CN112365025A - Renewable energy market long-term heat storage method supporting vector machine width learning - Google Patents

Renewable energy market long-term heat storage method supporting vector machine width learning Download PDF

Info

Publication number
CN112365025A
CN112365025A CN202011079092.9A CN202011079092A CN112365025A CN 112365025 A CN112365025 A CN 112365025A CN 202011079092 A CN202011079092 A CN 202011079092A CN 112365025 A CN112365025 A CN 112365025A
Authority
CN
China
Prior art keywords
class
heat storage
renewable energy
energy market
vector machine
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.)
Granted
Application number
CN202011079092.9A
Other languages
Chinese (zh)
Other versions
CN112365025B (en
Inventor
殷林飞
陶敏
徐紫东
杨凯
高放
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangxi University
Original Assignee
Guangxi University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangxi University filed Critical Guangxi University
Priority to CN202011079092.9A priority Critical patent/CN112365025B/en
Publication of CN112365025A publication Critical patent/CN112365025A/en
Application granted granted Critical
Publication of CN112365025B publication Critical patent/CN112365025B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/245—Classification techniques relating to the decision surface
    • G06F18/2451—Classification techniques relating to the decision surface linear, e.g. hyperplane
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/245—Classification techniques relating to the decision surface
    • G06F18/2453—Classification techniques relating to the decision surface non-linear, e.g. polynomial classifier
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/259—Fusion by voting

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Nonlinear Science (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • Quality & Reliability (AREA)
  • Primary Health Care (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a renewable energy market long-term heat storage method supporting vector machine width learning. The proposed method predicts long term consumption of distributed electrical heat storage in the renewable energy market through a combination of support vector machines and breadth learning. The support vector machine in the proposed method classifies the raw data, i.e. air temperature, and the corresponding long-term consumption of distributed electrical heat storage in the renewable energy market into three categories. And then, inputting various types of raw data into the width learning in the method for training. After the width learning in the method is finished, the long-term consumption of the distributed electric heat storage in the renewable energy market can be predicted by inputting the corresponding air temperature. The support vector machine in the method has better classification performance, and the width learning structure in the method is simple and has perfect performance. The method improves the accuracy of predicting the long-term consumption of distributed electric heat storage in the renewable energy market.

Description

Renewable energy market long-term heat storage method supporting vector machine width learning
Technical Field
The invention belongs to the field of scheduling of an integrated energy system, and relates to a method for deeply combining two types of artificial intelligence methods, which is suitable for predicting the long-term consumption of distributed electric heat storage in a renewable energy market.
Background
The national energy-saving emission-reducing policy regulates the temperature of the air conditioner as follows: the temperature of the indoor air conditioner in summer is set to be not lower than 26 ℃ and the temperature of the indoor air conditioner in winter is set to be not higher than 20 ℃ except for special units such as hospitals and the like and users which have specific requirements and are approved by the production industry, in all units in the public building, including state organs, social groups, enterprise and public institutions and individual industrial and commercial businesses. Therefore, people can generally use related electric loads for heating when the temperature is lower than 20 ℃; when the temperature is higher than 26 ℃, people generally use related electric loads for refrigeration; when the air temperature is between 20 ℃ and 26 ℃, people generally do not need to use electric loads to heat or cool.
With the development of economy, the daily electricity consumption of urban and rural residents is rapidly increased, and particularly under the influence of air temperature, the heating load in winter and the cooling load in summer are rapidly increased, so that the daily electricity consumption of the urban and rural residents is rapidly increased. In order to improve the stability of the power grid system and optimize the scheduling, the accuracy of load prediction is very important. The distributed electric heat storage system can realize peak clipping and valley filling of electric power. The invention provides a method for supporting vector machine width learning to predict the consumption of distributed electric heat storage in a renewable energy market. The deep neural network has a good effect on processing a large amount of data, but the structure of the deep neural network is complex. This complex structure makes it extremely difficult to theoretically analyze the structure of the deep neural network. In addition, in order to improve the accuracy of prediction in application, the deep neural network has to continuously increase the number of network layers or adjust the number of parameters. Therefore, a series of neural networks and corresponding combining methods have been developed to increase the training speed. Among them, the breadth learning system provides a substitute for the deep neural network. Meanwhile, if the network needs to be expanded, the accuracy of prediction is improved by a fast incremental learning method without retraining the complete network. The classification idea of the support vector machine in the method is simple, namely the interval between the sample and the decision surface is maximized, the classification problem of the nonlinear support vector machine can be solved by using the kernel function, and the classification effect is good. The width learning structure in the method is simpler and more perfect than a deep neural network structure. The proposed method of support vector machine width learning improves the accuracy of predicting long-term consumption of distributed electrical storage in the renewable energy market.
Disclosure of Invention
The invention provides a renewable energy market long-term heat storage method supporting vector machine width learning. The renewable energy long-term heat storage method supporting vector machine width learning can be applied to long-term consumption prediction of distributed electric heat storage in the renewable energy market. The consumption of the distributed electric heat accumulation in the renewable energy market is closely related to the temperature. T is the air temperature, P is the long-term consumption of distributed electric heat storage in the renewable energy market, and the classification is A, B and C; setting T as air temperature and P as long-term consumption of distributed electric heat storage in the renewable energy market; the A is the condition that T is less than 20 ℃ and P is more than or equal to 0; the B is the condition that T is more than or equal to 20 ℃ and less than or equal to 26 ℃ and P is more than or equal to 0; the C is the condition that T is more than or equal to 26 ℃ and P is more than or equal to 0. The support vector machine in the proposed method is first classified by the raw data, i.e. air temperature and the corresponding long-term consumption of distributed electrical heat storage in the renewable energy market. And then, inputting the original data of each type into the width learning in the method for training. Finally, after the width learning training in the method is finished, the required long-term consumption of the distributed electric heat storage in the renewable energy market can be predicted by inputting the corresponding air temperature. The specific method is as follows:
step 1: first, given input data and learning objectives: x ═ X1,...,XN},y＝{y1,...,yNAnd N is the number of samples. Each sample of input data contains two characteristics of air temperature and long term consumption of distributed electrical heat storage in the renewable energy market: xi＝[Ti,Pi]Wherein i 1. And the learning objective is a binary variable yiE { -1,1} represents a negative class and a positive class, where i ═ 1.
Step 2: the feature space where the input data is located has a hyperplane serving as a decision boundary, the learning target is separated according to a positive class and a negative class, and the distance between any sample point and the plane is more than or equal to 1:
decision boundary:
ωTX+b＝0 (1)
point-to-face distance:
yi(ωTXi+b)≥1 (2)
where ω and b are the normal vector and intercept of the hyperplane, respectively.
And step 3: the original problem of the nonlinear support vector machine is that the distance from the point closest to the plane is as large as possible, and the distance between two spaced boundaries is
Figure BDA0002718008640000021
Solving the maximum value of d can be converted into solving
Figure BDA0002718008640000022
