CN111028100A - Refined short-term load prediction method, device and medium considering meteorological factors - Google Patents
Refined short-term load prediction method, device and medium considering meteorological factors Download PDFInfo
- Publication number
- CN111028100A CN111028100A CN201911217442.0A CN201911217442A CN111028100A CN 111028100 A CN111028100 A CN 111028100A CN 201911217442 A CN201911217442 A CN 201911217442A CN 111028100 A CN111028100 A CN 111028100A
- Authority
- CN
- China
- Prior art keywords
- sequence
- layer
- short
- load
- decision tree
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000003066 decision tree Methods 0.000 claims abstract description 32
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 28
- 238000013528 artificial neural network Methods 0.000 claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 239000013598 vector Substances 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 7
- 238000013138 pruning Methods 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 238000007670 refining Methods 0.000 claims description 5
- 238000004445 quantitative analysis Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 125000004432 carbon atom Chemical group C* 0.000 claims description 3
- 238000007635 classification algorithm Methods 0.000 claims description 3
- 230000007423 decrease Effects 0.000 claims description 3
- 238000012854 evaluation process Methods 0.000 claims description 3
- 238000012886 linear function Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007621 cluster analysis Methods 0.000 abstract description 3
- 230000005611 electricity Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Biophysics (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method, a device and a medium for predicting refined short-term load by considering meteorological factors, wherein the method comprises the following steps: quantitatively analyzing the influence of meteorological factors on the power load characteristics based on a grey correlation degree analysis method, and selecting a plurality of key influence factors as input vectors of a decision tree algorithm; after the historical load data of the user is subjected to cluster analysis, establishing a category label for each daily load curve, establishing a classification rule through a decision tree algorithm, and classifying days to be predicted; and performing short-term load prediction on a user by adopting an Elman neural network. Compared with the prior art, the method and the device can achieve the purpose of user-level refined load prediction and improve the precision of the short-term load prediction by establishing the refined short-term load prediction model considering meteorological factors.
Description
Technical Field
The invention relates to the technical field of power systems, in particular to a method, a device and a medium for predicting a refined short-term load by considering meteorological factors.
Background
Short-term load prediction is an important function of a power grid energy management system and is a basis for safe, economic and reliable operation of a power system. The accuracy of load prediction directly affects the safety, economy and quality of power supply of the power system. Therefore, how to improve the prediction accuracy is the focus of the research of the short-term load prediction technology at present.
In the prior art, methods for short-term load prediction mainly comprise two main types of traditional prediction methods and modern prediction methods.
Conventional prediction methods include exponential smoothing, regression analysis, time series, grey prediction, etc. Among them, the time-series method is most widely used. The time series method is used for deducing the numerical change of the future load according to the historical load data aiming at the one-dimensional time series, and the influence of meteorological factors on the load is not considered, so that the loss of important information is caused. Even considering the influence of meteorological factors, some methods mostly analyze the relationship between daily load and single meteorological factors such as the highest temperature, the lowest temperature and the like, cannot reflect all meteorological information, and easily cause errors of analysis results, thereby influencing the precision of short-term load prediction and being difficult to meet the requirement of regional load prediction.
The modern prediction method mainly comprises an expert system method, a genetic algorithm, a neural network method, a support vector machine and the like. The neural network becomes an important method for predicting the short-term load due to the self-learning capability and the processing capability of the neural network on a complex nonlinear system. Because the structure and parameters of the neural network are mostly determined according to subjective experience, the prediction effect is difficult to ensure. The reasonable determination of the structure and parameters of the neural network can effectively improve the precision of load prediction.
For the defects of the traditional prediction method, the problem of user refined load prediction under the background of big data cannot be solved. With the continuous development and accumulation of the large power data, how to process the relationship between the large power data and the load characteristic influence factors is achieved, so that a refined short-term load prediction is obtained, and the method is worth further exploring. Factors such as weather and date types have different degrees of influence on load characteristics of different types of users, and short-term load prediction based on user load characteristic analysis needs to be researched.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method, a device and a medium for predicting a refined short-term load by considering meteorological factors, which can achieve the aim of predicting the refined load at a user level and improve the precision of the short-term load prediction.
In order to solve the above problem, an embodiment of the present invention provides a method for predicting a refined short-term load in consideration of meteorological factors, including:
s1, quantitatively analyzing the influence of meteorological factors on the power load characteristics based on a grey correlation degree analysis method, and selecting a plurality of key influence factors as input vectors of a decision tree algorithm;
s2, after the historical load data of the user are subjected to clustering analysis, establishing a category label for each daily load curve, establishing a classification rule through a decision tree algorithm, and classifying days to be predicted;
and S3, carrying out short-term load prediction on a certain user by adopting an Elman neural network.
Preferably, step S1 includes:
s11, determining a reference sequence and a comparison sequence which need to be used in the quantitative analysis process; wherein, the reference sequence is a system characteristic sequence, and the comparison sequence is an influence factor sequence;
let reference sequence be Y ═ Y1y2... yt)T,yi(i 1, 2.. t.) represents the value of each point in the reference sequence, t represents the total data volume, and the influencing factor X1,X2,...,XnThe constructed comparison sequence matrix can be represented in the following form:
in the above formula, n represents the number of the influence factors to be evaluated in the whole evaluation process;
s12, carrying out normalization processing on the data set by adopting a maximum value, wherein a calculation formula is as follows:
the normalized reference sequence and the normalized comparison sequence are respectively:
s13, regarding the reference sequence and the comparison sequence as points in a multi-dimensional space, researching the relation between each influence factor and system characteristics in a specific t-dimensional space, wherein the calculation formula of the difference sequence is as follows:
the difference matrix is then Δ:
s14, the correlation coefficient reflects the correlation between the system characteristic sequence and each influence factor sequence at different time points, and the calculation formula is as follows:
in the above equation, ρ ∈ (0, 1) is called a resolution coefficient, and as ρ decreases, the resolution increases, and ρ is taken to be 0.5; calculating the correlation coefficient of the system characteristic sequence and each influence factor sequence at each time point by the formula to obtain a correlation coefficient matrix as follows:
s15, carrying out weighted summation on the correlation coefficients of all time points, wherein the weighted summation is shown in the following formula:
in the above formula: l isjThe correlation degree of the system characteristic sequence and the jth influence factor is shown, and omega (i) is the correlation coefficient weight of the ith time point;
setting a threshold variable theta when Lj>Theta, it is determined that there is a correlation between the system signature sequence and the influencing factor j, and LjThe larger the correlation, the stronger the correlation.
Preferably, in step S2, specifically:
identifying the type of the load curve of the day to be predicted based on a CART decision tree algorithm; wherein,
the CART decision tree algorithm building model comprises a tree building process and a pruning process, the pruning process is used for preventing over-fitting of training data, the tree building process of the CART decision tree classification algorithm takes the minimum kiney coefficient of sample data as a target, the optimal class division is selected, a binary tree is built recursively, and each node division can generate two child nodes;
the stopping conditions of the CART decision tree algorithm are: the number of samples in a node is less than a given threshold or the kuney factor is less than a given threshold.
Preferably, the Elman neural network comprises an input layer, a hidden layer, a carrying layer and an output layer;
step S3, specifically:
the following Elman neural network nonlinear state space expression is constructed:
x(k)=f(w1xc(k)+w2(μ(k-1))+b1);
in the above formula, w1Representing the weight from the input layer to the hidden layer; x is the number ofc(k) Represents the output of the receiving layer; x (k) represents the output of the acceptor layer; w is a2Representing the weight from the bearer layer to the hidden layer; mu (k-1) An input representing a network model; b1Is the threshold of the input layer; f (-) is the transfer function of the hidden layer, usually sigmoid function;
xc(k)=x(k-1);
the above formula represents: the receiving layer memorizes the output x (k-1) of the hidden layer at the previous moment;
y(k)=g(w3x(k)+b2);
in the above formula, w3Representing the weight from the hidden layer to the output layer; b2A threshold value for the hidden layer; g (-) is the transfer function, usually a linear function, of the output layer;
the Elman neural network uses a sum of squared errors function as a learning index function, and the calculation formula is as follows:
in the above formula, the first and second carbon atoms are,an input vector corresponding to the target object.
The embodiment of the invention also provides a device for predicting the refined short-term load by considering meteorological factors, which comprises the following steps:
the input unit is used for quantitatively analyzing the influence of meteorological factors on the power load characteristics based on a grey correlation degree analysis method, and selecting a plurality of key influence factors as input vectors of a decision tree algorithm;
the classification unit is used for establishing a class label for each daily load curve after clustering analysis is carried out on the historical load data of the user, establishing a classification rule through a decision tree algorithm and classifying days to be predicted;
and the prediction unit is used for predicting the short-term load of a certain user by adopting the Elman neural network.
The embodiment of the invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, the device where the computer-readable storage medium is located is controlled to execute the above refined short-term load prediction method considering meteorological factors.
The embodiment of the invention has the following beneficial effects:
according to the method, based on the clustering analysis and the quantitative analysis result of the load characteristic influence factors, the classification rule is established according to algorithms such as a decision tree and the like, the data set of the same type of days to be predicted is obtained, and finally, the Elman neural network is applied to short-term load prediction and serves as a load model of the days to be predicted, so that the purpose of refining load prediction is achieved, and the accuracy of short-term load prediction is improved.
Drawings
Fig. 1 is a flowchart illustrating a method for refining short-term load prediction considering meteorological factors according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended 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 invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
Please refer to fig. 1.
The embodiment of the invention provides a method for predicting a refined short-term load by considering meteorological factors, which comprises the following steps:
s1, based on a grey correlation degree analysis method, quantitatively analyzing the influence of meteorological factors on the power load characteristics, and selecting a plurality of key influence factors as input vectors of a decision tree algorithm.
In a specific embodiment, for the daily load characteristic index in the user's historical power consumption: the daily maximum load, the daily minimum load, the daily peak-valley difference rate, the daily load rate and the daily minimum load rate are subjected to correlation analysis with meteorological factors such as air temperature (highest, lowest and average), daily average humidity, air pressure (highest, lowest and average) and daily maximum wind speed, and key influence factors influencing the electricity utilization characteristics of the user are obtained.
And S2, after the historical load data of the user is subjected to cluster analysis, establishing a class label for each daily load curve, establishing a classification rule through a decision tree algorithm, and classifying the days to be predicted.
In a specific embodiment, the predicted load values of the day to be predicted at 24 times a day are used for selecting historical electricity utilization data by using meteorological factors, and the historical electricity utilization data of the same category as the day to be predicted are input into the prediction model as a training set.
And S3, carrying out short-term load prediction on a certain user by adopting an Elman neural network.
With the development of intelligent measurement systems, the measurement dimension of the power grid operation data can be specific to users. In a specific embodiment, accurate description of load characteristics is achieved using user-level power data. In the technical scheme, a refined short-term load prediction model considering meteorological factors is established by integrating various methods such as cluster analysis, grey correlation degree analysis, CART decision tree and Elman neural network algorithm, and refined load prediction of a user level is realized.
The established refined load prediction model has reference and research values for user-level load prediction, and can also be popularized to load prediction of regional power grids (provincial and urban power grids). The embodiment of the invention has wider applicability and universality.
Step S1, including:
s11, determining a reference sequence and a comparison sequence which need to be used in the quantitative analysis process; wherein, the reference sequence is a system characteristic sequence, and the comparison sequence is an influence factor sequence;
let reference sequence be Y ═ Y1y2... yt)T,yi(i 1, 2.. t.) represents the value of each point in the reference sequence, t represents the total data volume, and the influencing factor X1,X2,...,XnThe constructed comparison sequence matrix can be represented in the following form:
in the above formula, n represents the number of influencing factors to be evaluated in the whole evaluation process.
The numerical difference of different sequences is large, and the dimensions of each type of data do not have comparability, which affects the accuracy of the final analysis result, so that the sequences need to be subjected to standard processing.
S12, carrying out normalization processing on the data set by adopting a maximum value, wherein a calculation formula is as follows:
the normalized reference sequence and the normalized comparison sequence are respectively:
s13, regarding the reference sequence and the comparison sequence as points in a multi-dimensional space, researching the relation between each influence factor and system characteristics in a specific t-dimensional space, wherein the calculation formula of the difference sequence is as follows:
the difference matrix is then Δ:
s14, the correlation coefficient reflects the correlation between the system characteristic sequence and each influence factor sequence at different time points, and the calculation formula is as follows:
in the above equation, ρ ∈ (0, 1) is called a resolution coefficient, and as ρ decreases, the resolution increases, and ρ is taken to be 0.5; calculating the correlation coefficient of the system characteristic sequence and each influence factor sequence at each time point by the formula to obtain a correlation coefficient matrix as follows:
the correlation coefficient is only the correlation value of the comparison sequence and the reference sequence at different time points, and is used for calculating the correlation degree between the system characteristic sequence and each influence factor subsequence.
S15, carrying out weighted summation on the correlation coefficients of all time points, wherein the weighted summation is shown in the following formula:
in the above formula: l isjAs the degree of association between the system feature sequence and the jth influencing factor, ω (i) is the relation of the ith time pointA joint coefficient weight;
setting a threshold variable theta when Lj>Theta, it is determined that there is a correlation between the system signature sequence and the influencing factor j, and LjThe larger the correlation, the stronger the correlation.
Preferably, in step S2, specifically:
identifying the type of the load curve of the day to be predicted based on a CART decision tree algorithm; wherein,
the CART decision tree algorithm building model comprises a tree building process and a pruning process, the pruning process is used for preventing over-fitting of training data, the tree building process of the CART decision tree classification algorithm takes the minimum kiney coefficient of sample data as a target, the optimal class division is selected, a binary tree is built recursively, and each node division can generate two child nodes;
the stopping conditions of the CART decision tree algorithm are: the number of samples in a node is less than a given threshold or the kuney factor is less than a given threshold.
Decision Tree algorithm (Decision Tree) is an analysis method for constructing a Decision Tree to obtain the probability that the expectation of the net present value is greater than or equal to zero on the basis of the known occurrence probability of sample data. According to the input data types and different prediction targets, classification trees and regression trees can be classified: the classification tree can realize the class prediction of the object with unknown type; the regression tree may enable prediction of continuous values. Therefore, the identification of the load curve type of the day to be predicted is realized based on the CART decision tree algorithm.
Preferably, the Elman neural network comprises an input layer, a hidden layer, a carrying layer and an output layer;
step S3, specifically:
the following Elman neural network nonlinear state space expression is constructed:
x(k)=f(w1xc(k)+w2(μ(k-1))+b1);
in the above formula, w1Representing the weight from the input layer to the hidden layer; x is the number ofc(k) Represents the output of the receiving layer; x (k) represents the output of the acceptor layer; w is a2Representing a bearer layer to a hidden layerThe weight of (2); μ (k-1) represents the input to the network model; b1Is the threshold of the input layer; f (-) is the transfer function of the hidden layer, usually sigmoid function;
xc(k)=x(k-1);
the above formula represents: the receiving layer memorizes the output x (k-1) of the hidden layer at the previous moment;
y(k)=g(w3x(k)+b2);
in the above formula, w3Representing the weight from the hidden layer to the output layer; b2A threshold value for the hidden layer; g (-) is the transfer function, usually a linear function, of the output layer;
the Elman neural network uses a sum of squared errors function as a learning index function, and the calculation formula is as follows:
in the above formula, the first and second carbon atoms are,an input vector corresponding to the target object.
The embodiment of the invention also provides a device for predicting the refined short-term load by considering meteorological factors, which comprises the following steps:
and the input unit is used for quantitatively analyzing the influence of meteorological factors on the power load characteristics based on the grey correlation degree analysis method, and selecting a plurality of key influence factors as input vectors of the decision tree algorithm.
And the classification unit is used for establishing a class label for each daily load curve after clustering analysis is carried out on the historical load data of the user, establishing a classification rule through a decision tree algorithm, and classifying the days to be predicted.
And the prediction unit is used for predicting the short-term load of a certain user by adopting the Elman neural network.
The embodiment of the invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, the device where the computer-readable storage medium is located is controlled to execute the above refined short-term load prediction method considering meteorological factors.
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 a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (6)
1. A refined short-term load prediction method considering meteorological factors is characterized by comprising the following steps:
s1, quantitatively analyzing the influence of meteorological factors on the power load characteristics based on a grey correlation degree analysis method, and selecting a plurality of key influence factors as input vectors of a decision tree algorithm;
s2, after the historical load data of the user are subjected to clustering analysis, establishing a category label for each daily load curve, establishing a classification rule through a decision tree algorithm, and classifying days to be predicted;
and S3, carrying out short-term load prediction on a certain user by adopting an Elman neural network.
2. The method for refining short-term load forecasting according to claim 1, wherein step S1 includes:
s11, determining a reference sequence and a comparison sequence which need to be used in the quantitative analysis process; wherein, the reference sequence is a system characteristic sequence, and the comparison sequence is an influence factor sequence;
let reference sequence be Y ═ Y1y2... yt)T,yi(i 1, 2.. t.) represents the value of each point in the reference sequence, t represents the total data volume, and the influencing factor X1,X2,...,XnThe constructed comparison sequence matrix can be represented in the following form:
in the above formula, n represents the number of the influence factors to be evaluated in the whole evaluation process;
s12, carrying out normalization processing on the data set by adopting a maximum value, wherein a calculation formula is as follows:
the normalized reference sequence and the normalized comparison sequence are respectively:
s13, regarding the reference sequence and the comparison sequence as points in a multi-dimensional space, researching the relation between each influence factor and system characteristics in a specific t-dimensional space, wherein the calculation formula of the difference sequence is as follows:
the difference matrix is then Δ:
s14, the correlation coefficient reflects the correlation between the system characteristic sequence and each influence factor sequence at different time points, and the calculation formula is as follows:
in the above equation, ρ ∈ (0, 1) is called a resolution coefficient, and as ρ decreases, the resolution increases, and ρ is taken to be 0.5; calculating the correlation coefficient of the system characteristic sequence and each influence factor sequence at each time point by the formula to obtain a correlation coefficient matrix as follows:
s15, carrying out weighted summation on the correlation coefficients of all time points, wherein the weighted summation is shown in the following formula:
in the above formula: l isjThe correlation degree of the system characteristic sequence and the jth influence factor is shown, and omega (i) is the correlation coefficient weight of the ith time point;
setting a threshold variable theta when Lj>Theta, it is determined that there is a correlation between the system signature sequence and the influencing factor j, and LjThe larger the correlation, the stronger the correlation.
3. The method for refining short-term load forecasting considering meteorological factors according to claim 1, wherein in step S2, specifically:
identifying the type of the load curve of the day to be predicted based on a CART decision tree algorithm; wherein,
the CART decision tree algorithm building model comprises a tree building process and a pruning process, the pruning process is used for preventing over-fitting of training data, the tree building process of the CART decision tree classification algorithm takes the minimum kiney coefficient of sample data as a target, the optimal class division is selected, a binary tree is built recursively, and each node division can generate two child nodes;
the stopping conditions of the CART decision tree algorithm are: the number of samples in a node is less than a given threshold or the kuney factor is less than a given threshold.
4. The method of claim 1, wherein the Elman neural network comprises an input layer, a hidden layer, a sink layer and an output layer;
step S3, specifically:
the following Elman neural network nonlinear state space expression is constructed:
x(k)=f(w1xc(k)+w2(μ(k-1))+b1);
in the above formula, w1Representing the weight from the input layer to the hidden layer; x is the number ofc(k) Represents the output of the receiving layer; x (k) represents the output of the acceptor layer; w is a2Representing the weight from the bearer layer to the hidden layer; μ (k-1) represents the input to the network model; b1Is the threshold of the input layer; f (-) is the transfer function of the hidden layer, usually sigmoid function;
xc(k)=x(k-1);
the above formula represents: the receiving layer memorizes the output x (k-1) of the hidden layer at the previous moment;
y(k)=g(w3x(k)+b2);
in the above formula, w3Representing the weight from the hidden layer to the output layer; b2A threshold value for the hidden layer; g (-) is the transfer function, usually a linear function, of the output layer;
the Elman neural network uses a sum of squared errors function as a learning index function, and the calculation formula is as follows:
5. A weather-based refined short-term load prediction device, comprising:
the input unit is used for quantitatively analyzing the influence of meteorological factors on the power load characteristics based on a grey correlation degree analysis method, and selecting a plurality of key influence factors as input vectors of a decision tree algorithm;
the classification unit is used for establishing a class label for each daily load curve after clustering analysis is carried out on the historical load data of the user, establishing a classification rule through a decision tree algorithm and classifying days to be predicted;
and the prediction unit is used for predicting the short-term load of a certain user by adopting the Elman neural network.
6. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the method for refining short-term load prediction considering meteorological factors according to any one of claims 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911217442.0A CN111028100A (en) | 2019-11-29 | 2019-11-29 | Refined short-term load prediction method, device and medium considering meteorological factors |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911217442.0A CN111028100A (en) | 2019-11-29 | 2019-11-29 | Refined short-term load prediction method, device and medium considering meteorological factors |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111028100A true CN111028100A (en) | 2020-04-17 |
Family
ID=70203952
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911217442.0A Pending CN111028100A (en) | 2019-11-29 | 2019-11-29 | Refined short-term load prediction method, device and medium considering meteorological factors |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111028100A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112200383A (en) * | 2020-10-28 | 2021-01-08 | 宁波立新科技股份有限公司 | Power load prediction method based on improved Elman neural network |
CN112348287A (en) * | 2020-11-26 | 2021-02-09 | 南方电网能源发展研究院有限责任公司 | Electric power system short-term load probability density prediction method based on LSTM quantile regression |
CN112446509A (en) * | 2020-11-10 | 2021-03-05 | 中国电子科技集团公司第三十八研究所 | Complex electronic equipment prediction maintenance method |
CN112734135A (en) * | 2021-01-26 | 2021-04-30 | 吉林大学 | Power load prediction method, intelligent terminal and computer readable storage medium |
CN112966868A (en) * | 2021-03-13 | 2021-06-15 | 山东大学 | Building load day-ahead prediction method and system |
CN113283774A (en) * | 2021-06-07 | 2021-08-20 | 润电能源科学技术有限公司 | Deep peak regulation method and device for heating unit, electronic equipment and storage medium |
CN114155038A (en) * | 2021-12-09 | 2022-03-08 | 国网河北省电力有限公司营销服务中心 | Method for identifying user affected by epidemic situation |
CN115085196A (en) * | 2022-08-19 | 2022-09-20 | 国网信息通信产业集团有限公司 | Power load predicted value determination method, device, equipment and computer readable medium |
CN115685924A (en) * | 2022-10-28 | 2023-02-03 | 中南大学 | Converter blowing end point forecasting method |
CN116502768A (en) * | 2023-05-23 | 2023-07-28 | 中国南方航空股份有限公司 | Civil aviation information post load early warning method, system and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104881706A (en) * | 2014-12-31 | 2015-09-02 | 天津弘源慧能科技有限公司 | Electrical power system short-term load forecasting method based on big data technology |
US20190081476A1 (en) * | 2017-09-12 | 2019-03-14 | Sas Institute Inc. | Electric power grid supply and load prediction |
-
2019
- 2019-11-29 CN CN201911217442.0A patent/CN111028100A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104881706A (en) * | 2014-12-31 | 2015-09-02 | 天津弘源慧能科技有限公司 | Electrical power system short-term load forecasting method based on big data technology |
US20190081476A1 (en) * | 2017-09-12 | 2019-03-14 | Sas Institute Inc. | Electric power grid supply and load prediction |
Non-Patent Citations (1)
Title |
---|
陈寒冬等: "考虑气象因素的精细化短期负荷预测模型研究", 《电力学报》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112200383A (en) * | 2020-10-28 | 2021-01-08 | 宁波立新科技股份有限公司 | Power load prediction method based on improved Elman neural network |
CN112200383B (en) * | 2020-10-28 | 2024-05-17 | 宁波立新科技股份有限公司 | Power load prediction method based on improved Elman neural network |
CN112446509B (en) * | 2020-11-10 | 2023-05-26 | 中国电子科技集团公司第三十八研究所 | Prediction maintenance method for complex electronic equipment |
CN112446509A (en) * | 2020-11-10 | 2021-03-05 | 中国电子科技集团公司第三十八研究所 | Complex electronic equipment prediction maintenance method |
CN112348287A (en) * | 2020-11-26 | 2021-02-09 | 南方电网能源发展研究院有限责任公司 | Electric power system short-term load probability density prediction method based on LSTM quantile regression |
CN112734135A (en) * | 2021-01-26 | 2021-04-30 | 吉林大学 | Power load prediction method, intelligent terminal and computer readable storage medium |
CN112734135B (en) * | 2021-01-26 | 2022-07-15 | 吉林大学 | Power load prediction method, intelligent terminal and computer readable storage medium |
CN112966868A (en) * | 2021-03-13 | 2021-06-15 | 山东大学 | Building load day-ahead prediction method and system |
CN113283774A (en) * | 2021-06-07 | 2021-08-20 | 润电能源科学技术有限公司 | Deep peak regulation method and device for heating unit, electronic equipment and storage medium |
CN114155038A (en) * | 2021-12-09 | 2022-03-08 | 国网河北省电力有限公司营销服务中心 | Method for identifying user affected by epidemic situation |
CN114155038B (en) * | 2021-12-09 | 2024-05-31 | 国网河北省电力有限公司营销服务中心 | Epidemic situation affected user identification method |
CN115085196A (en) * | 2022-08-19 | 2022-09-20 | 国网信息通信产业集团有限公司 | Power load predicted value determination method, device, equipment and computer readable medium |
CN115685924A (en) * | 2022-10-28 | 2023-02-03 | 中南大学 | Converter blowing end point forecasting method |
CN116502768A (en) * | 2023-05-23 | 2023-07-28 | 中国南方航空股份有限公司 | Civil aviation information post load early warning method, system and storage medium |
CN116502768B (en) * | 2023-05-23 | 2024-06-07 | 中国南方航空股份有限公司 | Civil aviation information post load early warning method, system and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111028100A (en) | Refined short-term load prediction method, device and medium considering meteorological factors | |
Dong et al. | Short-term load forecasting in smart grid: A combined CNN and K-means clustering approach | |
CN109063911B (en) | Load aggregation grouping prediction method based on gated cycle unit network | |
CN108734355B (en) | Short-term power load parallel prediction method and system applied to power quality comprehensive management scene | |
CN111105104A (en) | Short-term power load prediction method based on similar day and RBF neural network | |
CN110717535B (en) | Automatic modeling method and system based on data analysis processing system | |
CN107967542B (en) | Long-short term memory network-based electricity sales amount prediction method | |
Tak et al. | Recurrent type-1 fuzzy functions approach for time series forecasting | |
CN112734135B (en) | Power load prediction method, intelligent terminal and computer readable storage medium | |
CN111525587B (en) | Reactive load situation-based power grid reactive voltage control method and system | |
CN113408869A (en) | Power distribution network construction target risk assessment method | |
CN109325631A (en) | Electric car charging load forecasting method and system based on data mining | |
CN113361785A (en) | Power distribution network short-term load prediction method and device, terminal and storage medium | |
CN116186548B (en) | Power load prediction model training method and power load prediction method | |
CN114118596A (en) | Photovoltaic power generation capacity prediction method and device | |
Wang et al. | Short-term load forecasting with LSTM based ensemble learning | |
CN116169670A (en) | Short-term non-resident load prediction method and system based on improved neural network | |
CN116415732A (en) | User side power load data processing method based on improved ARNN | |
CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
CN115619028A (en) | Clustering algorithm fusion-based power load accurate prediction method | |
CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
CN118095570A (en) | Intelligent load prediction method and system for transformer area, electronic equipment, medium and chip | |
CN111027841A (en) | Low-voltage transformer area line loss calculation method based on gradient lifting decision tree | |
CN117913808A (en) | Distributed photovoltaic power generation prediction method and device | |
CN113837481A (en) | Financial big data management system based on block chain |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200417 |