CN115310982A - Electricity larceny prevention early warning data analysis method - Google Patents
Electricity larceny prevention early warning data analysis method Download PDFInfo
- Publication number
- CN115310982A CN115310982A CN202210339122.8A CN202210339122A CN115310982A CN 115310982 A CN115310982 A CN 115310982A CN 202210339122 A CN202210339122 A CN 202210339122A CN 115310982 A CN115310982 A CN 115310982A
- Authority
- CN
- China
- Prior art keywords
- data
- electricity
- analysis
- power consumption
- early warning
- 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 168
- 238000000034 method Methods 0.000 title claims abstract description 134
- 238000007405 data analysis Methods 0.000 title claims abstract description 56
- 230000002265 prevention Effects 0.000 title claims abstract description 46
- 238000004458 analytical method Methods 0.000 claims abstract description 68
- 230000006399 behavior Effects 0.000 claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 24
- 239000011159 matrix material Substances 0.000 claims abstract description 15
- 238000007418 data mining Methods 0.000 claims abstract description 14
- 238000010801 machine learning Methods 0.000 claims abstract description 10
- 230000008569 process Effects 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims description 51
- 238000012706 support-vector machine Methods 0.000 claims description 24
- 238000012544 monitoring process Methods 0.000 claims description 22
- 238000009826 distribution Methods 0.000 claims description 18
- 238000005516 engineering process Methods 0.000 claims description 17
- 239000000243 solution Substances 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 7
- 239000000203 mixture Substances 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013145 classification model Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 230000008878 coupling Effects 0.000 claims description 3
- 238000010168 coupling process Methods 0.000 claims description 3
- 238000005859 coupling reaction Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000005065 mining Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 239000002245 particle Substances 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000002347 injection Methods 0.000 claims description 2
- 239000007924 injection Substances 0.000 claims description 2
- 238000009394 selective breeding Methods 0.000 claims 1
- 238000003860 storage Methods 0.000 description 10
- 230000006872 improvement Effects 0.000 description 9
- 210000004556 brain Anatomy 0.000 description 4
- 238000007689 inspection Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013529 biological neural network Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000007123 defense Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000630 rising 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
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Educational Administration (AREA)
- Biophysics (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an electricity larceny prevention early warning data analysis method. The method comprises the following steps: writing an electricity larceny prevention early warning data analysis system program in mobile computer terminal equipment; loading a plurality of analysis methods including but not limited to state estimation, matrix decomposition, data mining, machine learning, game theory, AI recognition, and the like; connecting a terminal with electric energy data acquisition equipment; acquiring electric energy data, analyzing and summarizing; selecting and adjusting a method to analyze and process data; and making a corresponding electricity larceny prevention scheme according to the analysis result. The invention designs that the analysis system is arranged in the mobile computer terminal equipment and a plurality of data analysis methods are loaded, so that the mobile computer terminal equipment is convenient to be connected and applied with a plurality of different power management systems; analyzing and matching the electricity utilization data and calling an applicable method to analyze electricity stealing; the corresponding electricity larceny prevention scheme can be customized according to the analysis result, so that electricity larceny behavior is reduced, and electricity utilization safety is guaranteed.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an electricity larceny prevention early warning data analysis method.
Background
The electric energy is an important resource related to the national civilization, is inseparable from the production and the living of people, and along with the rapid development of the national economy, the demand of the production and the living of people on electric power resources is increasingly improved, so that the electric energy becomes an indispensable part for the development of social economy. In recent years, lawbreakers frequently find their hands black and time to power facilities, theft and damage cases are rising, and the problems of electricity stealing and illegal electricity utilization are not rare in electricity utilization business management. The existence of electricity stealing users seriously interferes with the normal market order of power supply and utilization, and causes huge economic loss for countries and power supply enterprises. The electric quantity is used as an intangible commodity, and in many times, the electricity utilization inspection personnel are often greatly limited in the process of carrying out electricity stealing prevention work, and in addition, the distribution range of power users is wide, the quantity is large, the quality of personnel is not uniform, after the inspection personnel arrive at the site, possible evidence does not exist, and the situation that evidence collection is difficult is formed. At present, the electricity utilization inspection system is not intelligentized, and the efficiency and the quality need to be improved. Therefore, people's air defense is emphasized, and technical air defense is also emphasized. Therefore, it is often necessary to analyze and determine the electricity stealing situation from a large amount of data based on a large amount of user electricity consumption data that can be collected by the smart meter, so as to derive a large amount of electricity stealing detection technologies. The purpose of electricity stealing detection is to determine whether an intelligent electric meter is attacked to cause electricity stealing behaviors, and through research of many experts for many years, various methods are applied to electricity stealing detection of an intelligent power grid. However, in the prior art, the electricity consumption data has various types, large data volume and complicated and disordered data structure, and different analysis methods are required to be applied to detect electricity stealing in different situations. However, the conventional electricity stealing detection device or system can only be loaded with a single detection method or analysis algorithm, and cannot be generally suitable for different power management systems. In view of this, we propose an electricity larceny prevention early warning data analysis method.
Disclosure of Invention
The invention aims to provide an electricity larceny prevention early warning data analysis method to solve the problems in the background technology.
