CN112101635A - Method and system for monitoring electricity utilization abnormity - Google Patents
Method and system for monitoring electricity utilization abnormity Download PDFInfo
- Publication number
- CN112101635A CN112101635A CN202010863258.XA CN202010863258A CN112101635A CN 112101635 A CN112101635 A CN 112101635A CN 202010863258 A CN202010863258 A CN 202010863258A CN 112101635 A CN112101635 A CN 112101635A
- Authority
- CN
- China
- Prior art keywords
- data
- user
- electricity consumption
- electricity
- model
- 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
- 230000005611 electricity Effects 0.000 title claims abstract description 223
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012544 monitoring process Methods 0.000 title claims abstract description 39
- 230000002159 abnormal effect Effects 0.000 claims abstract description 48
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 28
- 238000012795 verification Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 24
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 238000012706 support-vector machine Methods 0.000 claims abstract description 11
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 238000007637 random forest analysis Methods 0.000 claims abstract description 7
- 238000010219 correlation analysis Methods 0.000 claims description 20
- 230000005856 abnormality Effects 0.000 claims description 9
- 238000013075 data extraction Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000012550 audit Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for monitoring abnormal electricity utilization; the monitoring method of the electricity utilization abnormity comprises the following steps: step S1, acquiring power consumption data of a user; establishing a power utilization portrait model; step S2, judging whether the electricity data of the user is abnormal according to the electricity portrait model; step S1 includes the following steps: acquiring original power consumption data of a user, and filtering noise data in the original power consumption data to obtain filtered data; respectively extracting data from the filtered data to form training set data and verification set data; generating an initial model by utilizing a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm and a support vector machine algorithm based on training set data; and verifying the initial model by adopting the verification set data, judging whether the verification set data deviates from the initial model, and if not, determining the initial model as the power utilization portrait model. The method and the system for monitoring the electricity utilization abnormity are novel in design and high in practicability.
Description
Technical Field
The invention relates to the field of informatization of the power industry, in particular to a method and a system for monitoring power utilization abnormity.
Background
The traditional abnormal data detection of the electric energy meter mainly adopts a funnel method, namely, the abnormal data of line loss is taken as a starting point, layer-by-layer elimination is carried out, for example, data of a transformer and data of which the line loss is larger than a certain range are filtered, and then the data are manually checked and judged; or filtering data with the monthly line loss rate of more than 10% or < -10%, then manually screening the filtered data, confirming the correctness of the data, or screening by adopting other indexes and then checking and determining.
The method for carrying out layer-by-layer exclusion by taking the data with abnormal line loss as a starting point is easy to omit the behaviors of relatively hiding or always stealing electricity for a long time. And the analysis work links are many, the workload of manual audit is very large, and the efficiency is not high. It is easy to miss more concealed or long-term theft of electricity.
With the development and popularization of big data technology, enterprises urgently need a method and a system capable of rapidly analyzing and identifying electricity stealing and leaking to meet the needs of actual business.
Disclosure of Invention
The invention aims to provide a method and a system for monitoring power utilization abnormity aiming at the technical problems.
The technical scheme for solving the technical problem is as follows:
the invention provides a method for monitoring power utilization abnormity, which comprises the following steps:
step S1, acquiring power consumption data of a user; establishing a power utilization portrait model;
step S2, judging whether the electricity data of the user is abnormal according to the electricity portrait model;
wherein, step S1 includes the following steps:
step S11, acquiring original power consumption data of a user, and filtering noise data in the original power consumption data to obtain filtered data;
step S12, extracting data from the filtered data respectively to form training set data and verification set data;
s13, generating an initial model by using a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm and a support vector machine algorithm based on training set data;
step S14, verifying the initial model by adopting the verification set data, judging whether the verification set data deviates from the initial model, and if not, determining the initial model as the power utilization image model; if yes, the training set data is adjusted, and the process proceeds to step S13.
In the method for monitoring abnormal electricity consumption, historical electricity consumption data of a user is used as original electricity consumption data, so that an electricity consumption portrait model of the user is established;
step S2 includes:
step S21, acquiring the actual electricity consumption of the user at a specific time;
step S22, calculating the deviation rate between the actual power consumption of the user and the predicted power consumption predicted by the power consumption image model at the specific time;
and step S23, judging whether the calculated deviation rate is larger than a preset threshold value, if so, judging that the actual electricity consumption of the user at the specific time is abnormal data.
In the method for monitoring abnormal electricity consumption, historical electricity consumption data of users and units in the same industry are respectively used as original electricity consumption data, so that electricity consumption portrait models of the users and the units in the same industry are respectively established;
step S2 includes: whether the historical electricity consumption data of the user is abnormal is judged by judging whether the trend of the electricity consumption portrait model of the user is consistent with that of the electricity consumption portrait model of the same industry unit.
In the above method for monitoring abnormal power consumption of the present invention, step S1 further includes: performing correlation analysis on historical electricity utilization data of the user and the same industry unit of the user to obtain a correlation coefficient between the historical electricity utilization data of the user and the same industry unit of the user, and establishing an electricity utilization image model of the industry where the user is located by taking the correlation coefficient as original electricity utilization data;
step S2 further includes: and judging whether the historical electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model of the industry where the user is located.
In the above method for monitoring abnormal power consumption of the present invention, step S1 further includes: respectively carrying out correlation analysis on historical electricity consumption data of the user and the same industry unit of the user and meteorological data so as to obtain a correlation coefficient of the historical electricity consumption data of the user and the meteorological data and a correlation coefficient of the historical electricity consumption data of the same industry unit of the user and the meteorological data;
step S2 further includes: and judging whether the historical electricity utilization data of the user is abnormal or not by judging whether the correlation coefficient of the historical electricity utilization data of the user and the meteorological data is consistent or not and whether the correlation coefficient of the historical electricity utilization data of the same industry unit of the user and the meteorological data is consistent or not.
The invention also provides a system for monitoring abnormal electricity utilization, which comprises:
the data acquisition module is used for acquiring the electricity utilization data of the user;
the model establishing module is used for establishing an electricity utilization portrait model;
the judging module is used for judging whether the electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model;
wherein, the data acquisition module includes:
the original power consumption data acquisition module is used for acquiring original power consumption data of a user;
the filtering module is used for filtering noise data in the original power utilization data to obtain filtered data;
the model building module comprises:
the data extraction module is used for extracting data from the filtered data respectively so as to form training set data and verification set data;
the initial model generation module is used for generating an initial model by utilizing a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm and a support vector machine algorithm based on training set data;
the verification module is used for verifying the initial model by adopting the verification set data, judging whether the verification set data deviates from the initial model, and if not, determining the initial model as the power utilization portrait model; and if so, adjusting the training set data and sending the training set data to the initial model generation module to regenerate the initial model.
In the monitoring system for power consumption abnormity, the model establishing module is also used for utilizing historical power consumption data of a user as original power consumption data so as to establish a power consumption portrait model of the user;
the original electricity consumption data acquisition module is used for acquiring the actual electricity consumption of a user at a specific time;
the judging module is used for calculating the deviation rate between the actual power consumption of the user at the specific time and the predicted power consumption predicted by the power consumption image model; and judging whether the calculated deviation rate is larger than a preset threshold value or not, if so, judging that the actual electricity consumption of the user at the specific time is abnormal data.
In the monitoring system for power consumption abnormity, the model establishing module is also used for respectively utilizing historical power consumption data of users and units in the same industry as original power consumption data to respectively establish power consumption portrait models of the users and the units in the same industry;
and the judging module is used for judging whether the historical electricity consumption data of the user is abnormal or not by judging whether the trend of the user is consistent with that of the electricity consumption portrait model of the same industry unit.
In the above monitoring system for power consumption abnormality according to the present invention, the model building module further includes:
the correlation analysis module is used for carrying out correlation analysis on the historical electricity utilization data of the user and the same industry unit of the user so as to obtain a correlation coefficient between the historical electricity utilization data of the user and the same industry unit of the user, and the correlation coefficient is used as the original electricity utilization data to establish an electricity utilization portrait model of the industry where the user is located;
and the judging module is used for judging whether the historical electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model of the industry where the user is located.
In the monitoring system for power consumption abnormality, the relevance analysis module is used for respectively carrying out relevance analysis on historical power consumption data of the user and the same industry unit thereof and meteorological data so as to obtain a correlation coefficient of the historical power consumption data of the user and the meteorological data and a correlation coefficient of the historical power consumption data of the same industry unit of the user and the meteorological data;
and the judging module is used for judging whether the historical electricity utilization data of the user is abnormal or not by judging whether the correlation coefficient of the historical electricity utilization data of the user and the meteorological data is consistent or not and judging whether the correlation coefficient of the historical electricity utilization data of the same industry unit of the user and the meteorological data is consistent or not.
The monitoring method and the system for power utilization abnormity realize the following effective effects: 1) by utilizing a machine learning model and the strong computing power of a computer, the power utilization model of a single electric meter is trained and predicted once a month, and the abnormity is discovered, which cannot be realized in the manual era. 2) And (3) exploring a new service mode, pushing the predicted data of each electric meter to handheld equipment of a meter reader before reading the meter, and observing and visiting the electricity utilization condition of a user on the spot if the read data is far from the predicted data. 3) According to the invention, the weather information and the power utilization information are subjected to correlation analysis, and an analysis model is established, so that the power utilization abnormity can be accurately judged. The method and the system for monitoring the electricity utilization abnormity are novel in design and high in practicability.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
fig. 1 is a flowchart illustrating a step S1 of a power abnormality monitoring method according to a preferred embodiment of the present invention;
FIG. 2 is a trend chart of electricity consumption data of a user in a certain hospital after being processed by filtering and converting steps;
FIG. 3 is an elliptical graphical representation of the correlation coefficient between power usage and time and climate data for a particular hospital;
FIG. 4 is a schematic diagram showing factors related to electricity usage data for a particular hospital;
FIG. 5 is a diagram showing the relationship between the meter reading years of the information on the historical electricity consumption of the residential user and the prediction accuracy of the electricity consumption portrait model;
FIG. 6 is a diagram showing the relationship between the meter reading month of the resident user historical electricity consumption information and the prediction accuracy of the electricity consumption portrait model;
FIG. 7 is a schematic diagram of an industry profile based on correlation coefficients between historical electricity data for four leisure centers in different regions;
FIG. 8 shows a trend graph of a power usage profile model for a shore XX recreation center;
FIG. 9 shows a trend graph of the electrogram model for the royal XX center of relaxation;
FIG. 10 shows a trend graph of a power usage profile model for the Li XX casual center;
FIG. 11 shows a trend graph of a power usage profile model for a golden XX casual center;
fig. 12 is a functional block diagram of a power consumption abnormality monitoring system according to a preferred embodiment of the present invention.
Detailed Description
The technical problem to be solved by the invention is as follows: the traditional electric energy meter abnormal data detection method is easy to omit the hidden or long-term electricity stealing behaviors. And the analysis work links are many, the workload of manual audit is very large, and the efficiency is not high. It is easy to miss more concealed or long-term theft of electricity. The technical idea proposed by the invention for the technical problem is as follows: a monitoring method and a monitoring system for electricity consumption abnormity are constructed, big data are used for analysis, corresponding electricity consumption models are established according to the characteristics of different users, then the electricity consumption behaviors of the users are analyzed by the electricity consumption models, abnormal electricity consumption data are found quickly, and the situation of electricity stealing and leaking is screened quickly and effectively.
In order to make the technical purpose, technical solutions and technical effects of the present invention more clear and facilitate those skilled in the art to understand and implement the present invention, the present invention will be further described in detail with reference to the accompanying drawings and specific embodiments.
The invention provides a method for monitoring power utilization abnormity, which comprises the following steps:
step S1, acquiring power consumption data of a user; establishing a power utilization portrait model;
step S2, judging whether the electricity data of the user is abnormal according to the electricity portrait model;
as shown in fig. 1, fig. 1 is a flowchart illustrating step S1 of the method for monitoring abnormal power consumption according to the preferred embodiment of the present invention. Specifically, step S1 includes the steps of:
step S11, acquiring original power consumption data, and filtering noise data in the original power consumption data to obtain filtered data;
step S12, extracting data from the filtered data respectively to form training set data and verification set data;
s13, generating an initial model by using a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm and a support vector machine algorithm based on training set data;
step S14, verifying the initial model by adopting the verification set data, judging whether the verification set data deviates from the initial model, and if not, determining the initial model as the power utilization image model; if yes, the training set data is adjusted, and the process proceeds to step S13.
Further, the monitoring method of the electricity utilization abnormity can be realized by adopting a prediction deviation method or an outlier method. Specifically, the prediction deviation method is to use the historical electricity consumption data of the user as the original electricity consumption data so as to establish an electricity consumption portrait model of the user; thus, the power consumption of the user at a specific time can be predicted by using the power consumption portrait model of the user. Specifically, step S2 includes:
step S21, acquiring the actual electricity consumption of the user at a specific time;
step S22, calculating the deviation rate between the actual power consumption of the user and the predicted power consumption predicted by the power consumption image model at the specific time;
step S23, determining whether the calculated deviation ratio is greater than a preset threshold (for example, 30%), and if so, determining that the actual power consumption of the user at the specific time is abnormal data.
The prediction deviation method is suitable for detecting the electricity stealing of the user after the online.
The outlier method is to respectively use the historical electricity consumption data of the users and the units in the same industry as the original electricity consumption data to respectively establish an electricity consumption portrait model of the users and the units in the same industry; then, whether the historical electricity consumption data of the user is abnormal is judged by judging whether the trend of the electricity consumption portrait model of the user is consistent with that of the electricity consumption portrait model of the same industry unit. The outlier method is more suitable for the existing electricity stealing behaviors before the online, and is also helpful for finding the electricity stealing behaviors after the online.
Furthermore, in the outlier method, correlation analysis can be carried out on historical electricity utilization data of the users and the same industry units of the users, so that correlation coefficients among the historical electricity utilization data of the users and the same industry units of the users are obtained, and the correlation coefficients are used as original electricity utilization data to establish an electricity utilization image model of the industry where the users are located;
and then, judging whether the historical electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model of the industry where the user is located.
The historical electricity consumption data of the user and the same industry unit of the user can be respectively subjected to correlation analysis with the meteorological data, so that the correlation coefficient of the historical electricity consumption data of the user and the meteorological data and the correlation coefficient of the historical electricity consumption data of the same industry unit of the user and the meteorological data are obtained;
and then, judging whether the historical electricity utilization data of the user is abnormal or not by judging whether the correlation coefficient of the historical electricity utilization data of the user and the meteorological data is consistent or not and whether the correlation coefficient of the historical electricity utilization data of the same industry unit of the user and the meteorological data is consistent or not.
It can be understood that through the prediction deviation method and the outlier method, a data mining method can be adopted, and starting from the establishment of the general power utilization model, an industry model is respectively established, such as: on the basis of a salary household electricity utilization model, a heavy industrial electricity utilization model, a middle and primary school electricity utilization model and a large supermarket electricity utilization model, individual user models such as an A electricity meter model, a B electricity meter model and a C electricity meter model are further established and perfected, abnormal electricity utilization data are rapidly found through a big data analysis method, and therefore users who steal electricity are locked.
For the convenience of understanding, the following description will be given of the electricity usage profile model establishing method, its versatility, prediction bias method, and outlier method, by way of specific examples.
Power consumption portrait model
In the embodiment, the electricity utilization data of the users in the north hospital after the filtering and converting steps are selected as the research objects. The electricity consumption data after being processed by the filtering and converting steps is made into a trend chart, as shown in fig. 2. It is easy to see from the trend chart that the electricity utilization condition of a certain hospital in the north fluctuates along with the change of seasons, and a certain rule exists every year. The time series analysis is used for establishing a prediction model, and the specific method comprises the following steps: importing the electricity utilization data into a system, and analyzing historical data of a certain hospital in the north by using the ARIMA (4,0,0) through a time sequence so as to construct a time sequence model. The time sequence model is utilized to predict the electricity utilization condition of the first half year of 2020 in the north hospital, and the electricity utilization condition is compared with the actual electricity utilization quantity, and the following table shows that:
| meter reading month | Actual electricity consumption | Prediction of electricity | Deviation ratio | |
| 2020/1 | 1034498 | 1013359 | 2％ | |
| 2020/2 | 1049570 | 859896.8 | 18％ | |
| 2020/3 | 923871 | 731876.2 | 21％ | |
| 2020/4 | 1262611 | 693576.7 | 45％ | |
| 2020/5 | 1274319 | 812037.4 | 36％ | |
| 2020/6 | 1753360 | 957515.3 | 45％ |
The deviation rate gradually increases along with the month, which shows that the electricity consumption is related to other factors besides the influence of the time law. Through the collection and research of a large amount of data, the electricity consumption condition is found to be related to climate factors, and the correlation analysis is carried out on the electricity consumption and the meteorological data of the north hospital, so that the correlation coefficient between the electricity consumption, the time and the meteorological data of the north hospital can be obtained, and the correlation coefficient is shown in the following table:
visualizing the correlation coefficient, specifically displaying the correlation coefficient among the variables by an ellipsograph, as shown in fig. 3: the flatter the ellipse, the stronger the correlation. The right-oblique ellipse represents positive correlation, and the darker the color, the stronger the positive correlation; the left-slanted ellipse represents a negative correlation, and the darker the color, the stronger the negative correlation.
According to fig. 3, factors related to electricity consumption data of a certain hospital in north are determined, including year of meter reading, month of meter reading, precipitation, air temperature, humidity, evaporation, sunshine hours, average wind speed and electricity consumption days, as shown in fig. 4. Then, various regression models are established by using historical power consumption and climate information of a certain hospital, and an optimal model is selected through the standardized Mean Square Error (MSE) of the various models. The MSE (mean square error) of the support vector machine established model is minimum according to the comparison of the MSE values of the models, so that the model established by the support vector machine is an optimal model and is confirmed to be an electricity utilization image model of a certain hospital in the north.
Forecasting the electricity consumption of a certain northbound hospital by using the climate information of the first half year of 201X, and comparing the forecasting result with the actual electricity consumption:
in the comparison of the predicted results of the first half year of 201X, the predicted result of 4 months has higher accuracy, but the predicted deviation rate of 2 months of 201X is up to 15%.
Universality of electric portrait model
And (3) sorting historical power consumption information of residential users in a certain north area, then establishing a power consumption portrait model by using a support vector machine, predicting and checking the universality of the power consumption portrait model. The historical electricity consumption information of 604 residential users is selected to enter the research, and the relation between the meter reading years and the meter reading months of the historical electricity consumption information of the residential users and the prediction accuracy of the electricity consumption portrait model is respectively counted, as shown in fig. 5 and 6. It can be found that the accuracy of the general portrait obtained by machine learning is very high: 1) the data of about 91 percent of residential users reaches more than 90 percent of accuracy rate every year; 2) the prediction accuracy rates of months 2 and 4 are relatively low, and the accuracy rate higher than 90% accounts for about 84% of the data of the total resident users; 3) the prediction accuracy rate of 8 months and 10 months is ideal, and the data of 97 percent of residential users is higher than the accuracy rate of 90 percent.
Method of predicting deviation
The power consumption data and the climate information of the XX leisure center 2 months before 2019 are sorted, a power consumption portrait model is established by using a support vector machine in machine learning, a training set and a verification set are predicted, and the results are shown in the following table:
| year of year | Month of the year | Actual electricity consumption | Prediction of electricity | Error rate | |
| 2016 | 5 | 3542 | 3628 | -2％ | |
| 2016 | 7 | 6266 | 6180 | 1％ | |
| 2016 | 9 | 5805 | 5891 | -1％ | |
| 2016 | 11 | 6407 | 6321 | 1％ | |
| 2017 | 1 | 6437 | 6351 | 1％ | |
| 2017 | 3 | 7471 | 7385 | 1％ | |
| 2017 | 5 | 6927 | 6841 | 1％ | |
| 2017 | 7 | 6676 | 6590 | 1％ | |
| 2017 | 9 | 6852 | 6766 | 1％ | |
| 2017 | 11 | 6320 | 6234 | 1％ | |
| 2018 | 1 | 5875 | 5961 | -1％ | |
| 2018 | 3 | 5895 | 5981 | -1％ | |
| 2018 | 5 | 6570 | 6484 | 1％ | |
| 2018 | 7 | 6090 | 6176 | -1％ | |
| 2018 | 9 | 5673 | 5759 | -2％ | |
| 2018 | 11 | 5303 | 5389 | -2％ | |
| 2019 | 1 | 5474 | 5560 | -2％ |
The prediction of the training set (data 2 months before 2019) is very accurate in terms of error rate, which is between-2% and 1%.
The data of month 3 in 2019 are predicted by using a model trained before month 2 in 2019, and the results are shown in the following table:
the built electricity utilization portrait model predicts 3 months in 2019, the predicted electricity consumption is 6603, the electricity registered by the system is less than the predicted value and is close to 6000 degrees, the electricity utilization is abnormal, and suspicion of electricity stealing and electricity leakage exists.
Outlier method
The electricity usage of the customer typically fluctuates with the seasons and climate. Four leisure centers in different areas are selected for comparative analysis, and an industry portrait model is further established. The method comprises the following specific steps: and (4) sorting historical electric quantity data of the four leisure centers, and analyzing the correlation between the researches by utilizing the correlation. The sample cases and their correlation coefficients are shown in the following two tables:
as can be seen from the above two tables:
1. the correlation coefficient of the Li XX leisure center and the Huang XX leisure center is 0.62, which shows that the Li XX leisure center and the Huang XX leisure center have strong positive correlation;
2. the correlation coefficients of the golden XX, the beautiful XX, the shore XX and the royal XX leisure centers are negative numbers, which shows that the golden XX is inconsistent with the other three laws;
3. the correlation coefficient of the shore XX and the beautiful XX and the royal XX is less than-0.3, which shows that the rules of the shore XX and the two leisure centers are inconsistent.
An industry profile may be created based on the correlation coefficients of the samples, as shown in FIG. 7.
In addition, respective electricity use portrait models are established according to historical electricity quantity data of the four leisure centers, as shown in fig. 8-11. It can be seen that the golden XX leisure center is significantly different from the electricity consumption trend graphs of the other three families. Meanwhile, the correlation analysis is carried out on the electricity consumption of the four leisure centers and meteorological data, and the results are shown in the following table:
as can be seen from the table, the correlations of temperature, sunshine and the like influencing the electric quantity by the golden XX leisure center are obviously lower than the industry level, and the abnormality exists.
Further, as shown in fig. 12, fig. 12 is a functional module schematic diagram of a monitoring system for power consumption abnormality according to a preferred embodiment of the present invention. This monitoring system of power consumption anomaly includes:
the data acquisition module 100 is used for acquiring power consumption data of a user;
the model establishing module 200 is used for establishing an electricity utilization portrait model;
the judging module 300 is used for judging whether the power consumption data of the user is abnormal or not according to the power consumption portrait model;
the data acquisition module 100 includes:
an original power consumption data obtaining module 110, configured to obtain original power consumption data of a user;
the filtering module 120 is configured to filter noise data in the original power consumption data, so as to obtain filtered data;
the model building module 200 includes:
a data extraction module 210, configured to extract data from the filtered data, so as to form training set data and verification set data;
the initial model generation module 220 is configured to generate an initial model based on training set data by using a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm, and a support vector machine algorithm;
the verification module 230 is configured to verify the initial model by using the verification set data, determine whether the verification set data deviates from the initial model, and determine the initial model as the power consumption portrait model if the verification set data does not deviate from the initial model; if so, the training set data is adjusted and sent to the initial model generation module 220 to regenerate the initial model.
Further, the model establishing module 200 is further configured to utilize historical electricity consumption data of the user as original electricity consumption data, so as to establish an electricity consumption portrait model of the user;
the original power consumption data acquisition module 110 is configured to acquire an actual power consumption of a user at a specific time;
a judging module 300 for calculating a deviation ratio between an actual power consumption of the user at the specific time and a predicted power consumption predicted by a power consumption profile model thereof; and judging whether the calculated deviation rate is larger than a preset threshold value or not, if so, judging that the actual electricity consumption of the user at the specific time is abnormal data.
Further, the model establishing module 200 is further configured to respectively utilize historical electricity consumption data of the user and the same industry unit as original electricity consumption data, so as to respectively establish an electricity consumption portrait model of the user and the same industry unit;
the judging module 300 is used for judging whether the historical electricity consumption data of the user is abnormal or not by judging whether the trend of the user is consistent with that of the electricity consumption portrait model of the same industry unit.
Further, the model building module 200 further includes:
the correlation analysis module 240 is used for performing correlation analysis on the historical electricity consumption data of the user and the same industry unit thereof so as to obtain a correlation coefficient between the historical electricity consumption data of the user and the same industry unit thereof, and establishing an electricity consumption portrait model of the industry where the user is located by taking the correlation coefficient as the original electricity consumption data;
the judging module 300 is configured to judge whether the historical electricity consumption data of the user is abnormal according to the electricity consumption portrait model of the industry where the user is located.
Further, the correlation analysis module 240 is configured to perform correlation analysis on the historical electricity consumption data of the user and the same industry unit thereof and the meteorological data respectively, so as to obtain a correlation coefficient between the historical electricity consumption data of the user and the meteorological data, and a correlation coefficient between the historical electricity consumption data of the same industry unit of the user and the meteorological data;
the judging module 300 is configured to judge whether the historical electricity consumption data of the user is abnormal by judging whether the correlation coefficient between the historical electricity consumption data of the user and the meteorological data is consistent with the correlation coefficient between the historical electricity consumption data of the same industry unit of the user and the meteorological data.
The monitoring method and the system for power utilization abnormity realize the following effective effects:
1) by utilizing a machine learning model and the strong computing power of a computer, the power utilization model of a single electric meter is trained and predicted once a month, and the abnormity is discovered, which cannot be realized in the manual era.
2) And (3) exploring a new service mode, pushing the predicted data of each electric meter to handheld equipment of a meter reader before reading the meter, and observing and visiting the electricity utilization condition of a user on the spot if the read data is far from the predicted data.
3) According to the invention, the weather information and the power utilization information are subjected to correlation analysis, and an analysis model is established, so that the power utilization abnormity can be accurately judged.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (10)
1. A method for monitoring power utilization abnormity is characterized by comprising the following steps:
step S1, acquiring power consumption data of a user; establishing a power utilization portrait model;
step S2, judging whether the electricity data of the user is abnormal according to the electricity portrait model;
wherein, step S1 includes the following steps:
step S11, acquiring original power consumption data of a user, and filtering noise data in the original power consumption data to obtain filtered data;
step S12, extracting data from the filtered data respectively to form training set data and verification set data;
s13, generating an initial model by using a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm and a support vector machine algorithm based on training set data;
step S14, verifying the initial model by adopting the verification set data, judging whether the verification set data deviates from the initial model, and if not, determining the initial model as the power utilization image model; if yes, the training set data is adjusted, and the process proceeds to step S13.
2. The method for monitoring abnormal electricity consumption according to claim 1, wherein historical electricity consumption data of a user is used as original electricity consumption data, so as to establish an electricity consumption portrait model of the user;
step S2 includes:
step S21, acquiring the actual electricity consumption of the user at a specific time;
step S22, calculating the deviation rate between the actual power consumption of the user and the predicted power consumption predicted by the power consumption image model at the specific time;
and step S23, judging whether the calculated deviation rate is larger than a preset threshold value, if so, judging that the actual electricity consumption of the user at the specific time is abnormal data.
3. The method for monitoring abnormal electricity consumption according to claim 1, wherein historical electricity consumption data of users and the same industry units are respectively used as original electricity consumption data, so as to respectively establish electricity consumption portrait models of the users and the same industry units;
step S2 includes: whether the historical electricity consumption data of the user is abnormal is judged by judging whether the trend of the electricity consumption portrait model of the user is consistent with that of the electricity consumption portrait model of the same industry unit.
4. The method for monitoring power consumption abnormality according to claim 3, wherein the step S1 further includes: performing correlation analysis on historical electricity utilization data of the user and the same industry unit of the user to obtain a correlation coefficient between the historical electricity utilization data of the user and the same industry unit of the user, and establishing an electricity utilization image model of the industry where the user is located by taking the correlation coefficient as original electricity utilization data;
step S2 further includes: and judging whether the historical electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model of the industry where the user is located.
5. The method for monitoring power consumption abnormality according to claim 3, wherein the step S1 further includes: respectively carrying out correlation analysis on historical electricity consumption data of the user and the same industry unit of the user and meteorological data so as to obtain a correlation coefficient of the historical electricity consumption data of the user and the meteorological data and a correlation coefficient of the historical electricity consumption data of the same industry unit of the user and the meteorological data;
step S2 further includes: and judging whether the historical electricity utilization data of the user is abnormal or not by judging whether the correlation coefficient of the historical electricity utilization data of the user and the meteorological data is consistent or not and whether the correlation coefficient of the historical electricity utilization data of the same industry unit of the user and the meteorological data is consistent or not.
6. A system for monitoring power consumption anomalies, comprising:
the data acquisition module (100) is used for acquiring the electricity utilization data of a user;
the model building module (200) is used for building an electricity utilization portrait model;
the judging module (300) is used for judging whether the electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model;
wherein the data acquisition module (100) comprises:
the original electricity consumption data acquisition module (110) is used for acquiring original electricity consumption data of a user;
the filtering module (120) is used for filtering noise data in the original power utilization data to obtain filtered data;
the model building module (200) comprises:
a data extraction module (210) for extracting data from the filtered data, respectively, to form training set data and verification set data;
the initial model generation module (220) is used for generating an initial model by utilizing a Bagging algorithm, a random forest algorithm, an artificial neural network algorithm and a support vector machine algorithm based on training set data;
the verification module (230) is used for verifying the initial model by adopting the verification set data, judging whether the verification set data deviates from the initial model, and if not, determining the initial model as the power utilization portrait model; if yes, adjusting the training set data, and sending the training set data to an initial model generation module (220) to regenerate the initial model.
7. The system for monitoring abnormal electricity consumption according to claim 6, wherein the model establishing module (200) is further configured to utilize historical electricity consumption data of the user as original electricity consumption data to establish an electricity consumption portrait model of the user;
the original electricity consumption data acquisition module (110) is used for acquiring the actual electricity consumption of a user at a specific time;
a judging module (300) for calculating a deviation rate between an actual power consumption of the user at the specific time and a predicted power consumption predicted by a power consumption profile model thereof; and judging whether the calculated deviation rate is larger than a preset threshold value or not, if so, judging that the actual electricity consumption of the user at the specific time is abnormal data.
8. The system for monitoring the electricity consumption abnormity according to claim 6, wherein the model establishing module (200) is further used for respectively utilizing historical electricity consumption data of users and the units in the same industry as original electricity consumption data so as to respectively establish an electricity consumption portrait model of the users and the units in the same industry;
and the judging module (300) is used for judging whether the historical electricity consumption data of the user is abnormal or not by judging whether the trend of the user is consistent with that of the electricity consumption portrait model of the same industry unit.
9. The system for monitoring power consumption abnormality according to claim 8, characterized in that the model building module (200) further includes:
the correlation analysis module (240) is used for carrying out correlation analysis on historical electricity utilization data of the user and the same industry unit of the user so as to obtain a correlation coefficient between the historical electricity utilization data of the user and the same industry unit of the user, and the correlation coefficient is used as original electricity utilization data to establish an electricity utilization portrait model of the industry where the user is located;
and the judging module (300) is used for judging whether the historical electricity consumption data of the user is abnormal or not according to the electricity consumption portrait model of the industry where the user is located.
10. The system for monitoring power consumption abnormity according to claim 8, wherein the correlation analysis module (240) is used for performing correlation analysis on historical power consumption data of users and the same industry units thereof and meteorological data respectively, so as to obtain correlation coefficients of the historical power consumption data of the users and the meteorological data and correlation coefficients of the historical power consumption data of the same industry units of the users and the meteorological data;
and the judging module (300) is used for judging whether the historical electricity utilization data of the user is abnormal or not by judging whether the correlation coefficient of the historical electricity utilization data of the user and the meteorological data is consistent or not and whether the correlation coefficient of the historical electricity utilization data of the same industry unit of the user and the meteorological data is consistent or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010863258.XA CN112101635A (en) | 2020-08-25 | 2020-08-25 | Method and system for monitoring electricity utilization abnormity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010863258.XA CN112101635A (en) | 2020-08-25 | 2020-08-25 | Method and system for monitoring electricity utilization abnormity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112101635A true CN112101635A (en) | 2020-12-18 |
Family
ID=73753424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010863258.XA Pending CN112101635A (en) | 2020-08-25 | 2020-08-25 | Method and system for monitoring electricity utilization abnormity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112101635A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112633412A (en) * | 2021-01-05 | 2021-04-09 | 南方电网深圳数字电网研究院有限公司 | Abnormal electricity consumption detection method, equipment and storage medium |
| CN113223263A (en) * | 2021-04-25 | 2021-08-06 | 深圳市芯中芯科技有限公司 | Electrical fire monitoring system based on Internet of things |
| CN113269478A (en) * | 2021-07-21 | 2021-08-17 | 武汉中原电子信息有限公司 | Concentrator abnormal data reminding method and system based on multiple models |
| CN113362118A (en) * | 2021-07-08 | 2021-09-07 | 广东电网有限责任公司 | User electricity consumption behavior analysis method and system based on random forest |
| CN113377760A (en) * | 2021-07-06 | 2021-09-10 | 国网江苏省电力有限公司营销服务中心 | Method and system for establishing low-voltage resident feature portrait based on electric power data and multivariate data |
| CN113947504A (en) * | 2021-11-11 | 2022-01-18 | 国网辽宁省电力有限公司营销服务中心 | Electricity stealing analysis method and system based on random forest method |
| CN115170003A (en) * | 2022-09-08 | 2022-10-11 | 国网山东省电力公司兰陵县供电公司 | Electricity stealing monitoring method and system, storage medium and terminal |
| CN115905319A (en) * | 2022-11-16 | 2023-04-04 | 国网山东省电力公司营销服务中心(计量中心) | Automatic identification method and system for abnormal electricity charges of massive users |
| CN116662413A (en) * | 2023-07-25 | 2023-08-29 | 成都千嘉科技股份有限公司 | Industrial and commercial user business state change monitoring method based on gas consumption data disassembly image |
| CN116709062A (en) * | 2023-08-07 | 2023-09-05 | 安徽融兆智能有限公司 | Electricity consumption information acquisition equipment with detection function |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995161A (en) * | 2014-06-03 | 2014-08-20 | 深圳市康拓普信息技术有限公司 | Method and system for discriminating electricity stealing and leaking users |
| CN105630885A (en) * | 2015-12-18 | 2016-06-01 | 国网福建省电力有限公司泉州供电公司 | Abnormal power consumption detection method and system |
| CN106707099A (en) * | 2016-11-30 | 2017-05-24 | 国网上海市电力公司 | Monitoring and locating method based on abnormal electricity consumption detection module |
| CN107862347A (en) * | 2017-12-04 | 2018-03-30 | 国网山东省电力公司济南供电公司 | A kind of discovery method of the electricity stealing based on random forest |
| CN109614997A (en) * | 2018-11-29 | 2019-04-12 | 武汉大学 | A kind of stealing Risk Forecast Method and device based on deep learning |
| CN110991555A (en) * | 2019-12-16 | 2020-04-10 | 国网上海市电力公司 | Method for monitoring abnormal electricity consumption of user in typical industry |
-
2020
- 2020-08-25 CN CN202010863258.XA patent/CN112101635A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995161A (en) * | 2014-06-03 | 2014-08-20 | 深圳市康拓普信息技术有限公司 | Method and system for discriminating electricity stealing and leaking users |
| CN105630885A (en) * | 2015-12-18 | 2016-06-01 | 国网福建省电力有限公司泉州供电公司 | Abnormal power consumption detection method and system |
| CN106707099A (en) * | 2016-11-30 | 2017-05-24 | 国网上海市电力公司 | Monitoring and locating method based on abnormal electricity consumption detection module |
| CN107862347A (en) * | 2017-12-04 | 2018-03-30 | 国网山东省电力公司济南供电公司 | A kind of discovery method of the electricity stealing based on random forest |
| CN109614997A (en) * | 2018-11-29 | 2019-04-12 | 武汉大学 | A kind of stealing Risk Forecast Method and device based on deep learning |
| CN110991555A (en) * | 2019-12-16 | 2020-04-10 | 国网上海市电力公司 | Method for monitoring abnormal electricity consumption of user in typical industry |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112633412B (en) * | 2021-01-05 | 2024-05-14 | 南方电网数字平台科技(广东)有限公司 | Abnormal electricity utilization detection method, abnormal electricity utilization detection equipment and storage medium |
| CN112633412A (en) * | 2021-01-05 | 2021-04-09 | 南方电网深圳数字电网研究院有限公司 | Abnormal electricity consumption detection method, equipment and storage medium |
| CN113223263A (en) * | 2021-04-25 | 2021-08-06 | 深圳市芯中芯科技有限公司 | Electrical fire monitoring system based on Internet of things |
| CN113377760A (en) * | 2021-07-06 | 2021-09-10 | 国网江苏省电力有限公司营销服务中心 | Method and system for establishing low-voltage resident feature portrait based on electric power data and multivariate data |
| CN113362118A (en) * | 2021-07-08 | 2021-09-07 | 广东电网有限责任公司 | User electricity consumption behavior analysis method and system based on random forest |
| CN113269478A (en) * | 2021-07-21 | 2021-08-17 | 武汉中原电子信息有限公司 | Concentrator abnormal data reminding method and system based on multiple models |
| CN113947504A (en) * | 2021-11-11 | 2022-01-18 | 国网辽宁省电力有限公司营销服务中心 | Electricity stealing analysis method and system based on random forest method |
| CN115170003A (en) * | 2022-09-08 | 2022-10-11 | 国网山东省电力公司兰陵县供电公司 | Electricity stealing monitoring method and system, storage medium and terminal |
| CN115905319B (en) * | 2022-11-16 | 2024-04-19 | 国网山东省电力公司营销服务中心(计量中心) | Automatic identification method and system for abnormal electricity fees of massive users |
| CN115905319A (en) * | 2022-11-16 | 2023-04-04 | 国网山东省电力公司营销服务中心(计量中心) | Automatic identification method and system for abnormal electricity charges of massive users |
| CN116662413A (en) * | 2023-07-25 | 2023-08-29 | 成都千嘉科技股份有限公司 | Industrial and commercial user business state change monitoring method based on gas consumption data disassembly image |
| CN116662413B (en) * | 2023-07-25 | 2023-10-27 | 成都千嘉科技股份有限公司 | Industrial and commercial user business state change monitoring method based on gas consumption data disassembly image |
| CN116709062A (en) * | 2023-08-07 | 2023-09-05 | 安徽融兆智能有限公司 | Electricity consumption information acquisition equipment with detection function |
| CN116709062B (en) * | 2023-08-07 | 2023-10-20 | 安徽融兆智能有限公司 | Electricity consumption information acquisition equipment with detection function |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112101635A (en) | Method and system for monitoring electricity utilization abnormity | |
| CN110097297B (en) | Multi-dimensional electricity stealing situation intelligent sensing method, system, equipment and medium | |
| CN110634080B (en) | Abnormal electricity utilization detection method, device, equipment and computer readable storage medium | |
| CN107220732B (en) | Power failure complaint risk prediction method based on gradient lifting tree | |
| CN109583680B (en) | Power stealing identification method based on support vector machine | |
| Cody et al. | Decision tree learning for fraud detection in consumer energy consumption | |
| CN111882446B (en) | Abnormal account detection method based on graph convolution network | |
| CN106650797B (en) | Power distribution network electricity stealing suspicion user intelligent identification method based on integrated ELM | |
| CN110489785A (en) | A kind of online Source Apportionment of atmosphere pollution and system | |
| CN111861786A (en) | Special transformer electricity stealing identification method based on feature selection and isolated random forest | |
| CN104281779A (en) | Abnormal data judging and processing method and device | |
| CN111914942A (en) | Multi-table-combined one-use energy anomaly analysis method | |
| Madhure et al. | Cnn-lstm based electricity theft detector in advanced metering infrastructure | |
| CN111506635A (en) | System and method for analyzing residential electricity consumption behavior based on autoregressive naive Bayes algorithm | |
| CN112132210A (en) | Electricity stealing probability early warning analysis method based on customer electricity consumption behavior | |
| CN104657912A (en) | Method and system for detecting abnormal user based on water-electricity ratio and support vector clustering | |
| CN111967717A (en) | Data quality evaluation method based on information entropy | |
| CN112101471A (en) | Electricity stealing probability early warning analysis method | |
| CN109977984B (en) | Power stealing user judging method based on support vector machine | |
| CN116467551B (en) | Water supply pipe network leakage positioning method and system based on correlation coefficient | |
| CN111612054B (en) | User electricity stealing behavior identification method based on nonnegative matrix factorization and density clustering | |
| CN115905319B (en) | Automatic identification method and system for abnormal electricity fees of massive users | |
| CN112465397A (en) | Audit data analysis method and device | |
| CN115423383B (en) | Distributed village and town drinking water monitoring and regulation system and method based on artificial intelligence | |
| CN116205528A (en) | Illegal construction identification method and system based on construction site power data |
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: 20201218 |