The minimum value problem of (2); namely, the original problem can be converted into:
Figure BDA0002718008640000023
wherein, K (X)i,Xj)＝φ(Xi)Tφ(Xj) N, j is 1,2,. N is a selected kernel function. The constant G is a normalization coefficient. XiiIs an error function.
And 4, step 4: by lagrange multiplier: α ═ α1,...,αN},μ＝{μ1,...,μNIts lagrange function can be found:
Figure BDA0002718008640000031
wherein alpha isi≥0,μi≥0。
And 5: let the partial derivative of the lagrangian function on the optimization target ω, b, ξ be 0, an expression containing the lagrangian multiplier can be obtained:
Figure BDA0002718008640000032
step 6: after substituting the above formula into the Lagrangian function, the dual problem of the original problem can be obtained:
Figure BDA0002718008640000033
and 7: the method for solving the dual problem of the support vector machine in the method uses an iterative sequence minimum optimization method, and the design is to select two variables alpha in a Lagrange multiplier in each iteration stepi,αjAnd fixing other parameters, wherein the constraint conditions of the dual problem are as follows:
Figure BDA0002718008640000034
wherein Const is a constant. Substituting the right side of the above equation into the dual problem of the support vector machine and eliminating alpha in the summation termjCan be obtained with respect to only αiThe quadratic programming problem of (2).
And 8: finding the optimal solution alpha*. Find a 0 < alphaj< G. Then b is solved, and the optimal solution b is solved*Can be obtained by the following formula
Figure BDA0002718008640000035
And step 9: obtaining a classification decision function:
Figure BDA0002718008640000036
where f (x) sign [. cndot ] is a sign function. When · < 0, f (x) is-1; when ═ 0, f (x) is 0; when · > 0, f (x) is 1.
Step 10: three classification decision functions related to the A class and the B class, the B class and the C class, and the A class and the C class can be obtained according to the solving method, and three training results are obtained. During classification, the original data are respectively input into three training functions for judgment. Then, a voting form is adopted, and finally a group of results are obtained:
let NA，NBAnd NCThe ticket numbers of A class, B class and C class respectively;
initializing the ticket number: n is a radical ofA＝NB＝NC＝0；
Inputting the original data into a training function about class A and class B for judgment, and if the result is class A, NA＝NA+ 1; otherwise, NB＝NB+1；
Inputting the original data into training functions related to class B and class C for judgment, and if the result is class B, then NB＝NB+ 1; otherwise, NC＝NC+1；
Inputting the original data into a training function about class A and class C for judgment, and if the result is class A, NA＝NA+ 1; otherwise, NC＝NC+1；
NA，NBAnd NCThe category corresponding to the maximum value in (b) is the final classification result.
Step 11: and storing the classified original data belonging to the class A, the class B and the class C respectively while classifying.
Step 12: let the original data be X, input the X utilization function
Zi＝φ(XWei+βei),i＝1,...,n (10)
Generating ith set of mapping feature matrix Zi. n represents the number of mapping feature matrices, where WeiBeing a matrix of random weight coefficients, betaeiIs a random deviation matrix.
Step 13: mapping feature matrix ZnMultiplying by a random weight matrix WhjPlus a random deviation matrix betahjForming a jth enhanced node matrix HjI.e. by
Hj＝ξ(ZnWhj+βhj),j＝1,...,m (11)
Wherein Z isn＝[Z1,...,Zn]，Hm＝[H1,...,Hm]And m represents the number of enhanced node matrices.
Step 14: and all the mapping characteristic matrixes and the enhanced node matrixes form an augmented matrix, and the augmented matrix is multiplied by the random weight matrix W to form an output matrix Y. Wherein the random weight matrix W can be obtained by:
W＝A-1Y (12)
wherein A ═ Zn|Hm]。
Step 15: through the method, various kinds of original data are extracted and input into a width learning system in the extracted method for training; after the width learning in the method is finished, the long-term consumption of the distributed electric heat storage in the renewable energy market can be predicted by inputting the corresponding air temperature.
Drawings
FIG. 1 is a flowchart of a support vector machine width learning method according to the present invention.
Fig. 2 is a block diagram of the basic working principle of distributed electric heat storage in the renewable energy market according to the present invention.
Detailed Description
The invention provides a renewable energy market long-term heat storage method supporting vector machine width learning, which is described in detail in combination with the accompanying drawings as follows:
FIG. 1 is a flowchart of a support vector machine width learning method according to the present invention. The specific flow is as follows:
(1) inputting three training sets of class A and class B, class B and class C, and class A and class C.
(2) And respectively constructing an optimization function of the support vector machine in the method.
(3) And solving the parameters by using a sequence minimum optimization method.
(4) Three training results were obtained.
(5) Inputting original data for classification and obtaining a final classification result by adopting a voting form.
(6) And respectively extracting the raw data of each category, and respectively inputting the raw data into a width learning system in the method for training.
(7) After the width learning in the method is finished, the long-term consumption of the distributed electric heat storage in the corresponding renewable energy market can be predicted by inputting the air temperature.
Fig. 2 is a block diagram of the basic working principle of distributed electric heat storage in the renewable energy market according to the present invention. The electric energy generated by the wind driven generator and the light energy of sunlight absorbed by the solar photovoltaic device are converted into electric energy, then the electric energy is converted into heat energy, finally the heat energy is stored in the heat storage device, the long-term consumption of the distributed electric heat storage is predicted by the support vector machine width learning method, and uninterrupted heating is provided for customers in a heating season.

Claims (3)

1. A renewable energy market long-term heat storage method for support vector machine width learning is characterized in that the method combines the support vector machine and the width learning, and predicts the long-term consumption of distributed electric heat storage in the renewable energy market according to air temperature; the support vector machine in the proposed method is used to classify the raw data, i.e. air temperature, and the corresponding long-term consumption of distributed electrical heat storage in the renewable energy market; the classification categories are A, B and C; setting T as air temperature and P as long-term consumption of distributed electric heat storage in the renewable energy market; the A is the condition that T is less than 20 ℃ and P is more than or equal to 0; the B is the condition that T is more than or equal to 20 ℃ and less than or equal to 26 ℃ and P is more than or equal to 0; the C is the condition that T is more than or equal to 26 ℃ and P is more than or equal to 0; the width learning in the method is used for training the original data of each type respectively; after the width learning in the method is finished, the long-term consumption of the distributed electric heat storage in the renewable energy market can be predicted by inputting the corresponding air temperature.
2. The renewable energy market long-term heat storage method for support vector machine width learning of claim 1, wherein a kernel function is used to solve the support vector machine in the proposed method to obtain a classification decision function; the steps for solving the classification decision function are as follows:
(1) first, given input data and learning objectives: x ═ X1,...,XN},y＝{y1,...,yNN is the number of samples; each sample of input data contains two characteristics of air temperature and long term consumption of distributed electrical heat storage in the renewable energy market: xi＝[Ti,Pi]Wherein i 1.., N; and learning the target binary variable yiE { -1,1} represents a negative class and a positive class, where i ═ 1.., N;
(2) a hyperplane serving as a decision boundary exists in a feature space where input data are located, a learning target is separated according to a positive class and a negative class, and the distance from any sample point to the hyperplane is larger than or equal to 1;
decision boundary: omegaTX+b＝0；
Point-to-face distance: y isi(ωTXi+b)≥1；
In the formula, omega and b are respectively a normal vector and an intercept of the hyperplane;
(3) the original problem of the nonlinear support vector machine is that the distance from the point closest to the plane is as large as possible, and the distance between two spaced boundaries is
Figure FDA0002718008630000011
Solving the maximum value of d can be converted into solving
Figure FDA0002718008630000012
The minimum value problem of (2); namely, the original problem can be converted into:
Figure FDA0002718008630000013
s.t.yi[ωTφ(Xi)+b]≥1-ξi
ξi≥0,i＝1,2,...,N
wherein, K (X)i,Xj)＝φ(Xi)Tφ(Xj) N, j is 1,2,. N is a selected kernel function; the constant G is a normalization coefficient; xiiIs an error function;
(4) by lagrange multiplier: α ═ α1,...,αN},μ＝{μ1,...,μNIts lagrange function can be found:
Figure FDA0002718008630000021
wherein alpha isi≥0,μi≥0；
(5) Let the partial derivative of the lagrange function on the optimization target ω, b, ξ be 0, a series of expressions containing lagrange multipliers can be obtained:
Figure FDA0002718008630000022
(6) after substituting the above formula into the lagrange function, the dual problem of the original problem can be obtained as follows:
Figure FDA0002718008630000023
s.t.∑αiyi＝0,0≤αi≤G,i＝1,2,...,N,j＝1,2,...N
(7) the method for solving the dual problem of the support vector machine in the method uses an iterative sequence minimum optimization method, and the design is to select two variables alpha in a Lagrange multiplier in each iteration stepi,αjAnd fixing other parameters, wherein the constraint conditions of the dual problem are as follows:
Figure FDA0002718008630000024
wherein Const is a constant; substituting the right side of the above equation into the dual problem of the support vector machine and eliminating alpha in the summation termjCan be obtained with respect to only αiThe quadratic programming problem of (2);
(8) finding the optimal solution alpha*(ii) a Find a 0 < alphaj< G; then b is solved, and the optimal solution b is solved*Can be obtained by the following formula
Figure FDA0002718008630000025
(9) Obtaining a classification decision function:
Figure FDA0002718008630000026
wherein f (x) sign [ x ] is a sign function; when x < 0, f (x) is-1; when x is 0, f (x) is 0; when x > 0, f (x) is 1;
(10) three classification decision functions related to the A type, the B type, the C type and the A type and the C type can be obtained according to the solving method, and three training results are obtained; during classification, the original data are respectively input into three training functions for judgment; then, a voting form is adopted, and finally a group of results are obtained:
let NA，NBAnd NCThe ticket numbers of A class, B class and C class respectively;
initializing the ticket number: n is a radical ofA＝NB＝NC＝0；
Inputting the original data into a training function about class A and class B for judgment, and if the result is class A, NA＝NA+ 1; otherwise, NB＝NB+1；
Inputting the original data into training functions related to class B and class C for judgment, and if the result is class B, then NB＝NB+ 1; otherwise, NC＝NC+1；
Inputting the original data into a training function about class A and class C for judgment, and if the result is class A, NA＝NA+ 1; otherwise, NC＝NC+1；
NA，NBAnd NCThe category corresponding to the maximum value in the step (a) is the final classification result; and storing the classified original data belonging to the class A, the class B and the class C respectively while classifying.
3. The renewable energy market long-term heat storage method supporting vector machine width learning of claim 1, wherein after classification is finished, various types of raw data are respectively input into a width learning system in the method for training; the training process for breadth learning in the proposed method is as follows:
(1) let the original data be X, input the X utilization function
Zi＝φ(XWei+βei),i＝1,...,n
Generating ith set of mapping feature matrix ZiN represents the number of mapping feature matrices; wherein WeiBeing a matrix of random weight coefficients, betaeiIs a random deviation matrix;
(2) mapping feature matrix ZnMultiplying by a random weight matrix WhjPlus a random deviation matrix betahjForming a jth enhanced node matrix HjI.e. by
Hj＝ξ(ZnWhj+βhj),j＝1,...,m
Wherein Z isn＝[Z1,...,Zn]，Hm＝[H1,...,Hm]M represents the number of enhanced node matrices;
(3) all the mapping characteristic matrixes and the enhanced node matrixes form an augmented matrix, and the augmented matrix is multiplied by a random weight matrix W to form an output matrix Y; wherein the random weight matrix W can be obtained in the following manner;
W＝A-1Y
wherein A ═ Zn|Hm]；
By the method, various kinds of original data are extracted and input into a width learning system in the extracted method for training; after the width learning in the method is finished, the long-term consumption of the distributed electric heat storage in the renewable energy market can be predicted by inputting the corresponding air temperature.
CN202011079092.9A 2020-10-10 2020-10-10 Renewable energy market long-term heat storage method supporting vector machine width learning Active CN112365025B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011079092.9A CN112365025B (en) 2020-10-10 2020-10-10 Renewable energy market long-term heat storage method supporting vector machine width learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011079092.9A CN112365025B (en) 2020-10-10 2020-10-10 Renewable energy market long-term heat storage method supporting vector machine width learning

Publications (2)

Publication Number Publication Date
CN112365025A true CN112365025A (en) 2021-02-12
CN112365025B CN112365025B (en) 2022-03-25

Family

ID=74506611

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011079092.9A Active CN112365025B (en) 2020-10-10 2020-10-10 Renewable energy market long-term heat storage method supporting vector machine width learning

Country Status (1)

Country Link
CN (1) CN112365025B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103488869A (en) * 2013-08-23 2014-01-01 上海交通大学 Wind power generation short-term load forecast method of least squares support vector machine
CN105512768A (en) * 2015-12-14 2016-04-20 上海交通大学 User electricity consumption relevant factor identification and electricity consumption quantity prediction method under environment of big data
EP3130679A1 (en) * 2015-08-13 2017-02-15 Cladiac GmbH Method and test system for the detection and/or quantification of a target nucleic acid in a sample
CN107229993A (en) * 2017-05-03 2017-10-03 华北电力大学 A kind of Short-Term Load Forecasting Method based on KM APSO SVM models
CN107563539A (en) * 2017-07-24 2018-01-09 佛山市顺德区中山大学研究院 Short-term and long-medium term power load forecasting method based on machine learning model
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
CN110733038A (en) * 2019-09-30 2020-01-31 浙江工业大学 Industrial robot remote monitoring and data processing system

