CN115081948B - LinUCB-based resident load aggregation method - Google Patents

LinUCB-based resident load aggregation method Download PDF

Info

Publication number
CN115081948B
CN115081948B CN202210879707.9A CN202210879707A CN115081948B CN 115081948 B CN115081948 B CN 115081948B CN 202210879707 A CN202210879707 A CN 202210879707A CN 115081948 B CN115081948 B CN 115081948B
Authority
CN
China
Prior art keywords
residents
resident
load
matrix
linucb
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
Application number
CN202210879707.9A
Other languages
Chinese (zh)
Other versions
CN115081948A (en
Inventor
张伟椿
胡秦然
俞晓荣
丁一原
吴在军
周吉
钱俊良
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.)
Liyang Research Institute of Southeast University
Original Assignee
Liyang Research Institute of Southeast 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 Liyang Research Institute of Southeast University filed Critical Liyang Research Institute of Southeast University
Priority to CN202210879707.9A priority Critical patent/CN115081948B/en
Publication of CN115081948A publication Critical patent/CN115081948A/en
Application granted granted Critical
Publication of CN115081948B publication Critical patent/CN115081948B/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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
    • Y04S20/20—End-user application control systems
    • Y04S20/222—Demand response systems, e.g. load shedding, peak shaving

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a resident load aggregation method based on LinUCB, which comprises the following steps: step S1: acquisition of totality in the areanThe method comprises the following steps that (1) historical characteristic data of a demand response event is participated by residents, and normalization processing is carried out; step S2: calculating and calculating with each residentiRespectively corresponding matrixes; and step S3: when the system is intThe load is required to be reduced at any timeDAnd acquiring the feature vectors of all residents at the time t
Figure 100004_DEST_PATH_IMAGE001
Calculating each residentiCharacteristic weight coefficient of
Figure 36629DEST_PATH_IMAGE002
And desired value of power adjustment
Figure 100004_DEST_PATH_IMAGE003
(ii) a And step S4: desired value for the power adjustment
Figure 49716DEST_PATH_IMAGE003
Performing descending order and selecting in sequence
Figure 709324DEST_PATH_IMAGE004
Individual residents until the requirements are met
Figure 100004_DEST_PATH_IMAGE005
And sending load adjustment instructions to residents; step S5: according to the actual response condition of the selected residents, updating the two corresponding matrixes
Figure 666916DEST_PATH_IMAGE006
And
Figure 100004_DEST_PATH_IMAGE007
. The load aggregation method is beneficial to improving the utilization rate of resident load resources.

