CN116341788A - Accurate electric power fingerprint management method for power distribution network line loss analysis - Google Patents
Accurate electric power fingerprint management method for power distribution network line loss analysis Download PDFInfo
- Publication number
- CN116341788A CN116341788A CN202211396840.5A CN202211396840A CN116341788A CN 116341788 A CN116341788 A CN 116341788A CN 202211396840 A CN202211396840 A CN 202211396840A CN 116341788 A CN116341788 A CN 116341788A
- Authority
- CN
- China
- Prior art keywords
- line loss
- data
- fingerprint
- user
- analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 78
- 238000007726 management method Methods 0.000 title claims abstract description 12
- 230000002159 abnormal effect Effects 0.000 claims abstract description 79
- 230000005611 electricity Effects 0.000 claims abstract description 62
- 238000000034 method Methods 0.000 claims abstract description 46
- 230000008859 change Effects 0.000 claims abstract description 41
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000010219 correlation analysis Methods 0.000 claims description 18
- 238000011282 treatment Methods 0.000 claims description 17
- 238000012216 screening Methods 0.000 claims description 16
- 238000001228 spectrum Methods 0.000 claims description 15
- 230000005856 abnormality Effects 0.000 claims description 14
- 238000007781 pre-processing Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000008030 elimination Effects 0.000 claims description 5
- 238000003379 elimination reaction Methods 0.000 claims description 5
- 230000006399 behavior Effects 0.000 claims description 4
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012098 association analyses Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 description 4
- 238000011269 treatment regimen Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000002596 correlated effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012352 Spearman correlation analysis Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an accurate electric fingerprint management method for analyzing the line loss of a power distribution network, which comprises the following steps: s1, data acquisition; the acquired data comprise operation information data, and the required data are acquired according to a data application flow; s2, data processing: when the correlation between the line loss rate and the electricity consumption of the hanging user is analyzed, the sudden change data is processed; carrying out modal decomposition on the data signals of the line loss and the electric quantity, and taking out user data in advance; filling data of the missing electricity consumption data by using a Lagrangian interpolation method; s3, extracting and analyzing 'electric fingerprint'; s4, modeling analysis; s5, grading the suspicion degree of the abnormal user as a foothold point, and establishing a method strategy model.
Description
Technical Field
The invention belongs to the technical field of intelligent management of power grid line loss analysis, and particularly relates to an accurate power fingerprint management method for power distribution network line loss analysis.
Background
The power grid loss mostly occurs in the medium-voltage and low-voltage power distribution network, and the power consumption can be reduced through the management of the power distribution network line and the transformer area line loss, and various power consumption abnormal behaviors such as abnormal metering device, unbalanced three-phase load, electricity larceny and the like can be timely found. Therefore, the line loss management is directly related to the economic benefit of enterprises and implementation of national energy saving policies, and how to reduce the line loss has become an important work of power supply enterprises and an important research object of electric power workers.
The current intelligent technical means for line loss analysis in use by power grid companies mainly realizes line loss diagnosis through functional modules such as a synchronous line loss system, a transformer area physical examination and loss reduction closed loop management of an electricity consumption acquisition system and the like, locates the problem of abnormal line loss, but still has the limitations, lacks a set of relational model schemes for focusing the line loss and the electric quantity from a big data angle, and digs a line loss accurate treatment method for identifying the problem of line loss and electric quantity data, thereby improving the working efficiency of line loss treatment.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides the accurate electric fingerprint treatment method for the analysis of the line loss of the power distribution network, which is used for realizing the positioning of abnormal points of the line loss and the suspected diagnosis and early warning of the occurrence of the possible abnormality, and has the advantages of reducing the manual workload of the line loss treatment and improving the work efficiency of the line loss treatment.
In order to achieve the above purpose, the invention adopts the following technical scheme: a power fingerprint accurate management method for power distribution network line loss analysis comprises the following steps:
s1, data acquisition; the acquired data comprise operation information data, and the required data are acquired according to a data application flow;
s2, data processing: when the correlation between the line loss rate and the electricity consumption of the hanging user is analyzed, the sudden change data is processed; carrying out modal decomposition on the data signals of the line loss and the electric quantity, and taking out user data in advance; filling data of the missing electricity consumption data by using a Lagrangian interpolation method;
s3, extracting and analyzing 'electric fingerprint':
3.1 The fingerprint identification is carried into the power curve data generation, continuous data of all the level objects comprising the platform area are converted into graph textures, the graph textures are fitted, a true graph of the line loss change state of the power distribution network, namely a line loss 'power fingerprint', is formed, and a line loss fingerprint and an electric quantity fingerprint are respectively constructed;
3.2 Sampling and selecting the electric fingerprint of the line loss abnormal area, carrying out characteristic analysis on the extracted fingerprint from the descriptive statistical view angles of the centralized trend, the discrete degree and the distribution characteristics, and intuitively presenting the influence relation between the electric quantity and the line loss;
s4, modeling analysis:
4.1 Fingerprint correlation analysis model): positioning suspected users affecting abnormal fluctuation of the line loss of the platform area through fluctuation correlation analysis of the line loss fingerprint of the platform area and the electric quantity fingerprint of the hanging user;
4.2 Fingerprint fluctuation amount analysis model): the suspected users affecting the line loss abnormal fluctuation of the platform area are positioned through the change inflection point analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user;
4.3 A fingerprint signal empirical mode decomposition model): the suspected users affecting the abnormal fluctuation of the line loss of the platform area are positioned through the time-frequency signal analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user;
s5, grading the suspicion degree of the abnormal user as a foothold point, and establishing a method strategy model.
Further, in step S1, the operation information data includes archive information of the 10 kv distribution network line, the transformer area, the high-low voltage subscriber, daily frozen power, and line loss;
in the step S2, user data with average daily power consumption of 0 and average daily power consumption of less than 1 kW.h are taken out in advance;
in step S3, the continuous data of each hierarchical object includes data of a line, a station area, and a user.
Still further, the modeling process of step 4.1): (1) the data preprocessing comprises the steps of duplicate removal of user electricity consumption data, elimination of zero user of each daily electricity consumption and deletion of useless fields; (2) extracting a fingerprint spectrum, including a fingerprint curve spectrum of the line loss rate of the area and the electricity consumption of the hanging user; (3) calculating a correlation coefficient, and quantitatively analyzing the correlation degree of each user electric quantity fingerprint and the station area line loss rate fingerprint by using a Pearson equi-phase relation method; (4) and locating strongly-correlated users, and screening and locating abnormal suspected users with strong correlation between electric quantity fingerprints and line loss fingerprints by using correlation coefficient results of the electric quantity of each user and the line loss of the platform region.
Still further, the modeling process of step 4.2): (1) the data preprocessing comprises the steps of duplicate removal of the user electricity consumption data, invalid data removal and data wide table splicing; (2) locking the date of a change inflection point, and selecting the mutation point with the most obvious change of the line loss curve of the station area; (3) calculating an inflection point K value, and calculating a K value of a change speed of the line loss electric quantity of the station area caused by each user electric quantity change speed by using a defined model rule of K value = user electric quantity change quantity/station area loss electric quantity change quantity; (4) and (3) positioning abnormal users, and screening and positioning abnormal suspected users with large relative deviation degree to the line loss abnormality of the transformer area by using the K value result of the electricity consumption of each user and the line loss of the transformer area.
Still further, the modeling process of step 4.3): (1) the data preprocessing comprises the steps of user electricity consumption data de-duplication, lagrange interpolation, daily electricity consumption calculation and wide-scale splicing of the missing data; (2) calculating a correlation coefficient r of the user electricity consumption and the line loss rate of the station area, and sorting results based on the daily electricity consumption and the correlation coefficient r; (3) extracting 5% of users before sorting as primary screening users, carrying out signal modal decomposition on the user electricity consumption and the station area loss electricity consumption through an EMD algorithm, respectively extracting high frequency components (IMFs), and completing signal map fitting; (4) and marking abnormal users, and screening and positioning suspected line loss users by utilizing the empirical mode decomposition result of the fingerprint signals.
Still further, the step S5 includes:
and (3) constructing a strategy model: the method comprises the steps that a suspected user range which is strongly related to line loss abnormality is respectively positioned by using a fingerprint spectrum correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model and an abnormal user identification algorithm model of power fingerprint of three power distribution network line loss, and gradient label division is carried out for the suspected degree of an abnormal user; aiming at the dividing result of the user under each analysis model, carrying out analysis on the electric quantity change trend, the analysis on abnormal events and the analysis dimension of the equipment operation state by combining the operation data of the user on-line loss normal and abnormal time periods so as to assist in verifying specific abnormal electricity consumption behaviors and occurrence time periods;
the application flow comprises the following steps: (1) importing the file information of the 10 KV distribution network line, the station area, the high-low voltage users and the related data of the electric quantity and the line loss into a three-large line loss abnormal user identification algorithm model; (2) outputting a model result: based on the three models, respectively outputting suspected user ranges which are strongly related to line loss abnormality; (3) matching and fusing three model result labels, finishing comprehensive grading of suspicion degrees of abnormal users, and differentially dividing the suspicion users; (4) and classifying and executing the suspected users with different classifications until the damage reduction target is reached.
Further, in the step S4, modeling analysis is to perform service implementation logic and early-stage data processing for abnormal user positioning based on line loss and electric quantity association analysis, frame a machine learning and deep learning big data related algorithm range, perform quality comparison analysis on effects, efficiency and stability among algorithms through Python training, and finally complete training and construction of three big line loss abnormal user identification algorithm models based on electric quantity and line loss pattern fingerprints.
The invention has the technical effects that: the accurate power fingerprint treatment method for the power distribution network line loss analysis, disclosed by the invention, realizes line loss abnormal point positioning, performs suspicion diagnosis and early warning on the occurrence of possible abnormality, and has the advantages of reducing the manual workload of line loss treatment and improving the working efficiency of line loss treatment.
Drawings
FIG. 1 is a plot of a daily line loss rate fingerprint of a station area A in an embodiment of the present invention;
FIG. 2 is a graph of a daily power supply fingerprint for a bay area A according to an embodiment of the present invention;
fig. 3 is a chart illustrating a correlation analysis between the power and line loss of a subscriber a hanging under a transformer area according to an embodiment of the present invention;
FIG. 4 is a plot of a line loss change inflection point fingerprint for region A in accordance with an embodiment of the present invention;
FIG. 5 is a diagram of a high frequency component of a user b fingerprint signal hanging under a transformer area in an embodiment of the present invention;
FIG. 6 is a flowchart of an application of a method for precisely managing "power fingerprints" of power distribution network line loss analysis in an embodiment of the present invention;
FIG. 7 is a schematic diagram of a correlation modeling algorithm in accordance with an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A power fingerprint accurate management method for power distribution network line loss analysis comprises the following steps:
s1, data acquisition: the acquired data comprise operation information data such as file information of 10 kilovolt distribution network lines, transformer areas, high and low voltage users, daily frozen electric quantity, line loss and the like, and the required data are acquired according to a data application flow;
s2, data processing: when the correlation between the line loss rate and the electricity consumption of the hanging user is analyzed, the method has an analysis value for sudden change data such as sudden increase, sudden decrease and the like, and abnormal value processing is not needed; carrying out modal decomposition on data signals of line loss and electric quantity, and taking out user data such as zero electric quantity (daily average electric quantity is 0) and small electric quantity (daily average electric quantity is less than 1 kW.h) in advance so as to avoid influencing algorithm accuracy and applicability; filling data of the missing electricity consumption data by using a Lagrangian interpolation method;
s3, extracting and analyzing 'electric fingerprint':
3.1 The fingerprint identification is carried into the power curve data generation, continuous data of all levels of objects such as lines, areas, users and the like are converted into graph textures, the graph textures are fitted, a true graph of the power distribution network line loss change state, namely a line loss 'power fingerprint', is formed, and a line loss fingerprint and an electric quantity fingerprint are respectively constructed;
3.2 Sampling and selecting the electric fingerprint of the line loss abnormal area, and performing feature analysis on the extracted fingerprint from descriptive statistical view angles such as concentrated trend, discrete degree, distribution feature and the like to intuitively present the influence relation between electric quantity and line loss; the analysis method is also suitable for the analysis of other 10 KV distribution network lines and station areas;
s4, modeling analysis:
4.1 Fingerprint correlation analysis model): and positioning suspected users affecting abnormal fluctuation of the line loss of the platform area through fluctuation correlation analysis of the line loss fingerprint of the platform area and the electric quantity fingerprint of the hanging user.
Modeling: (1) the data preprocessing comprises the steps of duplicate removal of user electricity consumption data, elimination of zero user of each daily electricity consumption, deletion of useless fields and the like; (2) extracting a fingerprint spectrum, including a fingerprint curve spectrum of the line loss rate of the area and the electricity consumption of the hanging user; (3) calculating a correlation coefficient, and quantitatively analyzing the correlation degree of each user electric quantity fingerprint and the station area line loss rate fingerprint by using a Pearson equi-phase relation method; (4) locating strongly-correlated users, and screening abnormal suspected users with strongly-correlated locating electric quantity fingerprints and line loss fingerprints by using correlation coefficient results of the electric quantity of each user and the line loss of the platform area;
4.2 Fingerprint fluctuation amount analysis model): and (3) locating suspected users affecting the abnormal fluctuation of the line loss of the platform area through the change inflection point analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user.
Modeling: (1) the data preprocessing comprises the steps of duplicate removal of the power consumption data of a user, invalid data elimination, data wide table splicing and the like; (2) locking the date of a change inflection point, and selecting the mutation point with the most obvious change of the line loss curve of the station area; (3) calculating an inflection point K value, and calculating a K value of a change speed of the line loss electric quantity of the station area caused by each user electric quantity change speed by using a defined model rule of K value = user electric quantity change quantity/station area loss electric quantity change quantity; (4) positioning abnormal users, and screening and positioning abnormal suspected users with large relative deviation degree to the line loss of the transformer area by using the K value result of the electricity consumption of each user and the line loss of the transformer area;
4.3 A fingerprint signal empirical mode decomposition model): and (3) positioning suspected users affecting the abnormal fluctuation of the line loss of the platform area through the time-frequency signal analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user.
Modeling: (1) the data preprocessing comprises the steps of user electricity consumption data de-duplication, lagrange interpolation on the missing data, daily electricity consumption calculation, wide-scale splicing and the like; (2) calculating a correlation coefficient r of the user electricity consumption and the line loss rate of the station area, and sorting results based on the daily electricity consumption and the correlation coefficient r; (3) extracting 5% of users before sorting as primary screening users, carrying out signal modal decomposition on the user electricity consumption and the station area loss electricity consumption through an EMD algorithm, respectively extracting high frequency components (IMFs), and completing signal map fitting; (4) marking abnormal users, and screening and positioning suspected line loss users by utilizing the empirical mode decomposition result of the fingerprint signals;
s5, establishing an accurate treatment method strategy model, namely grading the suspicion degree of an abnormal user as a foothold point in the view of how a big data modeling analysis result is applied to the floor, and providing a set of differential treatment strategies (models) and model result application flow:
and (3) constructing a strategy model: and (3) respectively positioning suspicious user ranges which are strongly correlated with line loss anomalies by using three power distribution network line loss electric fingerprint abnormal user identification algorithm models (a fingerprint spectrum correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model), and carrying out gradient label division on the suspicious degree of the abnormal user, wherein the division rules are determined by the model rules. Aiming at the dividing results of the users under each analysis model, the method combines the operation data of the users in the normal and abnormal period of online loss, performs the dimension analysis of electricity change trend analysis, abnormal event analysis, equipment operation state analysis and the like, so as to assist in verifying specific electricity consumption abnormal behaviors and occurrence periods and assist in performing loss reduction treatment planning.
The application flow comprises the following steps: (1) importing the file information of the 10 kilovolt distribution network line, the transformer area, the high-low voltage users, the electric quantity, the line loss and other related data into a three-line loss abnormal user identification algorithm model (a fingerprint spectrum correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model); (2) outputting a model result: based on the three models, respectively outputting suspected user ranges which are strongly related to line loss abnormality; (3) the result labels of the three models are matched and fused, the suspicion degree comprehensive rating of the abnormal users is completed, and the suspicion users are differentially divided (such as being divided into important attention, general attention and non-important attention); (4) for suspected users of different classifications, (a differential accurate treatment strategy is adopted) classification (Shi Ce) is carried out until the damage reduction target is reached.
According to the power fingerprint accurate treatment method for the power distribution network line loss analysis, a set of power fingerprint big data analysis model and an accurate treatment method for power distribution network line loss analysis are constructed by means of a big data modeling analysis technology fused by multiple algorithms and deep analysis and research on the association relation between line loss and electric quantity time sequence fluctuation aiming at the problems that manual investigation efficiency is low, basic data quantity is large, abnormal root cause diagnosis difficulty is large and the like in the line loss treatment process of basic staff, and a differential accurate treatment strategy is provided for helping line loss suspected users to locate and analyze, so that manual work load of line loss treatment is reduced, and work efficiency of line loss treatment is improved.
The present invention is not limited to the preferred embodiments, and any person skilled in the art, based on the present invention, can apply to the present invention, and the technical solution and the inventive concept according to the present invention are equivalent or modified within the scope of the present invention.
Specifically, the accurate power fingerprint management method for the power distribution network line loss analysis comprises the following steps:
s1, data acquisition: acquiring required data, including file information of 10 kilovolt distribution network lines, transformer areas, high and low voltage users, daily frozen electric quantity, line loss and other operation information, and acquiring the required data according to a data application flow;
s2, data processing: when the correlation between the line loss rate and the electricity consumption of the hanging user is analyzed, the method has an analysis value for sudden change data such as sudden increase, sudden decrease and the like, and abnormal value processing is not needed; carrying out modal decomposition on data signals of line loss and electric quantity, and taking out user data such as zero electric quantity (daily average electric quantity is 0) and small electric quantity (daily average electric quantity is less than 1 kW.h) in advance so as to avoid influencing algorithm accuracy and applicability; filling data of the missing electricity consumption data by using a Lagrangian interpolation method;
s3, extracting and analyzing 'electric fingerprint':
3.1 The fingerprint identification is carried into the power curve data generation, continuous data of all levels of objects such as lines, areas, users and the like are converted into graph textures, the graph textures are fitted, a true graph of the power distribution network line loss change state, namely a line loss 'power fingerprint', is formed, and a line loss fingerprint and an electric quantity fingerprint are respectively constructed;
3.2 Sampling and selecting the electric fingerprint of the line loss abnormal area, and performing feature analysis on the extracted fingerprint from descriptive statistical view angles such as concentrated trend, discrete degree, distribution feature and the like to intuitively present the influence relation between electric quantity and line loss; the specific implementation results refer to fig. 1 and 2; the step is also applicable to the analysis of other 10 KV distribution network lines and station areas;
s4, modeling analysis: the method comprises the steps of carrying out business implementation logic and early-stage data processing for positioning abnormal users based on line loss and electric quantity association analysis, framing a machine learning and deep learning big data related algorithm range, carrying out quality comparison analysis on effects, efficiency and stability among algorithms through Python training, and finally completing training and construction of three line loss abnormal user identification algorithm models of a fingerprint pattern correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model based on electric quantity and line loss pattern fingerprints:
4.1 Fingerprint correlation analysis model): and positioning suspected users affecting abnormal fluctuation of the line loss of the platform area through fluctuation correlation analysis of the line loss fingerprint of the platform area and the electric quantity fingerprint of the hanging user. Modeling: (1) the data preprocessing comprises the steps of duplicate removal of user electricity consumption data, removal of zero users (8 users in total) in each daily electricity consumption, and deletion of useless fields; (2) extracting a fingerprint spectrum, including a fingerprint curve spectrum of the line loss rate of the area and the electricity consumption of the hanging user; (3) calculating a correlation coefficient, and quantitatively analyzing the correlation degree of each user electric quantity fingerprint and the station area line loss rate fingerprint by using a Pearson equi-phase relation method; (4) locating strongly-correlated users, and screening abnormal suspected users with strongly-correlated locating electric quantity fingerprints and line loss fingerprints by using correlation coefficient results of the electric quantity of each user and the line loss of the platform area; referring to fig. 3;
4.2 Fingerprint fluctuation amount analysis model): and (3) locating suspected users affecting the abnormal fluctuation of the line loss of the platform area through the change inflection point analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user. Modeling: (1) the data preprocessing comprises the steps of duplicate removal of the power consumption data of a user, invalid data elimination, data wide table splicing and the like; (2) locking the date of a change inflection point, and selecting the mutation point with the most obvious change of the line loss curve of the station area; (3) calculating an inflection point K value, and calculating a K value of the change speed of the line loss electric quantity of the station area caused by the change speed of the electric quantity of each user by using a defined model rule; (4) positioning abnormal users, and screening and positioning abnormal suspected users with large relative deviation degree to the line loss of the transformer area by using the K value result of the electricity consumption of each user and the line loss of the transformer area; referring to fig. 4;
4.3 A fingerprint signal empirical mode decomposition model): and (3) positioning suspected users affecting the abnormal fluctuation of the line loss of the platform area through the time-frequency signal analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user. Modeling: (1) the data preprocessing comprises the steps of user electricity consumption data de-duplication, lagrange interpolation on the missing data, daily electricity consumption calculation, wide-scale splicing and the like; (2) calculating a correlation coefficient r of the user electricity consumption and the line loss rate of the station area, and sorting results based on the daily electricity consumption and the correlation coefficient r; (3) extracting 5% of users before sorting as primary screening users, carrying out signal modal decomposition on the user electricity consumption and the station area loss electricity consumption through an EMD algorithm, respectively extracting high frequency components (IMFs), and completing signal map fitting; (4) marking abnormal users, and screening and positioning suspected line loss users by utilizing the empirical mode decomposition result of the fingerprint signals; referring to fig. 5;
s5, establishing an accurate treatment method strategy model: classifying the suspicion degree of the abnormal user as a footfall point, and establishing a method strategy model;
and (3) constructing a strategy model: and respectively positioning suspicious user ranges which are strongly associated with line loss anomalies by using a fingerprint spectrum correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model and an abnormal user identification algorithm model of the power fingerprint of the three power distribution network line losses, and carrying out gradient label division on the suspicious degree of the abnormal users.
Matching and fusing three model result labels to finish the comprehensive rating of the suspicion degree of the abnormal user; reference is made to the following table:
the specific application flow is as follows: (1) importing the file information of the 10 kilovolt distribution network line, the transformer area, the high-low voltage users, the electric quantity, the line loss and other related data into a fingerprint spectrum correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model; (2) outputting a model result: based on the three models, respectively outputting suspected user ranges which are strongly related to line loss abnormality; (3) the three model result labels are matched and fused, the suspicion degree comprehensive rating of the abnormal users is completed, and the suspicion users are differentially divided (specifically divided into 'important attention, general attention and non-important attention'); (4) classifying Shi Ce suspected users of different classifications, (providing accurate treatment strategies adopting differentiation) until the damage reduction target is reached; refer to fig. 6.
In the step S4, in order to find a user (taking the power consumption of the user Wang Mou as an example) with strong relevance to the line loss fluctuation of the platform, a pearson correlation coefficient and a spearman correlation analysis method are used to calculate the relevance between the power consumption of each user and the line loss rate of the platform, a correlation coefficient threshold is given, and if the absolute value of the correlation coefficient is greater than the given threshold, the correlation between the power consumption of the user and the line loss rate of the platform is defined as abnormal;
the correlation coefficient is a measure of the degree of linear correlation between the study variables, denoted by the letter r. The correlation table and the correlation diagram reflect the correlation between the two variables and the direction of the correlation thereof, but do not exactly indicate the degree of correlation between the two variables. The correlation coefficient is a statistical index for reflecting the degree of closeness of the correlation between the variables. The correlation coefficient is calculated by a product difference method, and the degree of correlation between two variables is reflected by multiplying the two dispersions based on the dispersion of the two variables from the respective average values, wherein the formula is as follows (wherein Cov (X, Y) is the covariance of X and Y, var [ X ] is the variance of X, var [ Y ] is the variance of Y).
The absolute value of r is between 0 and 1. The closer the r| is to 1, the stronger the degree of correlation between the two variables, whereas the closer the r| is to 0, the weaker the degree of correlation between the two variables; correlation coefficient r result classification table:
the user suspicion grading based on fingerprint correlation analysis is as follows:
based on the fingerprint correlation analysis model, the correlation coefficient analysis calculation between the line loss of the line/station area and the electricity consumption of the hanging user is realized, and the user suspicion differentiation rating is carried out according to the correlation coefficient value, which is specifically as follows:
the value of the r|is more than 0.9 and is 'extremely strong correlation', the value of the r|is between 0.7 and 0.9 and is 'strong correlation', the value of the r|is between 0.4 and 0.7 and is 'moderate correlation', the value of the r|is between 0.2 and 0.4 and is 'weak correlation', and the value of the r|is below 0.2 and is 'extremely weak correlation'.
Users whose labels are classified as "extremely strongly correlated", "moderately correlated" are concerned, and whether such users have abnormal electricity utilization conditions is checked.
The steps are based on the user suspicion grading of fingerprint fluctuation variation analysis:
based on the fingerprint fluctuation variable quantity analysis model, the K value calculation of the line loss and the change speed of the electric quantity of the hanging user caused by the change of the electric quantity of the line loss and the electric quantity of the station is realized. And performing user suspicion classification according to the K value, wherein the classification is as follows:
and when the K value is positive, the user adds the line loss abnormality to the line/station area, and the user is used as an abnormality suspected user to further diagnose and analyze whether abnormal electricity utilization behaviors exist. When the K value is larger, the relative deviation degree of the line loss abnormality is larger when the user electricity consumption is larger.
The steps are based on the user suspicion grading of fingerprint signal empirical mode decomposition:
based on the empirical mode decomposition model of the fingerprint signals, high-frequency signal analysis of line loss/station line loss fingerprint fluctuation and hanging user electric quantity fingerprint fluctuation is realized. And carrying out two classification of suspicions of the user according to the signal waveform coincidence degree, wherein the two classification is specifically as follows:
when the waveform coincidence degree of the high-frequency component of the power consumption of the user and the high-frequency component of the power consumption of the line/station area is relatively high, the user is used as an abnormal suspected user to further diagnose and analyze whether abnormal power consumption behaviors exist. The higher the overlap of the two waveforms, the higher the user's suspicion of abnormality.
It should be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made hereto without departing from the spirit and principles of the present invention.
The above embodiments are only preferred embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalent substitutions, improvements, etc. within the spirit and scope of the present invention should be included in the present invention.
Claims (7)
1. The accurate power fingerprint management method for the power distribution network line loss analysis is characterized by comprising the following steps of:
s1, data acquisition; the acquired data comprise operation information data, and the required data are acquired according to a data application flow;
s2, data processing: when the correlation between the line loss rate and the electricity consumption of the hanging user is analyzed, the sudden change data is processed; carrying out modal decomposition on the data signals of the line loss and the electric quantity, and taking out user data in advance; filling data of the missing electricity consumption data by using a Lagrangian interpolation method;
s3, extracting and analyzing 'electric fingerprint':
3.1 The fingerprint identification is carried into the power curve data generation, continuous data of all the level objects comprising the platform area are converted into graph textures, the graph textures are fitted, a true graph of the line loss change state of the power distribution network, namely a line loss 'power fingerprint', is formed, and a line loss fingerprint and an electric quantity fingerprint are respectively constructed;
3.2 Sampling and selecting the electric fingerprint of the line loss abnormal area, carrying out characteristic analysis on the extracted fingerprint from the descriptive statistical view angles of the centralized trend, the discrete degree and the distribution characteristics, and intuitively presenting the influence relation between the electric quantity and the line loss;
s4, modeling analysis:
4.1 Fingerprint correlation analysis model): positioning suspected users affecting abnormal fluctuation of the line loss of the platform area through fluctuation correlation analysis of the line loss fingerprint of the platform area and the electric quantity fingerprint of the hanging user;
4.2 Fingerprint fluctuation amount analysis model): the suspected users affecting the line loss abnormal fluctuation of the platform area are positioned through the change inflection point analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user;
4.3 A fingerprint signal empirical mode decomposition model): the suspected users affecting the abnormal fluctuation of the line loss of the platform area are positioned through the time-frequency signal analysis of the line loss fingerprint fluctuation of the platform area and the electric quantity fingerprint fluctuation of the hanging user;
s5, grading the suspicion degree of the abnormal user as a foothold point, and establishing a method strategy model.
2. The method for precisely managing the power fingerprint of the line loss analysis of the power distribution network according to claim 1, wherein in the step S1, the operation information data includes file information of 10 kv power distribution network lines, transformer areas, high-low voltage users, daily frozen power and line loss;
in the step S2, user data with average daily power consumption of 0 and average daily power consumption of less than 1 kW.h are taken out in advance;
in step S3, the continuous data of each hierarchical object includes data of a line, a station area, and a user.
3. The method for accurately managing the power fingerprints of the line loss analysis of the power distribution network according to claim 1 or 2, wherein the modeling process of the step 4.1) is as follows: (1) the data preprocessing comprises the steps of duplicate removal of user electricity consumption data, elimination of zero user of each daily electricity consumption and deletion of useless fields; (2) extracting a fingerprint spectrum, including a fingerprint curve spectrum of the line loss rate of the area and the electricity consumption of the hanging user; (3) calculating a correlation coefficient, and quantitatively analyzing the correlation degree of each user electric quantity fingerprint and the station area line loss rate fingerprint by using a Pearson equi-phase relation method; (4) and locating strongly-correlated users, and screening and locating abnormal suspected users with strong correlation between electric quantity fingerprints and line loss fingerprints by using correlation coefficient results of the electric quantity of each user and the line loss of the platform region.
4. The method for accurately managing the power fingerprints of the line loss analysis of the power distribution network according to claim 1 or 2, wherein the modeling process of the step 4.2) is as follows: (1) the data preprocessing comprises the steps of duplicate removal of the user electricity consumption data, invalid data removal and data wide table splicing; (2) locking the date of a change inflection point, and selecting the mutation point with the most obvious change of the line loss curve of the station area; (3) calculating an inflection point K value, and calculating a K value of a change speed of the line loss electric quantity of the station area caused by each user electric quantity change speed by using a defined model rule of K value = user electric quantity change quantity/station area loss electric quantity change quantity; (4) and (3) positioning abnormal users, and screening and positioning abnormal suspected users with large relative deviation degree to the line loss abnormality of the transformer area by using the K value result of the electricity consumption of each user and the line loss of the transformer area.
5. The method for accurately managing the power fingerprints of the line loss analysis of the power distribution network according to claim 1 or 2, wherein the modeling process of the step 4.3) is as follows: (1) the data preprocessing comprises the steps of user electricity consumption data de-duplication, lagrange interpolation, daily electricity consumption calculation and wide-scale splicing of the missing data; (2) calculating a correlation coefficient r of the user electricity consumption and the line loss rate of the station area, and sorting results based on the daily electricity consumption and the correlation coefficient r; (3) extracting 5% of users before sorting as primary screening users, carrying out signal modal decomposition on the user electricity consumption and the station area loss electricity consumption through an EMD algorithm, respectively extracting high frequency components (IMFs), and completing signal map fitting; (4) and marking abnormal users, and screening and positioning suspected line loss users by utilizing the empirical mode decomposition result of the fingerprint signals.
6. The method for precisely managing the power fingerprint of the line loss analysis of the power distribution network according to claim 2, wherein the step S5 includes:
and (3) constructing a strategy model: the method comprises the steps that a suspected user range which is strongly related to line loss abnormality is respectively positioned by using a fingerprint spectrum correlation analysis model, a fingerprint fluctuation variation analysis model and a fingerprint signal empirical mode decomposition model and an abnormal user identification algorithm model of power fingerprint of three power distribution network line loss, and gradient label division is carried out for the suspected degree of an abnormal user; aiming at the dividing result of the user under each analysis model, carrying out analysis on the electric quantity change trend, the analysis on abnormal events and the analysis dimension of the equipment operation state by combining the operation data of the user on-line loss normal and abnormal time periods so as to assist in verifying specific abnormal electricity consumption behaviors and occurrence time periods;
the application flow comprises the following steps: (1) importing the file information of the 10 KV distribution network line, the station area, the high-low voltage users and the related data of the electric quantity and the line loss into a three-large line loss abnormal user identification algorithm model; (2) outputting a model result: based on the three models, respectively outputting suspected user ranges which are strongly related to line loss abnormality; (3) matching and fusing three model result labels, finishing comprehensive grading of suspicion degrees of abnormal users, and differentially dividing the suspicion users; (4) and classifying and executing the suspected users with different classifications until the damage reduction target is reached.
7. The accurate power fingerprint treatment method for the line loss analysis of the power distribution network according to claim 1, wherein the modeling analysis in the step S4 is to perform business implementation logic and early-stage data processing for abnormal user positioning based on the line loss and electric quantity association analysis, frame a machine learning and deep learning big data related algorithm range, perform quality comparison analysis on the effect, efficiency and stability among algorithms through Python training, and finally complete training and construction of three large line loss abnormal user identification algorithm models based on electric quantity and line loss spectrum fingerprints.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211396840.5A CN116341788A (en) | 2022-11-09 | 2022-11-09 | Accurate electric power fingerprint management method for power distribution network line loss analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211396840.5A CN116341788A (en) | 2022-11-09 | 2022-11-09 | Accurate electric power fingerprint management method for power distribution network line loss analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116341788A true CN116341788A (en) | 2023-06-27 |
Family
ID=86888130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211396840.5A Pending CN116341788A (en) | 2022-11-09 | 2022-11-09 | Accurate electric power fingerprint management method for power distribution network line loss analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116341788A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562653A (en) * | 2023-06-28 | 2023-08-08 | 广东电网有限责任公司 | Distributed energy station area line loss monitoring method and system |
| CN116976707A (en) * | 2023-09-22 | 2023-10-31 | 安徽融兆智能有限公司 | User electricity consumption data anomaly analysis method and system based on electricity consumption data acquisition |
| CN118036905A (en) * | 2024-04-12 | 2024-05-14 | 国网山西省电力公司临汾供电公司 | Abnormal electricity utilization user detection method and device, storage medium and electronic equipment |
-
2022
- 2022-11-09 CN CN202211396840.5A patent/CN116341788A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562653A (en) * | 2023-06-28 | 2023-08-08 | 广东电网有限责任公司 | Distributed energy station area line loss monitoring method and system |
| CN116562653B (en) * | 2023-06-28 | 2023-11-28 | 广东电网有限责任公司 | Distributed energy station area line loss monitoring method and system |
| CN116976707A (en) * | 2023-09-22 | 2023-10-31 | 安徽融兆智能有限公司 | User electricity consumption data anomaly analysis method and system based on electricity consumption data acquisition |
| CN116976707B (en) * | 2023-09-22 | 2023-12-26 | 安徽融兆智能有限公司 | User electricity consumption data anomaly analysis method and system based on electricity consumption data acquisition |
| CN118036905A (en) * | 2024-04-12 | 2024-05-14 | 国网山西省电力公司临汾供电公司 | Abnormal electricity utilization user detection method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112488395B (en) | Method and system for predicting line loss of power distribution network | |
| CN116341788A (en) | Accurate electric power fingerprint management method for power distribution network line loss analysis | |
| Jiang et al. | Spatial-temporal synchrophasor data characterization and analytics in smart grid fault detection, identification, and impact causal analysis | |
| CN110223196A (en) | Analysis method of opposing electricity-stealing based on typical industry feature database and sample database of opposing electricity-stealing | |
| Li et al. | Damage localization of stacker’s track based on EEMD-EMD and DBSCAN cluster algorithms | |
| CN110288039A (en) | Based on user power utilization load characteristic stealing detection method | |
| CN104331532A (en) | Power transformer state evaluation method based on rough set-cloud model | |
| CN111856209A (en) | Power transmission line fault classification method and device | |
| CN111091223B (en) | Matching short-term load prediction method based on intelligent sensing technology of Internet of things | |
| CN113902062A (en) | Transformer area line loss abnormal reason analysis method and device based on big data | |
| Qiu et al. | Anomaly detection for power consumption patterns in electricity early warning system | |
| CN114266457A (en) | Method for detecting different loss inducement of distribution line | |
| CN110956281A (en) | Power equipment abnormity detection alarm system based on Log analysis | |
| CN113554229A (en) | Three-phase voltage unbalance abnormality detection method and device | |
| CN107274025B (en) | System and method for realizing intelligent identification and management of power consumption mode | |
| Wang et al. | Application of clustering technique to electricity customer classification for load forecasting | |
| Hu et al. | A data-driven method of users-transformer relationship identification in the secondary power distribution system | |
| CN115166625A (en) | Intelligent ammeter error estimation method and device | |
| Özdemir et al. | Harmonic estimation based support vector machine for typical power systems | |
| CN116136580A (en) | Line loss grading detection method and device based on sensing equipment | |
| Ugarte et al. | Characterization of load curves in a real distribution system based on K-MEANS algorithm with time-series data | |
| Li et al. | Semi-supervised learning for fault identification in electricity distribution networks | |
| Liu et al. | Short-term Load Forecasting Approach with SVM and Similar Days Based on United Data Mining Technology | |
| Mascali et al. | Synthetic ground truth generation of an electricity consumption dataset | |
| Zhao et al. | Computer System of Agricultural Economic Early Warning based on Bayesian Network Model |
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 |