CN111429045B - Energy internet clustering method based on region symmetry - Google Patents

Energy internet clustering method based on region symmetry Download PDF

Info

Publication number
CN111429045B
CN111429045B CN202010529891.5A CN202010529891A CN111429045B CN 111429045 B CN111429045 B CN 111429045B CN 202010529891 A CN202010529891 A CN 202010529891A CN 111429045 B CN111429045 B CN 111429045B
Authority
CN
China
Prior art keywords
clustering
symmetry
value
attribute
energy internet
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010529891.5A
Other languages
Chinese (zh)
Other versions
CN111429045A (en
Inventor
谢伟
明阳阳
曹军威
杨洁
黄旭东
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.)
Sichuan Huatai Electrical Co ltd
Original Assignee
Sichuan Huatai Electrical Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sichuan Huatai Electrical Co ltd filed Critical Sichuan Huatai Electrical Co ltd
Priority to CN202010529891.5A priority Critical patent/CN111429045B/en
Publication of CN111429045A publication Critical patent/CN111429045A/en
Application granted granted Critical
Publication of CN111429045B publication Critical patent/CN111429045B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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
    • 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Health & Medical Sciences (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides an energy internet clustering method based on regional symmetry, which comprises the following steps: acquiring an energy internet object to be clustered with two-dimensional attributes, and setting a clustering number range; pre-clustering the energy Internet objects to be clustered by using each clustering number in the clustering number range respectively to correspondingly obtain a plurality of clustering results; calculating the region symmetry corresponding to the plurality of clustering results, and determining the target clustering number according to the region symmetry; and taking the target clustering number and the clustering result under the target clustering number as a final clustering result. According to the invention, by combining the clustering of the regional symmetry, a clustering result with more symmetrical attribute distribution can be obtained, which is beneficial to realizing self-adaptive high-efficiency clustering and also improves the accuracy and the suitability of clustering; the typical mode of user production or consumption can be extracted in a clustering mode, the predictability of user energy production and consumption is improved, and therefore a foundation is laid for subsequent analysis and strategy generation.

Description

Energy internet clustering method based on region symmetry
Technical Field
The invention relates to the field of energy production, transmission and consumption, in particular to an object clustering method widely applied to the processes of energy internet power production, management and consumption.
Background
As the most advanced energy system at present, the energy Internet is based on ubiquitous and efficient information communication infrastructure and open and shared Internet concepts, and is combined with an Internet technology and an information physical fusion system, so that the overall coordination of source-network-load-storage can be realized, the energy utilization efficiency is maximized through energy cascade utilization and multi-energy complementation, the energy production and consumption cost is greatly reduced, the environmental protection is promoted, the exhaust emission is reduced, and a solid energy guarantee is provided for the social harmonious development and the human happy life.
According to the development trend of the energy internet, the large-scale access of distributed renewable energy sources is a typical characteristic and application of the energy internet. One significant drawback of distributed renewable energy power generation is the difficulty in ensuring the smoothness and continuity of power generation; the smooth and efficient operation of the energy internet, which is mainly characterized by wide access to distributed energy, is based on the precise understanding of the characteristics of power production and consumption.
In this regard, the inventors showed that: accurate understanding of the energy internet is premised on accurate clustering or classification of energy producing nodes and consuming nodes.
Disclosure of Invention
The present invention aims to address at least one of the above-mentioned deficiencies of the prior art. For example, it is an object of the present invention to provide a method for more accurate and sophisticated clustering of energy internet objects.
In order to achieve the above object, the present invention provides an energy internet clustering method based on region symmetry, which includes the following steps: acquiring an energy internet object to be clustered with two-dimensional attributes, and setting a clustering number range; pre-clustering the energy Internet objects to be clustered by using each clustering number in the clustering number range respectively to correspondingly obtain a plurality of clustering results; calculating the region symmetry corresponding to the plurality of clustering results, and determining the target clustering number according to the region symmetry; and taking the target clustering number and the clustering result under the target clustering number as a final clustering result.
In an exemplary embodiment of the present invention, the step of calculating the region symmetry corresponding to the plurality of clustering results may further include: for each clustering result in the plurality of clustering results, taking the normalized distance between the average value and the median of one-dimensional attributes of the clustering result as a first-dimensional attribute symmetry index, taking the normalized distance between the average value and the median of the other-dimensional attributes of the clustering result as a second-dimensional attribute symmetry index, jointly considering the first-dimensional attribute symmetry index and the second-dimensional symmetry index to determine an overall symmetry index, and expressing the region symmetry of the clustering result by the overall symmetry index.
In an exemplary embodiment of the present invention, the step of determining the target cluster number according to the region symmetry may use a cluster number of a clustering result having an optimal overall symmetry index as the target cluster number; or the overall symmetry index and other judgment attributes except the region symmetry can be comprehensively considered in a form of weighted sum to determine the target cluster number.
Compared with the prior art, the beneficial effects of the invention comprise one or more of the following:
the method has the advantages that the method can roughly have the characteristic of cycle repetition aiming at the mode and the operation trend followed by the energy production or the user power consumption of the energy internet, the typical mode of the user production or the user consumption is extracted in a clustering mode, the predictability of the user energy production and consumption is improved, and therefore a foundation is laid for the subsequent analysis and strategy generation;
by combining the clustering of the regional symmetry, a clustering result with more symmetrical attribute distribution can be obtained, the self-adaptive high-efficiency clustering is favorably realized, and the accuracy and the suitability of the clustering are improved, so that the service quality of customers can be further improved, the product marketing efficiency is improved, the system robustness and stable operation are ensured, and the superiority of the energy Internet is fully embodied.
Drawings
Fig. 1 is a flowchart illustrating an exemplary embodiment of an energy internet clustering method based on region symmetry according to the present invention.
Fig. 2 is a schematic diagram illustrating calculation of a first-dimension attribute symmetry indicator in an exemplary embodiment of the energy internet clustering method based on region symmetry according to the present invention.
Detailed Description
Hereinafter, the energy internet clustering method based on region symmetry according to the present invention will be described in detail with reference to the accompanying drawings and exemplary embodiments.
Example 1
Fig. 1 is a flowchart illustrating an exemplary embodiment of an energy internet clustering method based on region symmetry according to the present invention.
In an exemplary embodiment of the present invention, as shown in fig. 1, the energy internet clustering method based on the region symmetry may include the steps of:
step 1: and acquiring the energy internet object to be clustered, and setting a clustering number range.
Specifically, the energy internet objects to be clustered with two-dimensional attributes are obtained. The energy internet objects to be clustered can be energy production nodes and/or consumption nodes, management nodes and the like of the energy internet. For the energy Internet objects to be clustered, which have attribute dimensionality exceeding two dimensions, dimensionality reduction can be carried out in a mode of carrying out principal component analysis on the energy Internet objects to be clustered so as to reduce the energy Internet objects to have two-dimensional attributes. For example, the method of this embodiment may assume that the clustering objects are located in a two-dimensional plane, i.e., clustering is performed based on the two-dimensional attribute values of the objects. For the object with multi-dimensional attributes, the dimensionality can be reduced by a principal component analysis method before clustering processing, so that the method of the embodiment can be easily expanded to the multi-dimensional attribute clustering processing.
Then, the clustering number range is set for the object to be clustered. For example, the number of clusters may be set to 2 to N, N belonging to a natural number greater than 2. The setting of the clustering number range can be set manually or according to the estimation of the actual performance requirement of the related system.
Step 2: and pre-clustering to obtain a plurality of clustering results.
And (4) pre-clustering the energy internet objects to be clustered respectively according to each clustering number in the clustering number range in the step (1), so as to obtain a plurality of corresponding clustering results. For example, the pre-clustering process may include an initial position selection process, a center position calculation process, and a clustering range calculation process, and when the iteration converges or reaches a predetermined number, a clustering result is obtained. For example, the pre-cluster classes may be k-means (k-means) clusters, fuzzy (fuzzy) clusters, and the like.
And step 3: the target cluster number is determined in consideration of the area symmetry.
For the plurality of clustering results obtained in step 2, each clustering result corresponds to a known clustering number and a region symmetry (e.g., an overall symmetry index) to be determined. For each clustering result, region symmetry can be determined by calculation. Specifically, for each clustering result, a normalized distance between an average value of the one-dimensional attributes of the clustering result and a median value of the one-dimensional attributes may be used as a first-dimensional attribute symmetry index; taking the normalized distance between the average value of the other dimension attribute of the clustering result and the median of the other dimension attribute as a second dimension attribute symmetry index; and jointly considering the first dimension attribute symmetry index and the second dimension symmetry index by selecting a maximum value, a sum, a product or a ratio and the like, thereby determining an overall symmetry index, and expressing the regional symmetry of the clustering result by the overall symmetry index.
For example, for each clustering result, the overall symmetry index can be obtained by:
the first dimension attribute symmetry index and the second dimension attribute symmetry index are obtained by the following formulas (1) and (2), respectively.
Figure 269807DEST_PATH_IMAGE001
（1）
Wherein symvalue(x)Is a first dimension attribute symmetry index, x is a first dimension attribute, meanvalue(x)Is the average, mean, of the first dimension attributesvalue(x)The median of the first dimension attribute, range (x) function is the overall length of the value range of the first dimension attribute, and count (x) is the number of values of the first dimension attribute.
Figure 96948DEST_PATH_IMAGE002
（2）
Wherein symvalue(y)Is a second dimension attribute symmetry index, y is a second dimension attribute, meanvalue(y)Is the mean, of the second dimension attributesvalue(y)The range (y) function is the overall length of the value range of the second dimension attribute, and the count (y) is the number of values of the second dimension attribute.
Subsequently, the overall symmetry index can be obtained by equation (3), and the smaller the overall symmetry index is, the better the regional symmetry is. That is, the number of clusters corresponding to the clustering result having the smallest overall symmetry index may be used as the target cluster number. Wherein, formula (3) can be any one selected from the following three formulas:
Figure 38360DEST_PATH_IMAGE003
Figure 479705DEST_PATH_IMAGE004
（3）
Figure 404936DEST_PATH_IMAGE005
wherein,
Figure 629244DEST_PATH_IMAGE006
is an index of overall symmetry.
The overall symmetry index can also be obtained by equation (4), and the closer the overall symmetry index is to 1, the better the regional symmetry. That is, in
Figure 425162DEST_PATH_IMAGE007
If not, the cluster number corresponding to the cluster result with the overall symmetry index closest to 1 may be used as the target cluster number.
Figure 53720DEST_PATH_IMAGE008
（4）
Wherein,
Figure 528564DEST_PATH_IMAGE009
is an index of overall symmetry.
And then, selecting a clustering result corresponding to the optimal overall symmetry index from a set consisting of the overall symmetry indexes corresponding to each clustering result, and taking the clustering number of the clustering result as a target clustering number.
And 4, step 4: and determining a final clustering result according to the target clustering number.
And 3, the target clustering number determined in the step 3 and the clustering result under the target clustering number are used as final clustering results, so that a clustering result with better clustering performance is obtained under the condition of considering the area symmetry aiming at the energy internet objects waiting for clustering, such as energy production nodes, consumption nodes and/or management nodes, of the energy internet, further a foundation is laid for subsequent analysis and strategy generation, the customer service quality can be further improved, the product marketing efficiency is improved, the system robustness and stable operation are ensured, and the superiority of the energy internet is fully embodied.
Example 2
In another exemplary embodiment of the present invention, the energy internet clustering method based on the region symmetry may adopt steps 1 and 2 of the above exemplary embodiment to obtain a plurality of clustering results; and then, energy internet clustering based on region symmetry is realized through the following steps 3 'and 4'.
Step 3': and comprehensively considering the area symmetry and other judgment attributes to determine the target cluster number.
For the plurality of clustering results obtained in step 2, each clustering result corresponds to a known clustering number and a region symmetry (e.g., an overall symmetry index) to be determined. For each clustering result, region symmetry can be determined by calculation. Specifically, for each clustering result, a normalized distance between an average value of the one-dimensional attributes of the clustering result and a median value of the one-dimensional attributes may be used as a first-dimensional attribute symmetry index; taking the normalized distance between the average value of the other dimension attribute of the clustering result and the median of the other dimension attribute as a second dimension attribute symmetry index; and jointly considering the first dimension attribute symmetry index and the second dimension symmetry index by selecting a maximum value, a sum, a product or a ratio and the like, thereby determining an overall symmetry index, and expressing the regional symmetry of the clustering result by the overall symmetry index.
For example, for each clustering result, the overall symmetry index can be obtained by:
the first dimension attribute symmetry index and the second dimension attribute symmetry index are obtained by the above equations (1) and (2), respectively. Subsequently, the overall symmetry index can be obtained by the above formula (3) or the above formula (4). For formula (3), the smaller the overall symmetry index, the better the regional symmetry; for formula (4), in
Figure 290983DEST_PATH_IMAGE010
If the overall symmetry index is not equal to zero, the closer the overall symmetry index is to 1, the better the regional symmetry.
Then, the overall symmetry index and other judgment attributes except the region symmetry are comprehensively considered in a weighted sum form to obtain a weighted judgment index of each clustering result. The weight coefficient may be set based on the symmetry index and the degree of importance of the adopted other judgment attributes except for the region symmetry.
And then, selecting a clustering result corresponding to the optimal weighting judgment index from a set formed by the weighting judgment indexes corresponding to each clustering result, and taking the clustering number of the clustering result as a target clustering number.
Step 4': and determining a final clustering result according to the target clustering number.
And 3 ', taking the target clustering number determined in the step 3' and the clustering result under the target clustering number as a final clustering result, so that aiming at the energy Internet objects waiting for clustering, such as energy production nodes, consumption nodes and/or management nodes and the like, of the energy Internet, under the condition of comprehensively considering the area symmetry and other judgment attributes except the area symmetry, a clustering result with better clustering performance is obtained, a foundation is laid for subsequent analysis and strategy generation, the service quality of a client can be further improved, the product marketing efficiency is improved, the system robustness and stable operation are ensured, and the superiority of the energy Internet is fully embodied.
Example 3
In the present exemplary embodiment, the energy internet clustering method based on the region symmetry assumes that the clustering objects are located in one two-dimensional plane, i.e., clustering is performed based on the two-dimensional attribute values of the objects. For the object with multi-dimensional attributes, the dimensionality can be reduced by a principal component analysis method before clustering processing, so that the method of the embodiment can be easily expanded to the multi-dimensional attribute clustering processing.
Then, a cluster number range, such as 2 to N, may be set based on the system actual performance requirement estimate.
Next, each cluster number is pre-clustered by a k-means clustering algorithm. The k-means clustering process can comprise the processes of initial position selection, center position calculation, clustering range calculation iteration and the like, and when iteration converges or reaches a certain number of times, a final clustering result is obtained.
Subsequently, the region symmetry of each cluster number clustering result is calculated by the above-described formulas (1), (2) and (3) or by the formulas (1), (2) and (4), and the optimal cluster number is selected according to the symmetry. The associated schematic may be as shown in fig. 2. Fig. 2 shows a schematic diagram of the calculation of the first-dimension attribute symmetry indicator, which may also be referred to as a calculation of a distributed average indicator.
And then, taking the cluster number with the best symmetry and the best clustering result under the cluster number as a final clustering result.
Example 4
In another exemplary embodiment of the present invention, the energy internet clustering method based on the region symmetry may adopt steps 1 to 4 of the above embodiment 1, except that:
in step 2, for each cluster number, performing multiple (e.g., more than 3) random pre-clustering on the energy internet object to be clustered, for example, selecting an initial position of each pre-clustering by using a random sampling manner, so as to obtain a group (e.g., more than 3 per group) of clustering results for each cluster number;
in step 3, overall symmetry indexes of all the clustering results are obtained from the clustering results of each group corresponding to each clustering number according to the formula (1), the formula (2) and the formula (3) or the formula (1), the formula (2) and the formula (4), and the overall symmetry indexes are used for correspondingly representing the region symmetry of each clustering result. And then, taking the cluster number corresponding to the cluster result with the best region symmetry in all the cluster results as the target cluster number.
In step 4, the number of target clusters and the clustering result with the best region symmetry among all the clustering results are taken as the final clustering result.
In summary, the energy internet clustering method based on the region symmetry has the advantages that:
the method has the characteristics that the mode and the operation trend which are followed by the energy production or the user power consumption of the energy Internet are approximately repeated periodically, the typical mode of the user production or consumption is extracted in a clustering mode, the predictability of the user energy production and consumption is improved, and therefore a foundation is laid for the subsequent analysis and strategy generation;
by combining the clustering of the regional symmetry, a clustering result with more symmetrical attribute distribution can be obtained, the self-adaptive high-efficiency clustering is favorably realized, and the accuracy and the suitability of the clustering are improved, so that the service quality of customers can be further improved, the product marketing efficiency is improved, the system robustness and stable operation are ensured, and the superiority of the energy Internet is fully embodied.
In other words, the method can effectively identify the user consumption mode and the energy production mode through more accurate and suitable clustering, and excavate the potential characteristics of the user consumption mode and the energy production mode, thereby providing basic conditions for subsequent analysis and strategy formulation.
In addition, the method can be applied to providing client service and marketing activities in a targeted manner based on the user characteristic clustering result, and can also be applied to relevant strategies and applications which need to depend on a clustering algorithm in the energy Internet operation process, such as load modeling, load prediction, state evaluation, power quality monitoring and control, demand side management and response, distributed energy access, multi-energy scheduling planning, automatic fault positioning, system safety and situation perception and the like.
Although the present invention has been described above in connection with the exemplary embodiments and the accompanying drawings, it will be apparent to those of ordinary skill in the art that various modifications may be made to the above-described embodiments without departing from the spirit and scope of the claims.

Claims (7)

1. An energy internet clustering method based on region symmetry is characterized by comprising the following steps:
the method comprises the steps of obtaining an energy internet object to be clustered with two-dimensional attributes, and setting a clustering number range, wherein the energy internet object to be clustered is one or more of energy production nodes, consumption nodes and management nodes of an energy internet;
pre-clustering the energy Internet objects to be clustered by using each clustering number in the clustering number range respectively to correspondingly obtain a plurality of clustering results;
calculating the region symmetry corresponding to the plurality of clustering results, and determining the target clustering number according to the region symmetry;
taking the target clustering number and the clustering result under the target clustering number as final clustering results,
wherein, the step of calculating the region symmetry corresponding to the plurality of clustering results comprises: for each clustering result in the plurality of clustering results, taking the normalized distance between the average value and the median of one-dimensional attributes of the clustering result as a first-dimensional attribute symmetry index, taking the normalized distance between the average value and the median of the other-dimensional attributes of the clustering result as a second-dimensional attribute symmetry index, jointly considering the first-dimensional attribute symmetry index and the second-dimensional symmetry index to determine an overall symmetry index, and expressing the region symmetry of the clustering result by the overall symmetry index.
2. The energy internet clustering method based on region symmetry as claimed in claim 1, wherein the step of determining the target cluster number according to region symmetry takes the cluster number of the clustering result with the optimal overall symmetry index as the target cluster number.
3. The energy internet clustering method based on region symmetry as claimed in claim 1, wherein the step of determining the target cluster number according to region symmetry considers the overall symmetry index and other judgment attributes except for region symmetry in a weighted sum form to determine the target cluster number.
4. The energy internet clustering method based on the regional symmetry as claimed in any one of claims 1 to 3, wherein the first dimension attribute symmetry index and the second dimension attribute symmetry index are obtained by the following formulas 1 and 2, respectively, where formula 1 is:
symvalue(x)＝|meanvalue(x)-medianvalue(x)|/abs(range(x)/count(x))，
symvalue(x)is a first dimension attribute symmetry index, x is a first dimension attribute, meanvalue(x)Is the average, mean, of the first dimension attributesvalue(x)The median of the first dimension attribute, range (x) function is the total length of the value range of the first dimension attribute, and count (x) is the number of the values of the first dimension attribute;
the formula 2 is:
symvalue(y)＝|meanvalue(y)-medianvalue(y)|/abs(range(y)/count(y))，
symvalue(y)is a second dimension attribute symmetry indicator, y is a second dimension attribute, meanvalue(y)Is the mean, of the second dimension attributesvalue(y)The range (y) function is the overall length of the value range of the second dimension attribute, and the count (y) is the number of values of the second dimension attribute.
5. The energy internet clustering method based on regional symmetry as claimed in claim 4, wherein the overall symmetry index is obtained by equation 3, and the smaller the overall symmetry index is, the better the regional symmetry is, wherein equation 3 is selected from one of the following three equations:
symattribute＝max(symvalue(x)，symvalue(y))
symattribute＝symvalue(x)+symvalue(y)
symattribute＝symvalue(x)*symvalue(y)，
wherein symattributeIs an index of overall symmetry.
6. The energy internet clustering method based on regional symmetry as claimed in claim 4, wherein the overall symmetry index is obtained by equation 4, and the closer the overall symmetry index is to 1, the better the regional symmetry is, wherein equation 4 is:
symattribute＝symvalue(x)/symvalue(y)，
wherein symattributeIs an index of overall symmetry.
7. The energy internet clustering method based on region symmetry as claimed in claim 1, wherein the step of obtaining the energy internet objects to be clustered with two-dimensional attributes comprises: and carrying out principal component analysis on the to-be-clustered energy Internet object with the attribute not less than three dimensions so as to reduce the to-be-clustered energy Internet object to have the attribute of two dimensions.
CN202010529891.5A 2020-06-11 2020-06-11 Energy internet clustering method based on region symmetry Active CN111429045B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010529891.5A CN111429045B (en) 2020-06-11 2020-06-11 Energy internet clustering method based on region symmetry

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010529891.5A CN111429045B (en) 2020-06-11 2020-06-11 Energy internet clustering method based on region symmetry

Publications (2)

Publication Number Publication Date
CN111429045A CN111429045A (en) 2020-07-17
CN111429045B true CN111429045B (en) 2020-10-02

Family

ID=71551356

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010529891.5A Active CN111429045B (en) 2020-06-11 2020-06-11 Energy internet clustering method based on region symmetry

Country Status (1)

Country Link
CN (1) CN111429045B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107169511A (en) * 2017-04-27 2017-09-15 华南理工大学 Clustering ensemble method based on mixing clustering ensemble selection strategy
CN107767293A (en) * 2017-09-20 2018-03-06 国网浙江省电力公司电力科学研究院 A kind of larger power user divided method based on improvement AP and K means clusters
CN109117872A (en) * 2018-07-24 2019-01-01 贵州电网有限责任公司信息中心 A kind of user power utilization behavior analysis method based on automatic Optimal Clustering

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107169511A (en) * 2017-04-27 2017-09-15 华南理工大学 Clustering ensemble method based on mixing clustering ensemble selection strategy
CN107767293A (en) * 2017-09-20 2018-03-06 国网浙江省电力公司电力科学研究院 A kind of larger power user divided method based on improvement AP and K means clusters
CN109117872A (en) * 2018-07-24 2019-01-01 贵州电网有限责任公司信息中心 A kind of user power utilization behavior analysis method based on automatic Optimal Clustering

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
一种新聚类评价指标;谢娟英 等;《陕西师范大学学报（自然科学版）》;20151110(第6期);第1-8页 *
高效率的K-means 最佳聚类数确定算法;王勇 等;《计算机应用》;20140510(第5期);第1331-1335页 *

Also Published As

Publication number Publication date
CN111429045A (en) 2020-07-17

Similar Documents

Publication Publication Date Title
CN117078048B (en) Digital twinning-based intelligent city resource management method and system
CN111091247A (en) Power load prediction method and device based on deep neural network model fusion
CN114861788A (en) Load abnormity detection method and system based on DBSCAN clustering
CN112308341A (en) Power data processing method and device
CN111487873B (en) Energy internet energy dispersion cooperative control method
CN108829846B (en) Service recommendation platform data clustering optimization system and method based on user characteristics
CN114785824A (en) Intelligent Internet of things big data transmission method and system
CN111429045B (en) Energy internet clustering method based on region symmetry
CN117349687A (en) Daily load curve clustering method based on variable convolution self-encoder
CN115795368B (en) Enterprise internal training data processing method and system based on artificial intelligence
CN112182026A (en) Power grid section data retrieval method considering manifold sorting algorithm
CN112149052A (en) Daily load curve clustering method based on PLR-DTW
CN113763710B (en) Short-term traffic flow prediction method based on nonlinear adaptive system
CN114268625B (en) Feature selection method, device, equipment and storage medium
CN109685101B (en) Multi-dimensional data self-adaptive acquisition method and system
Li et al. High resolution radar data fusion based on clustering algorithm
CN115545107B (en) Cloud computing method and system based on mass power data
Lin et al. Load Data Analysis Based on Timestamp-Based Self-Adaptive Evolutionary Clustering
CN111127184A (en) Distributed combined credit evaluation method
Xiaoyun et al. PGMCLU: A novel parallel grid-based clustering algorithm for multi-density datasets
Gao et al. Daily power load curves analysis based on grey wolf optimization clustering algorithm
Peng et al. Community Detection Algorithm for Heterogeneous Networks Based on Central Node and Seed Community Extension
Yan Data mining method of false transaction in webcast platform based on Cluster Learning
CN114564623B (en) Knowledge graph embedding model based on entity and relation aggregation graph
CN118626892A (en) Power load curve clustering method, device, equipment and storage medium

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