CN110046801B - Typical scene generation method of power distribution network power system - Google Patents
Typical scene generation method of power distribution network power system Download PDFInfo
- Publication number
- CN110046801B CN110046801B CN201910227550.XA CN201910227550A CN110046801B CN 110046801 B CN110046801 B CN 110046801B CN 201910227550 A CN201910227550 A CN 201910227550A CN 110046801 B CN110046801 B CN 110046801B
- Authority
- CN
- China
- Prior art keywords
- clustering
- sample
- data
- distance
- power generation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000010248 power generation Methods 0.000 claims abstract description 47
- 238000003064 k means clustering Methods 0.000 claims abstract description 22
- 238000011156 evaluation Methods 0.000 claims description 11
- 230000001174 ascending effect Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 abstract description 2
- 238000007781 pre-processing Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 3
- 230000005622 photoelectricity Effects 0.000 description 2
- 241001497337 Euscorpius gamma Species 0.000 description 1
- 230000002902 bimodal effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000006471 dimerization reaction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000005829 trimerization reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Probability & Statistics with Applications (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a typical scene generation method of a power distribution network power system, which mainly comprises the following steps: carrying out normalization processing on the original data of the power grid according to time scales to obtain a clustering sample; preprocessing a clustering sample by adopting a density extremum method to obtain an initial clustering center; according to the initial clustering center, a K-means clustering method is adopted to synchronously cluster wind power generation, photovoltaic power generation, tidal power generation and time sequence loads, so that typical operation scenes of the power distribution network under different time scales are formed. The invention improves the K-means clustering by using the density peak value clustering, avoids the problem that the initial clustering center and the clustering number of the traditional K-means clustering can not be determined, and has stronger clustering effectiveness and convergence compared with the traditional K-means clustering after improvement.
Description
Technical Field
The invention relates to a method for generating a typical operation scene of a power distribution network, in particular to a method for generating a typical operation scene of a wind-solar-sea-oriented power distribution network based on a density peak value and a K-average value, and belongs to the field of power distribution network planning.
Background
For a power distribution system containing renewable energy sources, after a distributed power source is connected to a power grid, a power grid midrange Jing Duoyuan is formed, and complexity of planning and scheduling analysis is improved, so that a large number of scenes are required to be cut or clustered, and a typical scene is formed so as to be convenient for planning and scheduling.
The existing typical scene generation method of the related electric power system mainly comprises K-means clustering, hierarchical clustering, density clustering, fuzzy C-means clustering, scene subtraction and the like, the clustering results and the clustering efficiencies of different clustering methods are different, and the K-means clustering serving as a common clustering method is widely applied to the generation of the typical scene of the electric power system. However, the initial clustering center and the number of clusters of the traditional K-means clustering are required to be determined manually, so that the clustering effect is unstable. Aiming at the problem, part of documents adopt an enumeration method to enumerate the number of clustering centers so as to determine the optimal clustering number, but the method is complicated and cannot determine the initial clustering center position.
Disclosure of Invention
The invention provides a typical operation scene generation method of a power distribution network based on density peak values and K-means, which aims to solve the problems in the prior art, and comprises the following specific technical scheme:
the technical scheme provided by the invention is as follows: a typical scene generation method of a power distribution network power system is characterized by comprising the following steps:
1) Normalizing the original data of the power grid according to a time scale to form n sample points separated by the time scale and n samples to be clustered which correspond to each sample point and are composed of normalized data, wherein the n samples to be clustered are x 1 ,x 2 ,…,x n Form a sample set S to be clustered, S= { x 1 ,x 2 ,…,x n };
2) Calculating the distance d between any two sample points i and j in n sample points ij The n sample points are counted with M distance values, M=n (n-1)/2, the M distance values are arranged according to the order of magnitude to form a corresponding distance value array with the sequence numbers 1-M, the distance value dq corresponding to the q-th sequence number in the array is set as a cut-off distance dc, and q= [0.02M];
3) According toCalculating the local density ρ of the sample point i i And the local density values corresponding to all sample points are arranged in a descending order, and Q= { Q1, Q2, …, qn } is defined as a subscript sequence of the local density i arranged in a descending order, so as to form local density value arranged sequence Q1, Q2, … qn;
4) Calculating distance index delta of qi sample point qi Evaluation index gamma qi ,
5) Making a decision graph based on the evaluation index gamma qi Formed in a coordinate system with a ordinate and n sample points arranged in time sequence and a abscissa, gamma qi Forming a decision graph from high to low;
6) The clustering number T and the clustering center M are made according to the larger interval between the clustering center points and the non-clustering center points in the decision diagram k (k=1, 2, … T);
7) According to the cluster center M k And carrying out K-means clustering on the samples to be clustered.
The technical scheme is further designed as follows: the time scale is one hour or one day.
The samples to be clustered comprise: sample H to be clustered formed by normalized load data loadi, wind power generation data wpi, photoelectric power generation data pvi and tidal power generation data tpi per hour i ,H i =[load i ,wp i ,pv i ,tp i ] 1×4 ,H i There are 4 normalized data; or a sample D to be clustered formed by daily normalized load data Loadi, wind power generation data WPi, photoelectric power generation data PVi and tidal power generation data TPi i ,D i =[Load i ,WP i ,PV i ,TP i ] 1×96 ,D i There were 96 normalized data.
Distance d between any two sample points i and j ij , The corresponding c-th data in sample point i and sample point j, respectively.
The M distance values are arranged in order of magnitude including an ascending value arrangement or a descending value arrangement.
Said clustering center M k K-means clustering is carried out, which comprises the following steps:
1) Calculate each sample X i With each cluster center M k Distance D of (2) ik =|M k -X i I and sample X according to minimum distance i Dividing to form a number N containing objects k Cluster C of (C) k (k=1,2,…,T);
2) Calculating the mean value of each cluster as an updated cluster center:
3) Repeating the steps 1) and 2) until the cluster center is not changed.
The invention improves the K-mean value clustering by adopting the density peak value clustering, and determines the clustering center position and the number thereof by utilizing the characteristics that the local density of the clustering center point is high and the distance between the center point and the sample point with higher density is relatively larger, thereby solving the problem that the initial clustering center and the clustering number of the traditional K-mean value clustering are difficult to determine, and compared with the traditional K-mean value clustering, the improved clustering has stronger effectiveness and convergence.
Drawings
FIG. 1a is a normalized time series load data image of embodiment 1 of the present invention;
FIG. 1b is a normalized wind power generation data image of embodiment 1 of the present invention;
FIG. 1c is a normalized photovoltaic power generation data image of example 1 of the present invention;
FIG. 1d is a normalized tidal Power data image of embodiment 1 of the present invention;
FIG. 2 is a decision chart of evaluation indexes of the initial cluster center in example 1;
FIG. 3 is a typical scene clustering effect diagram in embodiment 1;
FIG. 4 is a graph of typical day-one clustering results in example 2;
FIG. 5 is a graph of typical daily dimerization results in example 2;
FIG. 6 is a graph of typical daily trimerization class results in example 2;
FIG. 7 is a graph of exemplary day four clustering results in example 2;
FIG. 8 is a graph of typical day five clustering results in example 2;
FIG. 9 is a graph of typical day six clustering results in example 2;
FIG. 10 is a graph of exemplary Japanese seven clustering results in example 2;
FIG. 11 is a graph of exemplary day eight clustering results in example 2;
FIG. 12 is a graph of exemplary Japanese nine clustering results in example 2;
FIG. 13 is a graph of exemplary day ten clustering results in example 2;
FIG. 14 is a graph of exemplary day eleven clustering results in example 2;
FIG. 15 is a graph of exemplary day twelve clustering results in example 2;
FIG. 16 is a graph of a typical day thirteen clustering result in example 2;
FIG. 17 is a graph of exemplary day fourteen clustering results in example 2;
FIG. 18 is a graph of exemplary day-fifteen clustering results in example 2;
fig. 19 is a flowchart of a scene generating method according to embodiment 1 of the present invention.
Detailed Description
The invention will now be described in detail with reference to the accompanying drawings and specific examples.
Example 1
The wind-solar-sea-oriented power distribution network comprises wind power generation, photovoltaic power generation, tidal power generation and time sequence load, so that the original data adopted in the embodiment comprise time sequence load data, wind power generation data, photovoltaic power generation data and tidal power generation data. The related wind power generation data and the photovoltaic power generation data are derived from the national renewable energy laboratory, the tidal power generation data are derived from the national atmospheric and ocean administration official website, and the load data adopt the typical time sequence load of the RBTS system.
The process for implementing the method of the present invention is shown in fig. 19, and the first step is: normalizing the original data of the power distribution network comprising wind power generation, photovoltaic power generation, tidal power generation and time sequence load according to a time scale of one hour to form load data Loadi, wind power generation data WPi, photovoltaic power generation data PVi and tidal power generation data TPi normalized per hour, wherein the normalized sample H is obtained by normalizing the original data of the power distribution network in one hour i ,H i =[load i ,wp i ,pv i ,tp i ] 1×4 . For a pair ofCorresponding to 8760 hours a year, 8760 sample points separated by a time scale of 1 hour are formed, each sample point is composed of corresponding normalized data and corresponds to a sample H to be clustered i =[load i ,wp i ,pv i ,tp i ] 1×4 Images corresponding to time series load data, wind power generation data, photovoltaic power generation data, and tidal power generation data for 8760 hours after normalization are shown in fig. 1a, 1b, 1c, and 1d, respectively. 8760 samples x to be clustered corresponding to 8760 sample points 1 ,x 2 ,…,x 8760 Form a sample set S to be clustered, S= { x 1 ,x 2 ,…,x n }。
And a second step of: according to Euclidean distance formulaCalculating the distance d between any two of 8760 sample points, e.g., the 6 th sample (i=6) and the 8 th sample (j=8) ij If the data arrangement of the sample points is sequentially load data Loadi, wind power generation data WPi, photovoltaic power generation data PVi and tidal power generation data TPi, the data arrangement sequence numbers are sequentially c=1, 2,3,4, m is the maximum sequence number, and m=4. M=n (n-1)/2= 38364420 distance values are counted between 8760 sample points, 38364420 distance values are arranged according to an ascending order to form an ascending ordered array (one-dimensional array) of distance values corresponding to sequence numbers 1-38364420, the distance value dq corresponding to the q-th sequence number in the array is set as a cut-off distance dc, and q= [0.02M is taken]= 767288, i.e.: the cut-off distance dc is a distance value corresponding to the sequence number 767288 in the ascending value sequence.
And a third step of: d obtained by the above calculation ij And dc, again according toCalculating the local density ρ of the sample point i i And the local density values corresponding to all the sample points are arranged in a descending order to form a corresponding local density array ρ q1 ,ρ q2 ,…ρ qn Definition q= { Q 1 ,q 2 ,…,q n Is the local density ρ i Descending order of arrangementIs a subscript sequence of (2).
Fourth step: calculating a distance index delta qi Evaluation index gamma qi . For the distance index delta corresponding to the local density value with the subscript sequence qi in the descending order of the local density values qi Evaluation index gamma qi The shortest distance between all sample points with higher density than the sample i and the sample i is calculated by the following formula qi 。
Fifth step: and (5) making a decision graph. The decision graph is based on the evaluation index gamma qi Formed in a coordinate system with ordinate and time-ordered scene time positions as abscissa, gamma qi The high to low ordering forms a decision graph, as in fig. 2.
Sixth step: the clustering number T and the clustering center M are made according to the larger interval between the clustering center points and the non-clustering center points in the decision diagram k (k=1, 2, … T). In fig. 2 there is a distinct break point (i.e. gamma between two adjacent points i Large difference) for a total of T break points. Adjacent points after the break point T are densely arranged, and gamma between the two points i The phase difference is small. Thus, the first T break points represent the number of clusters. First 5 points gamma i The difference is larger, and the rest point gamma i The phase difference is small, and the arrangement is dense. As can be seen from FIG. 2, the initial clustering number of typical hours is 5, so the first 5 evaluation indexes gamma i The corresponding scene is used as an initial clustering center.
Seventh step: according to the cluster center M k And carrying out K-means clustering on the samples to be clustered. Determining an initial cluster center M k Then, 8760 samples are clustered by K-means clustering, which specifically comprises the following steps:
1) Calculate each sample X i With each cluster center M k Distance D of (2) ik =|M k -X i I and sample X according to minimum distance i Dividing to form a number N containing objects k Cluster C of (C) k (k=1,2,…,T);
2) Calculating the mean value of each cluster as an updated cluster center:
3) Repeating the steps 1) and 2) until the cluster center is not changed.
Thus, 5 typical scenes were obtained, the probability of each scene is shown in table 1, and the clustering effect is shown in fig. 3.
Table 1 typical hour scene probability
Comparative examples
And (3) clustering the 8760-hour scene containing load, wind power, photoelectricity and tidal power by adopting the traditional K-means clustering, and comparing the K-means clustering with the K-means clustering based on density peak values, wherein CH (+) and DB (-) indexes are used as standards for judging the effectiveness of the clustering. Because the initial clustering center of the traditional K-means clustering algorithm is randomly generated, the clustering result is unstable, and each evaluation index is obtained after the average value of 100 times of clustering results is calculated.
TABLE 2K-means clustering contrast before and after improvement
As can be seen from Table 2, the improved K-means algorithm CH (+) and DB (-) has better indexes than the traditional K-means clustering algorithm, namely, the clustering effectiveness is high. Meanwhile, the iteration times of the improved K-means clustering algorithm are obviously lower than those of the traditional K-means clustering algorithm, and the reason is that the initial clustering center determined by the density peak clustering is near the final clustering center, so that the required iteration times are obviously reduced. The traditional K-means clustering initial clustering center is randomly generated, so that the needed iteration times are more.
Example 2
The original data and sources adopted in the embodiment are the same as those in the embodiment 1, and the difference is that the embodiment normalizes the original data of the power distribution network with wind power generation, photovoltaic power generation, tidal power generation and time sequence load according to a time scale of one day, and daily clusters the 365-day scene with load, wind power, photoelectricity and tidal power generation to form a typical scene with a 'day' as a time scale. The same as the time clustering, the first step is to determine the initial clustering center position and number by the density peak clustering, and the evaluation index gamma i The decision graph resulted in 15 typical days. And secondly, carrying out K-means clustering according to the initial clustering center, wherein the probability of finally obtaining 15 typical scenes is shown in table 3, and the clustering result of each scene is shown in fig. 4-18.
TABLE 3 probability of typical day scene
As can be seen from table 3 and fig. 4-18, the load was lower during the night and higher during the day. Wind power exhibits anti-peaking characteristics with load in most cases. The photoelectric has obvious regularity, and the peak value of the photoelectric reaches the peak value near noon in general, although the peak value of the photoelectric can be different in different running scenes. Tidal power generation shows a 'multimodal' characteristic because the tidal cycle at the research site belongs to a mixed tide, namely, part of the date in each month shows two high tides and two low tides in one solar day, and the rest date shows only one high tides and one low tides, so that the tidal power generation generally shows a 'bimodal' and a 'four peaks' in one solar day, meanwhile, 15 typical scenes are observed to find that the difference between different typical scenes is the largest difference of tidal power generation, because the daily fluctuation time of the tide changes along with the date, the corresponding moment of the peak of the tidal power generation changes, and the tidal power generation curve between different typical scenes is large. Therefore, the characteristics of the clustering result obtained by the scene generation method are consistent with the actual conditions, and the clustering result is effective.
The technical scheme of the invention is not limited to the embodiments, and all technical schemes obtained by adopting equivalent substitution modes fall within the scope of the invention.
Claims (5)
1. A typical scene generation method of a power distribution network power system is characterized by comprising the following steps of: the method comprises the following steps:
1) Normalizing the original data of the power grid according to a time scale to form n sample points separated by the time scale and n samples to be clustered which correspond to each sample point and are composed of normalized data, wherein the n samples to be clustered are x 1 ,x 2 ,…,x n Form a sample set S to be clustered, S= { x 1 ,x 2 ,…,x n };
The samples to be clustered comprise: sample H to be clustered formed by normalized load data loadi, wind power generation data wpi, photoelectric power generation data pvi and tidal power generation data tpi per hour i ,H i =[load i ,wp i ,pv i ,tp i ] 1×4 ,H i There are 4 normalized data; or a sample D to be clustered formed by daily normalized load data Loadi, wind power generation data WPi, photoelectric power generation data PVi and tidal power generation data TPi i ,D i =[Load i ,WP i ,PV i ,TP i ] 1×96 ,D i There were 96 normalized data;
2) Calculating the distance d between any two sample points i and j in n sample points ij The n sample points are counted with M distance values, M=n (n-1)/2, the M distance values are arranged according to the order of magnitude to form a corresponding distance value array with the sequence numbers 1-M, the distance value dq corresponding to the q-th sequence number in the array is set as a cut-off distance dc, and q= [0.02M];
3) According toCalculating the local density ρ of the sample point i i And all samples are takenThe local density values corresponding to the points are arranged in descending order, and Q= { Q1, Q2, …, qn } is defined as the local density ρ i A descending sequence of subscripts forming a descending sequence of partial density values q1, q2, … qn;
4) Calculating distance index delta of qi sample point qi Evaluation index gamma qi ,
5) Making a decision graph based on the evaluation index gamma qi Formed in a coordinate system with a ordinate and n sample points arranged in time sequence and a abscissa, gamma qi Forming a decision graph from high to low;
6) The clustering number T and the clustering center M are made according to the larger interval between the clustering center points and the non-clustering center points in the decision diagram k (k=1, 2, … T);
7) According to the cluster center M k And carrying out K-means clustering on the samples to be clustered.
2. A typical scenario generation method of a power distribution network power system according to claim 1, characterized in that: the time scale is one hour or one day.
3. A typical scenario generation method of a power distribution network power system according to claim 1, characterized in that: distance d between any two sample points i and j ij , Pairs of sample points i and j, respectivelyShould data c.
4. A typical scenario generation method of a power distribution network power system according to claim 1, characterized in that: the M distance values are arranged in order of magnitude including an ascending value arrangement or a descending value arrangement.
5. A typical scenario generation method of a power distribution network power system according to claim 1, characterized in that: said clustering center M k K-means clustering is carried out, which comprises the following steps:
1) Calculate each sample X i With each cluster center M k Distance D of (2) ik =|M k -X i I and sample X according to minimum distance i Dividing to form a number N containing objects k Cluster C of (C) k (k=1,2,…,T);
2) Calculating the mean value of each cluster as an updated cluster center:
3) Repeating the steps 1) and 2) until the cluster center is not changed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910227550.XA CN110046801B (en) | 2019-03-25 | 2019-03-25 | Typical scene generation method of power distribution network power system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910227550.XA CN110046801B (en) | 2019-03-25 | 2019-03-25 | Typical scene generation method of power distribution network power system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110046801A CN110046801A (en) | 2019-07-23 |
CN110046801B true CN110046801B (en) | 2023-07-21 |
Family
ID=67275095
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910227550.XA Active CN110046801B (en) | 2019-03-25 | 2019-03-25 | Typical scene generation method of power distribution network power system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110046801B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110516944B (en) * | 2019-08-20 | 2022-04-15 | 国网江苏省电力有限公司 | Power distribution network multi-stage typical operation scene generation method |
CN111523230B (en) * | 2020-04-22 | 2023-05-26 | 国网能源研究院有限公司 | Self-adaptive clustering method for wind-light load composite typical scene |
CN112966717A (en) * | 2021-02-04 | 2021-06-15 | 国网浙江省电力有限公司电力科学研究院 | Adjustable potential aggregation method for adjustable resources of power distribution network |
CN113190973A (en) * | 2021-04-09 | 2021-07-30 | 国电南瑞科技股份有限公司 | Bidirectional optimization method, device, equipment and storage medium for wind, light and load multi-stage typical scene |
CN113505939B (en) * | 2021-07-27 | 2023-01-17 | 云南电网有限责任公司电力科学研究院 | Energy system random planning method and system considering uncertainty factors |
CN117578434B (en) * | 2023-11-23 | 2024-07-23 | 南方电网能源发展研究院有限责任公司 | Power distribution network flexibility evaluation method and device considering flexible resource adjustability |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105914752A (en) * | 2016-05-25 | 2016-08-31 | 山东大学 | Pilot node selection method based on clustering by fast search and density peaks |
CN108764335A (en) * | 2018-05-28 | 2018-11-06 | 南方电网科学研究院有限责任公司 | Method and device for generating typical scene of multi-energy demand of comprehensive energy system |
CN108985318A (en) * | 2018-05-28 | 2018-12-11 | 中国地质大学(武汉) | A kind of global optimization K mean cluster method and system based on sample rate |
CN109325608A (en) * | 2018-06-01 | 2019-02-12 | 国网上海市电力公司 | Consider the distributed generation resource Optimal Configuration Method of energy storage and meter and photovoltaic randomness |
CN109508823A (en) * | 2018-11-06 | 2019-03-22 | 西安理工大学 | A kind of Distributed Generation in Distribution System planing method of method based on scene analysis |
-
2019
- 2019-03-25 CN CN201910227550.XA patent/CN110046801B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105914752A (en) * | 2016-05-25 | 2016-08-31 | 山东大学 | Pilot node selection method based on clustering by fast search and density peaks |
CN108764335A (en) * | 2018-05-28 | 2018-11-06 | 南方电网科学研究院有限责任公司 | Method and device for generating typical scene of multi-energy demand of comprehensive energy system |
CN108985318A (en) * | 2018-05-28 | 2018-12-11 | 中国地质大学(武汉) | A kind of global optimization K mean cluster method and system based on sample rate |
CN109325608A (en) * | 2018-06-01 | 2019-02-12 | 国网上海市电力公司 | Consider the distributed generation resource Optimal Configuration Method of energy storage and meter and photovoltaic randomness |
CN109508823A (en) * | 2018-11-06 | 2019-03-22 | 西安理工大学 | A kind of Distributed Generation in Distribution System planing method of method based on scene analysis |
Non-Patent Citations (1)
Title |
---|
基于密度期望和有效性指标的K-均值算法;何云斌等;《计算机工程与应用》;20131011;105-111 * |
Also Published As
Publication number | Publication date |
---|---|
CN110046801A (en) | 2019-07-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110046801B (en) | Typical scene generation method of power distribution network power system | |
CN114792156B (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
WO2023201552A1 (en) | County-wide photovoltaic prediction method based on cluster division and data enhancement | |
CN107220662A (en) | The hyperspectral image band selection method clustered based on global optimum | |
CN110852492A (en) | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance | |
CN113837426A (en) | Weather-typing-based photovoltaic power prediction method | |
CN105591407A (en) | Research method of renewable energy power plant active power prediction error correlation | |
CN115759389A (en) | Day-ahead photovoltaic power prediction method based on weather type similar day combination strategy | |
CN113270898A (en) | Distributed photovoltaic optimization configuration method based on two-stage planning | |
CN110795841B (en) | Mathematical modeling method for uncertainty of intermittent energy output | |
CN117349687A (en) | Daily load curve clustering method based on variable convolution self-encoder | |
CN108694475A (en) | Short-term time scale photovoltaic cell capable of generating power amount prediction technique based on mixed model | |
CN116995670A (en) | Photovoltaic power ultra-short-term prediction method based on multi-mode decomposition and multi-branch input | |
CN109214103B (en) | Gear reliability analysis system for generating countermeasure network based on boundary constraint | |
CN115392387B (en) | Low-voltage distributed photovoltaic power generation output prediction method | |
CN116664870A (en) | Power quality disturbance identification method based on transfer learning | |
CN114004393B (en) | Double-layer multi-time-scale operation optimization method for power distribution network-natural gas combined system | |
CN111523230B (en) | Self-adaptive clustering method for wind-light load composite typical scene | |
CN112365081A (en) | Photovoltaic power station power generation capacity prediction method and device | |
CN112633572A (en) | Day classification dual-model photovoltaic power generation combined prediction method based on prediction meteorological data | |
Yu et al. | Research on scene generation method of wind and solar active power output based on k-medoids clustering and generative adversarial networks | |
CN110990462A (en) | Time series data processing method for improving machine learning precision of ionosphere foF2 | |
CN108233373B (en) | Probability harmonic analysis method considering weather scene for distributed photovoltaic access power distribution network | |
CN113536664B (en) | Electric quantity balance calculation method, device and system for offshore wind power output | |
CN110008443A (en) | A kind of optimal quantile of the wind power probability based on EMD determines method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |