CN114336767A - Data-driven robust optimization scheduling implementation method based on multi-affine strategy - Google Patents

Data-driven robust optimization scheduling implementation method based on multi-affine strategy Download PDF

Info

Publication number
CN114336767A
CN114336767A CN202210146851.1A CN202210146851A CN114336767A CN 114336767 A CN114336767 A CN 114336767A CN 202210146851 A CN202210146851 A CN 202210146851A CN 114336767 A CN114336767 A CN 114336767A
Authority
CN
China
Prior art keywords
power
affine
constraint
scheduling
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210146851.1A
Other languages
Chinese (zh)
Inventor
徐潇源
严正
许少伦
陆建宇
徐超然
王晗
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.)
East China Branch Of State Grid Corp ltd
Shanghai Jiaotong University
Original Assignee
East China Branch Of State Grid Corp ltd
Shanghai Jiaotong 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 East China Branch Of State Grid Corp ltd, Shanghai Jiaotong University filed Critical East China Branch Of State Grid Corp ltd
Priority to CN202210146851.1A priority Critical patent/CN114336767A/en
Publication of CN114336767A publication Critical patent/CN114336767A/en
Pending legal-status Critical Current

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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/76—Power conversion electric or electronic aspects
    • 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/40—Arrangements for reducing harmonics
    • 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • 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

A data-driven robust optimization scheduling implementation method based on multiple affine strategies comprises the steps of establishing a two-stage robust optimization model and solving to obtain a scheduling strategy, clustering wind power sample data sets by adopting a self-organizing mapping neural network to obtain multiple uncertain sets, transforming an optimization model by adopting the multiple affine strategies through a robust optimization model solving method based on the multiple affine strategies, converting an original optimization model into a linear programming problem by combining a dual principle, and solving to achieve data-driven robust optimization scheduling. The invention is based on a robust optimization method, improves the conservatism of uncertainty factor modeling by establishing multiple uncertain sets, improves the economy of the obtained scheduling strategy by introducing multiple affine strategies, can effectively depict the fluctuation range of uncertainty factors while ensuring the solving efficiency of an optimization model, and improves the operation efficiency of a power grid while ensuring the safety of the power grid.

Description

Data-driven robust optimization scheduling implementation method based on multi-affine strategy
Technical Field
The invention relates to a technology in the field of power system scheduling, in particular to a data-driven robust optimization scheduling implementation method based on a multi-affine strategy.
Background
With the large-scale grid connection of renewable energy sources, a scheduling method based on uncertainty factors needs to be researched. With the large-scale grid connection of renewable energy sources, the output uncertainty of the renewable energy sources has a significant influence on the operation of a power grid, the output random characteristic of the renewable energy sources needs to be accurately depicted, and a reasonable scheduling strategy is arranged to improve the operation efficiency of the power grid. Meanwhile, the output fluctuation of the renewable energy sources puts higher requirements on the safe operation of a power grid, and a conventional thermal power generating unit needs to be ensured to be reserved with sufficient reserve so as to deal with the large fluctuation of the output of the renewable energy sources. Therefore, a scheduling method considering uncertainty factors needs to be studied.
Disclosure of Invention
The invention provides a data-driven robust optimization scheduling implementation method based on multiple affine strategies aiming at the problems that modeling of an uncertain set is conservative and the scheduling strategy obtained based on the affine strategies is poor in economy in the prior art.
The invention is realized by the following technical scheme:
the invention relates to a data-driven robust optimization scheduling implementation method based on multiple affine strategies, which comprises the steps of establishing a two-stage robust optimization model and solving to obtain a scheduling strategy, clustering wind power sample data sets by adopting a self-organizing mapping neural network to obtain multiple uncertain sets, transforming an optimization model by adopting the multiple affine strategies by adopting a robust optimization model solving method based on the multiple affine strategies, and solving by combining a dual principle to convert an original optimization model into a linear programming problem to realize data-driven robust optimization scheduling.
The establishing of the two-stage robust optimization model specifically comprises the following steps: establishing a first-stage model, wherein the optimization target is that the sum of the power generation cost and the standby cost of the conventional unit is minimum, and constructing a constraint condition required to be considered by the first-stage model according to the power balance constraint of the power system, the upper and lower limit constraints of the output force of the conventional unit, the line transmission power range constraint, the standby range constraint of the conventional unit and the climbing capacity; and establishing a second stage model, wherein the optimization target is that the sum of the re-scheduling cost, the wind abandoning cost and the load shedding cost of the conventional unit is minimum, and the constraint condition required to be considered by the second stage model is established according to the power balance constraint of the power system, the line transmission power range constraint, the wind abandoning range and the load shedding range constraint, the re-scheduling output range constraint of the conventional unit and the climbing capacity.
The objective function of the first-stage model is that the sum of the power generation cost and the standby cost of the conventional unit is minimum, namely:
Figure BDA0003509319860000021
wherein: p is a radical ofG,j,t、rGu,j,t、rGd,j,tRespectively the active output, the positive standby and the negative standby of the conventional unit j at the time t; a isG,j、bG,j、cG,jThe power generation cost coefficient of the conventional unit j; dGu,j、dGd,jRespectively is a positive standby cost coefficient and a negative standby cost coefficient of the conventional unit j; n is a radical ofGThe number of the conventional units is shown, and t is time.
The target function is a quadratic function, and is converted into a series of linear functions by adopting a piecewise linearization method. For the jth unit, the operation cost function at the time t is converted into a linear constraint, which specifically includes:
Figure BDA0003509319860000022
wherein: wG,j,tIs a newly introduced cost variable;
Figure BDA0003509319860000023
and
Figure BDA0003509319860000024
the i-th group of linearization coefficients are marked with linear function segment numbers; d is the number of segments, the objective function of the first stage model is updated as:
Figure BDA0003509319860000025
the constraint conditions to be considered when constructing the first-stage model include: the method comprises the following steps of power balance constraint of the power system, upper and lower limit constraint of the output of the conventional unit, line transmission power range constraint, standby range constraint of the conventional unit and climbing capacity, and specifically comprises the following steps:
Figure BDA0003509319860000027
wherein: n is a radical ofR、NLThe number of the wind turbine generators and the number of loads are respectively; p is a radical ofR,i,t、pL,k,tRespectively predicting active output of the wind turbine generator i at the moment t and active requirements of the load k; p is a radical ofG,j,min、pG,j,maxRespectively representing the lower limit and the upper limit of the output force of the conventional unit j; flMaximum transmission power for line l; piG,lj、πR,li、πL,lkRespectively are power transmission distribution factors among a conventional unit j, a wind turbine unit i, a load k and a line l; rGu,j、RGd,jRespectively is the maximum positive standby range and the maximum negative standby range of the conventional unit j; p is a radical ofG,j,rampThe maximum climbing power of the conventional unit j.
The objective function of the second stage model is that the sum of the re-dispatching cost, the wind abandoning cost and the load shedding cost of the conventional unit is minimum, namely:
Figure BDA0003509319860000028
wherein: p is a radical ofGu,j,t、pGd,j,tRespectively readjusting the output of the conventional unit j upwards and downwards at the time t; p is a radical ofRc,i,t、pLc,k,tRespectively discarding wind power and load k load shedding power for the wind turbine generator at the time t; c. CGu,j、cGd,jRespectively is the cost coefficient of the regular unit j for re-dispatching; c. CRc,i、cLc,kRespectively is the wind turbine generator i abandon wind cost coefficient and load k cut load cost coefficient.
The constraint conditions to be considered when constructing the second-stage model comprise: the method comprises the following steps of power balance constraint of the power system, line transmission power range constraint, wind abandoning range and load shedding range constraint, rescheduling output range constraint of the conventional unit and climbing capacity, and specifically comprises the following steps:
Figure BDA0003509319860000031
Figure BDA0003509319860000032
Figure BDA0003509319860000033
-pG，j，ramp≤(pG，j，t+1+pGu，j，t+1-pGd，j，t+1)-(pG，j，t+pGu，j，t-pGd，j，t)≤pG，j，rampwherein:
Figure BDA0003509319860000034
predicting an output prediction error of the wind turbine generator i at the time t; etaRc,i、ηLc,kThe maximum wind abandoning proportion of the wind turbine generator i and the maximum load cutting proportion of the load k are respectively.
The constructed two-stage economic scheduling problem considering the wind power prediction error can be expressed as the following abstract mathematical problem: min (a)Tx+bTy)，s.t.Ax≤c，
Figure BDA00035093198600000310
Wherein: x is stage 1 decision variables including pG, rGu, rGd, WG; y is decision variables of the 2 nd stage, including pGu, pGd, pRc and plcc; a. b, A, C, B, C, D and D are corresponding coefficient matrixes;
Figure BDA0003509319860000035
the method is a concrete implementation scene of wind power prediction errors.
The clustering process specifically comprises the following steps: for each subdata set, firstly, performing decorrelation processing on sample data by adopting a principal component analysis method, then, fitting by adopting a core density estimation method to obtain a probability density function, and finally, establishing multiple uncertain sets.
The establishing of the plurality of uncertain sets is as follows: processing a large amount of wind power output historical data by adopting a self-organizing mapping algorithm to obtain a plurality of subdata sets, analyzing the data of each data set by adopting a principal component analysis and kernel density estimation method, and respectively establishing an uncertain set, wherein the method specifically comprises the following steps:
1) and clustering the original wind power output data sets by adopting a self-organizing mapping algorithm to obtain N sub-data sets.
2) And for each subdata set, establishing an uncertain set by adopting a principal component analysis and kernel density estimation method.
3) Finally, obtaining N uncertain sets: u shapei＝{ξ|Hiξ≤hi}, i ∈ {1,.., N }, where: u shapeiIs the ith uncertain set, xi is the wind power output prediction error, Hi、hiCoefficient matrix for the ith uncertainty set.
The two-stage robust optimization model based on the multiple uncertain sets comprises the following steps:
Figure BDA0003509319860000036
the step of introducing a multi-affine strategy to transform the optimization model is as follows: for each sub uncertain set, when the actual wind power output prediction error belongs to the uncertain set range, the generator set reschedules the output, the wind abandoning power, the load shedding power and the wind power prediction error to meet the following requirements:
Figure BDA0003509319860000037
Figure BDA0003509319860000038
wherein:
Figure BDA0003509319860000039
and respectively re-scheduling positive output, re-scheduling negative output, load shedding power and affine coefficients between the abandoned wind power and the wind power prediction error for the w-th uncertain concentration unit.
The matrix form corresponding to the multiple affine strategies is as follows: y is Miξ+miWherein: mi，miCoefficient matrixes of the affine strategy in the ith subset are respectively.
The method for converting the original optimization model into the linear programming problem by combining the dual principle specifically comprises the following steps:
after a multi-affine strategy is introduced, an original two-stage robust optimization model becomes: objective function
Figure BDA0003509319860000041
Figure BDA0003509319860000042
s.t.ax ≦ c, where: robust Constraint (CM)i+D)ξ≤d-Bx-Cmi，
Figure BDA0003509319860000043
Figure BDA0003509319860000044
Ui＝{ξ|Hiξ≤hiThe robust constraint is equivalent to that:
Figure BDA0003509319860000045
Hi Tπi＝(CMi+D)T，πi≥0，πiis a dual variable; the dual problem that the objective function is equivalent to the inner-layer maximization problem is as follows:
Figure BDA0003509319860000046
Hi Tvi＝(bTMi)T，vi≥0，viis a dual variable; the converted linear programming model is:
Figure BDA0003509319860000047
Figure BDA0003509319860000048
the constraint conditions include the transformed constraint and Ax ≦ c.
Introducing a new variable to transform the maximization problem of the inner layer in the objective function:
Figure BDA0003509319860000049
wherein: z is a newly introduced variable; the constraint conditions comprise the converted constraint, Ax is less than or equal to c and Z is more than or equal to hi Tvi+bTmi，
Figure BDA00035093198600000410
The transformed model is a linear programming model.
The solution is calculated by a CPLEX solver without limitation, so that an optimized scheduling strategy is obtained.
Technical effects
Compared with the prior art, the method can accurately depict the fluctuation characteristics of the uncertain factors; meanwhile, a robust optimization model solving method based on a multi-affine strategy is provided, and the operation efficiency of the power grid can be effectively improved while the safety is guaranteed based on the scheduling strategy obtained by solving.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of the system of the present invention;
FIG. 3 is a schematic diagram illustrating the effects of the embodiment.
Detailed Description
In this embodiment, taking an IEEE 118 node system as an example: in the IEEE 118 node system, there are a total of 186 transmission lines and 54 generator sets, and in this embodiment, the 54 generator sets all have a rescheduling capability. The 4 wind turbines are respectively connected at the node 15, the node 49, the node 59 and the node 90, and the predicted output is 300 MW. The load data used in the embodiment is from a certain actual power grid, the wind power data is from 4 wind power plants in the Gansu region, and the load power and the output of the wind turbine generator are converted according to the system load of the IEEE 118 node.
As shown in fig. 1, the embodiment relates to a data-driven robust optimization scheduling implementation method based on a multi-affine strategy, and a deterministic optimization model is established according to power grid security constraints, unit operation constraints, wind curtailment and load shedding constraints; establishing multiple uncertain sets according to the renewable energy output prediction error; establishing a two-stage robust optimization model based on uncertainty factors based on a deterministic optimization model and multiple uncertain sets; and converting the model into a linear programming problem by adopting a multi-affine strategy and a dual principle, and solving to obtain a scheduling strategy.
As shown in fig. 2, a data-driven robust optimized scheduling system for implementing the above method according to an embodiment of the present invention includes: the device comprises a deterministic optimization model modeling module, a multiple uncertain set modeling module, a two-stage robust optimization model modeling module and a solving module, wherein: the deterministic optimization model modeling module and the multiple uncertain set modeling module respectively establish a deterministic optimization model and multiple uncertain sets according to renewable energy data of power grid data and output the deterministic optimization model and the multiple uncertain sets to the two-stage robust optimization model modeling module, the two-stage robust optimization model modeling module further establishes a robust optimization scheduling model based on uncertain factors, and a scheduling strategy is obtained through calculation of the solving module.
Through specific practical experiments, in an IEEE 118 node system, the results shown in table 1 and fig. 3 are obtained by using the above method according to system parameters, loads, and wind power data.
TABLE 1
Figure BDA0003509319860000051
As shown in Table 1, the mean total operating cost, the mean first stage operating cost, the mean second stage operating cost, and the standard deviation of the total operating cost of the three methods were compared. It can be seen from the table that the first stage operating cost of the method is significantly less than the first stage operating cost of the single affine strategy method, while the second stage operating costs of the two methods are very close. In the aspect of the total operation cost, the average value of the method is smaller than that of the single affine strategy method, and the standard deviation of the method is basically the same as that of the single affine strategy method. The method ensures the reliability of the scheduling scheme, and reduces the conservatism by adopting a multi-affine strategy method, thereby improving the operation efficiency of the system.
Meanwhile, as can be seen from fig. 3(a), the risk and reliability of the optimization method are comprehensively considered by taking the quartile as a standard, and the quartile of the method is obviously smaller than that of the single affine strategy method, i.e., the method can obtain a better scheduling result.
As shown in fig. 3(b), the total running cost of the method and the single affine strategy method is shown in 100 out-of-sample experiments. The 45 degree reference line in the figure means that the total operating costs of the two methods are equal, and the point above the reference line indicates that the total operating cost of the method is lower than that of the single affine strategy method. It can be clearly seen from the figure that the data points of the experiment outside the sample all fall above the reference line, i.e. the method can effectively improve the operating efficiency of the system.
In conclusion, the deterministic optimization model is established according to the operation constraints of the power grid and the units, multiple uncertain sets are established by adopting self-organizing mapping, principal component analysis and nuclear density estimation methods, the random characteristic of wind power output is accurately described, and a two-stage robust optimization model based on uncertainty is established; converting the optimization model into a linear programming problem which is easy to solve by adopting a multi-affine strategy and a dual principle; by establishing multiple uncertain sets and adopting multiple affine strategies, the safe operation of the power grid can be ensured, and meanwhile, the operation efficiency of the power grid can be effectively improved.
The foregoing embodiments may be modified in many different ways by those skilled in the art without departing from the spirit and scope of the invention, which is defined by the appended claims and all changes that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

Claims (10)

1. A data-driven robust optimization scheduling implementation method based on multiple affine strategies is characterized in that a scheduling strategy is obtained by establishing a two-stage robust optimization model and solving, a wind power sample data set is subjected to clustering processing by adopting a self-organizing mapping neural network to obtain multiple uncertain sets, then the optimization model is transformed by adopting the multiple affine strategies through a robust optimization model solving method based on the multiple affine strategies, and an original optimization model is converted into a linear programming problem by combining a dual principle to solve, so that the data-driven robust optimization scheduling is implemented.
2. The method for implementing the multi-affine-strategy-based data-driven robust optimization scheduling as claimed in claim 1, wherein the establishing of the two-stage robust optimization model specifically comprises: establishing a first-stage model, wherein the optimization target is that the sum of the power generation cost and the standby cost of the conventional unit is minimum, and constructing a constraint condition required to be considered by the first-stage model according to the power balance constraint of the power system, the upper and lower limit constraints of the output force of the conventional unit, the line transmission power range constraint, the standby range constraint of the conventional unit and the climbing capacity; and establishing a second stage model, wherein the optimization target is that the sum of the re-scheduling cost, the wind abandoning cost and the load shedding cost of the conventional unit is minimum, and the constraint condition required to be considered by the second stage model is established according to the power balance constraint of the power system, the line transmission power range constraint, the wind abandoning range and the load shedding range constraint, the re-scheduling output range constraint of the conventional unit and the climbing capacity.
3. The multi-affine-strategy-based data-driven robust optimization scheduling implementation method as claimed in claim 2, wherein an objective function of the first-stage model is that a sum of power generation and standby costs of a conventional unit is minimum, that is:
Figure FDA0003509319850000011
wherein: p is a radical ofG,j,t、rGu,j,t、rGd,j,tRespectively the active output, the positive standby and the negative standby of the conventional unit j at the time t; a isG,j、bG,j、cG,jThe power generation cost coefficient of the conventional unit j; dGu,j、dGd,jRespectively is a positive standby cost coefficient and a negative standby cost coefficient of the conventional unit j; n is a radical ofGThe number of the conventional units is adopted; t is time;
the target function is a quadratic function and is converted into a series of linear functions by adopting a piecewise linearization method; for the jth unit, the operation cost function at the time t is converted into a linear constraint, which specifically includes:
Figure FDA0003509319850000012
wherein: wG,j,tIs a newly introduced cost variable;
Figure FDA0003509319850000013
and
Figure FDA0003509319850000014
the i-th group of linearization coefficients are marked with linear function segment numbers; d is the number of segments, the objective function of the first stage model is updated as:
Figure FDA0003509319850000015
the constraint conditions to be considered when constructing the first-stage model include: power balance constraint of power system, upper and lower limit constraint of output of conventional unit and line transmission power rangeEnclose restraint, conventional unit reserve range restraint and climbing ability, specifically be:
Figure FDA0003509319850000021
Figure FDA0003509319850000022
wherein: n is a radical ofR、NLThe number of the wind turbine generators and the number of loads are respectively; p is a radical ofR,i,t、pL,k,tRespectively predicting active output of the wind turbine generator i at the moment t and active requirements of the load k; p is a radical ofG,j,min、pG,j,maxRespectively representing the lower limit and the upper limit of the output force of the conventional unit j; flMaximum transmission power for line l; piG,lj、πR,li、πL,lkRespectively are power transmission distribution factors among a conventional unit j, a wind turbine unit i, a load k and a line l; rGu,j、RGd,jRespectively is the maximum positive standby range and the maximum negative standby range of the conventional unit j; p is a radical ofG,j,rampThe maximum climbing power of the conventional unit j.
4. The method for realizing the data-driven robust optimization scheduling based on the multi-affine strategy as claimed in claim 2, wherein an objective function of the second-stage model is that a sum of a regular unit rescheduling cost, a wind curtailment cost and a load shedding cost is minimum, that is:
Figure FDA0003509319850000023
wherein: p is a radical ofGu,j,t、pGd,j,tRespectively readjusting the output of the conventional unit j upwards and downwards at the time t; p is a radical ofRc,i,t、pLc,k,tRespectively discarding wind power and load k load shedding power for the wind turbine generator at the time t; c. CGu,j、cGd,jRespectively is the cost coefficient of the regular unit j for re-dispatching; c. CRc,i、cLc,kRespectively obtaining a wind curtailment cost coefficient and a load k load shedding cost coefficient of the wind turbine generator i;
the said approximation to be considered when constructing the second stage modelThe bundle condition includes: the method comprises the following steps of power balance constraint of the power system, line transmission power range constraint, wind abandoning range and load shedding range constraint, rescheduling output range constraint of the conventional unit and climbing capacity, and specifically comprises the following steps:
Figure FDA0003509319850000024
Figure FDA0003509319850000025
Figure FDA0003509319850000026
-pG，j，ramp≤(pG，j，t+1+pGu，j，t+1-pGd，j，t+1)-(pG，j，t+pGu，j，t-pGd，j，t)≤pG，j，rampwherein
Figure FDA0003509319850000027
Predicting an output prediction error of the wind turbine generator i at the time t; etaRc,i、ηLc,kThe maximum wind abandoning proportion of the wind turbine generator i and the maximum load cutting proportion of the load k are respectively.
5. The method for realizing the multi-affine-strategy-based data-driven robust optimal scheduling as claimed in claim 1, wherein the constructed two-stage economic scheduling problem considering the wind power prediction error can be expressed as the following abstract mathematical problem: min (a)Tx+bTy)，s·t·Ax≤c，
Figure FDA0003509319850000028
Wherein: x is stage 1 decision variables including pG, rGu, rGd, WG; y is decision variables of the 2 nd stage, including pGu, pGd, pRc and plcc; a. b, A, C, B, C, D and D are corresponding coefficient matrixes;
Figure FDA0003509319850000031
the method is a concrete implementation scene of wind power prediction errors.
6. The multi-affine-strategy-based data-driven robust optimization scheduling implementation method according to claim 1, wherein the clustering process specifically comprises: for each subdata set, firstly, performing decorrelation processing on sample data by adopting a principal component analysis method, then, fitting by adopting a core density estimation method to obtain a probability density function, and finally, establishing multiple uncertain sets.
7. The multi-affine strategy-based data-driven robust optimization scheduling implementation method according to claim 1, wherein the establishing of the plurality of uncertainty sets is: processing a large amount of wind power output historical data by adopting a self-organizing mapping algorithm to obtain a plurality of subdata sets, analyzing the data of each data set by adopting a principal component analysis and kernel density estimation method, and respectively establishing an uncertain set, wherein the method specifically comprises the following steps:
1) clustering the original wind power output data sets by adopting a self-organizing mapping algorithm to obtain N sub-data sets;
2) for each subdata set, establishing an uncertain set by adopting a principal component analysis and kernel density estimation method;
3) finally, obtaining N uncertain sets: u shapei＝{ξ|Hiξ≤hi}, i ∈ {1,.., N }, where: u shapeiIs the ith uncertain set, xi is the wind power output prediction error, Hi、hiA coefficient matrix for the ith uncertainty set;
the two-stage robust optimization model based on the multiple uncertain sets comprises the following steps:
Figure FDA0003509319850000032
s.t.Ax≤c，
Figure FDA0003509319850000033
Ui＝{ξ|Hiξ≤hi}，i∈{1，...，N}。
8. the multi-affine based policy of claim 1The slight data-driven robust optimization scheduling implementation method is characterized in that the introduction of a multi-affine strategy to transform an optimization model refers to the following steps: for each sub uncertain set, when the actual wind power output prediction error belongs to the uncertain set range, the generator set reschedules the output, the wind abandoning power, the load shedding power and the wind power prediction error to meet the following requirements:
Figure FDA0003509319850000034
Figure FDA0003509319850000035
wherein:
Figure FDA0003509319850000036
and respectively re-scheduling positive output, re-scheduling negative output, load shedding power and affine coefficients between the abandoned wind power and the wind power prediction error for the w-th uncertain concentration unit.
9. The method for realizing the data-driven robust optimal scheduling based on the multi-affine strategy as claimed in claim 1, wherein the matrix form corresponding to the multi-affine strategy is as follows: y is Miξ+miWherein: mi，miCoefficient matrixes of affine strategies in the ith subset are respectively;
the method for converting the original optimization model into the linear programming problem by combining the dual principle specifically comprises the following steps:
after a multi-affine strategy is introduced, an original two-stage robust optimization model becomes: objective function
Figure FDA0003509319850000041
Figure FDA0003509319850000042
s.t.ax ≦ c, where: robust Constraint (CM)i+D)ξ≤d-Bx-Cmi，
Figure FDA0003509319850000043
Figure FDA0003509319850000044
Ui＝{ξ|Hiξ≤hiThe robust constraint is equivalent to that: i T i ihπ≤d-Bx-Cm，Hi Tπi＝(CMi+D)T，πi≥0，πiis a dual variable; the dual problem that the objective function is equivalent to the inner-layer maximization problem is as follows:
Figure FDA0003509319850000045
Hi Tvi＝(bTMi)T，vi≥0，viis a dual variable; the converted linear programming model is:
Figure FDA0003509319850000046
Figure FDA0003509319850000047
the constraint conditions comprise the transformed constraint and Ax is less than or equal to c;
introducing a new variable to transform the maximization problem of the inner layer in the objective function:
Figure FDA0003509319850000048
wherein: z is a newly introduced variable; the constraint conditions comprise the converted constraint, Ax is less than or equal to c and Z is more than or equal to hi Tvi+bTmi，
Figure FDA0003509319850000049
The transformed model is a linear programming model.
10. The system for realizing the multi-affine-strategy-based data-driven robust optimization scheduling according to the method of any one of claims 1-9, is characterized by comprising the following steps: the device comprises a deterministic optimization model modeling module, a multiple uncertain set modeling module, a two-stage robust optimization model modeling module and a solving module, wherein: the deterministic optimization model modeling module and the multiple uncertain set modeling module respectively establish a deterministic optimization model and multiple uncertain sets according to renewable energy data of power grid data and output the deterministic optimization model and the multiple uncertain sets to the two-stage robust optimization model modeling module, the two-stage robust optimization model modeling module further establishes a robust optimization scheduling model based on uncertain factors, and a scheduling strategy is obtained through calculation of the solving module.
CN202210146851.1A 2022-02-17 2022-02-17 Data-driven robust optimization scheduling implementation method based on multi-affine strategy Pending CN114336767A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210146851.1A CN114336767A (en) 2022-02-17 2022-02-17 Data-driven robust optimization scheduling implementation method based on multi-affine strategy

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210146851.1A CN114336767A (en) 2022-02-17 2022-02-17 Data-driven robust optimization scheduling implementation method based on multi-affine strategy

Publications (1)

Publication Number Publication Date
CN114336767A true CN114336767A (en) 2022-04-12

Family

ID=81029738

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210146851.1A Pending CN114336767A (en) 2022-02-17 2022-02-17 Data-driven robust optimization scheduling implementation method based on multi-affine strategy

Country Status (1)

Country Link
CN (1) CN114336767A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116316901A (en) * 2023-05-17 2023-06-23 广东电网有限责任公司佛山供电局 Micro-grid robust scheduling method and related device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116316901A (en) * 2023-05-17 2023-06-23 广东电网有限责任公司佛山供电局 Micro-grid robust scheduling method and related device

Similar Documents

Publication Publication Date Title
CN110298138B (en) Comprehensive energy system optimization method, device, equipment and readable storage medium
CN112186766B (en) Multistage decomposition method based on Benders decomposition
Zhao et al. Distributed risk-limiting load restoration for wind power penetrated bulk system
CN108599268B (en) Day-ahead optimization scheduling method considering wind power plant space-time association constraint
CN105790265A (en) AC power flow constraint-based uncertainty unit commitment model and solving method
CN107977744A (en) A kind of electric system based on traditional Benders decomposition methods Robust Scheduling method a few days ago
CN113241803B (en) Energy storage scheduling method based on new energy consumption and computer medium
Zhao et al. Robust distributed coordination of parallel restored subsystems in wind power penetrated transmission system
Singh et al. GA-based multi-objective optimization for distributed generations planning with DLMs in distribution power systems
He et al. Integrated scheduling of hydro, thermal and wind power with spinning reserve
CN114336767A (en) Data-driven robust optimization scheduling implementation method based on multi-affine strategy
Kurbanov et al. An Appropriate Wind Model for The Reliability Assessment of Incorporated Wind Power in Power Generation System
CN113344283B (en) Energy internet new energy consumption capability assessment method based on edge intelligence
CN105305485A (en) Large-scale intermittent energy consuming security constrained economic dispatch method
CN109726416B (en) Scheduling decision method based on new energy cluster prediction and load flow calculation
Lu et al. Clean generation mix transition: Large-scale displacement of fossil fuel-fired units to cut emissions
Hong et al. Short-term real-power scheduling considering fuzzy factors in an autonomous system using genetic algorithms
CN115377968A (en) Novel power distribution network sequence recovery optimization method considering renewable energy output fluctuation
CN114204552A (en) Comprehensive energy scheduling method for intelligent park
CN112003279B (en) Evaluation method for new energy consumption capability of hierarchical micro-grid
CN108039739A (en) A kind of active distribution network dynamic random economic load dispatching method
CN114676921A (en) Method for calculating wind power receptibility of system by considering source load storage coordination optimization
Kayalvizhi et al. Stochastic optimal power flow in presence of wind generations using harmony search algorithm
CN112365134A (en) Energy network random planning method based on point estimation method probability multi-energy flow
Zhang et al. Probabilistic transmission network expansion planning considering integration of wind power

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