Description

Resident load aggregation method based on LinUCB
Technical Field
The invention relates to the field of load aggregation of power systems, in particular to a residential load aggregation method based on LinUCB.
Background
With the continuous development of economy in China, the living standard of people is gradually improved, the electricity demand in China is continuously increased, and the problem of unbalanced power supply and demand is easily caused. In addition, the load of residents has become a major component of peak load in summer, wherein the air conditioning load accounts for 30%, and the local economic developed area even exceeds 50%. Therefore, if the load of the residents can be controlled by a proper demand response strategy to reduce the power, the power supply pressure of the power system can be greatly relieved, the electric energy consumption of the residents can be reduced, and the win-win situation is achieved.
Existing demand response contracts often provide the resident with the option of quitting, i.e., the resident will refuse to adjust his or her power during the present load aggregation event. However, such a response behavior of the residents has complicated uncertainties, and is affected by its own factors and the external environment. When the load needs to be reduced in the operation process of the power system, the load aggregator manages the residents in jurisdiction, and the appropriate residents are selected to send the adjusting instructions, so that the load aggregation is realized. In order to enable accurate and reliable load aggregation, it is a crucial requirement to fully evaluate the uncertainty of each resident so as to make a correct choice.
Currently, in the research on the load aggregation of residents, the uncertainty of residents is of little concern, or the uncertainty of residents is simply modeled without considering actual influencing factors, so that the actual power regulation result deviates from the regulation target required by the system, and the participation of the load resources of residents in the power grid scheduling is not facilitated.
Disclosure of Invention
The invention aims to provide a LinUCB-based resident load aggregation method, which considers the influence of various factors on the load aggregation behavior of residents, learns the demand response characteristic of the residents based on the LinUCB method, can avoid blindly sending an adjusting instruction in each event, improves the actual load aggregation effect and is beneficial to improving the utilization rate of resident load resources.
The purpose of the invention can be realized by the following technical scheme:
a resident load aggregation method based on LinUCB comprises the following steps:
step S1: acquisition of totality in the areanThe method comprises the following steps that (1) historical characteristic data of a demand response event is participated by residents, and normalization processing is carried out;
step S2: based on LinUCB algorithm, calculation is carried out on each residentiRespectively corresponding matrix
Figure 100002_DEST_PATH_IMAGE001
And
Figure 75958DEST_PATH_IMAGE002
wherein, in the step (A),
Figure 236812DEST_PATH_IMAGE001
is a matrix of coefficients in linear terms,
Figure 911507DEST_PATH_IMAGE002
is a bias matrix;
and step S3: when the system is intThe load is required to be reduced at any momentDObtaining the feature vectors of all residents at the time t
Figure 100002_DEST_PATH_IMAGE003
Calculating each residentiCharacteristic weight coefficient of
Figure 169313DEST_PATH_IMAGE004
And desired value of power adjustment
Figure 100002_DEST_PATH_IMAGE005
；
And step S4: desired value for the power adjustment
Figure 677786DEST_PATH_IMAGE005
Performing descending arrangement, and sequentially selecting
Figure 591516DEST_PATH_IMAGE006
Individual residents until the requirements are met
Figure 100002_DEST_PATH_IMAGE007
And sending a load adjustment instruction to residents;
step S5: according to the actual response condition of the selected residents, updating the two corresponding matrixes
Figure 335481DEST_PATH_IMAGE001
And
Figure 651056DEST_PATH_IMAGE002
。
in step S1, the whole is treatednThe performance of each demand response event of the residents in the past is obtained from the residents and external factors respectivelydHistorical feature data of the dimension; to pairdNormalizing the historical characteristic data of the dimension to an interval [0, 1]In, all residents' feature vectors are represented as
Figure 189484DEST_PATH_IMAGE008
。
In the step S2, the historical characteristic data is used as a sample set, and a LinUCB algorithm is adopted for training, wherein the specific method comprises the following steps:
1) Respectively initializing two corresponding matrixes for all residents
Figure 324930DEST_PATH_IMAGE001
And
Figure 607007DEST_PATH_IMAGE002
: for matrix
Figure 901722DEST_PATH_IMAGE001
Initialize it tod×dAn identity matrix of dimensions; for the
Figure 79894DEST_PATH_IMAGE002
Is initialized tod0 matrix of x 1 dimension;
2) Calculate each residentiCharacteristic weight coefficient of
Figure 100002_DEST_PATH_IMAGE009
；
3) Calculate each residentiExpected profit of
Figure 634460DEST_PATH_IMAGE010
WhereinαIs a constant for weighing the size of the two terms; t represents a transposed symbol of the matrix;
4) Based on each residentiHistory of response conditions
Figure 100002_DEST_PATH_IMAGE011
Updating the matrix of each resident:
Figure 985807DEST_PATH_IMAGE012
Figure 100002_DEST_PATH_IMAGE013
5) Repeating the above processes until all the resident historical data are utilized, and outputting the final matrix
Figure 682499DEST_PATH_IMAGE001
And
Figure 156205DEST_PATH_IMAGE002
。
in step S3, when the system is intThe load is required to be reduced at any timeDAnd acquiring the feature vectors of all residents at the time t
Figure 266244DEST_PATH_IMAGE003
Calculating each residentiCharacteristic weight coefficient of
Figure 624544DEST_PATH_IMAGE004
And desired value of power adjustment
Figure 503638DEST_PATH_IMAGE005
：
Figure 413825DEST_PATH_IMAGE014
Figure 100002_DEST_PATH_IMAGE015
。
In step S4, expected income to all residents
Figure 417685DEST_PATH_IMAGE005
Sorting in descending order, selecting in sequence
Figure 579676DEST_PATH_IMAGE006
Individual residents until the requirements are met
Figure 703489DEST_PATH_IMAGE007
And sends a load adjustment instruction to these residents.
In step S5, the actual power regulation condition of the selected residents is obtained
Figure 659944DEST_PATH_IMAGE016
If the load adjustment is completed, the value is 1, otherwise the value is 0; updating the matrix for each of the selected inhabitants
Figure 478996DEST_PATH_IMAGE001
And
Figure 179098DEST_PATH_IMAGE002
so as to facilitate the next load aggregation event call;
Figure 100002_DEST_PATH_IMAGE017
Figure 829523DEST_PATH_IMAGE018
。
the invention has the beneficial effects that:
the load aggregation method considers the influence of various factors on the load aggregation behavior of the user, and based on the LinUCB method to learn the demand response characteristic of the user, the blind sending of the adjusting instruction can be avoided in each event, the actual load aggregation effect is improved, and the utilization rate of resident load resources is improved.
Drawings
The invention will be further described with reference to the accompanying drawings.
FIG. 1 is a flow chart of the load polymerization process 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.
As shown in FIG. 1, the invention provides a resident load aggregation method based on LinUCB, which comprises the following steps:
step S1: acquisition of totality in the areanThe residents participate in the historical characteristic data of the demand response event and carry out normalizationC, trimming;
in particular, for the wholenThe performance of each demand response event of the residents in the past is obtained from the residents and external factors respectivelydHistorical feature data of the dimension; to pairdNormalizing the historical characteristic data of the dimension to an interval [0, 1]In, feature vectors of all residents are represented as
Figure 691299DEST_PATH_IMAGE008
。
Step S2: based on LinUCB algorithm, calculation is carried out on each residentiRespectively corresponding matrix
Figure 522946DEST_PATH_IMAGE001
And
Figure 885794DEST_PATH_IMAGE002
wherein
Figure 593987DEST_PATH_IMAGE001
Is a matrix of coefficients in linear terms,
Figure 626665DEST_PATH_IMAGE002
is a bias matrix;
specifically, training is performed by adopting a LinUCB algorithm based on past historical data as a sample set, and the method mainly comprises the following steps:
taking historical characteristic data as a sample set, and training by adopting a LinUCB algorithm, wherein the specific method comprises the following steps:
1) Respectively initializing two corresponding matrixes for all residents
Figure 685888DEST_PATH_IMAGE001
And
Figure 993373DEST_PATH_IMAGE002
: for matrix
Figure 946285DEST_PATH_IMAGE001
Initialize it tod×dAn identity matrix of dimensions; to pairIn the process
Figure 415444DEST_PATH_IMAGE002
Is initialized tod0 matrix of x 1 dimension;
2) Calculate each residentiCharacteristic weight coefficient of (2)
Figure 227542DEST_PATH_IMAGE009
；
3) Calculate each residentiExpected profit of
Figure 807559DEST_PATH_IMAGE010
In whichαIs a constant for weighing the size of the two terms;
4) Based on each residentiHistory of response conditions
Figure 490344DEST_PATH_IMAGE011
Updating the matrix for each resident:
Figure 395983DEST_PATH_IMAGE012
Figure 23274DEST_PATH_IMAGE013
5) Repeating the above processes until all the resident historical data are utilized, and outputting the final matrix
Figure 406982DEST_PATH_IMAGE001
And
Figure 475432DEST_PATH_IMAGE002
。
and step S3: when the system is intThe load is required to be reduced at any momentDObtaining the feature vectors of all residents at the time t
Figure 20814DEST_PATH_IMAGE003
Calculating each residentiCharacteristic weight coefficient of
Figure 135400DEST_PATH_IMAGE004
And desired value of power adjustment
Figure 588378DEST_PATH_IMAGE005
；
In particular, when the system is intThe load is required to be reduced at any momentDObtainingtFeature vectors of all residents at the moment
Figure 980177DEST_PATH_IMAGE003
Calculating each residentiCharacteristic weight coefficient of
Figure 962039DEST_PATH_IMAGE004
And desired value of power adjustment
Figure 563922DEST_PATH_IMAGE005
：
Figure 549152DEST_PATH_IMAGE014
Figure 795457DEST_PATH_IMAGE015
。
And step S4: desired value for the power adjustment
Figure 948221DEST_PATH_IMAGE005
Performing descending arrangement, and sequentially selecting
Figure 302979DEST_PATH_IMAGE006
Individual residents until they meet
Figure 566601DEST_PATH_IMAGE007
And sending a load adjustment instruction to residents;
step S5: according to the actual response condition of the selected residents, updating two matrixes corresponding to the selected residents
Figure 198570DEST_PATH_IMAGE001
And
Figure 522236DEST_PATH_IMAGE002
。
specifically, the actual power regulation condition of the selected residents is obtained
Figure 98710DEST_PATH_IMAGE016
If the load adjustment is completed, the value is 1, otherwise the value is 0; updating the matrix for each of the selected inhabitants
Figure 166023DEST_PATH_IMAGE001
And
Figure 652500DEST_PATH_IMAGE002
so as to facilitate the next load aggregation event call;
Figure 147066DEST_PATH_IMAGE017
Figure 210837DEST_PATH_IMAGE018
。
in the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed.

