CN115660385B - Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid - Google Patents

Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid Download PDF

Info

Publication number
CN115660385B
CN115660385B CN202211590666.8A CN202211590666A CN115660385B CN 115660385 B CN115660385 B CN 115660385B CN 202211590666 A CN202211590666 A CN 202211590666A CN 115660385 B CN115660385 B CN 115660385B
Authority
CN
China
Prior art keywords
convex hull
economic operation
layer optimization
power grid
double
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
CN202211590666.8A
Other languages
Chinese (zh)
Other versions
CN115660385A (en
Inventor
徐华廷
郭创新
冯华
章姝俊
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.)
Zhejiang University ZJU
State Grid Zhejiang Electric Power Co Ltd
Original Assignee
Zhejiang University ZJU
State Grid Zhejiang Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU, State Grid Zhejiang Electric Power Co Ltd filed Critical Zhejiang University ZJU
Priority to CN202211590666.8A priority Critical patent/CN115660385B/en
Publication of CN115660385A publication Critical patent/CN115660385A/en
Application granted granted Critical
Publication of CN115660385B publication Critical patent/CN115660385B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention discloses a power grid convex hull economic operation domain decomposition parallel solving method and device. Aiming at the problem of new energy consumption, the invention provides a method for solving the decomposition parallel of the convex hull economic operation domain for describing the influence of the uncertainty of the new energy output on the power grid dispatching plan. Firstly, constructing a first double-layer optimization model capable of being solved in parallel, and determining the dimension of the economic operation domain of the convex hull at each moment; then, initializing an initial convex hull containing initial economic operation points at each moment; and finally, constructing a second double-layer optimization model for expanding the top of the convex hull economic operation domain, and providing a double-layer iterative algorithm which can be executed in parallel to obtain the final convex hull economic operation domain. Compared with the traditional serial solving algorithm, the method has higher solving efficiency, and the obtained convex hull economic operation domain can be used for evaluating the safety and the economical efficiency of the operation of the power grid in real time on one hand, and can support intelligent fine scheduling of the power grid and support realization of an automatic cruising technology of the power grid on the other hand.

Description

Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid
Technical Field
The invention belongs to the technical field of large power grid refined intelligent scheduling, and particularly relates to a power grid convex hull economic operation domain decomposition parallel solving method and device.
Background
In recent years, excessive emission of greenhouse gases causes global climate change and extreme weather frequency. Therefore, in order to seek a sustainable development route, renewable energy sources need to be developed greatly, and carbon dioxide emission is reduced. However, the carbon emission of the energy industry in the current stage of China accounts for more than 80% of the total national quantity, wherein the carbon emission of the power industry accounts for more than 40% of the total national quantity. Therefore, the carbon emission of the power industry is reduced, the renewable energy consumption is improved, and the realization of the 'double carbon' target can be effectively supported.
In the process of implementing the present invention, the inventor finds that at least the following problems exist in the prior art:
in order to ensure safe and stable operation of the power grid, students in the field of power grid dispatching put forward a power grid security domain concept, but the power grid security domain only characterizes a power grid security operation boundary, and economy is not considered, so that carbon emission is not reduced.
Disclosure of Invention
Aiming at the defects of the prior art, the embodiment of the application aims to draw the economic operation boundary of the power grid on the basis of ensuring the safety of the power grid, and provides a decomposition parallel solving method and device of the economic operation domain of the convex hull of the power grid. On the basis of ensuring the safety of the power grid, new energy is maximally consumed, carbon emission is reduced, and realization of fine intelligent scheduling of a large power grid is supported.
According to a first aspect of an embodiment of the present application, there is provided a method for resolving and parallel solving an economic operation domain of a convex hull of a power grid, which is applied to the power grid, and includes:
constructing a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid at all times along with the change of the output of the new energy based on a given power grid foundation optimization scheduling model and prediction information of the new energy and the load, and determining the economic operation domain dimension of the convex hull of the power grid at all times based on the constructed first double-layer optimization model;
generating a polyhedral convex hull which meets the calculation precision requirement as small as possible and contains the initial economic operation point at each moment based on a generator number set and a corresponding convex hull economic operation domain dimension of the economic operation point which changes along with the new energy output at each moment, and obtaining a hyperplane set corresponding to the polyhedral convex hull;
constructing a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points or not based on a given power grid basic optimization scheduling model, prediction information of new energy and load, the polyhedral convex hull and a hyperplane set thereof, and converting the second double-layer optimization model into a corresponding second single-layer optimization model;
And constructing a double-layer iterative algorithm which can be decomposed and executed in parallel based on the second single-layer optimization model and the rapid convex hull algorithm, and gradually expanding the convex hulls in the polyhedral form at each moment until all economic operation points can be contained, so as to obtain a final convex hull economic operation domain.
Further, based on a given power grid basic optimization scheduling model and prediction information of new energy and load, a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid at all times along with the change of new energy output is constructed, and the dimension of the convex hull economic operation domain of the power grid at all times is determined based on the constructed first double-layer optimization model, and the method comprises the following steps:
step S11: converting a given grid base optimization scheduling model into an equivalent compact form thereof;
step S12: according to a compact form equivalent to the power grid basic optimization scheduling model, constructing a first double-layer optimization model A and a second double-layer optimization model B which can be decomposed and executed in parallel and are used for determining a generator number set of economic operation points of the power grid at each moment along with the change of new energy output;
step S13: converting the lower layer optimization problem of the first double-layer optimization model into a KKT condition form thereof;
step S14: converting the nonlinear non-convex constraint in the KKT conditional form of the lower-layer optimization problem into a linear mixed integer equivalent form;
Step S15: converting the first double-layer optimization models A and B into first single-layer optimization models C and D according to the KKT conditional form of the lower-layer optimization problem and the linear mixed integer equivalent form of nonlinear non-convex constraint;
step S16: at the position of
Figure 390212DEST_PATH_IMAGE001
Time of day for a generator
Figure 318985DEST_PATH_IMAGE002
Solving the first single-layer optimization model C and the first single-layer optimization model C based on a solverD, numbering the corresponding generators which are not equal to the objective function value of the first single-layer optimization model C and DiAs a generator number set of economic operating points varying with the output of new energy
Figure 237393DEST_PATH_IMAGE003
Wherein
Figure 593419DEST_PATH_IMAGE001
Will be
Figure 230068DEST_PATH_IMAGE003
The number of generators contained in (a)
Figure 228111DEST_PATH_IMAGE004
Is arranged as a power grid
Figure 797764DEST_PATH_IMAGE001
The moment convex hull is economical in terms of the dimension of the run domain.
Further, based on the generator number set and the corresponding convex hull economic operation domain dimension of the economic operation point changing with the new energy output at each moment, generating a polyhedral convex hull containing the initial economic operation point as small as possible and meeting the calculation precision requirement at each moment, and obtaining a hyperplane set corresponding to the polyhedral convex hull, wherein the method comprises the following steps:
step S21: by means oftNew energy unit at momentjIs the predicted force of (2)
Figure 777221DEST_PATH_IMAGE005
Solving a power grid foundation optimization scheduling model to obtain the optimal output column vector of each generator at each moment
Figure 697904DEST_PATH_IMAGE006
Wherein
Figure 234059DEST_PATH_IMAGE001
In the column vector
Figure 861480DEST_PATH_IMAGE006
In (3) take outGenerator numbering set with economic operating point changing along with new energy output
Figure 824888DEST_PATH_IMAGE003
The corresponding active output of the generator is used for obtaining the column vector
Figure 221148DEST_PATH_IMAGE007
；
Step S22: initializing a constant as small as possible under the condition of meeting the calculation accuracy requirement
Figure 498677DEST_PATH_IMAGE008
In the following
Figure 42922DEST_PATH_IMAGE001
At the moment, in sequence in the column vector
Figure 380493DEST_PATH_IMAGE007
Adding a constant to each dimension of (2)
Figure 744610DEST_PATH_IMAGE008
Obtaining
Figure 888146DEST_PATH_IMAGE004
Column vectors of linear uncorrelated
Figure 739428DEST_PATH_IMAGE009
Wherein
Figure 44638DEST_PATH_IMAGE010
Is at
Figure 161630DEST_PATH_IMAGE007
Is the first of (2)iAdding constants to dimensions
Figure 780961DEST_PATH_IMAGE008
The column vector obtained;
step S23: at the position of
Figure 299798DEST_PATH_IMAGE001
At the moment of time of day,
Figure 979172DEST_PATH_IMAGE011
individual operating points
Figure 583460DEST_PATH_IMAGE012
Is set of (a)
Figure 803220DEST_PATH_IMAGE013
Namely, a small enough polyhedral convex hull containing initial economic operation points is corresponding to the convex hull; obtaining the half-space representation form of the polyhedral convex hull by using a rapid convex hull algorithm, namely a hyperplane set
Figure 379826DEST_PATH_IMAGE014
。
Further, based on the given power grid basic optimization scheduling model, the prediction information of new energy and load, the polyhedral convex hull and the hyperplane set thereof, a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points is constructed, and the second double-layer optimization model is converted into a corresponding second single-layer optimization model, which comprises the following steps:
s31: based on a given power grid basic optimization scheduling model, prediction information of new energy and load, the polyhedral convex hull and a hyperplane set thereof, a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points is constructed as follows:
Objective function:
Figure 479369DEST_PATH_IMAGE015
(9a)
constraint conditions:
Figure 570953DEST_PATH_IMAGE016
(9b)
Figure 797666DEST_PATH_IMAGE017
(9c)
Figure 302814DEST_PATH_IMAGE018
(9d)
in the method, in the process of the invention,
Figure 855149DEST_PATH_IMAGE019
is thattMoment convex hull economic operation domain
Figure 637292DEST_PATH_IMAGE020
A hyperplane numbering set in the half-space representation;
Figure 198854DEST_PATH_IMAGE021
and
Figure 281211DEST_PATH_IMAGE022
respectively istMoment convex hull economic operation domain
Figure 738868DEST_PATH_IMAGE020
Numbered under the half-space representation form
Figure 991995DEST_PATH_IMAGE023
Is a super plane of (a)
Figure 91669DEST_PATH_IMAGE024
Coefficient column vectors and intercepts of (i) i.e
Figure 559691DEST_PATH_IMAGE025
；
Figure 188249DEST_PATH_IMAGE026
Is thattTime of day, collection
Figure 210563DEST_PATH_IMAGE003
The column vector formed by the active power output of the generator,
Figure 317190DEST_PATH_IMAGE027
output for schedulable machine set
Figure 639718DEST_PATH_IMAGE028
A column vector of components;
Figure 173599DEST_PATH_IMAGE029
to output new energy
Figure 214367DEST_PATH_IMAGE005
A column vector of components;
Figure 108374DEST_PATH_IMAGE030
and
Figure 19829DEST_PATH_IMAGE031
respectively charge power of the energy storage device
Figure 990191DEST_PATH_IMAGE032
And discharge power
Figure 975378DEST_PATH_IMAGE033
A column vector of components;
Figure 486125DEST_PATH_IMAGE034
discarding electric power for new energy
Figure 720928DEST_PATH_IMAGE035
A column vector of components;
Figure 658928DEST_PATH_IMAGE036
to discard load power
Figure 346393DEST_PATH_IMAGE037
A column vector of components;
Figure 129672DEST_PATH_IMAGE038
storing electrical energy for an energy storage device
Figure 281299DEST_PATH_IMAGE039
A column vector of components;
Figure 577151DEST_PATH_IMAGE040
、
Figure 814228DEST_PATH_IMAGE041
、
Figure 604461DEST_PATH_IMAGE042
、
Figure 610594DEST_PATH_IMAGE043
、
Figure 890397DEST_PATH_IMAGE044
、
Figure 83612DEST_PATH_IMAGE045
、
Figure 677536DEST_PATH_IMAGE046
and
Figure 803755DEST_PATH_IMAGE047
a corresponding coefficient matrix constrained by inequality;
Figure 192142DEST_PATH_IMAGE048
、
Figure 872653DEST_PATH_IMAGE049
、
Figure 519535DEST_PATH_IMAGE050
、
Figure 511979DEST_PATH_IMAGE051
、
Figure 868005DEST_PATH_IMAGE052
、
Figure 504654DEST_PATH_IMAGE053
、
Figure 237118DEST_PATH_IMAGE054
and
Figure 275612DEST_PATH_IMAGE055
corresponding coefficient matrixes respectively constrained by equations; superscriptTRepresenting a transpose of the matrix;
Figure 68119DEST_PATH_IMAGE056
and
Figure 926484DEST_PATH_IMAGE057
column vectors consisting of a lower limit and an upper limit of the predicted output error of the new energy unit are respectively formed;
s32: converting the second two-layer optimization model into a second equivalent single-layer optimization model according to steps S13, S14 and S15 as follows:
objective function:
Figure 462639DEST_PATH_IMAGE058
(10a)
constraint conditions:
Figure 604907DEST_PATH_IMAGE016
(10b)
Figure 568315DEST_PATH_IMAGE059
，
Figure 710715DEST_PATH_IMAGE060
，
Figure 253823DEST_PATH_IMAGE061
，
Figure 798068DEST_PATH_IMAGE062
，
Figure 870060DEST_PATH_IMAGE063
，
Figure 765335DEST_PATH_IMAGE064
，
Figure 112134DEST_PATH_IMAGE065
，
Figure 510885DEST_PATH_IMAGE066
，
Figure 3046DEST_PATH_IMAGE067
，
Figure 373899DEST_PATH_IMAGE068
(10c)
in the method, in the process of the invention,
Figure 993230DEST_PATH_IMAGE069
representing the optimal solution result of the second single-layer optimization model,
Figure 512067DEST_PATH_IMAGE070
And
Figure 191441DEST_PATH_IMAGE071
column vectors consisting of lagrangian multipliers;
Figure 530150DEST_PATH_IMAGE072
representing diagonal elements as
Figure 15489DEST_PATH_IMAGE070
Is a diagonal matrix of the (a),
Figure 326516DEST_PATH_IMAGE073
i.e.
Figure 239108DEST_PATH_IMAGE074
Is a column vector consisting of 0 or 1; m is a sufficiently large constant.
Further, constructing a double-layer iterative algorithm which can be decomposed and executed in parallel based on the second single-layer optimization model and the rapid convex hull algorithm, gradually expanding the polyhedral convex hulls at all times until all economic operation points are contained, and obtaining a final convex hull economic operation domain, wherein the method comprises the following steps:
s41: initialization ofRelevant calculation parameters: setting convergence criterion delta, and inputting an economic operation domain of a convex hull to be solved
Figure 517643DEST_PATH_IMAGE020
Number of time periods of (a)
Figure 541093DEST_PATH_IMAGE075
Fluctuation range data of new energy active output predicted value
Figure 972206DEST_PATH_IMAGE056
And
Figure 258962DEST_PATH_IMAGE057
convex hull economic operation domain at each moment
Figure 572263DEST_PATH_IMAGE020
Dimension of (2)
Figure 399404DEST_PATH_IMAGE004
Corresponding generator number set
Figure 950602DEST_PATH_IMAGE003
Half-space representation parameters of initial convex hull at each moment
Figure 939418DEST_PATH_IMAGE021
And
Figure 192545DEST_PATH_IMAGE022
initializing each time
Figure 557798DEST_PATH_IMAGE020
Is a hyperplane set of (1)
Figure 975222DEST_PATH_IMAGE076
Initializing each time
Figure 134939DEST_PATH_IMAGE020
Vertex set of (a)
Figure 626094DEST_PATH_IMAGE077
，
Figure 795038DEST_PATH_IMAGE078
；
S42: solving the expansion vertex of the current convex hull: based on the second monolayer optimization model, pair
Figure 320829DEST_PATH_IMAGE079
Time of day
Figure 651447DEST_PATH_IMAGE080
Hyperplane, solving and calculating optimization result
Figure 879166DEST_PATH_IMAGE081
Obtaining column vectors
Figure 320643DEST_PATH_IMAGE082
And update
Figure 700940DEST_PATH_IMAGE083
；
S43: updating extended vertices meeting the condition: if it is
Figure 468039DEST_PATH_IMAGE084
Then update
Figure 730524DEST_PATH_IMAGE085
The method comprises the steps of carrying out a first treatment on the surface of the If it is
Figure 975692DEST_PATH_IMAGE086
Step S45 is performed;
s44: updating the half-space representation of the current convex hull: based on a fast convex hull algorithm, calculation
Figure 272812DEST_PATH_IMAGE079
Time vertex set
Figure 148495DEST_PATH_IMAGE087
Is obtained in the form of a half-space representation of (a)
Figure 898276DEST_PATH_IMAGE019
And returns to step S42;
s45: outputting a convex hull economic operation domain: output of
Figure 884818DEST_PATH_IMAGE079
Time of day
Figure 239707DEST_PATH_IMAGE020
Form parameters of the half-space representation of (a)
Figure 71311DEST_PATH_IMAGE021
And
Figure 308388DEST_PATH_IMAGE022
thereby obtaining the final convex hull economic operation domain
Figure 347888DEST_PATH_IMAGE025
。
According to a second aspect of the embodiments of the present application, there is provided a decomposition parallel solving apparatus for a grid convex hull economic operation domain, applied to a grid, including:
the first construction module is used for constructing a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid at all times along with the change of the output of the new energy based on a given power grid foundation optimization scheduling model and the prediction information of the new energy and the load, and determining the dimension of the economic operation domain of the convex hull of the power grid at all times based on the constructed first double-layer optimization model;
the generation module is used for generating a polyhedral convex hull which meets the calculation precision requirement as small as possible and comprises an initial economic operation point at each moment and obtaining a hyperplane set corresponding to the polyhedral convex hull based on a generator number set of which the economic operation point at each moment changes along with the new energy output and the dimension of the corresponding convex hull economic operation domain;
The second construction module is used for constructing a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points or not based on a given power grid foundation optimization scheduling model, prediction information of new energy and load, the polyhedral convex hull and a hyperplane set thereof, and converting the second double-layer optimization model into a corresponding second single-layer optimization model;
and the third construction module is used for constructing a double-layer iterative algorithm which can be decomposed and executed in parallel based on the second single-layer optimization model and the rapid convex hull algorithm, and gradually expanding the polyhedral convex hulls at all times until all economic operation points are contained, so as to obtain a final convex hull economic operation domain.
According to a third aspect of embodiments of the present application, there is provided an electronic device, including:
one or more processors;
a memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of the first aspect.
According to a fourth aspect of embodiments of the present application, there is provided a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the method according to the first aspect.
The technical scheme provided by the embodiment of the application can comprise the following beneficial effects:
according to the embodiment, the power grid convex hull economic operation domain decomposition parallel solving method can decompose the solving process into a plurality of optimization sub-problems which can be executed in parallel, and compared with the original algorithm which can only be solved in series, the algorithm has higher solving efficiency; secondly, the power grid convex hull economic operation domain obtained by the method can be used for evaluating the safety and the economical efficiency of power grid operation in real time; finally, the power grid convex hull economic operation domain obtained by the method is provided with a power grid optimal scheduling plan set, and the power grid convex hull economic operation domain can be combined with a real-time scheduling algorithm based on artificial intelligence to realize high-resolution fine scheduling of a large power grid, so that the method has a good application prospect.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
FIG. 1 is a flow chart illustrating a grid convex hull economic run-domain decomposition parallel solution method, according to an example embodiment.
FIG. 2 is a block diagram illustrating a grid convex hull economic run-domain decomposition parallel solver in accordance with an exemplary embodiment.
Fig. 3 is a schematic diagram of an electronic device, according to an example embodiment.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application.
The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in this application 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 also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first message may also be referred to as a second message, and similarly, a second message may also be referred to as a first message, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
FIG. 1 is a flow chart illustrating a method of decomposition parallel solution of a grid convex hull economic run domain, as shown in FIG. 1, according to an exemplary embodiment, the method may include the steps of:
step S1: constructing a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid at all times along with the change of the output of the new energy based on a given power grid foundation optimization scheduling model and prediction information of the new energy and the load, and determining the economic operation domain dimension of the convex hull of the power grid at all times based on the constructed first double-layer optimization model;
step S2: generating a polyhedral convex hull which meets the calculation precision requirement as small as possible and contains the initial economic operation point at each moment based on a generator number set and a corresponding convex hull economic operation domain dimension of the economic operation point which changes along with the new energy output at each moment, and obtaining a hyperplane set corresponding to the polyhedral convex hull;
step S3: constructing a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points or not by utilizing the polyhedral convex hull and the hyperplane set thereof based on a given power grid foundation optimization scheduling model and prediction information of new energy and load, and converting the second double-layer optimization model into a corresponding second single-layer optimization model;
Step S4: and constructing a double-layer iterative algorithm which can be decomposed and executed in parallel based on the second single-layer optimization model and the rapid convex hull algorithm, and gradually expanding the convex hulls in the polyhedral form at each moment until all economic operation points can be contained, so as to obtain a final convex hull economic operation domain.
According to the embodiment, the power grid convex hull economic operation domain decomposition parallel solving method can decompose the solving process into a plurality of optimization sub-problems which can be executed in parallel, and compared with the original algorithm which can only be solved in series, the algorithm has higher solving efficiency; secondly, the power grid convex hull economic operation domain obtained by the method can be used for evaluating the safety and the economical efficiency of power grid operation in real time; finally, the power grid convex hull economic operation domain obtained by the method is provided with a power grid optimal scheduling plan set, and the power grid convex hull economic operation domain can be combined with a real-time scheduling algorithm based on artificial intelligence to realize high-resolution fine scheduling of a large power grid, so that the method has a good application prospect.
In the specific implementation of step S1, based on the given power grid basic optimization scheduling model and the prediction information of the new energy and the load, a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid varying with the output of the new energy is constructed, and based on the constructed first double-layer optimization model, the dimension of the economic operation domain of the convex hull of the power grid at each moment is determined, including:
Specifically, "economy" in the economic operation domain refers to generalized economy and can generally refer to any given scheduling target such as lowest power generation cost, highest new energy consumption, lowest carbon emission, etc., in the application, the scheduling target function needs to keep the grid base optimization scheduling model convex, and the target function also enables the charge and discharge variables to be [ + ]
Figure 354021DEST_PATH_IMAGE088
) On the basis of which the person skilled in the art can set the objective function according to the actual situation. This step may comprise the sub-steps of:
step S11: converting a given grid base optimization scheduling model into an equivalent compact form thereof;
in one embodiment, the grid-based optimal scheduling model is,
objective function:
Figure 571507DEST_PATH_IMAGE089
(1a)
constraint conditions:
Figure 295881DEST_PATH_IMAGE090
(1b)
Figure 889804DEST_PATH_IMAGE091
(1c)
Figure 750444DEST_PATH_IMAGE092
(1d)
Figure 138831DEST_PATH_IMAGE093
(1e)
Figure 350501DEST_PATH_IMAGE094
(1f)
Figure 731804DEST_PATH_IMAGE095
(1g)
Figure 712529DEST_PATH_IMAGE096
(1h)
Figure 271818DEST_PATH_IMAGE097
(1i)
Figure 970783DEST_PATH_IMAGE098
(1j)
in the method, in the process of the invention,
Figure 640930DEST_PATH_IMAGE075
for the number of time periods to be optimally calculated,
Figure 679424DEST_PATH_IMAGE099
the number of the schedulable generators;
Figure 471931DEST_PATH_IMAGE100
the number of the energy storage devices;
Figure 392614DEST_PATH_IMAGE101
is a new energy machineGroup number;
Figure 381298DEST_PATH_IMAGE102
is the number of loads;
Figure 82756DEST_PATH_IMAGE103
numbering a set for the generator;
Figure 249426DEST_PATH_IMAGE104
is that
Figure 595088DEST_PATH_IMAGE105
A set of time periods;
Figure 934933DEST_PATH_IMAGE106
is that
Figure 682440DEST_PATH_IMAGE107
A set of time periods;
Figure 816750DEST_PATH_IMAGE108
a line numbering set;
Figure 898975DEST_PATH_IMAGE109
numbering the new energy unit set;
Figure 776932DEST_PATH_IMAGE110
numbering a set for the energy storage device;
Figure 378946DEST_PATH_IMAGE111
numbering a set for a load;
Figure 949736DEST_PATH_IMAGE112
、
Figure 535569DEST_PATH_IMAGE113
and
Figure 889321DEST_PATH_IMAGE114
respectively, generatorsiIs a coefficient of power generation cost;
Figure 408158DEST_PATH_IMAGE028
Is thattTime generatoriIs an active force of (a);
Figure 336800DEST_PATH_IMAGE115
and
Figure 206667DEST_PATH_IMAGE116
respectively are energy storage devicessCharging and discharging cost coefficients of (a);
Figure 160848DEST_PATH_IMAGE032
is thattTime energy storage devicesIs set to the charging power of (a);
Figure 737454DEST_PATH_IMAGE033
is thattTime energy storage devicesIs set in the above range;
Figure 650046DEST_PATH_IMAGE117
is a new energy unitjA penalty coefficient of the electric power is abandoned;
Figure 464332DEST_PATH_IMAGE035
is thattNew energy unit at momentjIs not used for the power supply;
Figure 691045DEST_PATH_IMAGE118
is the loadrIs a load rejection penalty coefficient of (1);
Figure 122158DEST_PATH_IMAGE037
is thattTime loadrIs used for discarding the load power;
Figure 940072DEST_PATH_IMAGE119
and
Figure 971482DEST_PATH_IMAGE120
respectively, generatorsiMinimum and maximum active force of (2);
Figure 267465DEST_PATH_IMAGE121
and
Figure 880980DEST_PATH_IMAGE122
respectively, generatorsiMaximum landslide and climbing rate of (a);
Figure 338638DEST_PATH_IMAGE123
and
Figure 342497DEST_PATH_IMAGE124
respectively the lineslAn upper transmission power limit of (2);
Figure 442171DEST_PATH_IMAGE125
、
Figure 644613DEST_PATH_IMAGE126
、
Figure 7593DEST_PATH_IMAGE127
and
Figure 295486DEST_PATH_IMAGE128
the power transfer factors of the node branches corresponding to the generator, the new energy, the energy storage and the load are respectively;
Figure 916960DEST_PATH_IMAGE005
is thattNew energy unit at momentjIs an active force of (a);
Figure 973909DEST_PATH_IMAGE129
is thattTime loadrIs set to be a power demand of the engine;
Figure 304527DEST_PATH_IMAGE039
is thattTime energy storage devicesIs used for storing electric quantity;
Figure 548558DEST_PATH_IMAGE130
and
Figure 990035DEST_PATH_IMAGE131
respectively are energy storage devicessIs provided;
Figure 382050DEST_PATH_IMAGE132
is an energy storage devicesMaximum stored power of (2);
Figure 883570DEST_PATH_IMAGE133
is an energy storage devicesIs set to the maximum operating power of (a).
It should be noted that, the objective function (1 a) is an embodiment that uses as much new energy consumed by the lowest possible power generation cost as a scheduling objective, and in a specific implementation, a person skilled in the art may set a corresponding objective function according to an actual scheduling objective requirement, which is not described herein.
For convenience of description, the model is converted into its equivalent compact form, as follows:
objective function:
Figure 349317DEST_PATH_IMAGE134
(2a)
constraint conditions:
Figure 860064DEST_PATH_IMAGE135
(2b)
Figure 626026DEST_PATH_IMAGE136
(2c)
in the method, in the process of the invention,
Figure 750977DEST_PATH_IMAGE027
output for schedulable machine set
Figure 500758DEST_PATH_IMAGE028
A column vector of components;
Figure 18458DEST_PATH_IMAGE029
to output new energy
Figure 904506DEST_PATH_IMAGE005
A column vector of components;
Figure 951090DEST_PATH_IMAGE030
and
Figure 391430DEST_PATH_IMAGE031
respectively charge power of the energy storage device
Figure 978400DEST_PATH_IMAGE032
And discharge power
Figure 984534DEST_PATH_IMAGE033
A column vector of components;
Figure 716866DEST_PATH_IMAGE034
discarding electric power for new energy
Figure 175661DEST_PATH_IMAGE035
A column vector of components;
Figure 769584DEST_PATH_IMAGE036
to discard load power
Figure 833486DEST_PATH_IMAGE037
A column vector of components;
Figure 18611DEST_PATH_IMAGE038
storing electrical energy for an energy storage device
Figure 218562DEST_PATH_IMAGE039
A column vector of components;
Figure 350597DEST_PATH_IMAGE040
、
Figure 331323DEST_PATH_IMAGE041
、
Figure 139879DEST_PATH_IMAGE042
、
Figure 573265DEST_PATH_IMAGE043
、
Figure 305729DEST_PATH_IMAGE044
、
Figure 344223DEST_PATH_IMAGE045
、
Figure 871151DEST_PATH_IMAGE046
and
Figure 995096DEST_PATH_IMAGE047
a corresponding coefficient matrix constrained by inequality;
Figure 531250DEST_PATH_IMAGE048
、
Figure 424251DEST_PATH_IMAGE049
、
Figure 387659DEST_PATH_IMAGE050
、
Figure 982588DEST_PATH_IMAGE051
、
Figure 56855DEST_PATH_IMAGE052
、
Figure 804362DEST_PATH_IMAGE053
、
Figure 204250DEST_PATH_IMAGE054
and
Figure 833946DEST_PATH_IMAGE055
corresponding coefficient matrixes respectively constrained by equations; superscriptTRepresenting the transpose of the matrix.
Step S12: according to a compact form equivalent to the power grid basic optimization scheduling model, constructing a first double-layer optimization model A and a second double-layer optimization model B which can be decomposed and executed in parallel and are used for determining a generator number set of economic operation points of the power grid at each moment along with the change of new energy output;
specifically, i.e. by judging
Figure 649586DEST_PATH_IMAGE028
At the position of
Figure 528898DEST_PATH_IMAGE079
Whether the maximum and minimum values of the moment are equal or not is further judged whether the economic operation point of the moment is changed along with the fluctuation of the new energy output, and the first double-layer optimization models A and B are respectively shown as the following (3 a) - (3 d) and (4 a) - (4 d):
objective function:
Figure 99688DEST_PATH_IMAGE137
(3a)
Constraint conditions:
Figure 403630DEST_PATH_IMAGE016
(3b)
Figure 819699DEST_PATH_IMAGE017
(3c)
Figure 604116DEST_PATH_IMAGE138
(3d)
in the formula, subscript in objective function "
Figure 283490DEST_PATH_IMAGE139
"means a collection
Figure 91040DEST_PATH_IMAGE103
Any of (3)iSum set
Figure 310800DEST_PATH_IMAGE106
Any of (3)tLater similar subscripts all refer to the same combination meaning;
Figure 887406DEST_PATH_IMAGE056
and
Figure 534419DEST_PATH_IMAGE057
and respectively predicting column vectors consisting of a lower limit and an upper limit of the output error of the new energy unit.
Objective function:
Figure 78533DEST_PATH_IMAGE140
(4a)
constraint conditions:
Figure 101984DEST_PATH_IMAGE016
(4b)
Figure 798675DEST_PATH_IMAGE017
(4c)
Figure 616590DEST_PATH_IMAGE138
(4d)
step S13: converting the lower layer optimization problem of the first double-layer optimization model into a KKT condition form thereof;
specifically, since the double-layer optimization problem is a "NP" hard problem and is difficult to solve directly, for the convenience of solving, the lower-layer optimization problems of the double-layer optimization models a and B (corresponding to (3 c), (3 d) and (4 c), (4 d), respectively) are converted into their KKT conditional forms, as shown in (5 a) - (5 i).
Figure 398732DEST_PATH_IMAGE059
(5a)
Figure 694715DEST_PATH_IMAGE141
(5b)
Figure 245913DEST_PATH_IMAGE060
(5c)
Figure 234729DEST_PATH_IMAGE061
(5d)
Figure 487856DEST_PATH_IMAGE062
(5e)
Figure 841391DEST_PATH_IMAGE063
(5f)
Figure 512675DEST_PATH_IMAGE064
(5g)
Figure 875654DEST_PATH_IMAGE065
(5h)
Figure 429126DEST_PATH_IMAGE142
(5i)
In the method, in the process of the invention,
Figure 66912DEST_PATH_IMAGE070
and
Figure 327123DEST_PATH_IMAGE071
column vectors consisting of lagrangian multipliers;
Figure 923321DEST_PATH_IMAGE072
representing diagonal elements as
Figure 151040DEST_PATH_IMAGE070
Is a diagonal matrix of (a).
Step S14: converting the nonlinear non-convex constraint in the KKT conditional form of the underlying optimization problem into a linear mixed integer equivalent form:
Figure 858096DEST_PATH_IMAGE067
(6a)
Figure 769551DEST_PATH_IMAGE068
(6b)
in the method, in the process of the invention,
Figure 474333DEST_PATH_IMAGE073
i.e.
Figure 205660DEST_PATH_IMAGE074
Is a column vector consisting of 0 or 1; m is a sufficiently large constant, which is generally preferred
Figure 185248DEST_PATH_IMAGE143
。
This step can eliminate the difficult-to-solve nonlinear constraints.
Step S15: the first two-layer optimization models A and B can be converted into first single-layer optimization models C and D according to the KKT conditional form of the lower-layer optimization problem and the linear mixed integer equivalent form of nonlinear non-convex constraint, specifically as (7 a) - (7C) and (8 a) - (8C) respectively,
Objective function:
Figure 482368DEST_PATH_IMAGE137
(7a)
constraint conditions:
Figure 358052DEST_PATH_IMAGE016
(7b)
(5a)，(5c)~(5i)，(6a)~(6b) (7c)
objective function:
Figure 107833DEST_PATH_IMAGE140
(8a)
constraint conditions:
Figure 609221DEST_PATH_IMAGE016
(8b)
(5a)，(5c)~(5i)，(6a)~(6b) (8c)
it should be noted that the first single-layer optimization models C and D are each required to be
Figure 760848DEST_PATH_IMAGE028
Solving once, namely all that is needed to be solved respectively
Figure 615889DEST_PATH_IMAGE144
Since each solution does not affect each other, the solution can be decomposed to the following time
Figure 790650DEST_PATH_IMAGE145
And the calculation nodes are solved in parallel, so that the solving efficiency is improved.
Step S16: at the position of
Figure 580882DEST_PATH_IMAGE001
Time of day for a generator
Figure 587016DEST_PATH_IMAGE002
Solving the first single-layer optimization models C and D based on a solver respectively, and numbering corresponding generators unequal to objective function values of the first single-layer optimization models C and DiAs a generator number set of economic operating points varying with the output of new energy
Figure 866818DEST_PATH_IMAGE003
Wherein
Figure 528875DEST_PATH_IMAGE001
Will be
Figure 919536DEST_PATH_IMAGE003
The number of generators contained in (a)
Figure 232706DEST_PATH_IMAGE004
Is arranged as electricityNet
Figure 683410DEST_PATH_IMAGE001
Dimension of economic operation domain of moment convex hull;
in particular, determining an electrical grid
Figure 832763DEST_PATH_IMAGE001
The moment convex hull economic run domain dimension. At the position of
Figure 27115DEST_PATH_IMAGE001
Time of day for a generator
Figure 945523DEST_PATH_IMAGE002
Solving the models C and D based on solvers (gurobi, cplex, etc.), respectively, and numbering the corresponding generators with unequal objective function values of the models C and DiIs denoted as a collection of (2)
Figure 567129DEST_PATH_IMAGE003
Wherein
Figure 938198DEST_PATH_IMAGE001
. Thus, it can be considered thattTime of day collection
Figure 936241DEST_PATH_IMAGE003
The economic operating point of the generator can be along with the output of new energy
Figure 958424DEST_PATH_IMAGE029
Changes by changes in (a) and (b) a collection
Figure 750930DEST_PATH_IMAGE003
The economic operation point of the generator is not output along with new energy
Figure 140455DEST_PATH_IMAGE029
Is changed by a change in (a). So that the number of the parts to be processed,tonly the set needs to be considered in the moment power grid convex hull economic operation domain
Figure 399311DEST_PATH_IMAGE003
The generator in the inner part can be used for generating electricity,
Figure 26733DEST_PATH_IMAGE003
the number of generators contained in (a)
Figure 255720DEST_PATH_IMAGE004
Namely, the electric network
Figure 601382DEST_PATH_IMAGE001
The moment convex hull is economical in terms of the dimension of the run domain.
It should be noted that the number of the substrates,
Figure 675648DEST_PATH_IMAGE001
the moment of time of the power grid convex hull economic operation domain is characterized by aggregation
Figure 672423DEST_PATH_IMAGE003
In the power generator. For convenience of description,tsymbol for moment power grid convex hull economic operation domain
Figure 72312DEST_PATH_IMAGE020
To express%
Figure 639690DEST_PATH_IMAGE020
Representing both the convex hull economic operation domain in the solving process and the finally obtained convex hull economic operation domain), and introducing a column vector
Figure 986489DEST_PATH_IMAGE146
To represent convex hull economic run-time domain
Figure 385240DEST_PATH_IMAGE020
Is numbered inkEconomical operation point [ ]
Figure 690451DEST_PATH_IMAGE020
The economic operation points in the system are innumerable, and the serial numbers are introducedkFor convenience in describing the solution process), column vectors
Figure 10705DEST_PATH_IMAGE146
The dimension elements are set
Figure 426774DEST_PATH_IMAGE003
The active power output of the generator.
In the specific implementation of the step S2, based on the generator number set of economic operation points at each moment along with the change of new energy output and the dimension of the corresponding convex hull economic operation domain, generating a polyhedron convex hull containing initial economic operation points as small as possible and meeting the calculation precision requirement at each moment, and obtaining a hyperplane set corresponding to the polyhedron convex hull;
In particular, this step may comprise the sub-steps of:
step S21: by means oftNew energy unit at momentjIs the predicted force of (2)
Figure 398141DEST_PATH_IMAGE005
Solving the power grid basic optimization scheduling model to obtain the optimal output column vector of each generator at each moment
Figure 139832DEST_PATH_IMAGE006
Wherein
Figure 478541DEST_PATH_IMAGE001
In the column vector
Figure 901563DEST_PATH_IMAGE006
In the process, the generator number set of which the economic operation point is changed along with the output of new energy is taken out
Figure 540486DEST_PATH_IMAGE003
The corresponding active output of the generator is used for obtaining the column vector
Figure 746138DEST_PATH_IMAGE007
；
Specifically, an initial operating point for generating an economic operating domain for solving the convex hull is specified, new energy output is specified, a model is solved, and as can be easily seen,
Figure 306564DEST_PATH_IMAGE007
namely, the convex hull economic operation domain
Figure 64435DEST_PATH_IMAGE020
An economic operating point in the system
Figure 495548DEST_PATH_IMAGE007
As a solutiontMoment convex hull economic operation domain
Figure 313462DEST_PATH_IMAGE020
Is a starting point of the (c).
Step S22: initializing a constant as small as possible under the condition of meeting the calculation accuracy requirement
Figure 626763DEST_PATH_IMAGE008
In the following
Figure 640856DEST_PATH_IMAGE001
At the moment, in sequence in the column vector
Figure 988791DEST_PATH_IMAGE007
Adding a constant to each dimension of (2)
Figure 977607DEST_PATH_IMAGE008
Obtaining
Figure 981466DEST_PATH_IMAGE004
Column vectors of linear uncorrelated
Figure 346720DEST_PATH_IMAGE009
Wherein
Figure 752424DEST_PATH_IMAGE010
Is at
Figure 177721DEST_PATH_IMAGE007
Is the first of (2)iAdding constants to dimensions
Figure 403297DEST_PATH_IMAGE008
The resulting column vector.
Specifically, to generate an initial convex hull at each moment, the column vectors are vertices of the initial convex hull, each vector corresponds to a vertex, and the vertices directly form an initial convex hull, which is one of the representation forms of the convex hulls, namely, the vertex representation.
Step S23: at the position of
Figure 572241DEST_PATH_IMAGE001
At the moment of time of day,
Figure 81720DEST_PATH_IMAGE011
individual operating points
Figure 677917DEST_PATH_IMAGE012
Is set of (a)
Figure 656369DEST_PATH_IMAGE013
Namely, a small enough polyhedral convex hull containing initial economic operation points is corresponding to the convex hull; obtaining the half-space representation form of the polyhedral convex hull by using a rapid convex hull algorithm, namely a hyperplane set
Figure 351706DEST_PATH_IMAGE014
。
In particular, with the fast convex hull algorithm, it is these column vectors that are taken as input, another representation of the initial convex hull, i.e. a half-space representation, is obtained. Both representations correspond to the same convex hull.
The solving process of the initial convex hull at each moment is independent and can be decomposed to
Figure 466424DEST_PATH_IMAGE075
And executing in parallel on the computing nodes.
In the implementation of step S3, based on the given grid basic optimization scheduling model, the prediction information of the new energy and the load, the polyhedral convex hull and the hyperplane set thereof, a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points is constructed, and the second double-layer optimization model is converted into a corresponding second single-layer optimization model, which includes:
s31: based on the given power grid basic optimization scheduling model, the prediction information of new energy and load, the polyhedral convex hull and the hyperplane set thereof, a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points is constructed,
Specifically, the double-layer optimization model is a criterion for expanding the vertexes of the convex hull economic operation domain, and whether calculation is stopped is judged by judging whether the returned vertexes are close enough to the corresponding hyperplane or not; if the distance from the corresponding hyperplane is far, the vertex is used as an expansion vertex; otherwise, if the distance from the corresponding hyperplane is sufficiently close, the iterative process is stopped. The double-layer optimization model is specifically as follows:
objective function:
Figure 233523DEST_PATH_IMAGE015
(9a)
constraint conditions:
Figure 230429DEST_PATH_IMAGE016
(9b)
Figure 944438DEST_PATH_IMAGE017
(9c)
Figure 975979DEST_PATH_IMAGE138
(9d)
in the method, in the process of the invention,
Figure 100930DEST_PATH_IMAGE019
is thattMoment convex hull economic operation domain
Figure 850711DEST_PATH_IMAGE020
Hyperplane in a half-space representationA numbering set;
Figure 102832DEST_PATH_IMAGE021
and
Figure 988879DEST_PATH_IMAGE022
respectively istMoment convex hull economic operation domain
Figure 35464DEST_PATH_IMAGE020
Numbered under the half-space representation form
Figure 272541DEST_PATH_IMAGE023
Is a super plane of (a)
Figure 62774DEST_PATH_IMAGE024
Coefficient column vectors and intercepts of (i) i.e
Figure 334486DEST_PATH_IMAGE025
；
Figure 801240DEST_PATH_IMAGE026
Is thattTime of day, collection
Figure 260034DEST_PATH_IMAGE003
The column vector is composed of the active power output of the generator.
S32: converting the second two-layer optimization model into a second equivalent single-layer optimization model according to steps S13, S14 and S15 as follows:
in particular, the second two-layer optimization model is a "NP" hard problem that is difficult to solve directly. Thus, for ease of solution, the second two-layer optimization model can be converted to an equivalent second single-layer optimization model according to steps S13, S14 and S15, as follows:
Objective function:
Figure 853958DEST_PATH_IMAGE058
(10a)
constraint conditions:
Figure 980177DEST_PATH_IMAGE016
(10b)
(5a)，(5c)~(5i)，(6a)~(6b) (10c)
in the method, in the process of the invention,
Figure 380283DEST_PATH_IMAGE069
and representing the optimized solution result of the second single-layer optimization model.
The second single-layer optimization model is described as
Figure 326373DEST_PATH_IMAGE147
All or part of the combinations of (a) are solved, but the solutions of the combinations are relatively independent and can be decomposed to at most
Figure 255146DEST_PATH_IMAGE148
The solutions are performed in parallel on the individual compute nodes,
Figure 173554DEST_PATH_IMAGE149
is thattTime of day
Figure 529580DEST_PATH_IMAGE020
The number of hyperplanes contained.
In the specific implementation of the step S4, constructing a decomposable parallel-executed double-layer iterative algorithm based on the second single-layer optimization model and the rapid convex hull algorithm, and gradually expanding the polyhedral convex hulls at all times until all economic operation points can be contained, so as to obtain a final convex hull economic operation domain;
in particular, the double-layer iterative algorithm may comprise the sub-steps of:
s41: initializing relevant calculation parameters: setting convergence criterion delta, and inputting an economic operation domain of a convex hull to be solved
Figure 681076DEST_PATH_IMAGE020
Number of time periods of (a)
Figure 413540DEST_PATH_IMAGE075
New energyFluctuation range data of source active output predicted value
Figure 248772DEST_PATH_IMAGE056
And
Figure 244541DEST_PATH_IMAGE057
convex hull economic operation domain at each moment
Figure 102907DEST_PATH_IMAGE020
Dimension of (2)
Figure 373482DEST_PATH_IMAGE004
Corresponding generator number set
Figure 328800DEST_PATH_IMAGE003
Half-space representation parameters of initial convex hull at each moment
Figure 495470DEST_PATH_IMAGE021
And
Figure 90399DEST_PATH_IMAGE022
initializing each time
Figure 164666DEST_PATH_IMAGE020
Is a hyperplane set of (1)
Figure 177752DEST_PATH_IMAGE076
Initializing each time
Figure 312061DEST_PATH_IMAGE020
Vertex set of (a)
Figure 941757DEST_PATH_IMAGE077
，
Figure 11258DEST_PATH_IMAGE078
；
S42: solving the expansion vertex of the current convex hull: based on the second monolayer optimization model, pair
Figure 675589DEST_PATH_IMAGE079
Time of day
Figure 918482DEST_PATH_IMAGE080
Hyperplane, calculation of optimization results
Figure 301053DEST_PATH_IMAGE081
Obtaining column vectors
Figure 169652DEST_PATH_IMAGE082
And update
Figure 688489DEST_PATH_IMAGE083
；
S43: updating extended vertices meeting the condition: if it is
Figure 633443DEST_PATH_IMAGE084
Then update
Figure 972151DEST_PATH_IMAGE085
The method comprises the steps of carrying out a first treatment on the surface of the If it is
Figure 395173DEST_PATH_IMAGE086
Step S45 is performed;
s44: updating the half-space representation of the current convex hull: based on a fast convex hull algorithm, calculation
Figure 768517DEST_PATH_IMAGE079
Time vertex set
Figure 618792DEST_PATH_IMAGE087
Is obtained in the form of a half-space representation of (a)
Figure 710376DEST_PATH_IMAGE019
And returns to step S42;
s45: outputting a convex hull economic operation domain: output of
Figure 468248DEST_PATH_IMAGE079
Time of day
Figure 148628DEST_PATH_IMAGE020
Form parameters of the half-space representation of (a)
Figure 232122DEST_PATH_IMAGE021
And
Figure 545422DEST_PATH_IMAGE022
thereby obtaining the final convex hull economic operation domain
Figure 575826DEST_PATH_IMAGE025
。
Corresponding to the embodiment of the power grid convex hull economic operation domain decomposition parallel solving method, the application also provides an embodiment of the power grid convex hull economic operation domain decomposition parallel solving device.
FIG. 2 is a block diagram illustrating a grid convex hull economic run-domain decomposition parallel solver according to an exemplary embodiment. Referring to fig. 2, the apparatus may include:
the first construction module 21 is configured to construct a first double-layer optimization model of a generator number set, which is executed in parallel and can be used to determine the economic operation points of the power grid at each moment and changes with the output of the new energy, based on a given power grid base optimization scheduling model and the prediction information of the new energy and the load, and determine the economic operation domain dimension of the convex hull at each moment of the power grid based on the constructed first double-layer optimization model;
The generating module 22 is configured to generate, at each moment, a polyhedral convex hull containing an initial economic operation point as small as possible and that meets a calculation accuracy requirement, based on a generator number set and a corresponding convex hull economic operation domain dimension, where economic operation points at each moment change with new energy output, and obtain a hyperplane set corresponding to the polyhedral convex hull;
the second construction module 23 is configured to construct a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points based on the given power grid basic optimization scheduling model, the prediction information of new energy and load, the polyhedral convex hull and the hyperplane set thereof, and convert the second double-layer optimization model into a corresponding second single-layer optimization model;
and a third construction module 24, configured to construct a decomposable parallel-executed double-layer iterative algorithm based on the second single-layer optimization model and the fast convex hull algorithm, and gradually expand the polyhedral convex hulls at each moment until all economic operation points are included, so as to obtain a final convex hull economic operation domain.
The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.
For the device embodiments, reference is made to the description of the method embodiments for the relevant points, since they essentially correspond to the method embodiments. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purposes of the present application. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
Correspondingly, the application also provides electronic equipment, which comprises: one or more processors; a memory for storing one or more programs; and when the one or more programs are executed by the one or more processors, the one or more processors are enabled to realize the grid convex hull economic operation domain decomposition parallel solving method. As shown in fig. 3, a hardware structure diagram of an apparatus with data processing capability, where the method for resolving and parallel solving a convex hull economic operation domain of a power grid is provided in an embodiment of the present invention, except for a processor, a memory and a network interface shown in fig. 3, the apparatus with data processing capability in the embodiment is generally according to an actual function of the apparatus with data processing capability, and may further include other hardware, which is not described herein.
Correspondingly, the application also provides a computer readable storage medium, wherein computer instructions are stored on the computer readable storage medium, and the instructions realize the grid convex hull economic operation domain decomposition parallel solving method when being executed by a processor. The computer readable storage medium may be an internal storage unit, such as a hard disk or a memory, of any of the data processing enabled devices described in any of the previous embodiments. The computer readable storage medium may also be an external storage device, such as a plug-in hard disk, a Smart Media Card (SMC), an SD Card, a Flash memory Card (Flash Card), or the like, provided on the device. Further, the computer readable storage medium may include both internal storage units and external storage devices of any device having data processing capabilities. The computer readable storage medium is used for storing the computer program and other programs and data required by the arbitrary data processing apparatus, and may also be used for temporarily storing data that has been output or is to be output.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains.
It is to be understood that the present application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof.

Claims (8)

1. The utility model provides a parallel solving method of power grid convex hull economic operation domain decomposition, which is characterized in that the method is applied to a power grid and comprises the following steps:
constructing a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid at all times along with the change of the output of the new energy based on a given power grid foundation optimization scheduling model and prediction information of the new energy and the load, and determining the economic operation domain dimension of the convex hull of the power grid at all times based on the constructed first double-layer optimization model;
generating a polyhedral convex hull which meets the calculation precision requirement as small as possible and contains the initial economic operation point at each moment based on a generator number set and a corresponding convex hull economic operation domain dimension of the economic operation point which changes along with the new energy output at each moment, and obtaining a hyperplane set corresponding to the polyhedral convex hull;
constructing a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points or not based on a given power grid basic optimization scheduling model, prediction information of new energy and load, the polyhedral convex hull and a hyperplane set thereof, and converting the second double-layer optimization model into a corresponding second single-layer optimization model;
And constructing a double-layer iterative algorithm which can be decomposed and executed in parallel based on the second single-layer optimization model and the rapid convex hull algorithm, and gradually expanding the convex hulls in the polyhedral form at each moment until all economic operation points can be contained, so as to obtain a final convex hull economic operation domain.
2. The method of claim 1, wherein constructing a first double-layer optimization model of a generator number set that can be performed in parallel and that is used to determine a change in economic operating point of the grid with the output of the new energy, based on a given grid base optimization scheduling model and prediction information of the new energy and the load, and determining the economic operating domain dimension of the convex hull of the grid at each moment based on the constructed first double-layer optimization model, comprises:
step S11: converting a given grid base optimization scheduling model into an equivalent compact form thereof;
step S12: according to a compact form equivalent to the power grid basic optimization scheduling model, constructing a first double-layer optimization model A and a second double-layer optimization model B which can be decomposed and executed in parallel and are used for determining a generator number set of economic operation points of the power grid at each moment along with the change of new energy output;
step S13: converting the lower layer optimization problem of the first double-layer optimization model into a KKT condition form thereof;
Step S14: converting the nonlinear non-convex constraint in the KKT conditional form of the lower-layer optimization problem into a linear mixed integer equivalent form;
step S15: converting the first double-layer optimization models A and B into first single-layer optimization models C and D according to the KKT conditional form of the lower-layer optimization problem and the linear mixed integer equivalent form of nonlinear non-convex constraint;
step S16: at the position of
Figure 96890DEST_PATH_IMAGE001
Time of day for a generator
Figure 92659DEST_PATH_IMAGE002
Solving the first single-layer optimization models C and D based on a solver respectively, and numbering corresponding generators unequal to objective function values of the first single-layer optimization models C and DiAs a generator number set of economic operating points varying with the output of new energy
Figure 747762DEST_PATH_IMAGE003
Wherein
Figure 283917DEST_PATH_IMAGE001
Will be
Figure 426185DEST_PATH_IMAGE003
The number of generators contained in (a)
Figure 401312DEST_PATH_IMAGE004
Is arranged as a power grid
Figure 746974DEST_PATH_IMAGE001
The moment convex hull is economical in terms of the dimension of the run domain.
3. The method of claim 1, wherein generating as small as possible a polyhedral convex hull containing initial economic operating points at each moment and obtaining a hyperplane set corresponding thereto that meets the calculation accuracy requirement based on the generator number set and the corresponding convex hull economic operating domain dimension of the economic operating points as a function of new energy output at each moment, comprises:
Step S21: by means oftNew energy unit at momentjIs the predicted force of (2)
Figure 86819DEST_PATH_IMAGE005
Solving a power grid foundation optimization scheduling model to obtain the optimal output column vector of each generator at each moment
Figure 834326DEST_PATH_IMAGE006
Wherein
Figure 968636DEST_PATH_IMAGE001
In the column vector
Figure 801594DEST_PATH_IMAGE006
In the process, the generator number set of which the economic operation point is changed along with the output of new energy is taken out
Figure 679551DEST_PATH_IMAGE003
The corresponding active output of the generator is used for obtaining the column vector
Figure 530832DEST_PATH_IMAGE007
；
Step S22: initializing a constant as small as possible under the condition of meeting the calculation accuracy requirement
Figure 101622DEST_PATH_IMAGE008
In the following
Figure 218614DEST_PATH_IMAGE001
At the moment, in sequence in the column vector
Figure 837945DEST_PATH_IMAGE007
Adding a constant to each dimension of (2)
Figure 560044DEST_PATH_IMAGE008
Obtaining
Figure 36156DEST_PATH_IMAGE004
Column vectors of linear uncorrelated
Figure 843706DEST_PATH_IMAGE009
Wherein
Figure 63466DEST_PATH_IMAGE010
Is at
Figure 889340DEST_PATH_IMAGE007
Is the first of (2)iAdding constants to dimensions
Figure 536353DEST_PATH_IMAGE008
The column vector obtained;
step S23: at the position of
Figure 627937DEST_PATH_IMAGE001
At the moment of time of day,
Figure 577352DEST_PATH_IMAGE011
individual operating points
Figure 805202DEST_PATH_IMAGE012
Is set of (a)
Figure 888696DEST_PATH_IMAGE013
Namely, a small enough polyhedral convex hull containing initial economic operation points is corresponding to the convex hull; obtaining the half-space representation form of the polyhedral convex hull by using a rapid convex hull algorithm, namely a hyperplane set
Figure 405259DEST_PATH_IMAGE014
。
4. The method of claim 2, wherein constructing a second double-layer optimization model for determining whether the current convex hull already contains all possible economic operating points based on the given grid-based optimization scheduling model, the prediction information of new energy and load, the polyhedral convex hull and the hyperplane set thereof, and converting the second double-layer optimization model into a corresponding second single-layer optimization model, comprises:
S31: based on a given power grid basic optimization scheduling model, prediction information of new energy and load, the polyhedral convex hull and a hyperplane set thereof, a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points is constructed as follows:
objective function:
Figure 966821DEST_PATH_IMAGE015
(9a)
constraint conditions:
Figure 518020DEST_PATH_IMAGE016
(9b)
Figure 772415DEST_PATH_IMAGE017
(9c)
Figure 573011DEST_PATH_IMAGE018
(9d)
in the method, in the process of the invention,
Figure 390795DEST_PATH_IMAGE019
is thattMoment convex hull economic operation domain
Figure 530920DEST_PATH_IMAGE020
A hyperplane numbering set in the half-space representation;
Figure 956216DEST_PATH_IMAGE021
and
Figure 447372DEST_PATH_IMAGE022
respectively istMoment convex hull economic operation domain
Figure 616316DEST_PATH_IMAGE020
Numbered under the half-space representation form
Figure 673265DEST_PATH_IMAGE023
Is a super plane of (a)
Figure 941566DEST_PATH_IMAGE024
Coefficient column vectors and intercepts of (i) i.e
Figure 982335DEST_PATH_IMAGE025
；
Figure 876341DEST_PATH_IMAGE026
Is thattTime of day, collection
Figure 318955DEST_PATH_IMAGE003
The column vector formed by the active power output of the generator,
Figure 566614DEST_PATH_IMAGE027
output for schedulable machine set
Figure 32362DEST_PATH_IMAGE028
A column vector of components;
Figure 543109DEST_PATH_IMAGE029
to output new energy
Figure 512333DEST_PATH_IMAGE005
A column vector of components;
Figure 450333DEST_PATH_IMAGE030
and
Figure 403377DEST_PATH_IMAGE031
respectively charge power of the energy storage device
Figure 452235DEST_PATH_IMAGE032
And discharge power
Figure 790813DEST_PATH_IMAGE033
A column vector of components;
Figure 899714DEST_PATH_IMAGE034
discarding electric power for new energy
Figure 74475DEST_PATH_IMAGE035
A column vector of components;
Figure 927024DEST_PATH_IMAGE036
to discard load power
Figure 870840DEST_PATH_IMAGE037
A column vector of components;
Figure 150643DEST_PATH_IMAGE038
storing electrical energy for an energy storage device
Figure 812700DEST_PATH_IMAGE039
A column vector of components;
Figure 203361DEST_PATH_IMAGE040
、
Figure 516531DEST_PATH_IMAGE041
、
Figure 967235DEST_PATH_IMAGE042
、
Figure 913325DEST_PATH_IMAGE043
、
Figure 33642DEST_PATH_IMAGE044
、
Figure 14367DEST_PATH_IMAGE045
、
Figure 573655DEST_PATH_IMAGE046
and
Figure 7042DEST_PATH_IMAGE047
a corresponding coefficient matrix constrained by inequality;
Figure 942768DEST_PATH_IMAGE048
、
Figure 778000DEST_PATH_IMAGE049
、
Figure 757457DEST_PATH_IMAGE050
、
Figure 678140DEST_PATH_IMAGE051
、
Figure 417557DEST_PATH_IMAGE052
、
Figure 107295DEST_PATH_IMAGE053
、
Figure 70703DEST_PATH_IMAGE054
and
Figure 681944DEST_PATH_IMAGE055
corresponding coefficient matrixes respectively constrained by equations; superscriptTRepresenting a transpose of the matrix;
Figure 756211DEST_PATH_IMAGE056
and
Figure 238139DEST_PATH_IMAGE057
column vectors consisting of a lower limit and an upper limit of the predicted output error of the new energy unit are respectively formed;
S32: converting the second two-layer optimization model into a second equivalent single-layer optimization model according to steps S13, S14 and S15 as follows:
objective function:
Figure 638027DEST_PATH_IMAGE058
(10a)
constraint conditions:
Figure 720253DEST_PATH_IMAGE016
(10b)
Figure 598210DEST_PATH_IMAGE059
，
Figure 200224DEST_PATH_IMAGE060
，
Figure 782732DEST_PATH_IMAGE061
，
Figure 837407DEST_PATH_IMAGE062
，
Figure 519055DEST_PATH_IMAGE063
，
Figure 241155DEST_PATH_IMAGE064
，
Figure 717267DEST_PATH_IMAGE065
，
Figure 774084DEST_PATH_IMAGE066
，
Figure 993844DEST_PATH_IMAGE067
，
Figure 367188DEST_PATH_IMAGE068
(10c)
in the method, in the process of the invention,
Figure 217463DEST_PATH_IMAGE069
representing the optimal solution result of the second single-layer optimization model,
Figure 309047DEST_PATH_IMAGE070
and
Figure 535760DEST_PATH_IMAGE071
column vectors consisting of lagrangian multipliers;
Figure 966873DEST_PATH_IMAGE072
representing diagonal elements as
Figure 50367DEST_PATH_IMAGE070
Is a diagonal matrix of the (a),
Figure 816197DEST_PATH_IMAGE073
i.e.
Figure 377760DEST_PATH_IMAGE074
Is a column vector consisting of 0 or 1; m is a sufficiently large constant.
5. The method of claim 1, wherein constructing a decomposable parallel-executed double-layer iterative algorithm based on the second single-layer optimization model and the fast convex hull algorithm, gradually expanding the polyhedral convex hulls at each moment until all economic operation points can be included, and obtaining a final convex hull economic operation domain, comprises:
s41: initializing relevant calculation parameters: setting convergence criterion delta, and inputting an economic operation domain of a convex hull to be solved
Figure 928958DEST_PATH_IMAGE020
Number of time periods of (a)
Figure 917774DEST_PATH_IMAGE075
Fluctuation range data of new energy active output predicted value
Figure 718370DEST_PATH_IMAGE056
And
Figure 9588DEST_PATH_IMAGE057
convex hull economic operation domain at each moment
Figure 477610DEST_PATH_IMAGE020
Dimension of (2)
Figure 840589DEST_PATH_IMAGE004
Corresponding generator number set
Figure 394061DEST_PATH_IMAGE003
Half-space representation parameters of initial convex hull at each moment
Figure 484377DEST_PATH_IMAGE021
And
Figure 806905DEST_PATH_IMAGE022
initializing each time
Figure 340786DEST_PATH_IMAGE020
Is a hyperplane set of (1)
Figure 381554DEST_PATH_IMAGE076
Initializing each time
Figure 26293DEST_PATH_IMAGE020
Vertex set of (a)
Figure 937749DEST_PATH_IMAGE077
，
Figure 908110DEST_PATH_IMAGE078
；
S42: solving the expansion vertex of the current convex hull: based on the second monolayer optimization model, pair
Figure 436174DEST_PATH_IMAGE079
Time of day
Figure 415763DEST_PATH_IMAGE080
Hyperplane, solving and calculating optimization result
Figure 165413DEST_PATH_IMAGE081
Obtaining column vectors
Figure 41096DEST_PATH_IMAGE082
And update
Figure 525298DEST_PATH_IMAGE083
；
S43: updating extended vertices meeting the condition: if it is
Figure 839736DEST_PATH_IMAGE084
Then update
Figure 194625DEST_PATH_IMAGE085
The method comprises the steps of carrying out a first treatment on the surface of the If it is
Figure 49666DEST_PATH_IMAGE086
Step S45 is performed;
s44: updatingHalf-space representation of the current convex hull: based on a fast convex hull algorithm, calculation
Figure 490006DEST_PATH_IMAGE079
Time vertex set
Figure 263927DEST_PATH_IMAGE087
Is obtained in the form of a half-space representation of (a)
Figure 270060DEST_PATH_IMAGE019
And returns to step S42;
s45: outputting a convex hull economic operation domain: output of
Figure 753125DEST_PATH_IMAGE079
Time of day
Figure 211919DEST_PATH_IMAGE020
Form parameters of the half-space representation of (a)
Figure 602581DEST_PATH_IMAGE021
And
Figure 666483DEST_PATH_IMAGE022
thereby obtaining the final convex hull economic operation domain
Figure 117187DEST_PATH_IMAGE025
。
6. The utility model provides a power grid convex hull economic operation domain decomposes parallel solution device which characterized in that is applied to in the electric wire netting, includes:
the first construction module is used for constructing a first double-layer optimization model which can be decomposed and executed in parallel and is used for determining a generator number set of economic operation points of the power grid at all times along with the change of the output of the new energy based on a given power grid foundation optimization scheduling model and the prediction information of the new energy and the load, and determining the dimension of the economic operation domain of the convex hull of the power grid at all times based on the constructed first double-layer optimization model;
The generation module is used for generating a polyhedral convex hull which meets the calculation precision requirement as small as possible and comprises an initial economic operation point at each moment and obtaining a hyperplane set corresponding to the polyhedral convex hull based on a generator number set of which the economic operation point at each moment changes along with the new energy output and the dimension of the corresponding convex hull economic operation domain;
the second construction module is used for constructing a second double-layer optimization model for judging whether the current convex hull already contains all possible economic operation points or not based on a given power grid foundation optimization scheduling model, prediction information of new energy and load, the polyhedral convex hull and a hyperplane set thereof, and converting the second double-layer optimization model into a corresponding second single-layer optimization model;
and the third construction module is used for constructing a double-layer iterative algorithm which can be decomposed and executed in parallel based on the second single-layer optimization model and the rapid convex hull algorithm, and gradually expanding the polyhedral convex hulls at all times until all economic operation points are contained, so as to obtain a final convex hull economic operation domain.
7. An electronic device, comprising:
one or more processors;
a memory for storing one or more programs;
The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-5.
8. A computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the method according to any of claims 1-5.
CN202211590666.8A 2022-12-12 2022-12-12 Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid Active CN115660385B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211590666.8A CN115660385B (en) 2022-12-12 2022-12-12 Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211590666.8A CN115660385B (en) 2022-12-12 2022-12-12 Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid

Publications (2)

Publication Number Publication Date
CN115660385A CN115660385A (en) 2023-01-31
CN115660385B true CN115660385B (en) 2023-07-07

Family

ID=85019604

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211590666.8A Active CN115660385B (en) 2022-12-12 2022-12-12 Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid

Country Status (1)

Country Link
CN (1) CN115660385B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116029464B (en) * 2023-03-28 2023-07-18 浙江大学 Acceleration solving method for accurate box-type economic operation domain of power grid

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112381375A (en) * 2020-11-09 2021-02-19 浙江大学 Power grid economic operation domain rapid generation method based on power flow distribution matrix

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020235049A1 (en) * 2019-05-22 2020-11-26 日本電気株式会社 Optimization device, optimization method, and optimization program
CN113488987B (en) * 2021-05-17 2023-07-14 国网浙江杭州市余杭区供电有限公司 Power grid flexibility operation domain evaluation index calculation method considering source load fluctuation
CN114662798B (en) * 2022-05-17 2022-09-06 浙江大学 Scheduling method and device based on power grid economic operation domain and electronic equipment
CN114640138B (en) * 2022-05-18 2022-09-02 浙江大学 Method and device for solving convex hull economic operation domain and electronic equipment

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112381375A (en) * 2020-11-09 2021-02-19 浙江大学 Power grid economic operation domain rapid generation method based on power flow distribution matrix

Also Published As

Publication number Publication date
CN115660385A (en) 2023-01-31

Similar Documents

Publication Publication Date Title
Alhussan et al. Green hydrogen production ensemble forecasting based on hybrid dynamic optimization algorithm
Khan et al. Optimal sizing of a stand-alone photovoltaic, wind turbine and fuel cell systems
Kumar Khadanga et al. Frequency control in hybrid distributed power systems via type‐2 fuzzy PID controller
CN113890023B (en) Comprehensive energy micro-grid distributed economic dispatch optimization method and system
Zhang et al. Robust optimization for dynamic economic dispatch under wind power uncertainty with different levels of uncertainty budget
Dong et al. Machine-learning-based real-time economic dispatch in islanding microgrids in a cloud-edge computing environment
Chen et al. Effective load carrying capability evaluation of renewable energy via stochastic long-term hourly based SCUC
Zhao et al. Deep learning based model-free robust load restoration to enhance bulk system resilience with wind power penetration
Bornapour et al. An efficient scenario-based stochastic programming for optimal planning of combined heat, power, and hydrogen production of molten carbonate fuel cell power plants
CN113285490B (en) Power system scheduling method, device, computer equipment and storage medium
Song et al. Optimal parameter extraction of the proton exchange membrane fuel cells based on a new Harris Hawks Optimization algorithm
Deepanraj et al. Intelligent wild geese algorithm with deep learning driven short term load forecasting for sustainable energy management in microgrids
Sahoo et al. Modified Harris Hawks optimization-based fractional-order fuzzy PID controller for frequency regulation of multi-micro-grid
Miao et al. Metamodel based design optimization approach in promoting the performance of proton exchange membrane fuel cells
CN115660385B (en) Method and device for decomposing and parallelly solving economic operation domain of convex hull of power grid
CN109687448A (en) A kind of active power distribution network flexibility appraisal procedure based on uncertain domain
CN111509784B (en) Uncertainty-considered virtual power plant robust output feasible region identification method and device
Mumtaz et al. Adaptive control paradigm for photovoltaic and solid oxide fuel cell in a grid-integrated hybrid renewable energy system
Chen et al. Optimal control strategy for solid oxide fuel cell‐based hybrid energy system using deep reinforcement learning
CN117833285A (en) Micro-grid energy storage optimization scheduling method based on deep reinforcement learning
Zhang et al. Physical-model-free intelligent energy management for a grid-connected hybrid wind-microturbine-PV-EV energy system via deep reinforcement learning approach
Zhao et al. A data‐driven scheduling approach for integrated electricity‐hydrogen system based on improved DDPG
Liu et al. More efficient energy management for hybrid ac/dc microgrids with hard constrained neural network-based conversion loss surrogate models
Shi et al. Day‐ahead optimal dispatching of hybrid power system based on deep reinforcement learning
Sun et al. Optimal Scheduling of Wind-Photovoltaic-Pumped Storage Joint Complementary Power Generation System Based on Improved Firefly Algorithm

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