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103488869A (en) * 2013-08-23 2014-01-01 上海交通大学 Wind power generation short-term load forecast method of least squares support vector machine
EP3130679A1 (en) * 2015-08-13 2017-02-15 Cladiac GmbH Method and test system for the detection and/or quantification of a target nucleic acid in a sample
CN105512768A (en) * 2015-12-14 2016-04-20 上海交通大学 User electricity consumption relevant factor identification and electricity consumption quantity prediction method under environment of big data
CN107229993A (en) * 2017-05-03 2017-10-03 华北电力大学 A kind of Short-Term Load Forecasting Method based on KM APSO SVM models
CN107563539A (en) * 2017-07-24 2018-01-09 佛山市顺德区中山大学研究院 Short-term and long-medium term power load forecasting method based on machine learning model
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
CN110733038A (en) * 2019-09-30 2020-01-31 浙江工业大学 Industrial robot remote monitoring and data processing system

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
PEKKA KOPONENDENG: "Hybrid Model for Short-Term Forecasting of Loads and Load Control Responses", 《2016 IEEE PES INNOVATIVE SMART GRID TECHNOLOGIES CONFERENCE EUROPE(ISGT-EUROPE)》 *
XINYU LI等: "Prediction of Ship Fuel Consumption Based on Broad Learning System", 《2019 INTERNATIONAL CONFERENCE ON SECURITY, PATTERN ANALYSIS, AND CYBERNETICS (SPAC)》 *
刘晓娟: "基于气象因素分析的办公建筑动态短期负荷预测研究", 《中国优秀博硕士论文全文数据库（硕士）工程科技II辑》 *
古县医院办公室: "节能减排制度", 《网页在线公开：WENKU.BAIDU.COM/VIEW/7736D29A51E79B8968022646.HTML》 *
高学金等: "基于ISOA 的LS-SVM 地铁站空调***能耗预测模型", 《计算机与现代化》 *

Also Published As

Publication number Publication date
CN112365025B (en) 2022-03-25

Similar Documents

Publication Publication Date Title
Tang et al. Entropy method combined with extreme learning machine method for the short-term photovoltaic power generation forecasting
CN109508823B (en) Scene analysis method based distributed power supply planning method in power distribution network
CN109165774A (en) A kind of short-term photovoltaic power prediction technique
CN106251001A (en) A kind of based on the photovoltaic power Forecasting Methodology improving fuzzy clustering algorithm
CN102855412A (en) Wind electric power prediction method and device thereof
Guo et al. Study on short-term photovoltaic power prediction model based on the Stacking ensemble learning
CN108205717A (en) A kind of photovoltaic generation power Multiple Time Scales Forecasting Methodology
Wang et al. Self-adaptive discrete grey model based on a novel fractional order reverse accumulation sequence and its application in forecasting clean energy power generation in China
Shams et al. Artificial intelligence-based prediction and analysis of the oversupply of wind and solar energy in power systems
CN112070262B (en) Air conditioner load prediction method based on support vector machine
CN107609667B (en) Heat supply load prediction method and system based on Box _ cox transformation and UFCNN
Rao et al. An adaptive multi-team perturbation-guiding Jaya algorithm for optimization and its applications
Ibnelouad et al. Improved cooperative artificial neural network‐particle swarm optimization approach for solar photovoltaic systems using maximum power point tracking
CN112070311A (en) Day-ahead light power prediction method based on similar day clustering and meteorological factor weighting
CN113452033B (en) Method for controlling voltage of photovoltaic power distribution network with high proportion and partitioned and autonomous and storage medium
CN111401603A (en) Multi-type energy demand prediction method
CN112927097A (en) Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model
Dou et al. Elman neural network based short-term photovoltaic power forecasting using association rules and kernel principal component analysis
Li et al. Photovoltaic array prediction on short-term output power method in centralized power generation system
CN114706445B (en) DE-GWO algorithm-based photovoltaic maximum power point tracking method
CN116722540A (en) Cluster dynamic partitioning strategy based on distributed photovoltaic output prediction and clustering algorithm
CN114069687A (en) Distributed photovoltaic planning method considering reactive power regulation effect of inverter
Meng et al. A new PV generation power prediction model based on GA-BP neural network with artificial classification of history day
WO2023060579A1 (en) Solar cell structure performance prediction method
Quaiyum et al. Application of artificial neural network in forecasting solar irradiance and sizing of photovoltaic cell for standalone systems in Bangladesh

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