Claims (2)

1. A resident load aggregation method based on LinUCB is characterized by comprising the following steps:
step S1: acquire the whole of the regionnThe method comprises the following steps that (1) historical characteristic data of a demand response event is participated by residents, and normalization processing is carried out;
step S2: based on LinUCB algorithm, calculation is carried out on each residentiRespectively corresponding matrix
Figure DEST_PATH_IMAGE001
And
Figure DEST_PATH_IMAGE002
wherein, in the step (A),
Figure 243215DEST_PATH_IMAGE001
is a final coefficient matrix of the linear terms,
Figure 959499DEST_PATH_IMAGE002
is the final bias matrix;
and step S3: when the system is intThe load is required to be reduced at any momentDObtaining the feature vectors of all residents at the time t
Figure DEST_PATH_IMAGE003
Calculating each residentiCharacteristic weight coefficient of
Figure DEST_PATH_IMAGE004
And desired value of power adjustment
Figure DEST_PATH_IMAGE005
；
And step S4: desired value for the power adjustment
Figure 764643DEST_PATH_IMAGE005
Performing descending arrangement, and sequentially selecting
Figure DEST_PATH_IMAGE006
Individual residents until the requirements are met
Figure DEST_PATH_IMAGE007
And sending a load adjustment instruction to residents;
step S5: according to the actual response condition of the selected residents, updating two final matrixes corresponding to the selected residents
Figure 755733DEST_PATH_IMAGE001
And
Figure 329934DEST_PATH_IMAGE002
；
in step S1, the whole is treatednThe performance of each demand response event of the residents in the past is obtained from the residents and the external factors respectivelydHistorical feature data of the dimension; to pairdNormalizing the historical characteristic data of the dimension to an interval [0, 1]In, all residents' feature vectors are represented as
Figure DEST_PATH_IMAGE008
；
In step S2, the historical characteristic data is used as a sample set, and a LinUCB algorithm is adopted for training, wherein the specific method comprises the following steps:
1) Respectively initializing two corresponding matrixes for all residents
Figure 13856DEST_PATH_IMAGE001
And
Figure 40718DEST_PATH_IMAGE002
: for matrix
Figure 569920DEST_PATH_IMAGE001
Initialize it tod×dAn identity matrix of dimensions; for the
Figure 795365DEST_PATH_IMAGE002
Is initialized tod0 matrix of x 1 dimension;
2) Calculate each residentiCharacteristic weight coefficient of (2)
Figure DEST_PATH_IMAGE009
；
3) Calculate each residentiExpected profit of
Figure DEST_PATH_IMAGE010
In whichαIs a constant for weighing the size of the two terms; t represents a transposed symbol of the matrix;
4) Based on each residentiHistory of response conditions
Figure DEST_PATH_IMAGE011
Updating the matrix of each resident:
Figure DEST_PATH_IMAGE012
Figure DEST_PATH_IMAGE013
5) Repeating the processes of 2) to 4) until all the resident historical data are utilized, and outputting a final matrix
Figure 322292DEST_PATH_IMAGE001
And
Figure 102029DEST_PATH_IMAGE002
；
in step S3, when the system is intThe load is required to be reduced at any timeDObtaining the feature vectors of all residents at the time t
Figure 918411DEST_PATH_IMAGE003
Calculating each residentiCharacteristic weight coefficient of
Figure 263942DEST_PATH_IMAGE004
And desired value of power adjustment
Figure 758509DEST_PATH_IMAGE005
：
Figure 759963DEST_PATH_IMAGE009
Figure DEST_PATH_IMAGE014
；
In step S4, expected income to all residents
Figure 162125DEST_PATH_IMAGE005
Sorting in descending order, selecting in sequence
Figure 299845DEST_PATH_IMAGE006
Individual residents until the requirements are met
Figure DEST_PATH_IMAGE015
And sends the load adjustment instruction to the residents.
2. The method as claimed in claim 1, wherein in step S5, the actual power regulation status of the selected residents is obtained
Figure DEST_PATH_IMAGE016
If the load adjustment is finished, the value is 1, otherwise, the value is 0; updating the matrix for each of the selected inhabitants
Figure 496472DEST_PATH_IMAGE001
And
Figure 781959DEST_PATH_IMAGE002
so as to facilitate the next load aggregation event call;
Figure DEST_PATH_IMAGE017
Figure DEST_PATH_IMAGE018
。
CN202210879707.9A 2022-07-25 2022-07-25 LinUCB-based resident load aggregation method Active CN115081948B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210879707.9A CN115081948B (en) 2022-07-25 2022-07-25 LinUCB-based resident load aggregation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210879707.9A CN115081948B (en) 2022-07-25 2022-07-25 LinUCB-based resident load aggregation method

