CN113190973A - Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene - Google Patents
Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene Download PDFInfo
- Publication number
- CN113190973A CN113190973A CN202110380662.6A CN202110380662A CN113190973A CN 113190973 A CN113190973 A CN 113190973A CN 202110380662 A CN202110380662 A CN 202110380662A CN 113190973 A CN113190973 A CN 113190973A
- Authority
- CN
- China
- Prior art keywords
- scene
- stage
- typical
- scenes
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005457 optimization Methods 0.000 title claims abstract description 27
- 230000002457 bidirectional effect Effects 0.000 title claims abstract description 23
- 238000003860 storage Methods 0.000 title claims abstract description 20
- 239000013598 vector Substances 0.000 claims abstract description 54
- 230000009467 reduction Effects 0.000 claims abstract description 35
- 238000009826 distribution Methods 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 10
- 238000010248 power generation Methods 0.000 claims description 7
- 239000000126 substance Substances 0.000 claims description 7
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 239000004576 sand Substances 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000016507 interphase Effects 0.000 claims description 2
- 230000002123 temporal effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011946 reduction process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011234 economic evaluation Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000011541 reaction mixture Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Wind Motors (AREA)
Abstract
The invention discloses a bidirectional optimization method, a device, equipment and a storage medium for a wind-solar-load multi-stage typical scene, belonging to the technical field of power systems and comprising the following steps: generating a wind-solar load output sample set; generating a typical day scene in a phase; generating a multi-stage representative scene set containing more temporal scenes; constructing a scene feature vector; generating a correlation matrix; performing scene reduction on the multi-stage typical scene set by adopting an optimal reduction algorithm based on the correlation matrix until the number of the remaining scenes in the multi-stage typical scene set reaches a preset value, and acquiring typical scenes between stages; and generating an optimized multi-stage typical scene set according to the inter-stage typical scenes. The method can be used for solving the problem that the practicability of a single scene is limited from the multi-dimensional multivariable angle according to the uncertainty of the operation of the power distribution network, realizing the dimension reduction processing of multi-dimensional scene data and improving the generation efficiency of a typical scene.
Description
Technical Field
The invention relates to a bidirectional optimization method, device, equipment and storage medium for a wind-solar-load multi-stage typical scene, and belongs to the technical field of power systems.
Background
With the rapid development of new energy, the access of distributed power sources increases the uncertainty of the operation of the power distribution network. The uncertainty of the random variable has an important influence on coordinated planning, economic evaluation and the like of the power distribution network and the energy storage. With the gradual increase of the power generation proportion of renewable energy sources in the power system, the variation characteristics of the distributed power sources and the load need to be described by means of discrete scenes during power system planning decision, so that the generation of a scene set and the reduction based on scene features and scene probability are indispensable links.
In the process of simulating the operation of the power distribution network, long-time scale simulation is carried out according to actually measured data, so that the solution is difficult due to overlarge operation amount, and therefore a scene set is generated by means of describing the change characteristics of a distributed power supply and a load by means of a discrete scene and is reduced based on scene characteristics and scene probability. The typical scene refers to a discrete scene which is obtained by scene reduction and can sufficiently represent the distributed power supply and load change characteristics.
Currently, a large amount of research has been conducted by researchers for the generation of scene sets of wind, light and load characteristics and the reduction of scenes. However, in the aspect of generating a scene set, existing research only considers a single state variable, and a multi-dimensional scene is not discussed. In the aspect of simplifying a scene set, a clustering algorithm is commonly used for scene reduction. Aiming at the defects of the K-means clustering algorithm, improved clustering methods such as K-means and Clara are formed on the basis of the K-means clustering algorithm, but the method still has the defects of large calculation amount and influence of the size of a sample set on the result.
Therefore, in order to solve the above problems, the present application proposes a method, an apparatus, a device and a storage medium for bidirectional optimization of a wind-solar-charged multi-stage typical scene.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a bidirectional optimization method, device, equipment and storage medium for a wind-solar-charged multi-stage typical scene, which improve the characteristic diversification of a multi-stage wind-solar-charged scene set, reduce the calculation amount of scene reduction and improve the precision and the calculation efficiency of a generated scene by constructing a relevance matrix of a scene characteristic vector.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
in a first aspect, the invention provides a bidirectional optimization method for a wind-solar-load multi-stage typical scene, which comprises the following steps:
generating a wind and light load output sample set according to the wind turbine generator output model, the photovoltaic output model and the load data of the power distribution network;
synchronously clustering the samples in the sample set by adopting a clustering algorithm based on density peak values to generate a typical day scene in a stage;
carrying out Cartesian product fusion on adjacent typical daily scenes in the stages to generate a multi-stage typical scene set containing more scene segments at more moments;
constructing a scene feature vector according to the multi-stage typical scene set;
calculating the relevance between any two scene feature vectors to generate a relevance matrix;
performing scene reduction on the multi-stage typical scene set by adopting an optimal reduction algorithm based on the correlation matrix until the number of the remaining scenes in the multi-stage typical scene set reaches a preset value, and acquiring typical scenes between stages;
and generating an optimized multi-stage typical scene set according to the inter-stage typical scenes.
As an alternative embodiment:
the establishment of the wind turbine generator output model comprises the following steps:
describing the uncertainty of the wind speed at a single moment by using Weibull distribution, and a distribution function F thereofw(v)：
Wherein c is a wind speed parameter, k*The wind speed fluctuation shape parameter is shown, and v is the wind speed;
the establishment of the photovoltaic output model comprises the following steps:
and (3) describing a photovoltaic power continuous probability density function by adopting Beta distribution:
wherein alpha and Beta are Beta distribution shape parameters, PpTo photovoltaic power, PPmaxIs the maximum photovoltaic power;
the power distribution network load data comprises:
the load factor is used to describe the compliance characteristics as a function of:
Pl＝PlMPl rate
wherein: plFor the load demand at that moment, PlMIn order to be the peak value of the load,
the load factor at that time.
As an alternative embodiment: the method for synchronously clustering the samples in the sample set by adopting the clustering algorithm based on the density peak value comprises the following steps of:
calculating the distance d between two samples in the sample setij：
Wherein the content of the first and second substances,dij＝d(xi，xj) Represents the distance between the ith sample and the jth sample; x is the number ofiAnd xjRespectively the ith sample and the jth sample in the sample set,
and
respectively representing the kth attribute of the sample, wherein K represents the attribute dimension of each sample;
arranging the distances in ascending order to calculate the local density value rhoi：
Wherein d iscIs a preset truncation distance parameter;
obtaining a distance index according to the local density value
Wherein the content of the first and second substances,
is the q thiSample to qthjShortest distance of one sample, wheniWhen the local density maximum is obtained
Is the maximum distance, p, between the sample and other samplesiThe subscript sequence in descending order is Q ═ { Q ═ Q1,q2,…,qn}，qi、qjAll belong to a set Q;
obtaining a cluster center weight value gamma according to the local density value and the distance indexi：
γi＝ρiδi
Where ρ isiIs the local density value of the ith sample, deltaiThe distance index of the ith sample;
calculating cluster center weights of all sample points and arranging the cluster center weights in a descending order, wherein the sample points arranged in the descending order of the weights are used as cluster center points, namely cluster centers;
based on a clustering center, normalization processing is carried out on data samples of the wind turbine generator, the photovoltaic and the load, and synchronous clustering is carried out by taking the day as a unit to generate a typical day scene.
As an alternative embodiment: the constructing of the scene feature vector from the multi-stage typical scene set includes:
normalizing the data samples for each stage to construct a scene feature vector
And constructs feature vectors Eig of the multi-stage typical scene on the basis of the feature vectorsb：
Eigb＝[Wb,Pb,Lb]
Wherein, WbFor installed capacity of wind turbine, PbFor photovoltaic installed capacity, LbIn order to be the amount of the load,
for the installed capacity of the wind turbine at the u-th stage,
for the u-th stage photovoltaic installed capacity,
the u-th stage maximum load.
As an alternative embodiment: the calculating the relevance between any two scene feature vectors and the generating the relevance matrix comprises the following steps:
any two scene feature vectors Eig for a multi-phase sceneiAnd EigjIs the k-th dimension of (c) and a correlation coefficient ξ between the k-th dimensions of (a)ij(k) And degree of association ξijRespectively as follows:
ξij＝Πξij(k)
wherein rho is a resolution coefficient, and the interval range is (0, 1);
and constructing a relevance matrix xi according to the relevance:
as an alternative embodiment: the method for reducing scenes of the multi-stage typical scene set by adopting an optimal reduction algorithm based on the correlation matrix until the number of the remaining scenes in the multi-stage typical scene set reaches a preset value comprises the following steps:
scenes requiring downscaling
It satisfies the following formula:
wherein the content of the first and second substances,
for scenes requiring abatement
Corresponding probability, PsAs a scene XsCorresponding probability, PmIs a fieldScene XmCorresponding probability, PnAs a scene XnThe corresponding probability, N is the total number of scenes in the multi-stage scene set,
as a scene XsAnd scene
Degree of association of feature vector, ξ (X)m，Xn) As a scene XmAnd scene XnThe relevance of the feature vector;
changing the total number of scenes N-1, and finding out the scenes needing to be reduced
Scene with maximum correlation coefficient
It satisfies the following formula:
wherein, csFor scenes requiring reduction
Scene with maximum correlation coefficient
The data samples in the scene feature vector of (2),
for scenes requiring downscaling
The scene feature vector of (2);
updating a scene
Probability of (c):
wherein the content of the first and second substances,
for scenes requiring abatement
The corresponding probability;
judging whether the total number N of the current scene is less than or equal to a preset value Nc，
If yes, outputting a typical scene between stages;
if not, the steps are executed circularly.
In a second aspect, the present invention provides a bidirectional optimization apparatus for a multi-stage representative scenario, the bidirectional optimization apparatus comprising:
a sample set generation module: the wind power generation system is used for generating a wind-solar load output sample set according to the wind power generation set output model, the photovoltaic output model and the distribution network load data;
typical daily scene generation module: the method is used for synchronously clustering samples in a sample set by adopting a clustering algorithm based on density peak values to generate a typical day scene in a stage;
a multi-stage representative scene set generation module: the method comprises the steps of performing Cartesian product fusion on adjacent typical daily scenes in a stage to generate a multi-stage typical scene set containing more time scene segments;
a scene feature vector generation module: for constructing a scene feature vector from a multi-stage representative scene set;
the incidence matrix generation module: the method comprises the steps of calculating the relevance between any two scene feature vectors to generate a relevance matrix;
inter-phase typical scene acquisition module: the method comprises the steps of performing scene reduction on a multi-stage typical scene set by adopting an optimal reduction algorithm based on a correlation matrix until the number of remaining scenes in the multi-stage typical scene set reaches a preset value, and acquiring typical scenes between stages;
a multi-stage representative scene set generation module: and generating an optimized multi-stage typical scene set according to the inter-stage typical scenes.
In a third aspect, the invention provides a wind, light and load multi-stage typical scene bidirectional optimization device, which comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any of the above.
In a fourth aspect, the invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the methods described above.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a multi-stage typical scene bidirectional optimization method, a multi-stage typical scene bidirectional optimization device, equipment and a storage medium, 1) based on the lack of research on a multi-stage multi-index scene set in the traditional method, aiming at the uncertainty in the operation of a power distribution network, a wind-light-load multi-dimensional typical scene longitudinal generation strategy and a scene transverse reduction strategy based on a characteristic vector correlation coefficient are provided, the wind-light-load typical scene is optimized from the longitudinal direction and the transverse direction, a scene generation path is optimized, the difficulty of large-scale multi-stage scene calculation complexity is overcome, and the multi-dimensional scene data dimension reduction processing is realized.
2) The difference between multi-stage wind-light-load characteristic vector scenes is measured by introducing the correlation coefficient indexes, so that the correlation characteristics of the wind-light-load original scenes are well reflected in the reduction process, the scenes are reduced by utilizing an optimal reduction technology, a typical scene set of the wind-light-load of the power distribution network is formed, the scene generation precision is further improved, and the problem that the practicability of a single scene is limited is solved.
Drawings
Fig. 1 is a flowchart of a method for bidirectional optimization of a multi-stage typical scene according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The first embodiment is as follows:
a bidirectional optimization method of a wind, light and load multistage typical scene is characterized by comprising the following steps:
firstly, a wind-solar-load multi-dimensional typical scene longitudinal generation strategy:
step 1.1, generating a wind-solar load output sample set according to a wind turbine generator output model, a photovoltaic output model and distribution network load data;
step 1.1.1, the establishment of the wind turbine generator output model comprises the following steps:
the output of the fan mainly depends on the wind speed, the uncertainty of the wind speed at a single moment is described by adopting Weibull distribution, and the distribution function is as follows:
in the formula (1), c is a wind speed parameter, k*The wind speed fluctuation shape parameter is shown, and v is the wind speed;
step 1.1.2, the establishment of the photovoltaic output model comprises the following steps:
the photovoltaic output mainly depends on solar irradiance, Beta distribution can be well fitted with the solar irradiance in a certain time period of a fixed place, and a photovoltaic power continuous probability density function is described by adopting the Beta distribution:
in the formula (2), alpha and Beta are shape parameters of Beta distribution, PpTo photovoltaic power, PPmaxIs the maximum photovoltaic power;
step 1.1.3, the load data of the power distribution network comprises the following steps:
the load rate is adopted to describe the coincidence characteristic:
Pl＝PlMPl rate (3)
in equation (3): plFor the load demand at that moment, PlMIn order to be the peak value of the load,
the load factor at that time.
Step 1.2, synchronously clustering samples in a sample set by adopting a clustering algorithm based on density peak values to generate a typical day scene in a phase;
let the original data set S ═ { x1, x2, …, xn } contain n samples, each sample having K-dimensional attributes,
and synchronously clustering wind, light and load by adopting a clustering algorithm based on density peak values to form a typical scene in a stage.
Step 1.2.1, calculating pairwise distance d between samples in sample setij：
Wherein d isij＝d(xi，xj) Represents the distance between the ith sample and the jth sample; x is the number ofiAnd xjRespectively the ith sample and the jth sample in the sample set,
and
respectively representing the kth attribute of the sample, wherein K represents the attribute dimension of each sample;
step 1.2.2, arranging the distances in ascending order to calculate local density values:
in the formula (5), dcIs a preset truncation distance parameter;
step 1.2.3, obtaining a distance index according to the local density value:
in the formula (6), the first and second groups,
is the q thiSample to qthjShortest distance of one sample, wheniWhen the local density maximum is obtained
Is the maximum distance, p, between the sample and other samplesiThe subscript sequence in descending order is Q ═ { Q ═ Q1,q2,…,qn}，qi、qjAll belong to a set Q;
step 1.2.4, obtaining a cluster center weight according to the local density value and the distance index:
γi＝ρiδi (7)
in the formula (7), ρiIs the local density value of the ith sample, deltaiThe distance index of the ith sample;
step 1.2.5, calculating cluster center weights of all sample points and arranging the weights in a descending order, wherein the sample points with the weights in the descending order are taken as cluster center points, namely cluster centers;
step 1.2.6, based on the clustering center, carrying out normalization processing on data samples of the wind turbine generator, the photovoltaic and the load, and carrying out synchronous clustering by taking days as units to generate a typical day scene.
Secondly, a scene transverse reduction strategy based on the feature vector correlation coefficient:
carrying out Cartesian product fusion on the adjacent typical daily scenes generated in the step one to generate a multi-stage scene set containing more scene segments at moments; repeatedly performing subtraction fusion calculation to generate a classical scene set covering the whole operation interval; and a scene reduction method based on the correlation coefficient is provided for the scene multi-stage feature vector, so that an optimal simplified scene set is obtained.
Step 2.1, carrying out Cartesian product fusion on adjacent typical daily scenes in the stage to generate a multi-stage typical scene set containing more scene segments at more moments; in the classical scene set algorithm flow, Cartesian product fusion is carried out on adjacent typical daily scenes. Assuming that a wind power typical daily scene has X scene segments, a photovoltaic typical daily scene has Y scene segments, and a load typical daily scene has Z scene segments, sequentially combining the X segments of the wind power scene with the Y segments of the photovoltaic scene, and combining the X segments of the wind power scene with the Z segments of the load scene to form a multi-stage scene set with the size of X multiplied by Y multiplied by Z;
2.2, constructing a scene feature vector according to the multi-stage typical scene set;
normalizing the data samples for each stage to construct a scene feature vector
And constructs feature vectors Eig of the multi-stage typical scene on the basis of the feature vectorsb：
Eigb＝[Wb,Pb,Lb]
In the formula (8), WbFor installed capacity of wind turbine, PbFor photovoltaic installed capacity, LbIn order to be the amount of the load,
for the installed capacity of the wind turbine at the u-th stage,
for the u stage windThe capacity of the motor assembly is arranged,
for the u-th stage photovoltaic installed capacity,
the u-th stage maximum load.
2.3, calculating the relevance between any two scene feature vectors to generate a relevance matrix;
to measure the difference between the multi-stage scene feature vectors, a grey correlation coefficient is introduced as a measure of the correlation between the two scene vectors.
Any two scene feature vectors Eig for a multi-phase sceneiAnd EigjIs the k-th dimension of (c) and a correlation coefficient ξ between the k-th dimensions of (a)ij(k) And degree of association ξijRespectively as follows:
ξij＝∏ξij(k) (10)
in the formula (9), ρ is a resolution coefficient for reducing the influence of the maximum value on the distortion of the correlation coefficient, and the resolution between the correlation coefficients can be improved. The value interval of rho is (0,1), and rho usually takes 0.5;
2.4, constructing a relevance matrix according to the relevance:
2.5, performing scene reduction on the multi-stage typical scene set by adopting an optimal reduction algorithm based on the correlation matrix until the number of the remaining scenes in the multi-stage typical scene set reaches a preset value, and acquiring typical scenes between stages;
step 2.5.1, scenes that need to be reduced
It satisfies the following formula:
in the formula (12), the first and second groups,
for scenes requiring abatement
Corresponding probability, PsAs a scene XsCorresponding probability, PmAs a scene XmCorresponding probability, PnAs a scene XnThe corresponding probability, N is the total number of scenes in the multi-stage scene set,
as a scene XsAnd scene
Degree of association of feature vector, ξ (X)m，Xn) As a scene XmAnd scene XnThe relevance of the feature vector;
step 2.5.2, the total number N of the changed scenes is equal to N-1, and the scenes which need to be reduced are searched
Scene with maximum correlation coefficient
It satisfies the following formula:
in the formula (13), csFor scenes requiring reduction
Scene with maximum correlation coefficient
The data samples in the scene feature vector of (2),
for scenes requiring downscaling
The scene feature vector of (2);
step 2.5.3, updating scenes
Probability of (c):
in the formula (14), the reaction mixture,
for scenes requiring abatement
The corresponding probability;
step 2.5.4, judging whether the total number N of the current scenes is less than or equal to the total number N of the preset reserved scenesc，
If yes, outputting a scene reduction result;
if not, the steps 2.5.1-2.5.4 are executed in a circulating way.
And 2.6, generating an optimized multi-stage typical scene set according to the inter-stage typical scenes.
Step three, solving a multi-stage optimal scene set of the power distribution network by adopting an optimal reduction technology:
step 3.1, generating a wind-solar load output sample set:
generating a wind-solar load output sample set according to an output model of Weibull distribution of the wind turbine generator, an output model of photovoltaic Beta distribution and load data of the power distribution network;
step 3.2, generating a typical scene in the stage:
synchronously clustering samples containing wind, light and load three-dimensional attributes by adopting a clustering algorithm based on density peak values to form a typical scene in a stage;
3.3, carrying out Cartesian product fusion on adjacent typical day scenes to generate a multi-stage scene set containing more moment scene segments;
step 3.4, calculating a correlation coefficient between any two scene feature vectors to generate a correlation matrix;
and 3.5, utilizing an optimal reduction technology to reduce scenes: and circularly calculating the correlation coefficient between the scenes in each stage, and gradually deleting the scenes with higher correlation degree with other scenes until the total number of the reserved scenes reaches a preset value to obtain typical scenes in stages, thereby generating a multi-stage wind-solar-load typical scene set.
Example two:
a bidirectional optimization device for wind-solar-charged multi-stage typical scenes comprises:
a sample set generation module: the wind power generation system is used for generating a wind-solar load output sample set according to the wind power generation set output model, the photovoltaic output model and the distribution network load data;
typical daily scene generation module: the method is used for synchronously clustering samples in a sample set by adopting a clustering algorithm based on density peak values to generate a typical day scene in a stage;
a multi-stage representative scene set generation module: the method comprises the steps of performing Cartesian product fusion according to adjacent typical daily scenes in a stage to generate a multi-stage typical scene set containing more time scenes;
a scene feature vector generation module: for constructing a scene feature vector from a multi-stage representative scene set;
the incidence matrix generation module: the method comprises the steps of generating a relevance matrix according to a scene feature vector and introduced relevance;
a reduced result output module: the method is used for establishing a scene reduction method based on the correlation degree, reducing the multi-stage typical scene and outputting a reduction result;
a multi-stage representative scene set generation module: and obtaining an optimized multi-stage typical scene according to the reduction result and generating a multi-stage typical scene set.
It should be noted that: the wind, photovoltaic and load multi-stage typical scene bidirectional optimization device provided by the embodiment of the invention can be used for realizing the wind, photovoltaic and load multi-stage typical scene bidirectional optimization method described in the first embodiment, and the method steps for realizing the corresponding functions of each module in the device can be executed by referring to the first embodiment, which is not described herein again.
Example three:
the embodiment of the invention also provides wind, light and load multi-stage typical scene bidirectional optimization equipment, which comprises a processor and a storage medium;
a storage medium to store instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method of embodiment one.
Example four:
embodiments of the present invention also provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the method of an embodiment.
In summary, the bidirectional optimization method, device, equipment and storage medium for the multi-stage typical scene provided by the invention provide a wind-light-load multi-dimensional typical scene longitudinal generation strategy and a scene transverse reduction strategy based on the feature vector correlation coefficient aiming at the uncertainty in the operation of the power distribution network, optimize the wind-light-load typical scene from the longitudinal direction and the transverse direction, optimize the scene generation path, overcome the difficulty of complex large-scale multi-stage scene calculation, and realize the dimension reduction processing of multi-dimensional scene data; the difference between multi-stage wind-light-load characteristic vector scenes is measured by introducing the correlation coefficient indexes, so that the correlation characteristics of the wind-light-load original scenes are well reflected in the reduction process, the scenes are reduced by utilizing an optimal reduction technology, a typical scene set of the wind-light-load of the power distribution network is formed, the scene generation precision is further improved, and the problem that the practicability of a single scene is limited is solved.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (9)
1. A bidirectional optimization method of a wind-solar-charged multi-stage typical scene is characterized by comprising the following steps:
generating a wind and light load output sample set according to the wind turbine generator output model, the photovoltaic output model and the load data of the power distribution network;
synchronously clustering the samples in the sample set by adopting a clustering algorithm based on density peak values to generate a typical day scene in a stage;
carrying out Cartesian product fusion on adjacent typical daily scenes in the stages to generate a multi-stage typical scene set containing more scene segments at more moments;
constructing a scene feature vector according to the multi-stage typical scene set;
calculating the relevance between any two scene feature vectors to generate a relevance matrix;
performing scene reduction on the multi-stage typical scene set by adopting an optimal reduction algorithm based on the correlation matrix until the number of the remaining scenes in the multi-stage typical scene set reaches a preset value, and acquiring typical scenes between stages;
and generating an optimized multi-stage typical scene set according to the inter-stage typical scenes.
2. The bi-directional optimization method of claim 1,
the establishment of the wind turbine generator output model comprises the following steps:
describing the uncertainty of the wind speed at a single moment by using Weibull distribution, and a distribution function F thereofw(v)：
Wherein c is a wind speed parameter, k*The wind speed fluctuation shape parameter is shown, and v is the wind speed;
the establishment of the photovoltaic output model comprises the following steps:
and (3) describing a photovoltaic power continuous probability density function by adopting Beta distribution:
wherein alpha and Beta are Beta distribution shape parameters, PpTo photovoltaic power, PPmaxIs the maximum photovoltaic power;
the power distribution network load data comprises:
the load factor is used to describe the compliance characteristics as a function of:
Pl＝PlMPl rate
wherein: plFor the load demand at that moment, PlMIn order to be the peak value of the load,
the load factor at that time.
3. The bi-directional optimization method of claim 1, wherein said employing a density peak based clustering algorithm to cluster samples in the sample set synchronously, generating an intra-stage typical daily scenario comprises:
calculating the distance d between two samples in the sample setij：
Wherein d isij＝d(xi，xj) Represents the distance between the ith sample and the jth sample; x is the number ofiAnd xjRespectively the ith sample and the jth sample in the sample set,
and
respectively representing the kth attribute of the sample, wherein K represents the attribute dimension of each sample;
arranging the distances in ascending order to calculate the local density value rhoi：
Wherein d iscIs a preset truncation distance parameter;
obtaining a distance index according to the local density value
Wherein the content of the first and second substances,
is the q thiSample to qthjShortest distance of one sample, wheniWhen the local density maximum is obtained
Is the maximum distance, p, between the sample and other samplesiThe subscript sequence in descending order is Q ═ { Q ═ Q1,q2,…,qn}，qi、qjAll belong to a set Q;
obtaining a cluster center weight value gamma according to the local density value and the distance indexi：
γi＝ρiδi
Where ρ isiIs the local density value of the ith sample, deltaiThe distance index of the ith sample;
calculating cluster center weights of all sample points and arranging the cluster center weights in a descending order, wherein the sample points arranged in the descending order of the weights are used as cluster center points, namely cluster centers;
based on a clustering center, normalization processing is carried out on data samples of the wind turbine generator, the photovoltaic and the load, and synchronous clustering is carried out by taking the day as a unit to generate a typical day scene.
4. The bi-directional optimization method of claim 1, wherein said constructing scene feature vectors from a multi-stage representative scene set comprises:
normalizing the data samples for each stage to construct a scene feature vector
And constructs feature vectors Eig of the multi-stage typical scene on the basis of the feature vectorsb：
Eigb＝[Wb,Pb,Lb]
Wherein, WbFor installed capacity of wind turbine, PbFor photovoltaic installed capacity, LbIn order to be the amount of the load,
for the installed capacity of the wind turbine at the u-th stage,
for the u-th stage photovoltaic installed capacity,
the u-th stage maximum load.
5. The bi-directional optimization method of claim 4, wherein the calculating the correlation between any two scene feature vectors and the generating the correlation matrix comprises:
any two scene feature vectors Eig for a multi-phase sceneiAnd EigjIs the k-th dimension of (c) and a correlation coefficient ξ between the k-th dimensions of (a)ij(k) And degree of association ξijRespectively as follows:
ξij＝∏ξij(k)
wherein rho is a resolution coefficient, and the interval range is (0, 1);
and constructing a relevance matrix xi according to the relevance:
6. the bidirectional optimization method of claim 5, wherein the scene reduction is performed on the multi-stage typical scene set by using an optimal reduction algorithm based on the correlation matrix until the number of remaining scenes in the multi-stage typical scene set reaches a preset value, and the obtaining of the inter-stage typical scene comprises:
scenes requiring downscaling
It satisfies the following formula:
wherein the content of the first and second substances,
for scenes requiring abatement
Corresponding probability, PsAs a scene XsCorresponding probability, PmAs a scene XmCorresponding probability, PnAs a scene XnThe corresponding probability, N is the total number of scenes in the multi-stage scene set,
as a scene XsAnd scene
Degree of association of feature vector, ξ (X)m，Xn) As a scene XmAnd scene XnThe relevance of the feature vector;
changing the total number of scenes N-1, and finding out the scenes needing to be reduced
Scene with maximum correlation coefficient
It satisfies the following formula:
wherein, csFor scenes requiring reduction
Scene with maximum correlation coefficient
The data samples in the scene feature vector of (2),
for scenes requiring downscaling
The scene feature vector of (2);
updating a scene
Probability of (c):
wherein the content of the first and second substances,
for scenes requiring abatement
The corresponding probability;
judging whether the total number N of the current scene is less than or equal to a preset value Nc，
If yes, outputting a typical scene between stages;
if not, the steps are executed circularly.
7. A bidirectional optimization device for wind, light and load multi-stage typical scenes is characterized by comprising:
a sample set generation module: the wind power generation system is used for generating a wind-solar load output sample set according to the wind power generation set output model, the photovoltaic output model and the distribution network load data;
typical daily scene generation module: the method is used for synchronously clustering samples in a sample set by adopting a clustering algorithm based on density peak values to generate a typical day scene in a stage;
a multi-stage representative scene set generation module: the method comprises the steps of performing Cartesian product fusion on adjacent typical daily scenes in a stage to generate a multi-stage typical scene set containing more time scene segments;
a scene feature vector generation module: for constructing a scene feature vector from a multi-stage representative scene set;
the incidence matrix generation module: the method comprises the steps of calculating the relevance between any two scene feature vectors to generate a relevance matrix;
inter-phase typical scene acquisition module: the method comprises the steps of performing scene reduction on a multi-stage typical scene set by adopting an optimal reduction algorithm based on a correlation matrix until the number of remaining scenes in the multi-stage typical scene set reaches a preset value, and acquiring typical scenes between stages;
a multi-stage representative scene set generation module: and generating an optimized multi-stage typical scene set according to the inter-stage typical scenes.
8. A wind, light and load multi-stage typical scene bidirectional optimization device is characterized by comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any one of claims 1 to 6.
9. Computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110380662.6A CN113190973A (en) | 2021-04-09 | 2021-04-09 | Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110380662.6A CN113190973A (en) | 2021-04-09 | 2021-04-09 | Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113190973A true CN113190973A (en) | 2021-07-30 |
Family
ID=76975174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110380662.6A Pending CN113190973A (en) | 2021-04-09 | 2021-04-09 | Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113190973A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523351A (en) * | 2023-07-03 | 2023-08-01 | 广东电网有限责任公司湛江供电局 | Source-load combined typical scene set generation method, system and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012063544A1 (en) * | 2010-11-09 | 2012-05-18 | オリンパス株式会社 | Image processing device, image processing method, and recording medium |
| CN110046801A (en) * | 2019-03-25 | 2019-07-23 | 国网江苏省电力有限公司经济技术研究院 | A kind of typical scene generation method of power distribution network electric system |
| CN110516944A (en) * | 2019-08-20 | 2019-11-29 | 国网江苏省电力有限公司 | A kind of power distribution network multistage typical case's Run-time scenario generation method |
| CN111859283A (en) * | 2020-06-02 | 2020-10-30 | 天津大学 | Scene generation method considering multi-energy charge-time sequence and correlation |
| CN112541546A (en) * | 2020-12-14 | 2021-03-23 | 国网江苏省电力有限公司经济技术研究院 | Photovoltaic power station typical scene generation method based on multi-scene model |
-
2021
- 2021-04-09 CN CN202110380662.6A patent/CN113190973A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012063544A1 (en) * | 2010-11-09 | 2012-05-18 | オリンパス株式会社 | Image processing device, image processing method, and recording medium |
| CN110046801A (en) * | 2019-03-25 | 2019-07-23 | 国网江苏省电力有限公司经济技术研究院 | A kind of typical scene generation method of power distribution network electric system |
| CN110516944A (en) * | 2019-08-20 | 2019-11-29 | 国网江苏省电力有限公司 | A kind of power distribution network multistage typical case's Run-time scenario generation method |
| CN111859283A (en) * | 2020-06-02 | 2020-10-30 | 天津大学 | Scene generation method considering multi-energy charge-time sequence and correlation |
| CN112541546A (en) * | 2020-12-14 | 2021-03-23 | 国网江苏省电力有限公司经济技术研究院 | Photovoltaic power station typical scene generation method based on multi-scene model |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523351A (en) * | 2023-07-03 | 2023-08-01 | 广东电网有限责任公司湛江供电局 | Source-load combined typical scene set generation method, system and equipment |
| CN116523351B (en) * | 2023-07-03 | 2023-09-22 | 广东电网有限责任公司湛江供电局 | Source-load combined typical scene set generation method, system and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hu et al. | A new clustering approach for scenario reduction in multi-stochastic variable programming | |
| CN112564098B (en) | High-proportion photovoltaic power distribution network voltage prediction method based on time convolution neural network | |
| CN110247438B (en) | Active power distribution network resource optimization configuration based on longicorn whisker algorithm | |
| CN103972908A (en) | Multi-target reactive power optimization method based on adaptive chaos particle swarm algorithm | |
| CN110516944B (en) | Power distribution network multi-stage typical operation scene generation method | |
| CN111681130A (en) | Comprehensive energy system optimization scheduling method considering condition risk value | |
| CN110264012A (en) | Renewable energy power combination prediction technique and system based on empirical mode decomposition | |
| CN112271731B (en) | Method for generating and reducing wind power multi-period time sequence scene | |
| CN111144641A (en) | Improved particle swarm algorithm-based microgrid optimization scheduling method | |
| CN111027769A (en) | Distributed power supply site selection and volume fixing optimization method based on genetic-simulated annealing hybrid algorithm | |
| CN111404193A (en) | Data-driven-based microgrid random robust optimization scheduling method | |
| CN110661257A (en) | Water-fire combined power system optimal economic operation strategy analysis method based on longicorn swarm algorithm | |
| CN113190973A (en) | Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene | |
| CN109145503B (en) | High-precision dynamic modeling method for photovoltaic power station cluster | |
| Chen et al. | Output prediction of alpha-type Stirling engines using gradient boosted regression trees and corresponding heat recovery system optimization based on improved NSGA-II | |
| CN116865318A (en) | Power transmission network and energy storage joint planning method and system based on two-stage random optimization | |
| CN110765726B (en) | Intelligent generation system of energy network planning diagram | |
| CN110276478B (en) | Short-term wind power prediction method based on segmented ant colony algorithm optimization SVM | |
| CN116542385A (en) | Integrated learning method and system for wind and light power prediction | |
| CN107276093A (en) | The Probabilistic Load computational methods cut down based on scene | |
| CN114580298A (en) | RBF-ISBO-based microgrid optimization scheduling method and system | |
| CN110829440B (en) | Three-phase medium-low voltage integrated power distribution network voltage control method and system | |
| CN114204613A (en) | Reactive compensation method and system for offshore wind farm access power system | |
| CN113449466A (en) | Solar radiation prediction method and system for optimizing RELM based on PCA and chaos GWO | |
| CN109980649A (en) | It is a kind of meter and multiple stable point the probability load flow calculation method based on saddle point approximation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210730 |
|
| RJ01 | Rejection of invention patent application after publication |