CN110008276A - A kind of method, device and equipment detecting ammeter exception - Google Patents
A kind of method, device and equipment detecting ammeter exception Download PDFInfo
- Publication number
- CN110008276A CN110008276A CN201910346236.3A CN201910346236A CN110008276A CN 110008276 A CN110008276 A CN 110008276A CN 201910346236 A CN201910346236 A CN 201910346236A CN 110008276 A CN110008276 A CN 110008276A
- Authority
- CN
- China
- Prior art keywords
- ammeter
- layer
- error
- neural networks
- convolutional neural
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
-
- 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
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of method, device and equipments for detecting ammeter exception, belong to technical field of electric power, method includes: the input data set being converted into each ammeter data suitable for time series forecasting, error prediction is carried out to input data set using at least two layers of shot and long term memory network algorithm, obtains the prediction error on expected date;In the case of determining to have abnormal ammeter according to the prediction error on expected date, each ammeter is carried out abnormality detection using convolutional neural networks, obtains abnormal ammeter.The present invention carries out error prediction by least two layers of shot and long term memory network algorithm, in the case of judging to have abnormal ammeter, each ammeter is carried out abnormality detection using convolutional neural networks, effectively, the exception information of electric supply meter is accurately excavated, and abnormal ammeter is filtered out, it allows users to targetedly safeguard ammeter, replace, limit intelligent electric meter no longer by service life, for country and personal saving economic cost, economize on resources.
Description
Technical field
The present embodiments relate to technical field of electric power, and in particular to it is a kind of detect ammeter exception method, apparatus and set
It is standby.
Background technique
With the development of power technology, intelligent terminal and Data entries of the intelligent electric meter as smart grid, in order to adapt to
There is two-way a variety of rate meterings, user terminal real-time control, plurality of data transmission modes, intelligence to hand over for smart grid, intelligent electric meter
A variety of application functions such as mutual.
With the development of intelligent power grid technology, smart grid will cover the population in the whole world 80%, the construction of smart grid
The wide market demand is brought for global intelligent electric meter and power information acquisition, processing system product.Intelligent electric meter permeability
Reach 60%, and intelligent electric meter will play an increasingly important role in smart grid.
In the prior art, the service life of national regulation intelligent electric meter is 8 years, but actually intelligent electric meter is using 8 years
Afterwards, most of still to can be used normally, only intelligent electric meter is replaced only in accordance with service life, causes to replace intelligent electric meter
Economic cost it is higher, result in waste of resources.
Summary of the invention
For this purpose, the embodiment of the present invention provides a kind of method, device and equipment for detecting ammeter exception, to solve the prior art
The problems in.
To achieve the goals above, the embodiment of the present invention provides the following technical solutions:
According to a first aspect of the embodiments of the present invention, a kind of method detecting ammeter exception is provided, comprising:
Each ammeter data is converted into the input data set suitable for time series forecasting, using at least two layers of shot and long term
Memory network algorithm carries out error prediction to input data set, obtains the prediction error on expected date;
In the case of determining to have abnormal ammeter according to the prediction error on the expected date, using convolutional neural networks
Each ammeter is carried out abnormality detection, abnormal ammeter is obtained.
Further, the first layer of the shot and long term memory network algorithm includes 30 dimensions;The shot and long term memory network is calculated
The second layer of method includes 30 dimensions.
Further, the activation primitive that the convolutional neural networks use is sigmoid function.
Further, described that each ammeter is carried out abnormality detection using convolutional neural networks, abnormal ammeter is obtained, is wrapped
It includes:
Using first layer, that is, input layer of convolutional neural networks, reception is passed according to what the power information of each user was converted into
Gui Tu;Feature is extracted by the second layer of convolutional neural networks and the third layer of convolutional neural networks;The of convolutional neural networks
Four layers are maximum pond layer;By layer 5, that is, flat layer of convolutional neural networks, multidimensional input is converted to one-dimensional;According to volume
The Scalar operation prediction result that the layer 6 of product neural network and the layer 7 of convolutional neural networks provide, is sieved according to prediction result
Select abnormal ammeter.
Further, it is described each ammeter data is converted into the input data set suitable for time series forecasting before, also
Include:
Ammeter data is obtained, and carries out data analysis and feature fitting, error properties of distributions is obtained, in error properties of distributions
In the case of stabilization, abnormal ammeter is obtained according to error distribution situation.
In second aspect of the present invention, a kind of device for detecting ammeter exception is provided, comprising:
Error sensing module is adopted for each ammeter data to be converted into the input data set suitable for time series forecasting
Error prediction is carried out to input data set at least two layers of shot and long term memory network algorithm, the prediction for obtaining the expected date misses
Difference;
Abnormal ammeter detection module, connect with error sensing module, in the prediction error according to the expected date
Determine in the case of there is abnormal ammeter, each ammeter is carried out abnormality detection using convolutional neural networks, obtains abnormal ammeter.
Further, the convolutional neural networks include seven layers, wherein first layer is for receiving according to each user
The input layer for the recurrence plot that power information is converted into;The second layer and third layer are convolutional layers for extracting feature;4th layer is most
Great Chiization layer;Layer 5 is flat layer, one-dimensional for being converted to multidimensional input;Layer 6 and layer 7 are for providing scalar
Dense layer;Prediction result is calculated according to the output result of the dense layer, abnormal ammeter is filtered out according to prediction result.
Further, described device further includes the data analysis module connecting with the error sensing module;
The data analysis module for obtaining ammeter data, and carries out data analysis and feature fitting, obtains error point
Cloth attribute obtains abnormal ammeter according to error distribution situation, is distributed in the error in the case of error properties of distributions is stablized
In the case of attribute is unstable, the error sensing module is triggered.
In third aspect present invention, a kind of equipment for detecting ammeter exception is provided, comprising:
Processor, for each ammeter data to be converted into the input data set suitable for time series forecasting, using at least
Two layers of shot and long term memory network algorithm carries out error prediction to input data set, the prediction error on expected date is obtained, in root
Determine in the case of there is abnormal ammeter according to the prediction error on the expected date, using convolutional neural networks to each ammeter into
Row abnormality detection obtains abnormal ammeter.
In fourth aspect present invention, a kind of computer readable storage medium is provided, the computer readable storage medium is deposited
Program is contained, described program is for realizing the method as described above for detecting ammeter exception.
The embodiment of the present invention has the advantages that the embodiment of the present invention, when by the way that each ammeter data being converted into being suitable for
Between sequence prediction input data set, using at least two layers of shot and long term memory network algorithm to input data set carry out error it is pre-
Survey, obtain the prediction error on expected date, therefore, it is determined that there is abnormal ammeter, then using convolutional neural networks to each ammeter into
Row abnormality detection detects abnormal ammeter.The exception information of electric supply meter can effectively, be accurately excavated using this method,
And abnormal ammeter is filtered out, it allows users to targetedly safeguard ammeter, replace, use intelligent electric meter no longer
Time limit limitation avoids the occurrence of reaching time limit batch replacement ammeter, saves economic cost with personal for country, save money
Source.
Detailed description of the invention
It, below will be to embodiment party in order to illustrate more clearly of embodiments of the present invention or technical solution in the prior art
Formula or attached drawing needed to be used in the description of the prior art are briefly described.It should be evident that the accompanying drawings in the following description is only
It is merely exemplary, it for those of ordinary skill in the art, without creative efforts, can also basis
The attached drawing of offer, which is extended, obtains other implementation attached drawings.
Structure depicted in this specification, ratio, size etc., only to cooperate the revealed content of specification, for
Those skilled in the art understands and reads, and is not intended to limit the invention enforceable qualifications, therefore does not have technical
Essential meaning, the modification of any structure, the change of proportionate relationship or the adjustment of size are not influencing the function of the invention that can be generated
Under effect and the purpose that can reach, should all still it fall in the range of disclosed technology contents obtain and can cover.
Fig. 1 is a kind of method flow diagram for detection ammeter exception that one embodiment of the invention provides;
Fig. 2 be another embodiment of the present invention provides a kind of detection ammeter exception method flow diagram;
Fig. 3 be another embodiment of the present invention provides a kind of detection ammeter exception apparatus structure illustrate.
In figure: 301 be error sensing module, and 302 be abnormal ammeter detection module.
Specific embodiment
Embodiments of the present invention are illustrated by particular specific embodiment below, those skilled in the art can be by this explanation
Content disclosed by book is understood other advantages and efficacy of the present invention easily, it is clear that described embodiment is the present invention one
Section Example, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
In first aspect present invention, a kind of method for detecting ammeter exception is provided, as shown in Figure 1, comprising:
Step 101: each ammeter data being converted into the input data set suitable for time series forecasting, using at least two layers
Shot and long term memory network algorithm error prediction is carried out to input data set, obtain the prediction error on expected date;
In embodiments of the present invention, the data of ammeter are one group of time series datas, using the data of ammeter as original number
It is converted into the input data set suitable for time series forecasting according to collection, using at least two layers of shot and long term memory network algorithm to defeated
Enter data set and carry out error prediction, wherein shot and long term memory network algorithm is LSTM (Long Short-Term Memory) calculation
Method, at least two layers of LSTM, first layer has 30 dimensions, and the LSTM second layer also has 30 dimensions.
In embodiments of the present invention, activation primitive corresponding with LSTM algorithm uses sigmoid function.
Step 102: in the case of determining to have abnormal ammeter according to the prediction error on expected date, using convolutional Neural
Network carries out abnormality detection each ammeter, obtains abnormal ammeter.
In embodiments of the present invention, using seven layers of convolutional neural networks (CNN, Convolutional Neural
Networks the abnormality detection for) carrying out ammeter is received using first layer, that is, input layer of convolutional neural networks according to each user
The recurrence plot that is converted into of power information;It is extracted by the second layer of convolutional neural networks and the third layer of convolutional neural networks special
Sign;The 4th layer of convolutional neural networks is maximum pond layer;It is by layer 5, that is, flat layer of convolutional neural networks, multidimensional is defeated
Enter to be converted to one-dimensional;The Scalar operation prediction provided according to the layer 7 of the layer 6 of convolutional neural networks and convolutional neural networks
As a result, filtering out abnormal ammeter according to prediction result.
Further, seven layers of convolutional neural networks are respectively that first layer is to receive to be turned according to the power information of each user
The input layer of the recurrence plot changed into;The second layer and third layer are that convolutional layer is used to extract feature;4th layer is maximum pond layer;The
Five layers are flat layers, it is therefore an objective to are converted to multidimensional input one-dimensional;Layer 6 and layer 7 are the dense layers for providing scalar;Then
Prediction result is calculated according to the output result of dense layer, abnormal ammeter is filtered out according to prediction result.
In second aspect of the present invention, a kind of method for detecting ammeter exception is provided, as shown in Figure 2, comprising:
Step 201: obtaining ammeter data, and carry out data analysis and feature fitting, error properties of distributions is obtained, in error
In the case of properties of distributions is stablized, abnormal ammeter is obtained according to error distribution situation.
In embodiments of the present invention, it obtains respectively per user's table electricity W_sub (user data is more) for 24 hours, per cell for 24 hours
Summary table electricity W_super calculates error per ammeter electricity for 24 hours according to each user in cell summary table electricity and cell, and error is denoted as
E utilizes formula:
Further, obtained error is judged, deletes the error value that error is negative, utilizes integral point electric current and electricity
Pressure replaces missing values, completes data cleansing.
By drawing summary table and dividing table and scatter plot, absolute error curve graph and relative error curve graph, analytical error
Relationship between electricity judges ammeter abnormal conditions in the case of error is distributed and stablizes, for error distribution in normal state point
Cloth, but in the case of distribution is not fixed, abnormal point can not be determined by observation distribution.
Step 202: in the case of error properties of distributions is unstable, using the method for deep learning, being remembered using shot and long term
Recall network algorithm and error prediction is carried out to ammeter data, obtain the prediction error on expected date, in the prediction according to the expected date
Error determines in the case of there is abnormal ammeter, is carried out abnormality detection using convolutional neural networks to each ammeter, obtains exception
Ammeter.
In embodiments of the present invention, ammeter data is converted into the input data set suitable for time series forecasting, used
At least two layers of shot and long term memory network algorithm carries out error prediction to input data set, obtains the prediction error on expected date.
In embodiments of the present invention, the data of ammeter are one group of time series datas, using the data of ammeter as original number
It is converted into the input data set suitable for time series forecasting according to collection, using at least two layers of shot and long term memory network algorithm to defeated
Enter data set and carry out error prediction, wherein shot and long term memory network algorithm is LSTM (Long Short-Term Memory) calculation
Method, at least two layers of LSTM, first layer has 30 dimensions, and the LSTM second layer also has 30 dimensions.
In embodiments of the present invention, activation primitive corresponding with LSTM algorithm uses sigmoid function.
In the present embodiment, the abnormality detection of ammeter, further, seven layers of convolution are carried out using seven layers of convolutional neural networks
Neural network is respectively that first layer is the input layer for receiving the recurrence plot being converted into according to the power information of each user;Second
Layer and third layer are that convolutional layer is used to extract feature;4th layer is maximum pond layer;Layer 5 is flat layer, it is therefore an objective to by multidimensional
Input is converted to one-dimensional;Layer 6 and layer 7 are the dense layers for providing scalar;Then it is calculated according to the output result of dense layer
Prediction result filters out abnormal ammeter according to prediction result.
It should be noted that can also establish model in the embodiment of the present invention using the method for machine learning, utilize SVM
(Support Vector Machine, support vector machines) algorithm and SVM and EMD (Empirical Mode
Decomposition, empirical mode decomposition) mode that combines of algorithm carries out abnormal ammeter detection.
In third aspect present invention, a kind of device for detecting ammeter exception is provided, as shown in Figure 3, comprising:
Error sensing module 301, for each ammeter data to be converted into the input data set suitable for time series forecasting,
Error prediction is carried out to input data set using at least two layers of shot and long term memory network algorithm, the prediction for obtaining the expected date misses
Difference;
Abnormal ammeter detection module 302, connect with error sensing module 301, for pre- according to the expected date
It surveys error to determine in the case of there is abnormal ammeter, each ammeter is carried out abnormality detection using convolutional neural networks, is obtained different
Normal ammeter.
Further, convolutional neural networks include seven layers, wherein first layer is for receiving the electricity consumption according to each user
The input layer for the recurrence plot that information is converted into;The second layer and third layer are convolutional layers for extracting feature;4th layer is maximum pond
Change layer;Layer 5 is flat layer, one-dimensional for being converted to multidimensional input;Layer 6 and layer 7 are for providing the thick of scalar
Close layer;Prediction result is calculated according to the output result of the dense layer, abnormal ammeter is filtered out according to prediction result.
It should be noted that in embodiments of the present invention, device further includes the data analysis connecting with error sensing module
Module.
Data analysis module for obtaining ammeter data, and carries out data analysis and feature fitting, obtains error distribution and belongs to
Property, in the case of error properties of distributions is stablized, abnormal ammeter is obtained according to error distribution situation, and in error properties of distributions
In the case of unstable, trigger error detection module.
In fourth aspect present invention, a kind of equipment for detecting ammeter exception is provided, comprising:
Processor, for each ammeter data to be converted into the input data set suitable for time series forecasting, using at least
Two layers of shot and long term memory network algorithm carries out error prediction to input data set, the prediction error on expected date is obtained, in root
Determine in the case of there is abnormal ammeter according to the prediction error on the expected date, using convolutional neural networks to each ammeter into
Row abnormality detection obtains abnormal ammeter.
In sixth aspect present invention, a kind of computer readable storage medium is provided, the computer readable storage medium is deposited
Program is contained, described program is for realizing the method as described above for detecting ammeter exception.
Although above having used general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention are fallen within the scope of the claimed invention.
Claims (10)
1. a kind of method for detecting ammeter exception characterized by comprising
Each ammeter data is converted into the input data set suitable for time series forecasting, is remembered using at least two layers of shot and long term
Network algorithm carries out error prediction to input data set, obtains the prediction error on expected date;
In the case of determining to have abnormal ammeter according to the prediction error on the expected date, using convolutional neural networks to every
A ammeter carries out abnormality detection, and obtains abnormal ammeter.
2. the method as described in claim 1, which is characterized in that
The first layer of the shot and long term memory network algorithm includes 30 dimensions;The second layer of the shot and long term memory network algorithm includes
30 dimensions.
3. the method as described in claim 1, which is characterized in that
The activation primitive that the convolutional neural networks use is sigmoid function.
4. the method as described in claim 1, which is characterized in that
It is described that each ammeter is carried out abnormality detection using convolutional neural networks, obtain abnormal ammeter, comprising:
Using first layer, that is, input layer of convolutional neural networks, the recurrence being converted into according to the power information of each user is received
Figure;Feature is extracted by the second layer of convolutional neural networks and the third layer of convolutional neural networks;The 4th of convolutional neural networks
Layer is maximum pond layer;By layer 5, that is, flat layer of convolutional neural networks, multidimensional input is converted to one-dimensional;According to convolution
The Scalar operation prediction result that the layer 6 of neural network and the layer 7 of convolutional neural networks provide, is screened according to prediction result
Abnormal ammeter out.
5. the method as described in claim 1, which is characterized in that
It is described that each ammeter data is converted into before the input data set suitable for time series forecasting, further includes:
Ammeter data is obtained, and carries out data analysis and feature fitting, obtains error properties of distributions, is stablized in error properties of distributions
In the case of, abnormal ammeter is obtained according to error distribution situation.
6. a kind of device for detecting ammeter exception characterized by comprising
Error sensing module, for each ammeter data to be converted into the input data set suitable for time series forecasting, using extremely
Few two layers of shot and long term memory network algorithm carries out error prediction to input data set, obtains the prediction error on expected date;
Abnormal ammeter detection module, connect with error sensing module, for determining according to the prediction error on the expected date
There are in the case of abnormal ammeter, each ammeter is carried out abnormality detection using convolutional neural networks, obtains abnormal ammeter.
7. device as claimed in claim 6, which is characterized in that
The convolutional neural networks include seven layers, wherein first layer is to be converted for receiving according to the power information of each user
At recurrence plot input layer;The second layer and third layer are convolutional layers for extracting feature;4th layer is maximum pond layer;5th
Layer is flat layer, one-dimensional for being converted to multidimensional input;Layer 6 and layer 7 are the dense layers for providing scalar;According to
The output result of the dense layer calculates prediction result, filters out abnormal ammeter according to prediction result.
8. device as claimed in claim 6, which is characterized in that
Described device further includes the data analysis module connecting with the error sensing module;
The data analysis module for obtaining ammeter data, and carries out data analysis and feature fitting, obtains error distribution and belongs to
Property, in the case of error properties of distributions is stablized, abnormal ammeter is obtained according to error distribution situation;In the error properties of distributions
In the case of unstable, the error sensing module is triggered.
9. a kind of equipment for detecting ammeter exception characterized by comprising
Processor, for each ammeter data to be converted into the input data set suitable for time series forecasting, using at least two layers
Shot and long term memory network algorithm error prediction is carried out to input data set, the prediction error on expected date is obtained, according to institute
The prediction error for stating the expected date determines in the case of there is abnormal ammeter, is carried out using convolutional neural networks to each ammeter different
Often detection, obtains abnormal ammeter.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has program, institute
Program is stated for realizing the method for detection ammeter exception as described in any one in claim 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910346236.3A CN110008276A (en) | 2019-04-26 | 2019-04-26 | A kind of method, device and equipment detecting ammeter exception |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910346236.3A CN110008276A (en) | 2019-04-26 | 2019-04-26 | A kind of method, device and equipment detecting ammeter exception |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110008276A true CN110008276A (en) | 2019-07-12 |
Family
ID=67174798
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910346236.3A Pending CN110008276A (en) | 2019-04-26 | 2019-04-26 | A kind of method, device and equipment detecting ammeter exception |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110008276A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111142060A (en) * | 2019-12-02 | 2020-05-12 | 国网浙江省电力有限公司 | Self-adaptive threshold adjustment diagnosis method based on improved BP neural network |
CN111967512A (en) * | 2020-08-07 | 2020-11-20 | 国网江苏省电力有限公司电力科学研究院 | Abnormal electricity utilization detection method, system and storage medium |
CN112926645A (en) * | 2021-02-22 | 2021-06-08 | 国网四川省电力公司营销服务中心 | Electricity stealing detection method based on edge calculation |
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 |
CN113608163A (en) * | 2021-09-10 | 2021-11-05 | 天目数据(福建)科技有限公司 | Ammeter fault diagnosis method and device of stacked cyclic neural network |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE2237077B2 (en) * | 1971-08-06 | 1977-01-20 | VEB Energiekombinat Ost, DDR 8010 Dresden | Replacement time-cycle of electricity meters - is determined by random sample analysis to reduce maintenance time (OE151275) |
CN107818395A (en) * | 2017-09-05 | 2018-03-20 | 天津市电力科技发展有限公司 | A kind of electric energy meter error iterative calculation method based on uncertainty of measurement |
CN109597014A (en) * | 2018-11-30 | 2019-04-09 | 国网上海市电力公司 | A kind of electric energy meter error diagnostic method based on artificial intelligence technology |
-
2019
- 2019-04-26 CN CN201910346236.3A patent/CN110008276A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE2237077B2 (en) * | 1971-08-06 | 1977-01-20 | VEB Energiekombinat Ost, DDR 8010 Dresden | Replacement time-cycle of electricity meters - is determined by random sample analysis to reduce maintenance time (OE151275) |
CN107818395A (en) * | 2017-09-05 | 2018-03-20 | 天津市电力科技发展有限公司 | A kind of electric energy meter error iterative calculation method based on uncertainty of measurement |
CN109597014A (en) * | 2018-11-30 | 2019-04-09 | 国网上海市电力公司 | A kind of electric energy meter error diagnostic method based on artificial intelligence technology |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111142060A (en) * | 2019-12-02 | 2020-05-12 | 国网浙江省电力有限公司 | Self-adaptive threshold adjustment diagnosis method based on improved BP neural network |
CN111142060B (en) * | 2019-12-02 | 2023-11-07 | 国网浙江省电力有限公司 | Adaptive threshold adjustment diagnosis method based on improved BP neural network |
CN111967512A (en) * | 2020-08-07 | 2020-11-20 | 国网江苏省电力有限公司电力科学研究院 | Abnormal electricity utilization detection method, system and storage medium |
CN111967512B (en) * | 2020-08-07 | 2022-08-19 | 国网江苏省电力有限公司电力科学研究院 | Abnormal electricity utilization detection method, system and storage medium |
CN112926645A (en) * | 2021-02-22 | 2021-06-08 | 国网四川省电力公司营销服务中心 | Electricity stealing detection method based on edge calculation |
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 |
CN113608163A (en) * | 2021-09-10 | 2021-11-05 | 天目数据(福建)科技有限公司 | Ammeter fault diagnosis method and device of stacked cyclic neural network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110008276A (en) | A kind of method, device and equipment detecting ammeter exception | |
Bedi et al. | Deep learning framework to forecast electricity demand | |
CN103605103B (en) | Electric energy metrical intelligent fault diagnosis method based on S type curvilinear function | |
CN108875779A (en) | Training method, device and the terminal device of neural network | |
CN106777244A (en) | A kind of power customer electricity consumption behavior analysis method and system | |
CN108390393A (en) | Power distribution network multi-objective reactive optimization method and terminal device | |
CN109492048A (en) | A kind of extracting method, system and the terminal device of power consumer electrical characteristics | |
Yang et al. | Cuckoo search: state-of-the-art and opportunities | |
CN109871949A (en) | Convolutional neural networks accelerator and accelerated method | |
CN112566093B (en) | Terminal relation identification method and device, computer equipment and storage medium | |
CN110321934A (en) | Method and system for detecting abnormal data of user electricity consumption | |
CN107608748A (en) | Application program management-control method, device, storage medium and terminal device | |
CN107643948A (en) | Application program management-control method, device, medium and electronic equipment | |
CN108765194A (en) | A kind of effective residential electricity consumption behavior analysis system | |
CN107046557A (en) | The intelligent medical calling inquiry system that dynamic Skyline is inquired about under mobile cloud computing environment | |
CN110210723A (en) | A kind of stealing discrimination method based on analytic hierarchy process (AHP) and isolated forest | |
CN103366062A (en) | Method for constructing core backbone grid structure based on BBO algorithm and power grid survivability | |
Kumar et al. | An information theoretic approach for feature selection | |
CN112614004A (en) | Method and device for processing power utilization information | |
CN107844338A (en) | Application program management-control method, device, medium and electronic equipment | |
Abdulla et al. | Smart meter-based energy consumption forecasting for smart cities using adaptive federated learning | |
Li et al. | Twitter data mining for the social awareness of emerging technologies | |
Luo et al. | A comparison of three prediction models for predicting monthly precipitation in Liaoyuan city, China | |
CN117034046A (en) | Flexible load adjustable potential evaluation method based on ISODATA clustering | |
Pan et al. | Study on intelligent anti–electricity stealing early-warning technology based on convolutional neural networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190712 |
|
RJ01 | Rejection of invention patent application after publication |