Publications (2)

Publication Number Publication Date
CN115081948A CN115081948A (en) 2022-09-20
CN115081948B true CN115081948B (en) 2023-04-07

Family

ID=83242436

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210879707.9A Active CN115081948B (en) 2022-07-25 2022-07-25 LinUCB-based resident load aggregation method

Country Status (1)

Country Link
CN (1) CN115081948B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109812946A (en) * 2019-01-31 2019-05-28 河海大学 A kind of control method suitable for extensive residual air-conditioning load group demand response
CN113256031A (en) * 2021-06-25 2021-08-13 国网江西省电力有限公司供电服务管理中心 Self-learning optimization method based on resident demand response strategy
CN113468413A (en) * 2021-06-07 2021-10-01 南京邮电大学 Multi-user sharing-oriented multimedia network video recommendation method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109812946A (en) * 2019-01-31 2019-05-28 河海大学 A kind of control method suitable for extensive residual air-conditioning load group demand response
CN113468413A (en) * 2021-06-07 2021-10-01 南京邮电大学 Multi-user sharing-oriented multimedia network video recommendation method
CN113256031A (en) * 2021-06-25 2021-08-13 国网江西省电力有限公司供电服务管理中心 Self-learning optimization method based on resident demand response strategy

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
erwin.MAB系列2：Contextual Bandits: LinUCB.知乎.2021,第1-7页. *

Also Published As

Publication number Publication date
CN115081948A (en) 2022-09-20

Similar Documents

Publication Publication Date Title
CN110131843B (en) Intelligent air conditioner regulation and control method and system based on big data
CN109799851B (en) User side load response method based on load cluster temperature regulation and control
CN111561732B (en) Heat exchange station heat supply adjusting method and system based on artificial intelligence
CN113991655B (en) Method, device and medium for evaluating load aggregation demand response potential of fixed-frequency air conditioner
CN105790286A (en) Strategy for central air conditioner load aggregation and participation in distributed power supply output stabilization
CN108512258A (en) A kind of wind power plant active power dispatch method based on improvement multiple agent consistency algorithm
CN113255138B (en) Load distribution optimization method for power system
CN109391511B (en) Intelligent communication resource allocation strategy based on expandable training network
CN115882463A (en) Commercial building air conditioner load schedulable potential evaluation method
CN113972667A (en) Power distribution network reactive power optimization method based on one-dimensional convolutional neural network
CN116184828A (en) Online real-time optimization method and system for high-speed valve characteristic curve of self-adaptive steam turbine
CN113610330B (en) User experience-based user-side flexible resource energy utilization behavior optimization method
CN105719081A (en) Power system dynamic economic scheduling method based on improved artificial bee colony algorithm
CN115081948B (en) LinUCB-based resident load aggregation method
CN113888202A (en) Training method and application method of electricity price prediction model
CN116307074B (en) Method for acquiring real thermal data, system and method for constructing neural network model of thermal data, and method for predicting thermal load
CN112288161A (en) Method and device for optimizing peak-shifting electricity consumption of residents
CN108805445B (en) Grouping sequence scheduling method for providing rotary standby for air conditioner load group
CN108155639B (en) Method for identifying wide-area adjustable load
CN114943140B (en) Method for evaluating response potential of air conditioner clusters under grouping regulation considering user experience
CN115051374A (en) Control method and device for electric heating equipment to participate in electric power peak regulation and storage medium
CN115276002A (en) Intelligent power distribution property studying and judging method based on power load curve
CN115377963B (en) Novel temperature set value control method for heterogeneous cluster air conditioner load
CN113969873B (en) Ultra-short term output prediction method and system for wind turbine generator and storage medium
CN115130899B (en) Kmeas-GM-based air conditioner load day-ahead response capacity evaluation method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant