CN111967512A - Abnormal electricity utilization detection method, system and storage medium - Google Patents
Abnormal electricity utilization detection method, system and storage medium Download PDFInfo
- Publication number
- CN111967512A CN111967512A CN202010787989.0A CN202010787989A CN111967512A CN 111967512 A CN111967512 A CN 111967512A CN 202010787989 A CN202010787989 A CN 202010787989A CN 111967512 A CN111967512 A CN 111967512A
- Authority
- CN
- China
- Prior art keywords
- data
- electricity utilization
- abnormal
- abnormal electricity
- electric meter
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R11/00—Electromechanical arrangements for measuring time integral of electric power or current, e.g. of consumption
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R21/00—Arrangements for measuring electric power or power factor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R21/00—Arrangements for measuring electric power or power factor
- G01R21/001—Measuring real or reactive component; Measuring apparent energy
- G01R21/002—Measuring real component
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an abnormal electricity utilization detection method, system and storage medium, and belongs to the technical field of identification of abnormal electricity utilization behaviors of a smart grid. The abnormal electricity utilization detection method comprises the following steps: acquiring data of the intelligent ammeter to be detected; converting the data of the intelligent electric meter into multi-time sequence data; inputting the multi-time sequence data into a trained abnormal electricity utilization detection model to obtain a predicted electricity utilization type label; and taking the power utilization type with the highest probability in the predicted power utilization type labels as an abnormal power utilization detection result of the data of the intelligent electric meter to be detected. The invention can fully and comprehensively utilize the big data of the power utilization, deeply excavate the high-level characteristics of the abnormal power utilization behavior, improve the generalization performance of the abnormal power utilization detection, reduce the misjudgment proportion and efficiently and accurately detect the abnormal power utilization.
Description
Technical Field
The invention relates to an abnormal power utilization detection method, system and storage medium, and belongs to the technical field of identification of abnormal power utilization behaviors of a smart grid.
Background
For a long time, the abnormal electricity utilization such as electricity stealing, meter failure and installation error brings huge economic loss to the power grid operator every year. Moreover, because of a huge error in the collected power consumption data, the scheduling and management of the power grid and the operation safety are also affected. Therefore, the abnormal electricity utilization detection is one of important supports for safety electricity utilization in the operation and maintenance process of the smart grid, and has very important significance. By discriminating whether the electricity consumption behavior is normal or not, it is possible to compensate for a small amount of electricity charges due to abnormal electricity consumption. In addition, the distorted power consumption data can be corrected, the quality of power grid data is improved, and a guarantee is provided for more extensive power big data analysis. At present, industrial and commercial users in main countries of the world including China occupy a very large proportion of the electricity consumption in the whole society, and the proportion in China is over 80 percent, so that the detection of abnormal electricity consumption in the industrial and commercial users is very important.
More and more researchers are eager to search for abnormal electricity utilization behaviors by means of a big data analysis-based method and technical means such as statistics, data mining and machine learning. Compared with a hardware-based solution, the method has the advantages of high efficiency, high detection speed and the like; moreover, as the smart meter data accumulated by the power department is more and more, such a method gradually becomes mainstream. The abnormal electricity utilization detection method based on data analysis mainly depends on an implementation mode of artificial feature modeling and discrimination algorithm design; especially, the performance of the characteristic model directly determines the quality of the detection method. Often, the generalization capability of the feature model is insufficient, resulting in a large number of false positives. At present, the defects of the artificial feature modeling are expressed in the following aspects:
1) the data phenomenon of the abnormal electricity utilization behavior has great diversity, the data scale of the intelligent electric meter is extremely large, and all possible data phenomena cannot be traversed at all when the data characteristics are designed manually. Artificially designed data features are often based on a small amount of data, resulting in gradual degradation of the accuracy of the detection model over time;
2) the data characteristics of abnormal electricity utilization are various, and the abnormal electricity utilization data has short-term characteristics and long-term characteristics; both local features and global features; manually designing a characteristic model with excellent performance in a short time difficultly;
3) a large amount of various noises are contained in the data of the intelligent electric meter, so that human experts are interfered for data analysis and feature modeling;
4) the artificially designed data features often have strong limitations, are directly related to the technical level, experience and the like of human experts, and lead to strong subjectivity and poor scalability of the feature model, and cannot meet the application requirements.
Therefore, the traditional abnormal electricity utilization detection method cannot meet the operation, maintenance and development requirements of the smart power grid. How to fully utilize the big data of the power consumption, deeply mine the high-level characteristics of the abnormal power consumption behavior, and efficiently and accurately detect the abnormal power consumption becomes the main trend of the development of the smart grid.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a method, a system and a storage medium for detecting abnormal electricity utilization, which can fully and comprehensively utilize big data of electricity utilization, deeply excavate high-level characteristics of abnormal electricity utilization behaviors, improve generalization performance of abnormal electricity utilization detection, reduce misjudgment proportion and efficiently and accurately detect abnormal electricity utilization.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
in a first aspect:
an abnormal electricity utilization detection method comprises the following steps:
acquiring data of the intelligent ammeter to be detected;
converting the data of the intelligent electric meter into multi-time sequence data;
inputting the multi-time sequence data into a trained abnormal electricity utilization detection model to obtain a predicted electricity utilization type label;
and taking the power utilization type with the highest probability in the predicted power utilization type labels as an abnormal power utilization detection result of the data of the intelligent electric meter to be detected.
Further, the method for converting the data of the intelligent electric meter into the multi-time sequence data comprises the following steps:
converting the data of the intelligent electric meter into a multi-time sequence data form, and constructing and attaching a data quality sequence and a date type sequence;
filling missing data into the obtained sequence data, dividing the data according to a natural circumference, and constructing the data into multi-time sequence data;
and carrying out independent data normalization processing on the multi-time sequence data.
Further, the method for constructing the data quality sequence comprises the following steps:
and counting the missing number of the data of the intelligent electric meter according to the time stamp, endowing the counted missing number with the corresponding time stamp, and constructing a data quality sequence.
Further, the method of constructing a date type sequence includes the steps of:
and according to the attribute of the working day or the holiday of each natural day, giving date type data of one hour to each natural day to construct a date type sequence.
Further, the data normalization processing adopts the following classification normalization mode;
the voltage data were normalized as follows:
the current data were normalized as follows:
the normalization of the total active power is as follows:
the data quality was normalized as follows:
further, the training method of the abnormal electricity utilization detection model comprises the following steps:
acquiring intelligent electric meter data of normal electricity utilization cases and abnormal electricity utilization cases, and converting the intelligent electric meter data into multi-time sequence data samples;
calibrating the electricity type label of the multi-time sequence data sample, and dividing the electricity type label into a training sample set and a verification sample set;
randomly sampling a batch of data samples from a training sample set, inputting the data samples into a deep hybrid neural network, calculating loss and optimizing parameters of the deep hybrid neural network;
traversing the whole training sample set by using a parameter optimized deep hybrid neural network to complete a round of training;
after each training round is finished, evaluating the training effect of the deep hybrid neural network by using the verification sample set, and storing the abnormal power utilization detection model snapshot;
repeating the training to reach the preset number of rounds, so that the loss of the deep hybrid neural network tends to be stable, and finishing the training;
and determining the trained abnormal electricity utilization detection model according to the evaluation result after each round of training.
Further, the method for acquiring the power utilization type label comprises the following steps:
extracting local features of the data of the intelligent electric meter;
extracting global features of the data of the intelligent electric meter;
and splicing and combining the local features and the global features, and converting the local features and the global features into power utilization type labels.
Further, the data structure sequence of the multi-time sequence data sequentially comprises an A phase voltage, a B phase voltage, a C phase voltage, an A phase current, a B phase current, a C phase current, total active power, a total power factor, data quality and a date type.
In a second aspect:
an abnormal electricity usage detection system, the system comprising the following modules:
the data acquisition module is used for acquiring data of the intelligent ammeter to be detected;
the data conversion module is used for converting the data of the intelligent electric meter into multi-time sequence data;
the abnormal electricity utilization detection model is used for obtaining a predicted electricity utilization type label according to the input multi-time sequence data;
and the detection output module is used for taking the electricity utilization type with the highest probability in the predicted electricity utilization type labels as an abnormal electricity utilization detection result of the data of the intelligent electric meter to be detected.
Further, the abnormal electricity utilization detection model adopts a deep hybrid neural network, and the deep hybrid neural network comprises a local feature network, a global feature network and a classification network;
the local feature network is used for extracting local features of the data of the intelligent electric meter;
the global feature network is used for extracting global features of the data of the intelligent electric meter;
and the classification network is used for splicing and combining the local features and the global features and converting the local features and the global features into power utilization type labels.
Further, the local feature network comprises a plurality of cascaded one-dimensional convolution modules and a Flatten layer;
the one-dimensional convolution module is used for learning and extracting local features of the data of the intelligent ammeter; the Flatten layer is used for expanding the local features extracted by the one-dimensional convolution module and outputting the local features as one-dimensional vectors;
the one-dimensional convolution module comprises a one-dimensional convolution layer and a one-dimensional average pooling layer;
the one-dimensional convolution layer is used for keeping the dimension of the output time axis consistent with the dimension of the input time axis;
and the one-dimensional average pooling layer is used for reducing the dimension of the input time axis on the time axis by adopting an average operator.
Further, the global feature network comprises a Flatten layer and a plurality of cascaded fully connected layers;
the Flatten layer is used for unfolding the data of the intelligent electric meter with multiple time sequences into one-dimensional vectors;
and the full connection layer is used for learning and extracting the global features of the intelligent electric meter data from the one-dimensional vectors output by the Flatten layer of the global feature network.
Further, the classification network comprises a splicing layer and a plurality of cascaded fully-connected layers;
the splicing layer is used for splicing the local features and the global features output by the local feature network and the global feature network into a one-dimensional feature vector and inputting the one-dimensional feature vector into a plurality of stacked full-connection layers;
and the full connection layer is used for calculating and outputting type labels of normal electricity utilization and abnormal electricity utilization.
In a third aspect:
an abnormal electricity utilization detection system comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate according to the instructions to perform the steps of any one of the abnormal electricity usage detection methods according to the first aspect.
In a fourth aspect:
a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of any one of the abnormal electricity usage detection methods of the first aspect.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides an abnormal electricity utilization detection method based on a deep hybrid neural network, which can more comprehensively extract the high-level characteristics of electricity utilization behaviors by learning the local characteristics and the global characteristics of intelligent electric meter data, thereby accurately identifying the abnormal electricity utilization behaviors, and better ensuring the long-term stability of the abnormal electricity utilization detection performance and the generalization capability under different application scenes;
by adopting multi-time-series data and taking a natural week as unit time span, the periodicity of the power utilization behavior of the user can be better described; by supplementing the date type and the data quality, the information quantity of the data sample is further enriched, so that the interference of data missing on abnormal power utilization identification is reduced, the influence of dynamic adjustment on abnormal power utilization detection in working days and holidays is improved, the abnormal power utilization detection performance is more stable, the misjudgment rate is reduced, and the operation efficiency of an electric power company is improved.
Drawings
Fig. 1 is a flowchart of an abnormal electricity usage detection method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for constructing multi-time series data according to an embodiment of the present invention;
FIG. 3 is a block diagram of multi-time series data according to an embodiment of the present invention;
FIG. 4 is a flow chart illustrating the training of an abnormal electricity usage detection model according to an embodiment of the present invention;
FIG. 5 is a flow chart for obtaining a power usage type tag according to an embodiment of the invention;
fig. 6 is a schematic diagram of a framework of an abnormal electricity usage detection system according to an embodiment of the present invention;
FIG. 7 is a block diagram of an abnormal electricity usage detection model according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a deep hybrid neural network according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The first embodiment is as follows:
as shown in fig. 1, based on the requirements of efficiently and accurately detecting abnormal power consumption and operation, maintenance and development of the smart grid, the embodiment of the present invention provides an abnormal power consumption detection method, which includes the following steps:
and 104, identifying the power utilization type with the highest probability in the predicted power utilization type labels as an abnormal power utilization detection result of the data of the intelligent electric meter to be detected.
The modern smart power grid collects massive smart meter data, and the data have the characteristics of large scale, high latitude, various types and the like. Aiming at the characteristics of the data of the intelligent electric meter and aiming at improving the data quality of the intelligent electric meter of the power grid, the embodiment of the invention provides a method for constructing multi-time-series data as shown in fig. 2, namely a method for converting the data of the intelligent electric meter into the multi-time-series data, which specifically comprises the following steps:
the normalization mode of the three-phase current data is as follows:
the normalization mode of the total active power is as follows:
wherein, the standard range of the total power factor is between [0, 1], so normalization processing is not needed;
the normalization mode of the data quality is as follows:
wherein, the numerical value of the date type data only comprises 0 and 1, so the normalization processing is not needed.
Specifically, as shown in fig. 3, an embodiment of the present invention provides a multi-channel time series data structure, where the data structure includes: 10 time sequences such as A phase voltage, B phase voltage, C phase voltage, A phase current, B phase current, C phase current, total active power, total power factor, data quality, date type and the like; the data acquisition frequency is 1 data/hour, and the time stamps of all channels are strictly aligned; by supplementing the date type and data quality, the information content of the data sample is further enriched.
As shown in fig. 3, the multi-channel time series is arranged in the following order: phase-A voltage, phase-B voltage, phase-C voltage, phase-A current, phase-B current, phase-C current, total active power, total power factor, data quality, and date type. The time span of the multi-channel time series data sample structure is a natural week, starts from Monday 0 and ends at Sunday 23.
Specifically, as shown in fig. 4, in order to improve the stability and reliability of the abnormal electricity detection model, the training method for the abnormal electricity detection model provided in the embodiment of the present invention specifically includes the following steps:
409, if the number of training rounds has reached the preset maximum number of training rounds, executing 410, otherwise executing 403, wherein the maximum number of training rounds is optionally 50 as an embodiment of the present invention;
and step 410, finishing the training, and taking the stored optimal snapshot file as a trained abnormal electricity utilization detection model.
The trained abnormal electricity utilization detection model can extract the advanced characteristics of electricity utilization behaviors more comprehensively by learning the local characteristics and the global characteristics of the data of the intelligent electric meter, and has higher abnormal electricity utilization evaluation efficiency and higher abnormal electricity utilization detection accuracy, so that a large amount of resources such as manpower and material resources are saved.
Specifically, as shown in fig. 5, an embodiment of the present invention provides a method for acquiring an electricity consumption type tag, which specifically includes the following steps:
and 503, splicing and combining the local features and the global features, and converting the local features and the global features into power utilization type labels.
It should be noted that, in the electricity consumption type tag obtaining process, there is no strict sequential limitation on the extraction of the local features and the global features of the smart meter data, as long as the relevant features can be extracted smoothly.
As can be seen from the above, the abnormal electricity utilization detection method provided in the embodiments of the present invention can extract advanced features (local features and global features) of electricity utilization behavior more comprehensively, reduce interference of data loss on abnormal electricity utilization identification, improve influence of dynamic adjustment of working days and holidays on abnormal electricity utilization detection, ensure more stable abnormal electricity utilization detection performance, reduce misjudgment rate, and improve operation efficiency of electric power companies and generalization capability in different application scenarios.
Example two:
as shown in fig. 6, an embodiment of the present invention provides an abnormal electricity usage detection system, which includes the following modules:
the data acquisition module is used for acquiring data of the intelligent ammeter to be detected;
the data conversion module is used for converting the data of the intelligent electric meter into multi-time sequence data;
the abnormal electricity utilization detection model is used for obtaining a predicted electricity utilization type label according to the input multi-time sequence data;
and the detection output module is used for taking the electricity utilization type with the highest probability in the predicted electricity utilization type labels as an abnormal electricity utilization detection result of the data of the intelligent electric meter to be detected.
Specifically, as shown in fig. 7 and 8, the abnormal electricity utilization detection model of the abnormal electricity utilization detection system employs a deep hybrid neural network, wherein the deep hybrid neural network includes a local feature network, a global feature network, and a classification network;
the local feature network is used for extracting local features of the data of the intelligent electric meter; the global feature network is used for extracting local features of the data of the intelligent electric meter; the classification network is used for splicing and combining the local features and the global features and converting the local features and the global features into power utilization type labels.
Specifically, the local feature network processes the data of the intelligent electric meter along a time axis based on a plurality of cascaded one-dimensional convolution modules, and learns and extracts local features of the data of the intelligent electric meter by using local experience characteristics of the one-dimensional convolution units; and before output, a Flatten layer is used for unfolding local features output by the one-dimensional convolution module into one-dimensional vectors. The one-dimensional convolution module comprises a one-dimensional convolution layer and a one-dimensional average pooling layer, the dimension of a time axis output by the one-dimensional convolution layer is consistent with that of an input, and the dimension of the input is reduced on the time axis by the one-dimensional average pooling layer through an average operator.
As shown in fig. 8, in the present embodiment, the local feature network includes 3 cascaded 1-dimensional convolution modules and one scatter layer. Wherein, the convolution kernel size of the 1-dimensional convolution layer-1 of the 1-dimensional convolution module 1 is 1 x 5, the filter number is 16, and the pooling kernel size of the 1-dimensional average pooling layer-1 is 1 x 4; the convolution kernel size of the 1-dimensional convolution layer-2 of the 1-dimensional convolution module 2 is 1 x 3, the filter number is 32, and the pooling kernel size of the 1-dimensional average pooling layer-2 is 1 x 3; the convolution kernel size of the 1-dimensional convolution layer-3 of the 1-dimensional convolution module 3 is 1 x 3, the filter count is 64, and the pooling kernel size of the 1-dimensional average pooling layer-3 is 1 x 2; and the activation functions of the 3 1-dimensional convolutional layers are all relus.
Specifically, the global feature network comprises one Flatten layer and a plurality of cascaded fully connected layers. Firstly, unfolding multi-time-series intelligent electric meter data into a one-dimensional vector by a Flatten layer, then inputting the one-dimensional vector into a stacked full-connection layer, and learning and extracting global features of the intelligent electric meter data by utilizing the characteristic that each neuron of the full-connection layer is connected with the input.
As shown in fig. 8, in this embodiment, the global feature network includes one flat layer and 2 cascaded fully connected layers. Wherein the number of the neurons of the full connection layer-1 is 1000 and the number of the neurons of the full connection layer-2 is 500; and the activation functions of the 2 fully connected layers are all relus.
Specifically, the classification network is composed of a splicing layer and a plurality of cascaded fully-connected layers. The splicing layer splices the local features and the global features output by the local feature network and the global feature network into a larger one-dimensional feature vector, then inputs the larger one-dimensional feature vector into a plurality of stacked full-connection layers, and finally calculates and outputs the type labels of normal electricity and abnormal electricity and obtains the probability of each type label.
As shown in fig. 8, in the present embodiment, the classification network includes one splice layer and 2 cascaded fully-connected layers. Wherein, the number of the neurons of the full connection layer-3 is 1000 and the number of the neurons of the full connection layer-4 is 100. The activation function of the fully-connected layer-3 is ReLU, and the activation function of the fully-connected layer-4 is Softmax.
From the above, except that the activation function of the last fully-connected layer of the classification network is the Softmax function, the activation functions of the one-dimensional convolution layers and the fully-connected layers of all the other networks are the ReLU functions.
The abnormal electricity utilization detection system and the abnormal electricity utilization detection model thereof provided by the embodiment of the invention can learn local characteristics and global characteristics of intelligent electric meter data, and extract advanced characteristics of electricity utilization behaviors more comprehensively, so that the abnormal electricity utilization behaviors are accurately identified, and the long-term stability of the abnormal electricity utilization detection performance and the generalization capability under different application scenes can be better ensured.
Example three:
the method for detecting abnormal electricity consumption provided by the invention is explained by specific experimental examples as follows:
training and verification are carried out by taking the intelligent electric meter data of 461 three-phase four-wire 10kV users of a certain power-saving company as cases.
The abnormal electricity utilization detection method as shown in the figure 1 provided by the invention is adopted to obtain the data of the intelligent electric meter and convert the data into a multi-time series data form; the data acquisition frequency of each user is 96 acquisition points/day, and each data acquisition point comprises eight items of data such as A phase voltage, B phase voltage, C phase voltage, A phase current, B phase current, C phase current, total active power, total power factor and the like.
The data sample structure shown in fig. 3 is obtained by adopting the data sample construction method shown in fig. 2 provided by the invention, the data is down-sampled, the frequency is changed into 24 acquisition points/day, and the data items are unchanged; after the data of all users are divided, 51289 samples are obtained, wherein 35506 normal electricity utilization samples and 15783 abnormal electricity utilization samples are obtained.
The method comprises the steps of adopting a structure of a deep hybrid neural network shown in fig. 8, and carrying out detection model training according to a training method shown in fig. 4, wherein the size of a training sample batch is 200 samples/batch, the learning rate is fixed to be 0.001, an optimizer is Adam, and the training round is 50 rounds, so that a well-trained abnormal electricity utilization detection model is obtained.
The abnormal electricity utilization detection model provided by the invention and shown in fig. 7 is subjected to 5-fold cross validation, and each fold of training sample and test sample classification is divided into 80% and 20% of the normal electricity utilization classification and the abnormal electricity utilization classification. The 5-fold cross validation performance of the abnormal electricity utilization detection model provided by the invention is shown in the following table:
as can be seen from the results in the above table, the abnormal electricity usage detection result performed by the abnormal electricity usage detection method according to the first embodiment and the abnormal electricity usage detection system according to the second embodiment has a better accuracy in identifying abnormal electricity usage, and shows stability of the identification performance in the 5-fold cross validation experiment, thereby obtaining a better effect overall.
Example four:
the embodiment of the invention provides an abnormal electricity utilization detection system, which comprises a processor and a storage medium, wherein the processor is used for processing abnormal electricity utilization;
the storage medium is used for storing instructions;
the processor is used for operating according to the instruction to execute any one of the abnormal electricity utilization detection method according to the first embodiment.
Example five:
the embodiment of the invention provides a computer-readable storage medium, on which a computer program is stored, and the computer program is used for implementing the steps of any abnormal electricity utilization detection method of the first embodiment when being executed by a processor.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (15)
1. An abnormal electricity consumption detection method, characterized by comprising the steps of:
acquiring data of the intelligent ammeter to be detected;
converting the data of the intelligent electric meter into multi-time sequence data;
inputting the multi-time sequence data into a trained abnormal electricity utilization detection model to obtain a predicted electricity utilization type label;
and taking the power utilization type with the highest probability in the predicted power utilization type labels as an abnormal power utilization detection result of the data of the intelligent electric meter to be detected.
2. The abnormal electricity utilization detection method according to claim 1, wherein the method for converting the smart meter data into the multi-time series data comprises the steps of:
converting the data of the intelligent electric meter into a multi-time sequence data form, and constructing and attaching a data quality sequence and a date type sequence;
filling missing data into the obtained sequence data, dividing the data according to a natural circumference, and constructing the data into multi-time sequence data;
and carrying out independent data normalization processing on the multi-time sequence data.
3. The abnormal electricity utilization detection method according to claim 2, wherein the method of constructing the data quality sequence comprises the steps of:
and counting the missing number of the data of the intelligent electric meter according to the time stamp, endowing the counted missing number with the corresponding time stamp, and constructing a data quality sequence.
4. The abnormal electricity usage detection method according to claim 2, wherein the method of constructing a date type sequence includes the steps of:
and according to the attribute of the working day or the holiday of each natural day, giving one-per-hour date type data to the natural day to construct a date type sequence.
5. The abnormal electricity utilization detection method according to claim 2, wherein the data normalization processing is performed in a classified normalization manner;
the voltage data were normalized as follows:
the current data were normalized as follows:
the normalization of the total active power is as follows:
the data quality was normalized as follows:
6. the abnormal electricity usage detection method according to claim 1, wherein the training method of the abnormal electricity usage detection model includes the steps of:
acquiring intelligent electric meter data of normal electricity utilization cases and abnormal electricity utilization cases, and converting the intelligent electric meter data into multi-time sequence data samples;
calibrating the electricity type label of the multi-time sequence data sample, and dividing the electricity type label into a training sample set and a verification sample set;
randomly sampling a batch of data samples from a training sample set, inputting the data samples into a deep hybrid neural network, calculating loss and optimizing parameters of the deep hybrid neural network;
traversing the whole training sample set by using a parameter optimized deep hybrid neural network to complete a round of training;
after each training round is finished, evaluating the training effect of the deep hybrid neural network by using the verification sample set, and storing the abnormal power utilization detection model snapshot;
repeating the training to reach the preset number of rounds, so that the loss of the deep hybrid neural network tends to be stable, and finishing the training;
and determining the trained abnormal electricity utilization detection model according to the evaluation result after each round of training.
7. The abnormal electricity usage detection method according to claim 1, wherein the electricity usage type tag acquisition method includes the steps of:
extracting local features of the data of the intelligent electric meter;
extracting global features of the data of the intelligent electric meter;
and splicing and combining the local features and the global features, and converting the local features and the global features into power utilization type labels.
8. The abnormal electricity usage detection method according to any one of claims 1 to 7, wherein the data structure sequence of the multi-time series data includes A-phase voltage, B-phase voltage, C-phase voltage, A-phase current, B-phase current, C-phase current, total active power, total power factor, data quality, and date type.
9. An abnormal electricity utilization detection system, characterized in that the system comprises the following modules:
the data acquisition module is used for acquiring data of the intelligent ammeter to be detected;
the data conversion module is used for converting the data of the intelligent electric meter into multi-time sequence data;
the abnormal electricity utilization detection model is used for obtaining a predicted electricity utilization type label according to the input multi-time sequence data;
and the detection output module is used for taking the electricity utilization type with the highest probability in the predicted electricity utilization type labels as an abnormal electricity utilization detection result of the data of the intelligent electric meter to be detected.
10. The abnormal electricity utilization detection system of claim 9, wherein the abnormal electricity utilization detection model employs a deep hybrid neural network, the deep hybrid neural network comprising a local feature network, a global feature network, and a classification network;
the local feature network is used for extracting local features of the data of the intelligent electric meter;
the global feature network is used for extracting global features of the data of the intelligent electric meter;
and the classification network is used for splicing and combining the local features and the global features and converting the local features and the global features into electricity utilization type labels.
11. The abnormal electricity utilization detection system according to claim 10, wherein the local feature network comprises a plurality of cascaded one-dimensional convolution modules and a Flatten layer;
the one-dimensional convolution module is used for learning and extracting local features of the data of the intelligent ammeter; the Flatten layer is used for expanding the local features extracted by the one-dimensional convolution module and outputting the local features as one-dimensional vectors;
the one-dimensional convolution module comprises a one-dimensional convolution layer and a one-dimensional average pooling layer;
the one-dimensional convolution layer is used for keeping the dimension of the output time axis consistent with the dimension of the input time axis;
and the one-dimensional average pooling layer is used for reducing the dimension of the input time axis on the time axis by adopting an average operator.
12. The abnormal electricity utilization detection system according to claim 10, wherein the global feature network comprises a scatter layer and a plurality of cascaded fully-connected layers;
the Flatten layer is used for unfolding the data of the intelligent electric meter with multiple time sequences into one-dimensional vectors;
and the full connection layer is used for learning and extracting the global features of the intelligent electric meter data from the one-dimensional vectors output by the Flatten layer of the global feature network.
13. The abnormal electricity usage detection system of claim 10, wherein the classification network includes a splice layer and a plurality of cascaded fully-connected layers;
the splicing layer is used for splicing the local features and the global features output by the local feature network and the global feature network into a one-dimensional feature vector and inputting the one-dimensional feature vector into a plurality of stacked full-connection layers;
and the full connection layer is used for calculating and outputting type labels of normal electricity utilization and abnormal electricity utilization.
14. An abnormal electricity utilization detection system is characterized by comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any one of claims 1 to 8.
15. Computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010787989.0A CN111967512B (en) | 2020-08-07 | 2020-08-07 | Abnormal electricity utilization detection method, system and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010787989.0A CN111967512B (en) | 2020-08-07 | 2020-08-07 | Abnormal electricity utilization detection method, system and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111967512A true CN111967512A (en) | 2020-11-20 |
CN111967512B CN111967512B (en) | 2022-08-19 |
Family
ID=73364876
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010787989.0A Active CN111967512B (en) | 2020-08-07 | 2020-08-07 | Abnormal electricity utilization detection method, system and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111967512B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112200160A (en) * | 2020-12-02 | 2021-01-08 | 成都信息工程大学 | Deep learning-based direct-reading water meter reading identification method |
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 |
CN112798888A (en) * | 2020-12-30 | 2021-05-14 | 中南大学 | Non-invasive fault diagnosis method for vehicle-mounted electrical system of unmanned train |
CN113011530A (en) * | 2021-04-29 | 2021-06-22 | 国网新疆电力有限公司营销服务中心(资金集约中心、计量中心) | Intelligent ammeter fault prediction method based on multi-classifier fusion |
CN113033089A (en) * | 2021-03-23 | 2021-06-25 | 中国联合网络通信集团有限公司 | Method and device for identifying abnormal power consumption user |
CN113033414A (en) * | 2021-03-26 | 2021-06-25 | 深圳供电局有限公司 | Power consumption data anomaly detection method and device, computer equipment and storage medium |
CN113361608A (en) * | 2021-06-08 | 2021-09-07 | 太原科技大学 | Hidden electricity stealing detection method based on transverse pair bit sum and neural network |
CN114580467A (en) * | 2022-02-22 | 2022-06-03 | 国网山东省电力公司信息通信公司 | Power data anomaly detection method and system based on data enhancement and Tri-tracing |
CN116150616A (en) * | 2022-12-12 | 2023-05-23 | 东北电力大学 | Electric charge abnormity accounting method and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108985330A (en) * | 2018-06-13 | 2018-12-11 | 华中科技大学 | A kind of autoencoder network and its training method abnormal use electric detection method and system |
CN109034127A (en) * | 2018-08-31 | 2018-12-18 | 中国电子科技集团公司第三十六研究所 | A kind of frequency spectrum method for detecting abnormality, device and electronic equipment |
CN110008276A (en) * | 2019-04-26 | 2019-07-12 | 清湖光旭数据科技(北京)有限公司 | A kind of method, device and equipment detecting ammeter exception |
-
2020
- 2020-08-07 CN CN202010787989.0A patent/CN111967512B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108985330A (en) * | 2018-06-13 | 2018-12-11 | 华中科技大学 | A kind of autoencoder network and its training method abnormal use electric detection method and system |
CN109034127A (en) * | 2018-08-31 | 2018-12-18 | 中国电子科技集团公司第三十六研究所 | A kind of frequency spectrum method for detecting abnormality, device and electronic equipment |
CN110008276A (en) * | 2019-04-26 | 2019-07-12 | 清湖光旭数据科技(北京)有限公司 | A kind of method, device and equipment detecting ammeter exception |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112200160A (en) * | 2020-12-02 | 2021-01-08 | 成都信息工程大学 | Deep learning-based direct-reading water meter reading identification method |
CN112798888A (en) * | 2020-12-30 | 2021-05-14 | 中南大学 | Non-invasive fault diagnosis method for vehicle-mounted electrical system of unmanned train |
CN112798888B (en) * | 2020-12-30 | 2021-12-17 | 中南大学 | Non-invasive fault diagnosis method for vehicle-mounted electrical system of unmanned train |
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 |
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 |
CN113033089A (en) * | 2021-03-23 | 2021-06-25 | 中国联合网络通信集团有限公司 | Method and device for identifying abnormal power consumption user |
CN113033414A (en) * | 2021-03-26 | 2021-06-25 | 深圳供电局有限公司 | Power consumption data anomaly detection method and device, computer equipment and storage medium |
CN113011530A (en) * | 2021-04-29 | 2021-06-22 | 国网新疆电力有限公司营销服务中心(资金集约中心、计量中心) | Intelligent ammeter fault prediction method based on multi-classifier fusion |
CN113361608A (en) * | 2021-06-08 | 2021-09-07 | 太原科技大学 | Hidden electricity stealing detection method based on transverse pair bit sum and neural network |
CN114580467A (en) * | 2022-02-22 | 2022-06-03 | 国网山东省电力公司信息通信公司 | Power data anomaly detection method and system based on data enhancement and Tri-tracing |
CN114580467B (en) * | 2022-02-22 | 2023-11-17 | 国网山东省电力公司信息通信公司 | Power data anomaly detection method and system based on data enhancement and Tri-Training |
CN116150616A (en) * | 2022-12-12 | 2023-05-23 | 东北电力大学 | Electric charge abnormity accounting method and system |
Also Published As
Publication number | Publication date |
---|---|
CN111967512B (en) | 2022-08-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111967512B (en) | Abnormal electricity utilization detection method, system and storage medium | |
CN110097297B (en) | Multi-dimensional electricity stealing situation intelligent sensing method, system, equipment and medium | |
CN106845717B (en) | Energy efficiency evaluation method based on multi-model fusion strategy | |
CN110503256A (en) | Short-term load forecasting method and system based on big data technology | |
CN105005825B (en) | Method and system for predicting photovoltaic power based on dynamic neural network | |
CN111008726B (en) | Class picture conversion method in power load prediction | |
CN110321934A (en) | Method and system for detecting abnormal data of user electricity consumption | |
CN116842459B (en) | Electric energy metering fault diagnosis method and diagnosis terminal based on small sample learning | |
CN112001409A (en) | Power distribution network line loss abnormity diagnosis method and system based on K-means clustering algorithm | |
CN114004296A (en) | Method and system for reversely extracting monitoring points based on power load characteristics | |
CN112257784A (en) | Electricity stealing detection method based on gradient boosting decision tree | |
CN109359665A (en) | A kind of family's electric load recognition methods and device based on support vector machines | |
CN110188221A (en) | A kind of load curve hierarchy clustering method based on shape distance | |
CN116205355B (en) | Power load prediction method, device and storage medium | |
CN110390440B (en) | Clustering and deep neural network-based intelligent ammeter user aggregate load prediction method | |
CN115758151A (en) | Combined diagnosis model establishing method and photovoltaic module fault diagnosis method | |
CN115758249A (en) | Non-intrusive load identification method and device | |
CN117725437A (en) | Machine learning-based data accurate matching analysis method | |
CN111612054B (en) | User electricity stealing behavior identification method based on nonnegative matrix factorization and density clustering | |
CN112595918A (en) | Low-voltage meter reading fault detection method and device | |
CN117331017A (en) | Method and system for studying and judging misconnection of three-phase four-wire electric energy meter | |
CN112083275A (en) | Distribution network fault type identification method and system | |
CN116933201A (en) | Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile | |
CN116502149A (en) | Low-voltage power distribution network user-transformation relation identification method and system based on current characteristic conduction | |
CN114676931B (en) | Electric quantity prediction system based on data center technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |