CN111144440A - Method and device for analyzing daily power load characteristics of special transformer user - Google Patents

Method and device for analyzing daily power load characteristics of special transformer user Download PDF

Info

Publication number
CN111144440A
CN111144440A CN201911191506.4A CN201911191506A CN111144440A CN 111144440 A CN111144440 A CN 111144440A CN 201911191506 A CN201911191506 A CN 201911191506A CN 111144440 A CN111144440 A CN 111144440A
Authority
CN
China
Prior art keywords
power load
daily
special transformer
transformer user
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201911191506.4A
Other languages
Chinese (zh)
Inventor
易姝慧
殷小东
刁赢龙
汪根荣
姜春阳
周峰
雷民
王斌武
刘俊杰
熊博
刘浩
刘俭
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.)
State Grid Corp of China SGCC
China Electric Power Research Institute Co Ltd CEPRI
Original Assignee
State Grid Corp of China SGCC
China Electric Power Research Institute Co Ltd CEPRI
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 State Grid Corp of China SGCC, China Electric Power Research Institute Co Ltd CEPRI filed Critical State Grid Corp of China SGCC
Priority to CN201911191506.4A priority Critical patent/CN111144440A/en
Publication of CN111144440A publication Critical patent/CN111144440A/en
Pending legal-status Critical Current

Links

Images

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
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/80—Management or planning
    • Y02P90/82—Energy audits or management systems therefor

Landscapes

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

Abstract

The invention discloses a method for analyzing daily power load characteristics of a special transformer user, which comprises the following steps: eliminating and correcting daily power load abnormal data of the special transformer user; processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user; the load characteristic pattern classification is carried out on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, the optimal clustering number of fuzzy clustering is automatically determined based on a clustering effectiveness evaluation index, the typical pattern of daily power load data characteristics of the special transformer user is obtained, and the problem that the prior art cannot be suitable for daily power load analysis of the special transformer user with high-dimensional characteristics is solved.

Description

Method and device for analyzing daily power load characteristics of special transformer user
Technical Field
The application relates to the field of analysis and evaluation of daily power load of a special transformer user, in particular to an analysis method of daily power load characteristics of the special transformer user, and also relates to an analysis device of the daily power load characteristics of the special transformer user.
Background
The special transformer users (power users using special transformers) are used as important components of the power users, the power load characteristics of the special transformer users are in a high-proportion and diversified development trend, and the special transformer users have important influence on the upgrading, operation and maintenance of power grid sources, power grids, loads and power storage. Meanwhile, with the rapid development of smart power grids and ubiquitous power internet of things, real-time monitoring and big data acquisition means of power loads of users by power grids are increasingly abundant, and in order to meet information requirements of power grid dispatching automation, marketing specialization and overhaul rationalization, daily power load characteristic analysis of special transformer users is the basis and core of various relevant researches.
The daily power load characteristic analysis comprises load characteristic index calculation, typical type division and pattern recognition. The time sequence set of the special variable user daily power load has chaotic, random and time-varying characteristics due to the complex nonlinearity, and the completely accurate load mode category is difficult to determine. Most of existing daily power load characteristic analysis methods are based on specific industries, load characteristic types are relatively fixed, analysis is carried out by methods such as a load averaging method and a K-means method, however, with the development of a power grid, for special transformer users with more industry types, unknown load characteristic types and long daily power load characteristic sequence sets, the traditional analysis method is poor in effect, and needs to be improved on the basis of the original analysis method, so that the method is suitable for daily power load analysis of the special transformer users with high-dimensional characteristics.
Disclosure of Invention
The application provides a method and a device for analyzing daily power load characteristics of a special transformer user, which are used for solving the problem that the prior art cannot be suitable for daily power load analysis of the special transformer user with high-dimensional characteristics.
The application provides an analysis method for daily power load characteristics of a special transformer user, which comprises the following steps:
eliminating and correcting daily power load abnormal data of the special transformer user;
processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user;
and carrying out load characteristic pattern classification on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, automatically determining the optimal clustering number of fuzzy clustering based on a clustering effectiveness evaluation index, and acquiring a typical pattern of daily power load data characteristics of the special transformer user.
Preferably, before the step of eliminating and correcting the daily power load abnormal data of the special transformer user, the method further comprises the following steps:
longitudinal and transverse abnormal data judgment is carried out on the sequence set of daily power load abnormal data of the special transformer user; the basis of the longitudinal judgment of the sequence set is that,
Figure BDA0002293693580000021
in the above formula, xn,iCollecting data of the ith point on the nth day for a sequence of abnormal data of a specific transformer user load, wherein 96 points are collected in one day because the load collection interval of the specific transformer user is 15min, and i is 1,2, … and 96;
Figure BDA0002293693580000022
the above formula is used to calculate the vertical variance of the sequence set;
Figure BDA0002293693580000023
judging abnormal data according to the formula, wherein epsilon is a preset threshold value and is 1.0-1.2, and if the formula is met, judging the abnormal data to be longitudinal abnormal data;
the basis for the horizontal decision of the sequence set is,
Figure BDA0002293693580000024
the above equation is used to calculate the horizontal mean of the sequence set,
σn，i＝|xn，i-x'n，i|
the above equation is used to calculate the lateral error of the sequence set,
σn,i＞λx′n,i
the above formula is used for judging abnormal data, lambda is a preset threshold value and is 0.1-0.2, and if the formula is met, the data are judged to be transverse data.
Preferably, the deletion and correction of the daily power load abnormal data of the special transformer user comprises the following steps:
based on interpolation method, the deletion and correction of longitudinal and transverse data are carried out on the sequence set of daily power load abnormal data of the special transformer user,
xn,i＝α∑xn±1,i+β∑xn,i±1
in the above formula xn±1,iIs the nearest 2 longitudinal load points, xn,i±1For the last 2 lateral load points,wherein α + β is 1.
Preferably, the processing is performed on the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user, and the processing includes:
taking the holidays as special days, and deleting the special day characteristic load sequences within the calculation time range;
the daily power load sequence set of the special transformer user is subjected to transverse normalization processing,
x′n,i＝xn,i/xn,max
in the above formula xn,maxIs the transverse maximum in the sequence set;
carrying out longitudinal mean processing on the normalized load sequence set to obtain a daily typical load sequence of the special transformer user,
Figure BDA0002293693580000031
in the formula, m is the total number of days in the longitudinal direction of a daily power load sequence set of a special transformer user, and i is 1,2, … and 96;
collecting the daily typical load sequences of the special transformer users to obtain daily typical load data sets of the special transformer users,
Figure BDA0002293693580000032
preferably, the load characteristic pattern classification is performed on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, the optimal clustering number of the fuzzy clustering is automatically determined based on the clustering effectiveness evaluation index, and the typical pattern of the daily power load data characteristic of the special transformer user is obtained, and the method comprises the following steps:
for a daily typical load data set X of a specific transformer user ═ X1，X2，…，XNAnd setting a clustering number range c belonging to [2,8 ]]Initializing a membership degree matrix U, selecting a kernel function as the weight parameter m is 2,
Figure BDA0002293693580000033
in the feature space, the objective function of the kernel fuzzy clustering algorithm is,
Figure BDA0002293693580000041
in the above formula, U is a membership matrix, V is a clustering center, and UikThe membership degree of the kth sample belonging to the ith class;
taking the minimum value of the objective function as the optimal solution, calculating a clustering center,
Figure BDA0002293693580000042
updating the membership degree matrix according to the calculated clustering center,
Figure BDA0002293693580000043
calculating to obtain an objective function according to the membership function and the clustering center, and setting the objective function value after the (k-1) th iteration as Jm k-1The objective function value after the kth iteration is Jm kIf Jm k-Jm k-1If | > epsilon, enabling k ← k +1, returning to the feature space again, carrying out load feature pattern classification on the daily typical power load sequence of the special transformer user through a target function of a kernel fuzzy clustering algorithm, and otherwise, terminating iteration to obtain a final clustering center and a membership matrix, wherein epsilon is an allowable error;
determining the optimal clustering number by taking the inter-class distance variance and the intra-class average distance sum as evaluation indexes,
Figure BDA0002293693580000044
in the above formula
Figure BDA0002293693580000045
Is the integral data set mean, xjIs the jth sample of class iN isiIs the number of samples in category i, c is the number of clusters of the dataset;
calculating kernel fuzzy clustering evaluation indexes of different clustering numbers in a set range, selecting the clustering number corresponding to the maximum value of the indexes as the optimal clustering number, wherein the corresponding clustering center is the extracted typical characteristic of the daily power load of the special transformer user.
This application provides an analytical equipment of special transformer user day power load characteristic simultaneously, its characterized in that includes:
the deletion and correction unit is used for deleting and correcting the daily power load abnormal data of the special transformer user;
the typical power load sequence acquisition unit is used for processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user;
the typical pattern obtaining unit of the power load data features classifies load feature patterns of a daily typical power load sequence of a special transformer user based on an improved kernel fuzzy clustering algorithm, automatically determines the optimal clustering number of fuzzy clustering based on a clustering effectiveness evaluation index, and obtains the typical pattern of the daily power load data features of the special transformer user.
The application provides an analysis method and device for daily power load characteristics of a special transformer user, which are used for eliminating and correcting abnormal data of the daily power load of the special transformer user; processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user; the load characteristic pattern classification is carried out on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, the optimal clustering number of fuzzy clustering is automatically determined based on a clustering effectiveness evaluation index, the typical pattern of daily power load data characteristics of the special transformer user is obtained, and the problem that the prior art cannot be suitable for daily power load analysis of the special transformer user with high-dimensional characteristics is solved.
Drawings
Fig. 1 is a schematic flowchart of an analysis method for daily power load characteristics of a specific transformer user provided by the present application;
FIG. 2 is a schematic diagram of a fuzzy clustering algorithm for daily load characteristics of a specially-changed user based on an improved kernel according to the present application;
fig. 3 is a schematic diagram of an analysis apparatus for daily power load characteristics of a specific transformer user according to the present application.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.
Fig. 1 is a schematic flow chart of an analysis method for daily power load characteristics of a specific transformer user provided by the present application, and the method provided by the present application is described in detail below with reference to fig. 1.
And step S101, eliminating and correcting daily power load abnormal data of the special transformer user.
Before the step of eliminating and correcting the abnormal data of the daily power load of the special transformer user, judging the longitudinal and transverse abnormal data of a sequence set of the abnormal data of the daily power load of the special transformer user; the basis of the longitudinal judgment of the sequence set is that,
Figure BDA0002293693580000061
Figure BDA0002293693580000062
Figure BDA0002293693580000063
x in the formula (1)n,iData of the ith point of the nth day is concentrated for a certain special variable user load sequence, 96 points are collected in one day because the special variable user load collection interval is 15min, and i is 1,2, … and 96. And (2) calculating the longitudinal variance of the sequence set, judging abnormal data according to the formula (3), wherein epsilon is a preset threshold and is 1.0-1.2, and if the formula (3) is satisfied, judging the abnormal data to be longitudinal abnormal data.
The basis for the horizontal decision of the sequence set is,
Figure BDA0002293693580000064
σn,i＝|xn,i-x′n,i| (5)
σn,i＞λx′n,i(6)
and (3) calculating a transverse average value of the sequence set by using the formula (4), calculating a transverse error of the sequence set by using the formula (5), judging abnormal data according to the formula (6), wherein lambda is a preset threshold and is 0.1-0.2, and judging the transverse and longitudinal data if the formula (6) is met.
Then, based on interpolation method, the deletion and correction of longitudinal and horizontal data are carried out on the sequence set of daily power load abnormal data of the special transformer user,
xn,i＝α∑xn±1,i+β∑xn,i±1(7)
x in formula (7)n±1,iIs the nearest 2 longitudinal load points, xn,i±1The nearest 2 lateral load points, wherein α + β is 1.
And step S102, processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user.
Taking holidays as special days, and deleting the special day characteristic load sequences in the calculation time range because the daily power load characteristics of the special days are not representative;
carrying out transverse normalization processing on the daily power load sequence set of the special transformer user:
x′n,i＝xn,i/xn,max(8)
x in formula (8)n,maxIs the transverse maximum in the sequence set.
Carrying out longitudinal mean processing on the normalized load sequence set to obtain a daily typical load sequence of the special transformer user,
Figure BDA0002293693580000071
in the formula (9), m is the total number of days in the daily power load sequence set of a specific transformer user, and i is 1,2, … and 96.
Collecting the daily typical load sequences of the special transformer users to obtain daily typical load data sets of the special transformer users,
Figure BDA0002293693580000072
and S103, carrying out load characteristic pattern classification on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, automatically determining the optimal clustering number of fuzzy clustering based on a clustering effectiveness evaluation index, and acquiring a typical pattern of daily power load data characteristics of the special transformer user.
For a daily typical load data set X of a specific transformer user ═ X1，X2，…，XNAnd setting a clustering number range c belonging to [2,8 ]]Initializing a membership degree matrix U, selecting a kernel function as the weight parameter m is 2,
Figure BDA0002293693580000073
in the feature space, the objective function of the kernel fuzzy clustering algorithm is,
Figure BDA0002293693580000074
in the formula (12), U is a membership matrix, V is a clustering center, and UikThe membership degree of the kth sample belonging to the ith class;
taking the minimum value of the objective function as the optimal solution, calculating a clustering center,
Figure BDA0002293693580000081
updating the membership degree matrix according to the calculated clustering center,
Figure BDA0002293693580000082
calculating to obtain an objective function according to the membership function and the clustering center, and setting the objective function value after the (k-1) th iteration as Jm k-1The objective function value after the kth iteration is Jm kIf Jm k-Jm k-1If | > epsilon, enabling k ← k +1, returning to the feature space again, carrying out load feature pattern classification on the daily typical power load sequence of the special transformer user through a target function of a kernel fuzzy clustering algorithm, and otherwise, terminating iteration to obtain a final clustering center and a membership matrix, wherein epsilon is an allowable error;
determining the optimal clustering number by taking the inter-class distance variance and the intra-class average distance sum as evaluation indexes,
Figure BDA0002293693580000083
in the formula (15)
Figure BDA0002293693580000084
Is the integral data set mean, xjIs the jth sample of class i, niIs the number of samples in category i, c is the number of clusters of the dataset;
calculating kernel fuzzy clustering evaluation indexes of different clustering numbers in a set range, selecting the clustering number corresponding to the maximum value of the indexes as the optimal clustering number, wherein the corresponding clustering center is the extracted typical characteristic of the daily power load of the special transformer user.
FIG. 2 is a schematic diagram of a fuzzy clustering algorithm for daily load characteristics of a specially-changed user based on an improved kernel according to the present application. As can be seen from the figure, the flow of the principle includes: data set import, which is to obtain daily power load data of a special transformer user, then judge longitudinal and transverse abnormal data of a sequence set of the daily power load abnormal data of the special transformer user, then eliminate and correct the daily power load abnormal data of the special transformer user, and then perform a series of processing on the eliminated and corrected daily power load data of the special transformer user, and the processing method comprises the following steps: the method comprises the steps of special day processing, transverse normalization processing, longitudinal mean processing and parameter initialization, then load characteristic pattern classification is carried out on a typical power load sequence of a special transformer user day based on an improved kernel fuzzy clustering algorithm, the optimal clustering number of fuzzy clustering is automatically determined based on a clustering effectiveness evaluation index, and a typical pattern of the typical power load data characteristics of the special transformer user day is obtained.
The present application also provides an analysis apparatus 300 for daily power load characteristics of a specific transformer user, as shown in fig. 3, including:
the deletion and correction unit is used for deleting and correcting the daily power load abnormal data of the special transformer user;
the typical power load sequence acquisition unit is used for processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user;
the typical pattern obtaining unit of the power load data features classifies load feature patterns of a daily typical power load sequence of a special transformer user based on an improved kernel fuzzy clustering algorithm, automatically determines the optimal clustering number of fuzzy clustering based on a clustering effectiveness evaluation index, and obtains the typical pattern of the daily power load data features of the special transformer user.
The application provides an analysis method and device for daily power load characteristics of a special transformer user, which are used for eliminating and correcting abnormal data of the daily power load of the special transformer user; processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user; the load characteristic pattern classification is carried out on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, the optimal clustering number of fuzzy clustering is automatically determined based on a clustering effectiveness evaluation index, the typical pattern of daily power load data characteristics of the special transformer user is obtained, and the problem that the prior art cannot be suitable for daily power load analysis of the special transformer user with high-dimensional characteristics is solved.
Finally, it should be noted that: although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the spirit and scope of the invention.

Claims (6)

1. A method for analyzing daily power load characteristics of a specific transformer user is characterized by comprising the following steps:
eliminating and correcting daily power load abnormal data of the special transformer user;
processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user;
and carrying out load characteristic pattern classification on the daily typical power load sequence of the special transformer user based on an improved kernel fuzzy clustering algorithm, automatically determining the optimal clustering number of fuzzy clustering based on a clustering effectiveness evaluation index, and acquiring a typical pattern of daily power load data characteristics of the special transformer user.
2. The method of claim 1, further comprising, prior to the step of pruning and correcting the data for the daily power load anomalies of the dedicated transformer customer:
longitudinal and transverse abnormal data judgment is carried out on the sequence set of daily power load abnormal data of the special transformer user; the basis of the longitudinal judgment of the sequence set is that,
Figure FDA0002293693570000011
in the above formula, xn,iThe data of the ith point at the nth day is concentrated for a sequence of abnormal data of the load of a specific variable user,
because the load collection interval of the special transformer user is 15min, 96 points are collected in one day, and i is 1,2, … and 96;
Figure FDA0002293693570000012
the above formula is used to calculate the vertical variance of the sequence set;
Figure FDA0002293693570000013
judging abnormal data according to the formula, wherein epsilon is a preset threshold value and is 1.0-1.2, and if the formula is met, judging the abnormal data to be longitudinal abnormal data;
the basis for the horizontal decision of the sequence set is,
Figure FDA0002293693570000014
the above equation is used to calculate the horizontal mean of the sequence set,
σn,i＝|xn,i-x′n,i|
the above equation is used to calculate the lateral error of the sequence set,
σn,i＞λx′n,i
the above formula is used for judging abnormal data, lambda is a preset threshold value and is 0.1-0.2, and if the formula is met, the data are judged to be transverse data.
3. The method of claim 1, wherein the eliminating and correcting of the daily power load anomaly data of the specific transformer users comprises:
based on interpolation method, the deletion and correction of longitudinal and transverse data are carried out on the sequence set of daily power load abnormal data of the special transformer user,
xn,i＝α∑xn±1,i+β∑xn,i±1
in the above formula xn±1,iIs the nearest 2 longitudinal load points, xn,i±1The nearest 2 lateral load points, wherein α + β is 1.
4. The method of claim 1, wherein processing the deleted and corrected daily power load data of the special transformer users to obtain a daily typical power load sequence of the special transformer users comprises:
taking the holidays as special days, and deleting the special day characteristic load sequences within the calculation time range;
the daily power load sequence set of the special transformer user is subjected to transverse normalization processing,
x′n,i＝xn,i/xn,max
in the above formula xn,maxIs the transverse maximum in the sequence set;
carrying out longitudinal mean processing on the normalized load sequence set to obtain a daily typical load sequence of the special transformer user,
Figure FDA0002293693570000021
in the formula, m is the total number of days in the longitudinal direction of a daily power load sequence set of a special transformer user, and i is 1,2, … and 96;
collecting the daily typical load sequences of the special transformer users to obtain daily typical load data sets of the special transformer users,
Figure FDA0002293693570000022
5. the method according to claim 1, wherein the load characteristic pattern classification is carried out on the daily typical power load sequence of the special transformer users based on an improved kernel fuzzy clustering algorithm, the optimal clustering number of the fuzzy clustering is automatically determined based on the clustering effectiveness evaluation index, and the typical pattern of the daily power load data characteristics of the special transformer users is obtained, and the method comprises the following steps:
for a daily typical load data set X of a specific transformer user ═ X1，X2，…，XNAnd setting a clustering number range c belonging to [2,8 ]]Initializing a membership degree matrix U, selecting a kernel function as the weight parameter m is 2,
Figure FDA0002293693570000031
in the feature space, the objective function of the kernel fuzzy clustering algorithm is,
Figure FDA0002293693570000032
in the above formula, U is a membership matrix, V is a clustering center, and UikThe membership degree of the kth sample belonging to the ith class;
taking the minimum value of the objective function as the optimal solution, calculating a clustering center,
Figure FDA0002293693570000033
updating the membership degree matrix according to the calculated clustering center,
Figure FDA0002293693570000034
calculating to obtain an objective function according to the membership function and the clustering center, and setting the objective function value after the (k-1) th iteration as Jm k-1The objective function value after the kth iteration is Jm kIf Jm k-Jm k-1If | > epsilon, enabling k ← k +1, returning to the feature space again, carrying out load feature pattern classification on the daily typical power load sequence of the special transformer user through a target function of a kernel fuzzy clustering algorithm, and otherwise, terminating iteration to obtain a final clustering center and a membership matrix, wherein epsilon is an allowable error;
determining the optimal clustering number by taking the inter-class distance variance and the intra-class average distance sum as evaluation indexes,
Figure FDA0002293693570000041
in the above formula
Figure FDA0002293693570000042
Is the integral data set mean, xjIs the jth sample of class i, niIs the number of samples in category i, c is the number of clusters of the dataset;
calculating kernel fuzzy clustering evaluation indexes of different clustering numbers in a set range, selecting the clustering number corresponding to the maximum value of the indexes as the optimal clustering number, wherein the corresponding clustering center is the extracted typical characteristic of the daily power load of the special transformer user.
6. An analysis device for daily power load characteristics of a specific transformer user, comprising:
the deletion and correction unit is used for deleting and correcting the daily power load abnormal data of the special transformer user;
the typical power load sequence acquisition unit is used for processing the deleted and corrected daily power load data of the special transformer user to obtain a daily typical power load sequence of the special transformer user;
the typical pattern obtaining unit of the power load data features classifies load feature patterns of a daily typical power load sequence of a special transformer user based on an improved kernel fuzzy clustering algorithm, automatically determines the optimal clustering number of fuzzy clustering based on a clustering effectiveness evaluation index, and obtains the typical pattern of the daily power load data features of the special transformer user.
CN201911191506.4A 2019-11-28 2019-11-28 Method and device for analyzing daily power load characteristics of special transformer user Pending CN111144440A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911191506.4A CN111144440A (en) 2019-11-28 2019-11-28 Method and device for analyzing daily power load characteristics of special transformer user

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911191506.4A CN111144440A (en) 2019-11-28 2019-11-28 Method and device for analyzing daily power load characteristics of special transformer user

Publications (1)

Publication Number Publication Date
CN111144440A true CN111144440A (en) 2020-05-12

Family

ID=70517304

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911191506.4A Pending CN111144440A (en) 2019-11-28 2019-11-28 Method and device for analyzing daily power load characteristics of special transformer user

Country Status (1)

Country Link
CN (1) CN111144440A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111724278A (en) * 2020-06-11 2020-09-29 国网吉林省电力有限公司 Fine classification method and system for power multi-load users
CN112308109A (en) * 2020-09-08 2021-02-02 国网浙江省电力有限公司宁波供电公司 Special transformer user health condition analysis method based on BIRCH clustering and Gaussian cloud distribution
CN112307435A (en) * 2020-10-30 2021-02-02 三峡大学 Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend
CN112862535A (en) * 2021-02-25 2021-05-28 国网河北省电力有限公司营销服务中心 Method for determining power price responsiveness of power-dedicated transformer client and terminal equipment
CN114676955A (en) * 2021-12-30 2022-06-28 国网山东省电力公司济南供电公司 Fuzzy C-means clustering feeder load property analysis method
CN117113159A (en) * 2023-10-23 2023-11-24 国网山西省电力公司营销服务中心 Deep learning-based power consumer side load classification method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104200275A (en) * 2014-06-24 2014-12-10 国家电网公司 Power utilization mode classification and control method based on user behavior characteristics
CN107247969A (en) * 2017-06-02 2017-10-13 常州工学院 The Fuzzy c-Means Clustering Algorithm of core is induced based on Gauss
CN107273911A (en) * 2017-04-24 2017-10-20 国网江西省电力公司赣州供电分公司 A kind of platform area load precise classification method based on fuzzy C-mean clustering analysis
CN107657266A (en) * 2017-08-03 2018-02-02 华北电力大学（保定） A kind of load curve clustering method based on improvement spectrum multiple manifold cluster

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104200275A (en) * 2014-06-24 2014-12-10 国家电网公司 Power utilization mode classification and control method based on user behavior characteristics
CN107273911A (en) * 2017-04-24 2017-10-20 国网江西省电力公司赣州供电分公司 A kind of platform area load precise classification method based on fuzzy C-mean clustering analysis
CN107247969A (en) * 2017-06-02 2017-10-13 常州工学院 The Fuzzy c-Means Clustering Algorithm of core is induced based on Gauss
CN107657266A (en) * 2017-08-03 2018-02-02 华北电力大学（保定） A kind of load curve clustering method based on improvement spectrum multiple manifold cluster

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111724278A (en) * 2020-06-11 2020-09-29 国网吉林省电力有限公司 Fine classification method and system for power multi-load users
CN112308109A (en) * 2020-09-08 2021-02-02 国网浙江省电力有限公司宁波供电公司 Special transformer user health condition analysis method based on BIRCH clustering and Gaussian cloud distribution
CN112307435A (en) * 2020-10-30 2021-02-02 三峡大学 Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend
CN112307435B (en) * 2020-10-30 2024-05-31 三峡大学 Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend
CN112862535A (en) * 2021-02-25 2021-05-28 国网河北省电力有限公司营销服务中心 Method for determining power price responsiveness of power-dedicated transformer client and terminal equipment
CN114676955A (en) * 2021-12-30 2022-06-28 国网山东省电力公司济南供电公司 Fuzzy C-means clustering feeder load property analysis method
CN117113159A (en) * 2023-10-23 2023-11-24 国网山西省电力公司营销服务中心 Deep learning-based power consumer side load classification method and system

Similar Documents

Publication Publication Date Title
CN111144440A (en) Method and device for analyzing daily power load characteristics of special transformer user
CN112699913B (en) Method and device for diagnosing abnormal relationship of household transformer in transformer area
CN111222290B (en) Multi-parameter feature fusion-based method for predicting residual service life of large-scale equipment
WO2018045642A1 (en) A bus bar load forecasting method
CN106055918B (en) Method for identifying and correcting load data of power system
CN109740648B (en) Method and device for identifying abnormal data of power load and computer equipment
CN111199016A (en) DTW-based improved K-means daily load curve clustering method
CN108376262B (en) Analytical model construction method for typical characteristics of wind power output
CN111539657B (en) Typical power industry load characteristic classification and synthesis method combined with user daily electricity quantity curve
CN116523320B (en) Intellectual Property Risk Intelligent Analysis Method Based on Internet Big Data
CN116522268B (en) Line loss anomaly identification method for power distribution network
CN113837488B (en) Method, system and equipment for predicting energy consumption data
CN113393028A (en) Load prediction method based on data mining technology
CN112418476A (en) Ultra-short-term power load prediction method
CN115422788B (en) Power distribution network line loss analysis management method, device, storage medium and system
CN116448161A (en) Artificial intelligence-based environment monitoring equipment fault diagnosis method
CN116187861A (en) Isotope-based water quality traceability monitoring method and related device
CN116187836A (en) Electric power material quality evaluation and spot check method
CN115358481A (en) Early warning and identification method, system and device for enterprise ex-situ migration
CN114118401A (en) Neural network-based power distribution network flow prediction method, system, device and storage medium
CN110991723A (en) Application method of artificial intelligence in seasonal load prediction
CN108827905B (en) near-infrared model online updating method based on local weighting L asso
CN114926075B (en) Machine part production scheduling method based on man-hour prediction
CN111028086A (en) Enhanced index tracking method based on clustering and LSTM network
CN116561569A (en) Industrial power load identification method based on EO feature selection and AdaBoost algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20200512

RJ01 Rejection of invention patent application after publication