In order to solve the above technical problems, one of the objectives of the present invention is to provide an electricity larceny prevention early warning data analysis method, which includes the following steps:
s1, designing and compiling an electricity larceny prevention early warning data analysis system program in mobile computer terminal equipment;
s2, loading a plurality of analysis methods including but not limited to state estimation, matrix decomposition, data mining, machine learning, chess playing theory, AI identification and the like into the system program;
s3, connecting the mobile computer terminal equipment with electric energy data acquisition equipment;
s4, acquiring electric energy data, and analyzing and summarizing the data quantity of the electric energy data, the data composition/distribution condition, whether a large amount of historical data exists, the analysis requirement and other conditions;
s5, selecting and adjusting one or more of the analysis methods to analyze and process the data according to the acquired data and the analysis requirements;
and S6, early warning is carried out on the electricity stealing situation according to the data analysis result, and a corresponding electricity stealing prevention scheme is formulated.
As a further improvement of the present technical solution, in S5, according to the acquired data and the analysis requirement, a specific method for adjusting one or more of the above analysis methods to perform data analysis processing includes the following steps:
s5.1, when the data volume is small and high-accuracy early warning analysis is required, a state estimation method can be selected for analysis processing;
s5.2, when the data volume is not large, the detection overhead budget is low and the requirement on the detection precision is not high, analyzing and processing can be carried out by selecting and adjusting a matrix decomposition method;
s5.3, when the data volume is large and the detection investment budget is low, a data mining method can be selected and adjusted for analysis processing;
s5.4, when the data volume is large and the data is complex, a machine learning method can be selected and adjusted for analysis and processing;
s5.5, when a perfect monitoring system exists and a large amount of image data can be obtained, an AI identification method can be selected for analysis and processing;
and S5.6, when the analysis requirement is mainly to reduce the power loss, the method of the game theory can be selected and adjusted to carry out analysis processing and make a corresponding measure scheme.
As a further improvement of the technical solution, in S5.1, an application method of the state estimation method includes the following steps:
a state estimation theory, namely a state observer, can be introduced into static safety monitoring of the power system, wherein the state observer is a system for estimating the internal state of a given actual system by measuring the input and the output of the actual system;
a Kalman filtering algorithm can be introduced to establish a dynamic state estimation theory, meanwhile, a solution scheme based on state estimation can be realized by introducing some integrated distributed state estimation techniques, and the state estimation is carried out by a weighted least square method, so that the deviation obtained by the electricity consumption data is compared with the actual deviation, and a thief is obtained by analysis;
decomposing the power consumption into two parallel loose coupling filters through a Kalman filter, carrying out deviation estimation on the power consumption to carry out electricity stealing detection analysis, utilizing AMI and SCADA measurement, adopting an integrated distribution state estimation method to analyze MW and LV distribution networks, and analyzing the power consumption behavior of the intelligent power grid from reliable measurement data which can be recorded by the intelligent power meter instead of the traditional priori measurement value with high error;
and integrally fitting the estimated value of the injection power consumption of the pseudo feeder bus according to the power consumption data of users on the distribution transformer so as to detect whether the power consumption is regular or not, creating a suspicious user list by using variance analysis after the state estimation result, and evaluating the use condition of the power consumption data and estimating the actual use condition.
Wherein, the state estimation utilizes the power network to estimate to obtain the deviation so as to realize the electricity stealing detection; the method is high in accuracy, but is not suitable for detecting mass data, and meanwhile, the method is lack of plasticity, namely different state estimation algorithm models are difficult to establish according to the difference of power utilization behaviors caused by different power utilization areas.
As a further improvement of the technical solution, in S5.2, matrix decomposition is performed in the mathematical discipline of linear algebra, that is, the matrix is decomposed into a product of matrices, and the specific application method includes:
firstly, through LU decomposition and detection of electricity stealing behaviors, a thief can be determined in a small-scale network, but the thief is unstable in a large-scale network;
secondly, local rotation LU decomposition is improved on the basis, so that thieves can be detected in a large network, but more execution time is needed;
meanwhile, under the condition that other electric meter data are not known, distributed QR (orthogonal triangle) decomposition is utilized to enable each intelligent electric meter to calculate the honesty coefficient of the intelligent electric meter; and if the integrity coefficient of the normal user is 1 and the actual power consumption of the electricity stealing user is k times of the reported power consumption, k is the integrity coefficient of the user, and whether the electricity stealing behavior exists in the user can be judged according to the k value.
The method is simple and easy to implement, but if a large amount of power consumption data are detected, the detection cost is too high, and the analysis result obtained by using matrix decomposition to perform electricity stealing detection cannot give too much information to inspectors, so that the workload is increased.
As a further improvement of the technical solution, in S5.3, the specific application method of data mining includes the following steps:
s5.3.1, carrying out semi-automatic or automatic analysis on a large amount of data to extract a previously unknown and interesting mode, such as abnormal record (abnormal condition detection);
s5.3.2, decomposing the total power consumption information of the user into information of each power consumption device by adopting non-invasive load monitoring (NILM) which is substantially load decomposition, thereby obtaining the power consumption information such as the consumption condition of the power consumption device, the power consumption behavior of the user and the like;
s5.3.3, collecting network intrusion and physical intrusion logs, and analyzing a power consumption curve of a user;
s5.3.4, listing a list which is possible to be an illegal user with the minimum error report quantity based on an attack graph fusion algorithm;
and S5.3.5, finally determining the electricity utilization behavior of the user from the integrated net load curve by adopting non-invasive load monitoring.
Among them, based on data mining technique can be called as non-intrusive load monitoring (NILM) detection electricity stealing, compare with intrusive load monitoring, NILM's economic investment is littleer, and the practicality is stronger.
As a further improvement of the technical solution, in S5.4, the machine learning includes two types, namely, a support vector machine and an artificial neural network, which can be selected according to specific data distribution conditions and application requirements.
As a further improvement of the technical solution, a specific application method of the support vector machine includes:
firstly, adopting a method for creating a nonlinear classifier by applying a kernel technique to a maximum edge hyperplane;
secondly, a power stealing detection technology based on classification of a Support Vector Machine (SVM) is used for classifying legal and suspicious samples of a power consumption database to be detected so as to detect power stealing, and the specific method flow comprises the following steps:
firstly training a support vector machine SVW from a historical data set, and then testing the support vector machine under different data sets to find out abnormal electricity utilization of customers;
on the basis of analyzing a support vector machine technology and analyzing the electricity utilization behavior of a user, introducing a 0ne-class SVM algorithm into electricity stealing detection, and providing an electricity stealing detection model formed by combining the curve fluctuation characteristic of electricity consumption and One-class SVM;
the improved power consumption fluctuation coefficient is used for representing the fluctuation condition of the power consumption data of the user, the power consumption fluctuation coefficient is used as an index for judging whether electricity is stolen or not, the index is used as a training sample, a classification model can be obtained after training, and the obtained model is used for analyzing the power consumption data to detect whether the power consumption data are abnormal or not.
However, electricity theft detection based on SVW techniques typically requires a large amount of historical power usage data collected from smart meters as training data to obtain characteristics.
Wherein, when creating a nonlinear classifier by applying a kernel technique to a maximum-edge hyperplane, a good separation is achieved by the hyperplane, which is the largest distance from the nearest training data point of any class; and in general, the larger the margin, the smaller the generalization error of the classifier.
As a further improvement of the technical solution, the specific application method of the artificial neural network includes:
firstly, a plurality of parameters required by a model are automatically set by using a hybrid neural network model and a coding technology, and a hierarchical model for classifying data is provided, so that illegal consumers are identified;
secondly, analyzing a fine-grained load curve of the user smart meter by using a neural network classifier which is searched by using a system including but not limited to a charging system and the like to find a thief;
then, a convolutional neural network model is established based on the periodicity of the power utilization behavior of the user to detect the power utilization condition of the user;
then, by collecting and storing long-time power data and mining the power consumption behavior pattern of the user based on a large amount of data, a normal power consumption behavior model based on a power consumption sample library is established so as to identify the electricity stealing behavior.
Wherein the Artificial Neural Network (ANN) is a computing system designed inspired by biological neural networks constituting animal brains; and the deep learning is particularly suitable for constructing complex models of mass data and is increasingly popular in data mining/anomaly detection.
As a further improvement of the technical solution, in S5.5, a specific application method of AI identification includes the following steps:
s5.5.1, using different image recognition models including a template matching model to compile a computer program for simulating human image recognition activities;
s5.5.2, storing face image information of a worker with the authority of operating the terminal equipment in advance;
s5.5.3, distributing a monitoring system in an operation terminal equipment area and loading a corresponding AI identification program;
s5.5.4, acquiring the face image information of each person operating the terminal equipment from the real-time monitoring system;
and S5.5.5, finally, identifying and confirming whether the personnel operating the terminal equipment is a power stealing personnel through AI.
Among them, AI identification technology is an important field of artificial intelligence.
Furthermore, the template matching model considers that a certain image is recognized, and a memory mode of the image is required to be available in past experience and is called a template; if the current stimulus matches the template in the brain, the image is identified.
Meanwhile, the pattern recognition in the image recognition is a process of automatically completing recognition and evaluation on shapes, patterns, curves, numbers, character formats and graphs by using a computer and a mathematical reasoning method on the basis of expert experience and existing recognition from a large amount of information and data.
As a further improvement of the technical solution, in S5.6, a specific application method of the game theory includes the following steps:
firstly, a game model for determining the electricity stealing behavior is based on electricity stealing users and electric power companies;
secondly, establishing a game model of the power company and the electricity stealing users by formulating a statistical anomaly detection scheme; in the game model, the purpose of the power company is to improve profit to the maximum extent and reduce the cost for detecting the electricity stealing users to the greatest extent, and the aim of the electricity stealing users is to reduce the possibility of being detected due to the constraint related to the electricity stealing amount to the greatest extent;
then researching the phenomenon of electricity stealing of the user based on the basic principle of the game theory;
secondly, constructing payment matrixes under different model combinations through particle hypothesis, and finding Nash balance by using a 'drawing bar method';
finally, an effective measure for preventing the user from stealing electricity is obtained, namely: the power company can charge the electricity fee according to the calculation formula of ai + kx (k > 1), so that the user can consciously execute the electricity utilization regulation and legally use the electricity.
The game theory is a research on a mathematical model of strategic interaction among rational decision makers, and the power theft detection technology based on the game theory is a novel power theft detection technology. When there is an objective power stealing behavior, the game-based model can give a stable and reasonable (even if not optimal) solution to reduce power loss.
The invention also aims to provide an electricity larceny prevention early warning data analysis system, which comprises a data acquisition and storage unit, a data analysis and classification unit, a method matching and calling unit, an early warning data analysis unit and an electricity larceny prevention early warning application unit; the data acquisition and storage unit, the data analysis and classification unit, the method matching and calling unit, the early warning data analysis unit and the electricity stealing early warning application unit are sequentially connected through network communication; wherein:
the data acquisition and storage unit is used for directly acquiring the electricity utilization data in the functional electric meter or importing the electricity utilization data through other data acquisition modules;
the data analysis and classification unit is used for detecting and analyzing the conditions of the composition type/distribution condition, the data volume and the like of the power utilization data and classifying the power utilization data according to a preset rule;
the method matching and calling unit is used for matching and calling an appropriate method from a plurality of pre-loaded data analysis methods according to the data condition;
the early warning data analysis unit is used for analyzing and processing the electricity stealing early warning data based on the acquired electricity utilization data by calling an applicable analysis method;
and the electricity stealing early warning application unit is used for further analyzing and formulating a corresponding electricity stealing prevention measure scheme according to the electricity stealing early warning analysis result.
The invention also provides an operation device of the electricity larceny prevention early warning data analysis system, which comprises a processor, a memory and a computer program which is stored in the memory and operated on the processor, wherein the processor is used for realizing any step of the electricity larceny prevention early warning data analysis method when executing the computer program.
It is a fourth object of the present invention to provide a computer-readable storage medium, wherein a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above electricity larceny prevention early warning data analysis methods are implemented.
Compared with the prior art, the invention has the following beneficial effects:
1. the electricity larceny prevention early warning data analysis method is convenient for being connected with various different power management systems for application by arranging an electricity larceny prevention early warning data analysis system in the mobile computer terminal equipment and loading various electricity larceny detection data analysis methods;
2. according to the electricity larceny prevention early warning data analysis method, the electricity larceny detection analysis is performed on the data by preliminarily analyzing the electricity consumption data of the users acquired in the power management system and classifying the data according to the composition, the data size and the like of the data, so that the data are matched and called by an applicable method according to the requirements of the users;
3. the electricity larceny prevention early warning data analysis method aims at different electricity utilization data structures, obtains the analysis result of the electricity larceny prevention early warning data through corresponding method analysis, and can customize the corresponding electricity larceny prevention scheme according to the analysis result in a targeted mode, so that electricity larceny behaviors are effectively reduced, and electricity utilization safety is guaranteed.
Drawings
FIG. 1 is an overall process flow diagram of the present invention;
FIG. 2 is a flow chart of a partial method of the present invention;
FIG. 3 is a second flowchart of a partial method according to the present invention;
FIG. 4 is a third flowchart of a partial method according to the present invention;
FIG. 5 is a block diagram of an exemplary electronic computer platform assembly according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Example 1
As shown in fig. 1-a, the present embodiment provides a method for analyzing early warning data of electricity larceny prevention, which includes the following steps:
s1, designing and compiling an electricity larceny prevention early warning data analysis system program in mobile computer terminal equipment;
s2, loading a plurality of analysis methods including but not limited to state estimation, matrix decomposition, data mining, machine learning, game theory, AI identification and the like into a system program;
s3, connecting the mobile computer terminal equipment with electric energy data acquisition equipment;
s4, acquiring electric energy data, and analyzing and summarizing the data quantity of the electric energy data, the data composition/distribution condition, whether a large amount of historical data exists, the analysis requirement and other conditions;
s5, selecting and adjusting one or more of the analysis methods to analyze and process data according to the acquired data and analysis requirements;
and S6, early warning the electricity stealing situation according to the data analysis result, and formulating a corresponding electricity stealing prevention scheme.
In this embodiment, in S5, a specific method for tuning one or more of the above analysis methods to perform data analysis processing according to the acquired data and the analysis requirements includes the following steps:
s5.1, when the data volume is small and high-accuracy early warning analysis is required, a state estimation method can be selected for analysis processing;
s5.2, when the data volume is not large, the detection overhead budget is low and the requirement on the detection precision is not high, analyzing and processing can be carried out by selecting and adjusting a matrix decomposition method;
s5.3, when the data volume is large and the detection investment budget is low, a data mining method can be selected and adjusted for analysis processing;
s5.4, when the data volume is large and the data is complex, a machine learning method can be selected and adjusted for analysis and processing;
s5.5, when a perfect monitoring system exists and a large amount of image data can be obtained, an AI identification method can be selected for analysis and processing;
and S5.6, when the analysis requirement is mainly to reduce the power loss, the method of the game theory can be selected and adjusted to carry out analysis processing and make a corresponding measure scheme.
In this embodiment, in S5.1, the application method of the state estimation method includes the following steps:
a state estimation theory, namely a state observer, can be introduced into static safety monitoring of the power system, wherein the state observer is a system for estimating the internal state of a given actual system by measuring the input and the output of the actual system;
a Kalman filtering algorithm can be introduced to establish a dynamic state estimation theory, meanwhile, a solution scheme based on state estimation can be realized by introducing some integrated distributed state estimation techniques, and the state estimation is carried out by a weighted least square method, so that the deviation obtained by the electricity consumption data is compared with the actual deviation, and a thief is obtained by analysis;
decomposing a Kalman filter into two parallel loose coupling filters, performing deviation estimation on power consumption to perform electricity stealing detection analysis, measuring by utilizing AMI and SCADA, analyzing MW and LV distribution networks by adopting an integrated distribution state estimation method, and analyzing power consumption behavior of a smart grid from reliable measurement data which can be recorded by the smart meter instead of the traditional priori measurement value with high error;
and performing overall fitting on the estimated value of the power consumption injected into the simulated feeder bus according to the power consumption data of the users on the distribution transformer so as to detect whether the power consumption is regular or not, creating a suspicious user list by using variance analysis after a state estimation result, and evaluating the use condition of the power consumption data and estimating the actual use condition.
Specifically, the state estimation utilizes a power network to estimate to obtain deviation so as to realize electricity stealing detection; the method has high accuracy, but is not suitable for detecting mass data, and meanwhile, the method is lack of plasticity, namely different state estimation algorithm models are difficult to establish according to the difference of power utilization behaviors caused by different power utilization areas.
In this embodiment, in S5.2, the matrix decomposition is in the mathematical discipline of linear algebra, that is, the matrix is decomposed into the product of matrices, and the specific application method includes:
firstly, through LU decomposition and detection of electricity stealing behaviors, a thief can be determined in a small-scale network, but the thief is unstable in a large-scale network;
secondly, local rotation LU decomposition is improved on the basis, so that thieves can be detected in a large network, but more execution time is needed;
meanwhile, under the condition that data of other electric meters are unknown, each intelligent electric meter calculates the honesty coefficient by utilizing distributed QR (orthogonal triangle) decomposition; and if the integrity coefficient of the normal user is 1 and the actual power consumption of the electricity stealing user is k times of the reported power consumption, k is the integrity coefficient of the user, and whether the electricity stealing behavior exists in the user can be judged according to the k value.
The method is simple and easy to implement, but if a large amount of power consumption data are detected, the detection cost is too high, and the analysis result obtained by using matrix decomposition to perform electricity stealing detection cannot give too much information to inspectors, so that the workload is increased.
In this embodiment, in S5.3, the specific application method of data mining includes the following steps:
s5.3.1, carrying out semi-automatic or automatic analysis on a large amount of data to extract a previously unknown and interesting mode, such as abnormal record (abnormal condition detection);
s5.3.2, decomposing the total power consumption information of the user into information of each power consumption device by adopting non-invasive load monitoring (NILM) which is load decomposition in essence, thereby obtaining the power consumption information such as the consumption condition of the power consumption device, the power consumption behavior of the user and the like;
s5.3.3, collecting network intrusion and physical intrusion logs, and analyzing a power consumption curve of a user;
s5.3.4, listing a list which can be illegal users with the minimum error report quantity based on an attack graph fusion algorithm;
and S5.3.5, finally determining the electricity utilization behavior of the user from the integrated net load curve by adopting non-invasive load monitoring.
Among them, based on data mining technique can be called as non-intrusive load monitoring (NILM) detection electricity stealing, compare with intrusive load monitoring, NILM's economic investment is littleer, and the practicality is stronger.
In this embodiment, in S5.4, the machine learning includes two types, namely, a support vector machine and an artificial neural network, and may be selected according to a specific data distribution situation and an application requirement.
Further, the specific application method of the support vector machine comprises the following steps:
firstly, adopting a method for creating a nonlinear classifier by applying a kernel technique to a maximum edge hyperplane;
secondly, a power stealing detection technology based on classification of a Support Vector Machine (SVM) is used for classifying legal and suspicious samples of a power consumption database to be detected so as to detect power stealing, and the specific method flow comprises the following steps:
firstly training a support vector machine SVW from a historical data set, and then testing the support vector machine under different data sets to find out abnormal electricity utilization of customers;
on the basis of analyzing a support vector machine technology and analyzing the electricity utilization behavior of a user, introducing a 0ne-class SVM algorithm into electricity stealing detection, and providing an electricity stealing detection model formed by combining the curve fluctuation characteristic of electricity consumption and One-class SVM;
the improved power consumption fluctuation coefficient is used for representing the fluctuation condition of the power consumption data of the user, the power consumption fluctuation coefficient is used as an index for judging whether electricity is stolen or not, the index is used as a training sample, a classification model can be obtained after training, and the obtained model is used for analyzing the power consumption data so as to detect whether the power consumption data is abnormal or not.
However, electricity theft detection based on SVW techniques typically requires a large amount of historical power usage data collected from smart meters as training data to obtain characteristics.
Wherein, when creating a nonlinear classifier by applying a kernel technique to a maximum-edge hyperplane, a good separation is achieved by the hyperplane, which is the largest distance from the nearest training data point of any class; and in general, the larger the margin, the smaller the generalization error of the classifier.
Further, the specific application method of the artificial neural network comprises the following steps:
firstly, a plurality of parameters required by a model are automatically set by utilizing a hybrid neural network model and a coding technology, and a hierarchical model for classifying data is provided, so that illegal consumers are identified;
secondly, analyzing a fine-grained load curve of the user smart meter by using a neural network classifier which is searched by using a system including but not limited to a charging system and the like to find a thief;
then, a convolutional neural network model is established based on the periodicity of the power utilization behavior of the user to detect the power utilization condition of the user;
then, by collecting and storing long-time power data and mining the power consumption behavior pattern of the user based on a large amount of data, a normal power consumption behavior model based on a power consumption sample library is established so as to identify the electricity stealing behavior.
Wherein the Artificial Neural Network (ANN) is a computing system designed inspired by biological neural networks constituting the animal brain; and the deep learning is particularly suitable for constructing complex models of mass data and is increasingly popular in data mining/anomaly detection.
In this embodiment, in S5.5, a specific application method of AI identification includes the following steps:
s5.5.1, compiling a computer program for simulating human image recognition activities by using different image recognition models including a template matching model;
s5.5.2, storing face image information of a worker with the authority of operating the terminal equipment in advance;
s5.5.3, distributing a monitoring system in an operation terminal equipment area and loading a corresponding AI identification program;
s5.5.4, acquiring the face image information of the personnel of each operation terminal device from the real-time monitoring system;
and S5.5.5, finally, identifying and confirming whether the personnel operating the terminal equipment is a power stealing personnel through AI.
Among them, AI identification technology is an important field of artificial intelligence.
Furthermore, the template matching model considers that a certain image is recognized, and a memory mode of the image, namely a template, is required in past experience; if the current stimulus matches the template in the brain, the image is identified.
Meanwhile, pattern recognition in image recognition is a process of automatically completing recognition and evaluation on shapes, patterns, curves, numbers, character formats and graphs by using a computer and a mathematical reasoning method on the basis of expert experience and existing recognition from a large amount of information and data.
In this embodiment, in S5.6, a specific application method of the game theory includes the following steps:
firstly, a game model for determining the electricity stealing behavior is based on electricity stealing users and an electric power company;
secondly, establishing a game model of the power company and the electricity stealing users by formulating a statistical anomaly detection scheme; in the game model, the purpose of the power company is to improve profit to the maximum extent and reduce the cost for detecting the electricity stealing users to the greatest extent, and the aim of the electricity stealing users is to reduce the possibility of being detected due to the constraint related to the electricity stealing amount to the greatest extent;
then researching the phenomenon of electricity stealing of the user based on the basic principle of the game theory;
secondly, constructing payment matrixes under different model combinations through particle hypothesis, and finding Nash balance by using a 'drawing bar method';
finally, an effective measure for preventing the user from stealing electricity is obtained, namely: the power company can charge the electricity fee according to the calculation formula of ai + kx (k > 1), so that the user can consciously execute the electricity utilization regulation and legally use the electricity.
The game theory is a research on a mathematical model of strategic interaction among rational decision makers, and the power theft detection technology based on the game theory is a novel power theft detection technology. When there is an objective power stealing behavior, the game-based model can give a stable and reasonable (even if not optimal) solution to reduce power loss.
The embodiment also provides an electricity larceny prevention early warning data analysis system which comprises a data acquisition and storage unit, a data analysis and classification unit, a method matching and calling unit, an early warning data analysis unit and an electricity larceny early warning application unit; the data acquisition and storage unit, the data analysis and classification unit, the method matching and calling unit, the early warning data analysis unit and the electricity stealing early warning application unit are sequentially connected through network communication; wherein:
the data acquisition and storage unit is used for directly acquiring the electricity utilization data in the functional electric meter or importing the electricity utilization data through other data acquisition modules;
the data analysis and classification unit is used for detecting and analyzing the conditions of the composition type/distribution condition, the data volume and the like of the power consumption data and classifying the power consumption data according to a preset rule;
the method matching and calling unit is used for matching and calling an applicable method from a plurality of pre-loaded data analysis methods according to the data condition;
the early warning data analysis unit is used for analyzing and processing the electricity stealing early warning data based on the acquired electricity utilization data by calling an applicable analysis method;
and the electricity stealing early warning application unit is used for further analyzing and formulating a corresponding electricity stealing prevention measure scheme according to the electricity stealing early warning analysis result.
As shown in fig. 5, the present embodiment further provides an operating apparatus of an electricity larceny prevention early warning data analysis system, which includes a processor, a memory, and a computer program stored in the memory and running on the processor.
The processor comprises one or more than one processing core, the processor is connected with the memory through the bus, the memory is used for storing program instructions, and the steps of the electricity larceny prevention early warning data analysis method are realized when the processor executes the program instructions in the memory.
Alternatively, the memory may be implemented by any type or combination of volatile and non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
In addition, the invention also provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps of the electricity larceny prevention early warning data analysis method are realized.
Optionally, the present invention further provides a computer program product containing instructions, which when run on a computer, causes the computer to perform the steps of the above-mentioned electricity larceny prevention early warning data analysis method in each aspect.
It will be understood by those skilled in the art that the processes for implementing all or part of the steps of the above embodiments may be implemented by hardware, or by a program instructing relevant hardware to implement, and the program may be stored in a computer-readable storage medium, where the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and the above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention, and that various changes and modifications may be made without departing from the spirit and scope of the present invention, which fall within the scope of the claimed invention. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. An electricity larceny prevention early warning data analysis method is characterized by comprising the following steps: the method comprises the following steps:
s1, designing and compiling an electricity larceny prevention early warning data analysis system program in mobile computer terminal equipment;
s2, loading a plurality of analysis methods including but not limited to state estimation, matrix decomposition, data mining, machine learning, game theory, AI identification and the like into a system program;
s3, connecting the mobile computer terminal equipment with electric energy data acquisition equipment;
s4, acquiring electric energy data, and analyzing and summarizing the data quantity of the electric energy data, the data composition/distribution condition, whether a large amount of historical data exists, the analysis requirement and other conditions;
s5, selecting and adjusting one or more of the analysis methods to analyze and process data according to the acquired data and analysis requirements;
and S6, early warning the electricity stealing situation according to the data analysis result, and formulating a corresponding electricity stealing prevention scheme.
2. The electricity larceny prevention early warning data analysis method of claim 1, wherein: in S5, according to the acquired data and the analysis requirement, a specific method for tuning one or more of the above analysis methods to perform data analysis processing includes the following steps:
s5.1, when the data volume is small and high-accuracy early warning analysis is required, a state estimation method can be selected for analysis processing;
s5.2, when the data volume is not large, the detection overhead budget is low and the requirement on the detection precision is not high, a matrix decomposition method can be selected and adjusted for analysis;
s5.3, when the data volume is large and the detection investment budget is low, a data mining method can be selected and adjusted for analysis processing;
s5.4, when the data volume is large and the data is complex, a machine learning method can be selected and adjusted for analysis and processing;
s5.5, when a perfect monitoring system exists and a large amount of image data can be obtained, an AI identification method can be selected for analysis and processing;
and S5.6, when the analysis requirement is mainly to reduce the power loss, the method of the game theory can be selected and adjusted to carry out analysis processing and make a corresponding measure scheme.
3. The electricity larceny prevention early warning data analysis method of claim 2, wherein: in S5.1, the application method of the state estimation method includes the following steps:
a state estimation theory, namely a state observer, can be introduced into static safety monitoring of the power system, wherein the state observer is a system for estimating the internal state of a given actual system by measuring the input and the output of the actual system;
a Kalman filtering algorithm can be introduced to establish a dynamic state estimation theory, meanwhile, a solution scheme based on state estimation can be realized by introducing some integrated distributed state estimation techniques, and the state estimation is carried out by a weighted least square method, so that the deviation obtained by the electricity consumption data is compared with the actual deviation, and a thief is obtained by analysis;
decomposing the power consumption into two parallel loose coupling filters through a Kalman filter, carrying out deviation estimation on the power consumption to carry out electricity stealing detection analysis, utilizing AMI and SCADA measurement, adopting an integrated distribution state estimation method to analyze MW and LV distribution networks, and analyzing the power consumption behavior of the intelligent power grid from reliable measurement data which can be recorded by the intelligent power meter;
and integrally fitting the estimated value of the injection power consumption of the pseudo feeder bus according to the power consumption data of the users on the distribution transformer, creating a suspicious user list by using variance analysis after the state estimation result, evaluating the use condition of the power consumption data and estimating the actual use condition.
4. The electricity larceny prevention early warning data analysis method of claim 2, wherein: in S5.2, the matrix decomposition is in the mathematical discipline of linear algebra, that is, the matrix is decomposed into the product of matrices, and the specific application method includes:
firstly, electricity stealing behavior is detected through LU decomposition, and a thief can be determined in a small-scale network;
secondly, local rotation LU decomposition is improved on the basis, so that thieves can be detected in a large network;
meanwhile, under the condition that data of other electric meters are not known, each intelligent electric meter calculates the honest coefficient by utilizing distributed QR decomposition; and if the integrity coefficient of the normal user is 1 and the actual power consumption of the electricity stealing user is k times of the reported power consumption, k is the integrity coefficient of the user, and whether the electricity stealing behavior exists in the user can be judged according to the k value.
5. The electricity larceny prevention early warning data analysis method of claim 2, wherein: in S5.3, the specific application method of data mining includes the following steps:
s5.3.1, carrying out semi-automatic or automatic analysis on a large amount of data to extract an unknown and interesting mode in the past;
s5.3.2, decomposing the total power consumption information of the user into information of each power consumption device by adopting non-invasive load monitoring which is essentially load decomposition, thereby obtaining the power consumption information such as the consumption condition of the power consumption device, the power consumption behavior of the user and the like;
s5.3.3, collecting network intrusion and physical intrusion logs, and analyzing a power consumption curve of a user;
s5.3.4, listing a list which can be illegal users with the minimum error report quantity based on an attack graph fusion algorithm;
and S5.3.5, determining the electricity utilization behavior of the user from the integrated net load curve by adopting non-intrusive load monitoring.
6. The electricity larceny prevention early warning data analysis method of claim 2, wherein: in S5.4, machine learning comprises a support vector machine and an artificial neural network, and selection can be performed according to specific data distribution conditions and application requirements.
7. The electricity larceny prevention early warning data analysis method of claim 6, wherein: the specific application method of the support vector machine comprises the following steps:
firstly, adopting a method for creating a nonlinear classifier by applying a kernel technique to a maximum edge hyperplane;
secondly, classifying legal and suspicious samples of the power consumption database to be detected by using a classification-based electricity stealing detection technology of a support vector machine to detect electricity stealing, wherein the specific method flow comprises the following steps:
firstly training a support vector machine SVW from a historical data set, and then testing the support vector machine under different data sets to find out abnormal electricity utilization of customers;
on the basis of analyzing a support vector machine technology and analyzing the electricity utilization behavior of a user, introducing a 0ne-class SVM algorithm into electricity stealing detection, and providing an electricity stealing detection model formed by combining the curve fluctuation characteristic of electricity consumption and One-class SVM;
the improved power consumption fluctuation coefficient is used for representing the fluctuation condition of the power consumption data of the user, the power consumption fluctuation coefficient is used as an index for judging whether electricity is stolen or not, the index is used as a training sample, a classification model can be obtained after training, and the power consumption data is analyzed from the obtained model so as to detect whether the power consumption data is abnormal or not.
8. The electricity larceny prevention early warning data analysis method of claim 6, wherein: the specific application method of the artificial neural network comprises the following steps:
firstly, a plurality of parameters required by a model are automatically set by using a hybrid neural network model and a coding technology, and a hierarchical model for classifying data is provided, so that illegal consumers are identified;
secondly, analyzing a fine-grained load curve of the user smart meter by using a neural network classifier which is searched by using a system including but not limited to a charging system and the like to find a thief;
then, a convolutional neural network model is established based on the periodicity of the power utilization behavior of the user to detect the power utilization condition of the user;
then, by collecting and storing long-time power data and mining the power consumption behavior pattern of the user based on a large amount of data, a normal power consumption behavior model based on a power consumption sample library is established so as to identify the electricity stealing behavior.
9. The electricity larceny prevention early warning data analysis method as claimed in claim 2, wherein: in S5.5, the specific application method of AI identification includes the following steps:
s5.5.1, compiling a computer program for simulating human image recognition activities by using different image recognition models including a template matching model;
s5.5.2, face image information of workers with the authority of operating the terminal equipment is stored in advance;
s5.5.3, distributing a monitoring system in an operation terminal equipment area and loading a corresponding AI identification program;
s5.5.4, acquiring the face image information of each person operating the terminal equipment from the real-time monitoring system;
and S5.5.5, finally, identifying and confirming whether the personnel operating the terminal equipment is a power stealing personnel through AI.
10. The electricity larceny prevention early warning data analysis method of claim 2, wherein: in S5.6, the specific application method of the game theory includes the following steps:
firstly, a game model for determining the electricity stealing behavior is based on electricity stealing users and an electric power company;
secondly, establishing a game model of the power company and the electricity stealing users by formulating a statistical anomaly detection scheme; wherein, in the game model, the purpose of the power company is to improve profit to the utmost extent and reduce the cost for detecting the electricity stealing users as much as possible, and the goal of the electricity stealing users is to reduce the possibility of being detected due to the restriction related to the electricity stealing amount as much as possible;
researching the phenomenon of electricity stealing of the user based on the basic principle of the game theory;
secondly, constructing payment matrixes under different model combinations through particle hypothesis, and finding Nash equilibrium by using a 'drawing bar method';
finally, an effective measure for preventing the user from stealing electricity is obtained, namely: the electric power company can charge the electric power fee according to the calculation formula of ai + kx (k > 1).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210339122.8A CN115310982A (en) | 2022-04-01 | 2022-04-01 | Electricity larceny prevention early warning data analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210339122.8A CN115310982A (en) | 2022-04-01 | 2022-04-01 | Electricity larceny prevention early warning data analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115310982A true CN115310982A (en) | 2022-11-08 |
Family
ID=83854345
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210339122.8A Pending CN115310982A (en) | 2022-04-01 | 2022-04-01 | Electricity larceny prevention early warning data analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115310982A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117421610A (en) * | 2023-12-19 | 2024-01-19 | 山东德源电力科技股份有限公司 | Data anomaly analysis method for electric energy meter running state early warning |
-
2022
- 2022-04-01 CN CN202210339122.8A patent/CN115310982A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117421610A (en) * | 2023-12-19 | 2024-01-19 | 山东德源电力科技股份有限公司 | Data anomaly analysis method for electric energy meter running state early warning |
| CN117421610B (en) * | 2023-12-19 | 2024-03-15 | 山东德源电力科技股份有限公司 | Data anomaly analysis method for electric energy meter running state early warning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110097297B (en) | Multi-dimensional electricity stealing situation intelligent sensing method, system, equipment and medium | |
| Zheng et al. | A novel combined data-driven approach for electricity theft detection | |
| Angelos et al. | Detection and identification of abnormalities in customer consumptions in power distribution systems | |
| Xia et al. | Detection methods in smart meters for electricity thefts: A survey | |
| Jiang et al. | Wavelet based feature extraction and multiple classifiers for electricity fraud detection | |
| Zheng et al. | Electricity theft detecting based on density-clustering method | |
| CN111178396A (en) | Method and device for identifying abnormal electricity consumption user | |
| Feng et al. | Data mining for abnormal power consumption pattern detection based on local matrix reconstruction | |
| Shehzad et al. | A robust hybrid deep learning model for detection of non-technical losses to secure smart grids | |
| Babu et al. | Detecting unusual customer consumption profiles in power distribution systems—APSPDCL | |
| CN115222303B (en) | Industry risk data analysis method and system based on big data and storage medium | |
| CN105574666A (en) | Method and device for evaluating credit level of enterprise based on key data modeling | |
| CN111612371A (en) | Intelligent ammeter quality evaluation method based on analytic hierarchy process | |
| CN113256324A (en) | Data asset pricing method, device, computer equipment and storage medium | |
| Xia et al. | A control-chart-based detector for small-amount electricity theft (SET) attack in smart grids | |
| Yip et al. | Detection of energy theft and metering defects in advanced metering infrastructure using analytics | |
| CN116668039A (en) | Computer remote login identification system and method based on artificial intelligence | |
| CN115310982A (en) | Electricity larceny prevention early warning data analysis method | |
| CN112614004A (en) | Method and device for processing power utilization information | |
| CN115965237A (en) | Electric power market risk prevention and control method, device and storage medium | |
| CN107274025B (en) | System and method for realizing intelligent identification and management of power consumption mode | |
| CN112511503A (en) | Method, system, equipment and storage medium for improving safety of intelligent electric meter | |
| CN116561659A (en) | Method, device and system for detecting electricity stealing behavior | |
| Emadaleslami et al. | A Machine Learning Approach to Detect Energy Fraud in Smart Distribution Network | |
| CN115166625A (en) | Intelligent ammeter error estimation method and device |
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 |