support vector machine power demand prediction model by using a verification set to obtain the pre-established support vector machine power demand prediction model.
5. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 2 when executing the computer program.
6. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program for executing the method of any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011124185.9A CN112365280B (en) | 2020-10-20 | 2020-10-20 | Electric power demand prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011124185.9A CN112365280B (en) | 2020-10-20 | 2020-10-20 | Electric power demand prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112365280A CN112365280A (en) | 2021-02-12 |
| CN112365280B true CN112365280B (en) | 2024-04-19 |
Family
ID=74506958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011124185.9A Active CN112365280B (en) | 2020-10-20 | 2020-10-20 | Electric power demand prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112365280B (en) |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318332A (en) * | 2014-10-29 | 2015-01-28 | 国家电网公司 | Power load predicting method and device |
| CN105184424A (en) * | 2015-10-19 | 2015-12-23 | 国网山东省电力公司菏泽供电公司 | Mapreduced short period load prediction method of multinucleated function learning SVM realizing multi-source heterogeneous data fusion |
| CN105305426A (en) * | 2015-10-20 | 2016-02-03 | 国网山东省电力公司菏泽供电公司 | Mapreduce two-step short-period load prediction method based on deviation control mechanism |
| CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学（保定） | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
| CN108197764A (en) * | 2017-12-22 | 2018-06-22 | 浙江中易和节能技术有限公司 | Predict the method and its equipment of electric power enterprise comprehensive energy consumption |
| CN108334988A (en) * | 2018-02-08 | 2018-07-27 | 吕欣 | A kind of short-term Load Forecasting based on SVM |
| CN108416366A (en) * | 2018-02-06 | 2018-08-17 | 武汉大学 | A kind of power-system short-term load forecasting method of the weighting LS-SVM based on Meteorological Index |
| CN108932557A (en) * | 2018-04-28 | 2018-12-04 | 云南电网有限责任公司临沧供电局 | A kind of Short-term Load Forecasting Model based on temperature cumulative effect and grey relational grade |
| CN109325880A (en) * | 2018-09-21 | 2019-02-12 | 国网天津市电力公司 | A kind of Mid-long term load forecasting method based on Verhulst-SVM |
| CN109345027A (en) * | 2018-10-25 | 2019-02-15 | 国网江苏省电力有限公司盐城供电分公司 | Micro-capacitance sensor short-term load forecasting method based on independent component analysis and support vector machines |
| CN110033134A (en) * | 2019-04-09 | 2019-07-19 | 国网安徽省电力有限公司 | A kind of short-term load forecasting algorithm of segmentation day by day considering meteorologic factor |
| KR20190095586A (en) * | 2018-01-23 | 2019-08-16 | 한국전력정보(주) | Smart power demand management system |
| CN110309965A (en) * | 2019-06-26 | 2019-10-08 | 国家电网有限公司 | A kind of electric grid investment needing forecasting method and system based on improvement support vector machines |
| CN110322062A (en) * | 2019-06-27 | 2019-10-11 | 国网宁夏电力有限公司电力科学研究院 | Short-Term Load Forecasting Method |
| JP2019204458A (en) * | 2018-05-25 | 2019-11-28 | 清水建設株式会社 | Electric power demand prediction system, learning device, and electric power demand prediction method |
| CN110826789A (en) * | 2019-10-30 | 2020-02-21 | 深圳市康必达控制技术有限公司 | Power load prediction method and device based on power system and terminal equipment |
| CN110874671A (en) * | 2019-10-24 | 2020-03-10 | 腾讯科技（深圳）有限公司 | Power load prediction method and device of power distribution network and storage medium |
| CN111311025A (en) * | 2020-03-17 | 2020-06-19 | 南京工程学院 | Load prediction method based on meteorological similar days |
| CN111598303A (en) * | 2020-04-20 | 2020-08-28 | 广东电网有限责任公司电力调度控制中心 | Summer short-term load prediction method based on meteorological component decomposition |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4202890B2 (en) * | 2003-10-30 | 2008-12-24 | 株式会社東芝 | Weather forecasting system, power demand forecasting system, weather forecasting method, and power demand forecasting method |
-
2020
- 2020-10-20 CN CN202011124185.9A patent/CN112365280B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318332A (en) * | 2014-10-29 | 2015-01-28 | 国家电网公司 | Power load predicting method and device |
| CN105184424A (en) * | 2015-10-19 | 2015-12-23 | 国网山东省电力公司菏泽供电公司 | Mapreduced short period load prediction method of multinucleated function learning SVM realizing multi-source heterogeneous data fusion |
| CN105305426A (en) * | 2015-10-20 | 2016-02-03 | 国网山东省电力公司菏泽供电公司 | Mapreduce two-step short-period load prediction method based on deviation control mechanism |
| CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学（保定） | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
| CN108197764A (en) * | 2017-12-22 | 2018-06-22 | 浙江中易和节能技术有限公司 | Predict the method and its equipment of electric power enterprise comprehensive energy consumption |
| KR20190095586A (en) * | 2018-01-23 | 2019-08-16 | 한국전력정보(주) | Smart power demand management system |
| CN108416366A (en) * | 2018-02-06 | 2018-08-17 | 武汉大学 | A kind of power-system short-term load forecasting method of the weighting LS-SVM based on Meteorological Index |
| CN108334988A (en) * | 2018-02-08 | 2018-07-27 | 吕欣 | A kind of short-term Load Forecasting based on SVM |
| CN108932557A (en) * | 2018-04-28 | 2018-12-04 | 云南电网有限责任公司临沧供电局 | A kind of Short-term Load Forecasting Model based on temperature cumulative effect and grey relational grade |
| JP2019204458A (en) * | 2018-05-25 | 2019-11-28 | 清水建設株式会社 | Electric power demand prediction system, learning device, and electric power demand prediction method |
| CN109325880A (en) * | 2018-09-21 | 2019-02-12 | 国网天津市电力公司 | A kind of Mid-long term load forecasting method based on Verhulst-SVM |
| CN109345027A (en) * | 2018-10-25 | 2019-02-15 | 国网江苏省电力有限公司盐城供电分公司 | Micro-capacitance sensor short-term load forecasting method based on independent component analysis and support vector machines |
| CN110033134A (en) * | 2019-04-09 | 2019-07-19 | 国网安徽省电力有限公司 | A kind of short-term load forecasting algorithm of segmentation day by day considering meteorologic factor |
| CN110309965A (en) * | 2019-06-26 | 2019-10-08 | 国家电网有限公司 | A kind of electric grid investment needing forecasting method and system based on improvement support vector machines |
| CN110322062A (en) * | 2019-06-27 | 2019-10-11 | 国网宁夏电力有限公司电力科学研究院 | Short-Term Load Forecasting Method |
| CN110874671A (en) * | 2019-10-24 | 2020-03-10 | 腾讯科技（深圳）有限公司 | Power load prediction method and device of power distribution network and storage medium |
| CN110826789A (en) * | 2019-10-30 | 2020-02-21 | 深圳市康必达控制技术有限公司 | Power load prediction method and device based on power system and terminal equipment |
| CN111311025A (en) * | 2020-03-17 | 2020-06-19 | 南京工程学院 | Load prediction method based on meteorological similar days |
| CN111598303A (en) * | 2020-04-20 | 2020-08-28 | 广东电网有限责任公司电力调度控制中心 | Summer short-term load prediction method based on meteorological component decomposition |
Non-Patent Citations (3)
| Title |
|---|
| Photovoltaic System Output Power Forecasting Based on Weather Type Clustering and LS-SVM;Zhang Huabin; Yang Mingyu;《 Electric Power Science and Engineering》;20150507;第30卷(第10期);第42-47页 * |
| 基于小波分解与随机森林的短期负荷预测;黄青平 邹晓明 刘楚群 叶明武 黄祺珺;《电力信息与通信技术》;20190925;第17卷(第9期);第24-29页 * |
| 智能小区中空调负荷分析预测方法研究;李颖;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑 (月刊)》;20150315(第3期);C042-853 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112365280A (en) | 2021-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wen et al. | Big data driven marine environment information forecasting: a time series prediction network | |
| Cai et al. | Predicting the energy consumption of residential buildings for regional electricity supply-side and demand-side management | |
| CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
| WO2022135265A1 (en) | Failure warning and analysis method for reservoir dispatching rules under effects of climate change | |
| Räsänen et al. | Data-based method for creating electricity use load profiles using large amount of customer-specific hourly measured electricity use data | |
| Hu et al. | Research and application of a hybrid model based on Meta learning strategy for wind power deterministic and probabilistic forecasting | |
| Dong et al. | Applying the ensemble artificial neural network-based hybrid data-driven model to daily total load forecasting | |
| CN109325607A (en) | A kind of short-term wind power forecast method and system | |
| CN116186548B (en) | Power load prediction model training method and power load prediction method | |
| Prasanna et al. | An analysis on stock market prediction using data mining techniques | |
| CN113361785A (en) | Power distribution network short-term load prediction method and device, terminal and storage medium | |
| CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
| Zhao et al. | Short-term microgrid load probability density forecasting method based on k-means-deep learning quantile regression | |
| Bao-De et al. | Improved genetic algorithm-based research on optimization of least square support vector machines: an application of load forecasting | |
| CN114117852B (en) | Regional heat load rolling prediction method based on finite difference working domain division | |
| CN116169670A (en) | Short-term non-resident load prediction method and system based on improved neural network | |
| CN105488598A (en) | Medium-and-long time electric power load prediction method based on fuzzy clustering | |
| Zhang et al. | Application of decision trees to the determination of the year-end level of a carryover storage reservoir based on the iterative dichotomizer 3 | |
| Tang et al. | Predicting housing price based on ensemble learning algorithm | |
| Zhang et al. | Short-term load forecasting based on big data technologies | |
| CN112365280B (en) | Electric power demand prediction method and device | |
| Sarmas et al. | Baseline energy modeling for improved measurement and verification through the use of ensemble artificial intelligence models | |
| Mao et al. | Naive Bayesian algorithm classification model with local attribute weighted based on KNN | |
| Zhang et al. | Building Energy Consumption Prediction Based on Temporal-Aware Attention and Energy Consumption States | |
| Chen et al. | Spatial–temporal information model-based load current interval prediction for transmission lines |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |