CN110968703A - Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm - Google Patents
Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm Download PDFInfo
- Publication number
- CN110968703A CN110968703A CN201911238633.5A CN201911238633A CN110968703A CN 110968703 A CN110968703 A CN 110968703A CN 201911238633 A CN201911238633 A CN 201911238633A CN 110968703 A CN110968703 A CN 110968703A
- Authority
- CN
- China
- Prior art keywords
- metering point
- information
- abnormal
- knowledge base
- lstm
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 133
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000000605 extraction Methods 0.000 title claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 19
- 230000005611 electricity Effects 0.000 claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 8
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 7
- 230000008014 freezing Effects 0.000 claims description 29
- 238000007710 freezing Methods 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000009411 base construction Methods 0.000 claims description 6
- 238000000819 phase cycle Methods 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 abstract description 10
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000012216 screening Methods 0.000 abstract 1
- 238000003745 diagnosis Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 238000004458 analytical method Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention relates to a method for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm, which is characterized by comprising the following steps of: the method comprises the following steps: (1) extracting abnormal metering point information in an electricity utilization information acquisition system; (2) based on the abnormal metering point information, calculating the probability of the abnormal metering point knowledge information through an LSTM network unit model to form a multi-dimensional knowledge label; (3) and extracting the formed multi-dimensional knowledge labels based on a bidirectional LSTM training network, establishing abnormal metering point knowledge identification labels aiming at the knowledge characteristic values with the maximum probability, and outputting a multi-dimensional abnormal metering point knowledge base. The invention identifies the metering abnormality through the matching of the knowledge base, thereby screening out the problem electric energy meter, and providing accurate guidance for the maintenance, the service life and the like of the electric energy meter.
Description
Technical Field
The invention belongs to the field of electric energy meter metering error analysis, and particularly relates to an abnormal metering point knowledge base construction method and system based on an LSTM end-to-end extraction algorithm.
Background
Since 2009, national grid companies have vigorously built electricity consumption information acquisition systems, and currently, the operation of 4.5 hundred million electric meters in the universe is realized. After years of operation, the system accumulates massive electricity data. Through data analysis, effective power utilization information such as operation errors of an electric energy meter and power utilization behavior patterns of users are mined, the potential of mass data can be developed, the operation cost can be greatly reduced, and decision support is provided for power grid companies.
However, the mass data collected by the electricity consumption information collection system is a large amount of various data collected from the real world, and the quality of the original data is affected by diversity, uncertainty and complexity, so that the collected actual data is messy, has the phenomena of deficiency, abnormality and the like, and does not meet the standard requirement of knowledge acquisition of a data mining tool under many conditions. The traditional knowledge base triple extraction model is limited by the sample scale, and the effect on the long-tail relationship is difficult to meet the requirements of practical application and the requirement of low-voltage transformer area management refinement.
Therefore, a knowledge base needs to be established for the related abnormal conditions so as to guide the maintenance, service life and the like of the electric energy meter, provide clean, concise and accurate data, enable the mining process to be more effective and easier, and improve the mining efficiency and accuracy.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a method and a system for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
an abnormal metering point knowledge base construction method based on an LSTM end-to-end extraction algorithm comprises the following steps:
(1) extracting abnormal metering point information in an electricity utilization information acquisition system;
(2) based on the abnormal metering point information, calculating the probability of the abnormal metering point knowledge information through an LSTM network unit model to form a multi-dimensional knowledge label;
(3) and extracting the formed multi-dimensional knowledge labels based on a bidirectional LSTM training network, establishing abnormal metering point knowledge identification labels aiming at the knowledge characteristic values with the maximum probability, and outputting a multi-dimensional abnormal metering point knowledge base.
And the abnormal metering point information comprises instantaneous over-range metering point information, stable over-range metering point information, suspected electricity stealing metering point information, wiring error metering point information, collected abnormal metering point information, clock error metering point information, transformer overload metering point information, user variable relation error metering point information, transformer light load metering point information and collector abnormal information.
The instantaneous over-range metering point information comprises field check information and power utilization user file information, the stable over-range metering point information comprises the field check information and the power utilization user file information, the suspected electricity stealing metering point information comprises the field check information, running error calculation information, voltage and current information and power factor information, the wiring error metering point information comprises information such as reversed power flow, abnormal phase sequence and reversed electric quantity, the collected abnormal metering point information comprises daily freezing data, abnormal event information and high-frequency collected 96-point voltage and current data information, the clock error metering point information comprises daily freezing data and abnormal event information, the transformer overload metering point information comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the user variable relation error metering point information comprises load environment, load environment and load environment, The system comprises a power utilization curve, a geographical position and other information, the light-load metering point information of the mutual inductor comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the collector abnormal information comprises daily freezing data, 96-point voltage and current data information and electric energy meter archive information.
Furthermore, the LSTM network element model is
Where p (wj) represents the prediction probability of the j-th knowledge point.
And the multi-dimensional abnormal metering point knowledge base comprises an instantaneous over-range metering point knowledge base, a stable over-range metering point knowledge base, a suspected electricity stealing metering point knowledge base, a wiring error metering point knowledge base, an acquisition abnormal metering point knowledge base, a clock error metering point knowledge base, a mutual inductor overload metering point knowledge base, a user-variant relation error metering point knowledge base, a mutual inductor underload metering point knowledge base and an acquisition device abnormal knowledge base.
An abnormal metering point knowledge base construction system based on LSTM end-to-end extraction algorithm comprises
The abnormal metering point information extraction module is used for extracting the abnormal metering point information from the power utilization information acquisition system;
the abnormal metering point knowledge information probability calculation module is used for calculating the abnormal metering point knowledge information probability through an LSTM network unit model based on the abnormal metering point information to form a multi-dimensional knowledge label;
and the multi-dimensional abnormal metering point knowledge base construction module is used for extracting the formed multi-dimensional knowledge labels based on the bidirectional LSTM training network, establishing abnormal metering point knowledge identification labels aiming at the knowledge characteristic values with the maximum probability, and outputting the multi-dimensional abnormal metering point knowledge base.
And the abnormal metering point information comprises instantaneous over-range metering point information, stable over-range metering point information, suspected electricity stealing metering point information, wiring error metering point information, collected abnormal metering point information, clock error metering point information, transformer overload metering point information, user variable relation error metering point information, transformer light load metering point information and collector abnormal information.
The instantaneous over-range metering point information comprises field check information and power utilization user file information, the stable over-range metering point information comprises the field check information and the power utilization user file information, the suspected electricity stealing metering point information comprises the field check information, running error calculation information, voltage and current information and power factor information, the wiring error metering point information comprises information such as reversed power flow, abnormal phase sequence and reversed electric quantity, the collected abnormal metering point information comprises daily freezing data, abnormal event information and high-frequency collected 96-point voltage and current data information, the clock error metering point information comprises daily freezing data and abnormal event information, the transformer overload metering point information comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the user variable relation error metering point information comprises load environment, load environment and load environment, The system comprises a power utilization curve, a geographical position and other information, the light-load metering point information of the mutual inductor comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the collector abnormal information comprises daily freezing data, 96-point voltage and current data information and electric energy meter archive information.
Furthermore, the LSTM network element model is
Where p (wj) represents the prediction probability of the j-th knowledge point.
And the multi-dimensional abnormal metering point knowledge base comprises an instantaneous over-range metering point knowledge base, a stable over-range metering point knowledge base, a suspected electricity stealing metering point knowledge base, a wiring error metering point knowledge base, an acquisition abnormal metering point knowledge base, a clock error metering point knowledge base, a mutual inductor overload metering point knowledge base, a user-variant relation error metering point knowledge base, a mutual inductor underload metering point knowledge base and an acquisition device abnormal knowledge base.
The invention has the advantages and positive effects that:
the system for constructing the abnormal metering point knowledge base based on the LSTM end-to-end extraction algorithm improves a traditional triple knowledge extraction model, accurately extracts abnormal metering information of a long-tail relationship based on the LSTM end-to-end extraction algorithm, effectively utilizes information contained in rich samples of a head relationship to form a knowledge base in ten aspects of instantaneous over-range, stable over-range, suspected electricity stealing, wiring error, acquisition abnormality, clock error, mutual inductor overload, user-variant relationship error, mutual inductor light load, collector abnormality and the like, effectively identifies and calculates the abnormality of the metering point of the electric energy meter, and helps to accurately guide the maintenance, service life and the like of the electric energy meter.
Drawings
FIG. 1 is a schematic block diagram of the present invention;
FIG. 2 is a schematic diagram of the LSTM knowledge probability computation model of the present invention (where "< S >" and "</S >" denote the beginning and end of a statement, respectively).
Detailed Description
The embodiments of the invention are described in further detail below with reference to the following figures:
the method comprises the following steps: abnormal metering point information collection
And extracting information of the abnormal metering points in the electricity utilization information acquisition system.
Step two: and calculating the probability of knowledge information of the abnormal metering points through an LSTM network unit model based on the information of the abnormal metering points to form a multi-dimensional knowledge label.
The LSTM language model uses the previous word sequence to predict the probability of current knowledge of the anomaly metric points. After mapping the input abnormal metering point information to a pre-trained word vector with fixed dimensions (the word vector of 'NE' is initially a zero vector, but the word vector is also updated in the training process), the abnormal metering point information is input into the LSTM model. And (3) sending the output of each LSTM unit into a softmax classifier, and finally outputting the knowledge point distribution of each piece of information in the current position vocabulary:
where p (wj) represents the predicted probability of the jth knowledge point in the question. The smaller the LP (S), the higher the probability that statement S will meet the requirements.
Furthermore, aiming at the knowledge characteristic value with the maximum probability, an abnormal metering point knowledge identification label is established, and the multi-dimensional abnormal metering knowledge point is output.
Step three: knowledge label extraction is carried out based on bidirectional LSTM training network to form a multi-dimensional abnormal metering point knowledge base
The knowledge extraction is carried out on the abnormal metering points of the long-tail relationship through the bidirectional LSTM training network, information contained in rich samples of the head relationship is effectively utilized, a triple is extracted from the unstructured text, labeled data are automatically constructed for each relationship through the idea of weak supervision labeling, and an abnormal metering point knowledge base is formed.
1. Constructing instantaneous over-range metering point knowledge base
The method comprises the steps of establishing an instantaneous over-range expert knowledge base based on an electric energy meter instantaneous over-range abnormity diagnosis model, site checking information fed back by a work order and power utilization user file information, analyzing power utilization behaviors of users according to calculation data of the diagnosis model and instantaneous over-range power utilization rules and characteristics, diagnosing users with suspected instantaneous over-range abnormity, and giving power utilization time periods of suspected instantaneous over-range of the power utilization users. Meanwhile, according to the site information fed back by the work order, the knowledge base is continuously perfected, enriched and optimized.
2. Construction of stable over-range metering point knowledge base
And establishing a stable over-range expert knowledge base based on the electric energy meter stable over-range abnormal diagnosis model, the field check information fed back by the work order and the power utilization user profile information. And analyzing the power consumption behavior of the user according to the data calculated by the diagnosis model and the power consumption characteristics and rules of the stable over-range, and diagnosing the abnormal user with suspected stable over-range.
One possible determination method is that the current average value is greater than Imax, and the absolute value of the calculation error of the electric energy meter exceeds the limit value, then the type of the stable over-range abnormality is determined.
Meanwhile, according to the site information fed back by the work order, the knowledge base is continuously perfected and optimized.
3. Construction of suspected electricity stealing metering point knowledge base
And establishing a knowledge base based on-site checking information, operation error calculation information, voltage and current information, power factor information and the like. The knowledge base of suspected electricity stealing is based on comprehensive analysis of various information, such as:
(1) electric quantity is abnormal:
when a user has a condition of not having a value of 0 for a long time, negative growth, zero growth or abnormal growth, the situation is probably caused by electricity stealing;
(2) abnormal line loss:
the line loss of the transformer area exceeds a threshold value, or is compared with the line loss in the same period of the last year, and whether electricity stealing possibly exists under the transformer area is jointly judged;
(3) three-phase imbalance analysis:
the fluctuation change of the three-phase unbalance rate can also be characterized as an electricity stealing phenomenon;
(4) and (3) abnormal event analysis:
based on the obtained abnormal events of the metering points, such as the uncovering of the electric energy meter and the opening and closing of the metering door, the possibility of electricity stealing is reported by the abnormal events related to electricity stealing.
And according to the conditions and the error analysis result, establishing a suspected electricity stealing expert knowledge base. Providing a diagnosis basis for a suspected electricity stealing conclusion.
4. Construction of wiring error metering point knowledge base
Through theoretical derivation in the aspect of electricity, the influence on the metering of the ammeter under different wiring conditions is obtained, meanwhile, abnormal events reported by wiring error metering points, such as flow reversal, phase sequence abnormality, reverse electric quantity abnormality and the like, are recorded, and the method can be used for correlation analysis of the abnormal events in the later period, so that the wrong wiring judgment accuracy is improved. And (4) sorting the related contents into a wiring error metering point knowledge base, and diagnosing whether wiring errors exist or not through the input given by a wiring error abnormity diagnosis model.
5. Constructing and acquiring abnormal metering point knowledge base
And constructing an expert knowledge base for collecting abnormal metering points based on the daily freezing data, the abnormal event information and the high-frequency collected 96-point voltage and current data information of the electric energy meter. The phenomenon that the electricity consumption information data cannot be reported appears at the abnormal collecting metering points, the data of certain data points are null values when the electricity consumption information data are reflected in the daily freezing and voltage and current data, the electric meter is subjected to statistical analysis that the data points are null values, the collecting success rate of the electric meter is calculated, and the abnormal collecting metering points are found. In addition, comprehensive judgment can be performed by means of collected abnormal information provided by abnormal events so as to improve the diagnosis success rate.
After the diagnosis result is given, the diagnosis logic of abnormal acquisition can be continuously perfected and optimized according to the result of on-site check feedback.
6. Constructing a knowledge base of clock error metering points
And establishing a clock error metering point expert knowledge base based on the daily freezing data and the abnormal event information of the electric energy meter. On one hand, the knowledge base can diagnose whether the clock error occurs by utilizing the abnormity reported by the electric energy meter when the clock error occurs, on the other hand, when the clock error exists, daily frozen data can deviate, the translation of the electricity consumption of a certain block of meter can be tried, and whether the line loss is greatly improved is detected, so that the clock error is diagnosed.
The knowledge base stores the measuring points with problems and measures error information so as to facilitate the calling of the error analysis module.
7. Constructing a mutual inductor overload metering point knowledge base
And constructing a transformer overload metering point knowledge base based on high-frequency acquisition 96-point voltage and current data, daily freezing data and electric energy meter specification data of the electric energy meter. And estimating daily average working current based on daily freezing data and 96-point current data, and judging that the transformer is overloaded if the average current is greater than the maximum current of the electric energy meter.
And (5) feeding back an abnormal result by the on-site inspection work order to classify and store the abnormal result so as to facilitate automatic comparison and judgment at a later period.
8. Construction of knowledge base of error measurement points of user-variable relationship
The abnormal line loss caused by the disorder of the station area house-to-house transformation relationship influences error calculation and analysis and is not beneficial to the data management work of the station area. The method comprises the steps of recording information such as load environment, power utilization curve and geographical position of metering points with the wrong subscriber variable relationship, establishing a subscriber variable relationship metering point knowledge base, introducing relevant judgment indexes during station area error analysis, preprocessing data first, and greatly improving model error calculation accuracy. Meanwhile, according to the knowledge base of the metering point with the wrong user variable relationship, a comparison basis can be provided for diagnosis and treatment of the user variable relationship of similar metering points.
9. Constructing a mutual inductor light-load metering point knowledge base
And constructing a mutual inductor light-load metering point knowledge base based on high-frequency acquisition 96-point voltage and current data, daily freezing data and electric energy meter specification data of the electric energy meter. And estimating daily average working current based on daily freezing data and 96-point current data, setting a light load threshold, and judging that the mutual inductor is light load if the average current is less than the rated current set threshold of the electric energy meter.
And (5) feeding back an abnormal result by the on-site inspection work order to classify and store the abnormal result so as to facilitate automatic comparison and judgment at a later period.
10. Constructing collector abnormal knowledge base
And constructing an collector abnormal expert knowledge base based on the daily freezing data of the electric energy meter, the high-frequency collected 96-point voltage and current data information and the electric meter archive information. The collector abnormality can cause the phenomenon that batch electricity consumption information data of the electric meters cannot be reported, data of data points of a plurality of meters are null values when the electricity consumption information data of the electric meters are reflected to be in daily freezing and voltage and current data, statistical analysis is carried out on adjacent electric meters under a distribution area, the data points are null values, the collection success rate of the electric meters under the area is calculated, and therefore the collector abnormality is found.
After the diagnosis result is given, the abnormal diagnosis logic of the collector can be continuously perfected and optimized according to the result of the on-site check feedback.
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.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A method for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm is characterized by comprising the following steps: the method comprises the following steps:
(1) extracting abnormal metering point information in an electricity utilization information acquisition system;
(2) based on the abnormal metering point information, calculating the probability of the abnormal metering point knowledge information through an LSTM network unit model to form a multi-dimensional knowledge label;
(3) and extracting the formed multi-dimensional knowledge labels based on a bidirectional LSTM training network, establishing abnormal metering point knowledge identification labels aiming at the knowledge characteristic values with the maximum probability, and outputting a multi-dimensional abnormal metering point knowledge base.
2. The method for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 1, wherein the method comprises the following steps: the abnormal metering point information comprises instantaneous over-range metering point information, stable over-range metering point information, suspected electricity stealing metering point information, wiring error metering point information, collected abnormal metering point information, clock error metering point information, transformer overload metering point information, user variable relation error metering point information, transformer light load metering point information and collector abnormal information.
3. The method for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 2, wherein the method comprises the following steps: the instantaneous over-range metering point information comprises field check information and power utilization user file information, the stable over-range metering point information comprises the field check information and the power utilization user file information, the suspected electricity stealing metering point information comprises the field check information, running error calculation information, voltage and current information and power factor information, the wiring error metering point information comprises information such as reversed trend, abnormal phase sequence and reversed electric quantity, the collected abnormal metering point information comprises daily freezing data, abnormal event information and high-frequency collected 96-point voltage and current data information, the clock error metering point information comprises daily freezing data and abnormal event information, the transformer overload metering point information comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the user variable relation error metering point information comprises load environment, load environment and power consumption meter specification data, and the like, The system comprises a power utilization curve, a geographical position and other information, the light-load metering point information of the mutual inductor comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the collector abnormal information comprises daily freezing data, 96-point voltage and current data information and electric energy meter archive information.
4. The method for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 1, wherein the method comprises the following steps: the LSTM network element model is
Where p (wj) represents the prediction probability of the j-th knowledge point.
5. The method for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 1, wherein the method comprises the following steps: the multi-dimensional abnormal metering point knowledge base comprises an instantaneous over-range metering point knowledge base, a stable over-range metering point knowledge base, a suspected electricity stealing metering point knowledge base, a wiring error metering point knowledge base, an acquisition abnormal metering point knowledge base, a clock error metering point knowledge base, a mutual inductor overload metering point knowledge base, a user-variable relation error metering point knowledge base, a mutual inductor light-load metering point knowledge base and an acquisition device abnormal knowledge base.
6. An abnormal metering point knowledge base construction system based on an LSTM end-to-end extraction algorithm is characterized in that: comprises that
The abnormal metering point information extraction module is used for extracting the abnormal metering point information from the power utilization information acquisition system;
the abnormal metering point knowledge information probability calculation module is used for calculating the abnormal metering point knowledge information probability through an LSTM network unit model based on the abnormal metering point information to form a multi-dimensional knowledge label;
and the multi-dimensional abnormal metering point knowledge base construction module is used for extracting the formed multi-dimensional knowledge labels based on the bidirectional LSTM training network, establishing abnormal metering point knowledge identification labels aiming at the knowledge characteristic values with the maximum probability, and outputting the multi-dimensional abnormal metering point knowledge base.
7. The system for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 6, wherein: the abnormal metering point information comprises instantaneous over-range metering point information, stable over-range metering point information, suspected electricity stealing metering point information, wiring error metering point information, collected abnormal metering point information, clock error metering point information, transformer overload metering point information, user variable relation error metering point information, transformer light load metering point information and collector abnormal information.
8. The system for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 7, wherein: the instantaneous over-range metering point information comprises field check information and power utilization user file information, the stable over-range metering point information comprises the field check information and the power utilization user file information, the suspected electricity stealing metering point information comprises the field check information, running error calculation information, voltage and current information and power factor information, the wiring error metering point information comprises information such as reversed trend, abnormal phase sequence and reversed electric quantity, the collected abnormal metering point information comprises daily freezing data, abnormal event information and high-frequency collected 96-point voltage and current data information, the clock error metering point information comprises daily freezing data and abnormal event information, the transformer overload metering point information comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the user variable relation error metering point information comprises load environment, load environment and power consumption meter specification data, and the like, The system comprises a power utilization curve, a geographical position and other information, the light-load metering point information of the mutual inductor comprises 96-point voltage and current data, daily freezing data and electric energy meter specification data, and the collector abnormal information comprises daily freezing data, 96-point voltage and current data information and electric energy meter archive information.
9. The system for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 6, wherein: the LSTM network element model is
Where p (wj) represents the prediction probability of the j-th knowledge point.
10. The system for constructing an abnormal metering point knowledge base based on an LSTM end-to-end extraction algorithm according to claim 6, wherein: the multi-dimensional abnormal metering point knowledge base comprises an instantaneous over-range metering point knowledge base, a stable over-range metering point knowledge base, a suspected electricity stealing metering point knowledge base, a wiring error metering point knowledge base, an acquisition abnormal metering point knowledge base, a clock error metering point knowledge base, a mutual inductor overload metering point knowledge base, a user-variable relation error metering point knowledge base, a mutual inductor light-load metering point knowledge base and an acquisition device abnormal knowledge base.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911238633.5A CN110968703B (en) | 2019-12-06 | 2019-12-06 | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911238633.5A CN110968703B (en) | 2019-12-06 | 2019-12-06 | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110968703A true CN110968703A (en) | 2020-04-07 |
| CN110968703B CN110968703B (en) | 2024-01-02 |
Family
ID=70033120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911238633.5A Active CN110968703B (en) | 2019-12-06 | 2019-12-06 | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110968703B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469253A (en) * | 2021-07-02 | 2021-10-01 | 河海大学 | Electricity stealing detection method based on triple twin network |
| CN115292387A (en) * | 2022-08-18 | 2022-11-04 | 北京志翔科技股份有限公司 | Energy overrange use monitoring method based on multidimensional data analysis |
| CN115358347A (en) * | 2022-09-30 | 2022-11-18 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080300659A1 (en) * | 2007-05-17 | 2008-12-04 | Matos Jeffrey A | Implantable medical device which may be controlled from central station |
| CN105915398A (en) * | 2016-06-24 | 2016-08-31 | 广西电网有限责任公司南宁供电局 | Rapid fault detection and power recovery system based on rural power network, and concentrator detection method thereof |
| CN106161138A (en) * | 2016-06-17 | 2016-11-23 | 贵州电网有限责任公司贵阳供电局 | A kind of intelligence automatic gauge method and device |
| CN106199494A (en) * | 2016-07-25 | 2016-12-07 | 国网上海市电力公司 | A kind of intelligent diagnosis system based on metering device fault |
| CN106772205A (en) * | 2016-11-30 | 2017-05-31 | 广东电网有限责任公司电力科学研究院 | A kind of automatic power-measuring system terminal unit exception monitoring method and device |
| CN107703383A (en) * | 2017-09-21 | 2018-02-16 | 国网上海市电力公司 | A kind of method for building up of information acquisition system fault diagnosis knowledge base |
| CN108336725A (en) * | 2016-12-16 | 2018-07-27 | 泰豪软件股份有限公司 | The management of dispatching of power netwoks monitoring of tools and intelligent analysis system |
| CN109858679A (en) * | 2018-12-30 | 2019-06-07 | 国网浙江省电力有限公司 | A kind of opposing electricity-stealing for the man-machine object of combination checks monitoring system and its working method |
| CN109886475A (en) * | 2019-01-24 | 2019-06-14 | 广西电网有限责任公司电力科学研究院 | The information security Situation Awareness System of metering automation system based on AI |
| CN109934337A (en) * | 2019-03-14 | 2019-06-25 | 哈尔滨工业大学 | A kind of detection method of the spacecraft telemetry exception based on integrated LSTM |
| CN110009525A (en) * | 2019-04-02 | 2019-07-12 | 国网新疆电力有限公司电力科学研究院 | Power information acquisition system and application method |
| CN110046145A (en) * | 2019-04-02 | 2019-07-23 | 国网安徽省电力有限公司 | Expert intelligence Analysis Service platform based on electric flux big data research |
| CN110109771A (en) * | 2019-05-07 | 2019-08-09 | 北京恒泰实达科技股份有限公司 | The abnormal key diagnostic method of a kind of area's line loss |
-
2019
- 2019-12-06 CN CN201911238633.5A patent/CN110968703B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080300659A1 (en) * | 2007-05-17 | 2008-12-04 | Matos Jeffrey A | Implantable medical device which may be controlled from central station |
| CN106161138A (en) * | 2016-06-17 | 2016-11-23 | 贵州电网有限责任公司贵阳供电局 | A kind of intelligence automatic gauge method and device |
| CN105915398A (en) * | 2016-06-24 | 2016-08-31 | 广西电网有限责任公司南宁供电局 | Rapid fault detection and power recovery system based on rural power network, and concentrator detection method thereof |
| CN106199494A (en) * | 2016-07-25 | 2016-12-07 | 国网上海市电力公司 | A kind of intelligent diagnosis system based on metering device fault |
| CN106772205A (en) * | 2016-11-30 | 2017-05-31 | 广东电网有限责任公司电力科学研究院 | A kind of automatic power-measuring system terminal unit exception monitoring method and device |
| CN108336725A (en) * | 2016-12-16 | 2018-07-27 | 泰豪软件股份有限公司 | The management of dispatching of power netwoks monitoring of tools and intelligent analysis system |
| CN107703383A (en) * | 2017-09-21 | 2018-02-16 | 国网上海市电力公司 | A kind of method for building up of information acquisition system fault diagnosis knowledge base |
| CN109858679A (en) * | 2018-12-30 | 2019-06-07 | 国网浙江省电力有限公司 | A kind of opposing electricity-stealing for the man-machine object of combination checks monitoring system and its working method |
| CN109886475A (en) * | 2019-01-24 | 2019-06-14 | 广西电网有限责任公司电力科学研究院 | The information security Situation Awareness System of metering automation system based on AI |
| CN109934337A (en) * | 2019-03-14 | 2019-06-25 | 哈尔滨工业大学 | A kind of detection method of the spacecraft telemetry exception based on integrated LSTM |
| CN110009525A (en) * | 2019-04-02 | 2019-07-12 | 国网新疆电力有限公司电力科学研究院 | Power information acquisition system and application method |
| CN110046145A (en) * | 2019-04-02 | 2019-07-23 | 国网安徽省电力有限公司 | Expert intelligence Analysis Service platform based on electric flux big data research |
| CN110109771A (en) * | 2019-05-07 | 2019-08-09 | 北京恒泰实达科技股份有限公司 | The abnormal key diagnostic method of a kind of area's line loss |
Non-Patent Citations (1)
| Title |
|---|
| 罗慧，刘梅招，周钰山: "基于长短期记忆网络的智能用电数据甄别方法", vol. 32, no. 2, pages 47 - 54 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469253A (en) * | 2021-07-02 | 2021-10-01 | 河海大学 | Electricity stealing detection method based on triple twin network |
| CN113469253B (en) * | 2021-07-02 | 2024-05-14 | 河海大学 | Electric larceny detection method based on triple twinning network |
| CN115292387A (en) * | 2022-08-18 | 2022-11-04 | 北京志翔科技股份有限公司 | Energy overrange use monitoring method based on multidimensional data analysis |
| CN115292387B (en) * | 2022-08-18 | 2023-05-23 | 北京志翔科技股份有限公司 | Multi-dimensional data analysis-based energy source overscan use monitoring method |
| CN115358347A (en) * | 2022-09-30 | 2022-11-18 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
| CN115358347B (en) * | 2022-09-30 | 2023-01-31 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110968703B (en) | 2024-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108375715B (en) | Power distribution network line fault risk day prediction method and system | |
| CN106054104B (en) | A kind of intelligent electric meter failure real-time predicting method based on decision tree | |
| Zhu et al. | Time series shapelet classification based online short-term voltage stability assessment | |
| CN111103565B (en) | Data transformation method and system based on intelligent electric energy meter metering error analysis | |
| CN106154209A (en) | Electrical energy meter fault Forecasting Methodology based on decision Tree algorithms | |
| CN110968703B (en) | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm | |
| CN109472241A (en) | Combustion engine bearing remaining life prediction technique based on support vector regression | |
| CN108830406A (en) | A kind of main distribution based on data mining has a power failure configuration method in advance | |
| CN105759201A (en) | High voltage circuit breaker self-diagnosis method based on abnormal sample identification | |
| CN111612019A (en) | Method for identifying and analyzing fault abnormality of intelligent electric meter based on big data model | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| CN116432123A (en) | Electric energy meter fault early warning method based on CART decision tree algorithm | |
| CN114399076A (en) | Method for screening electricity stealing suspicion users based on big data analysis | |
| CN117390546A (en) | Multimode database fusion calculation model for instant anti-electricity-theft detection | |
| CN118194202A (en) | Transverse federal-based electricity stealing identification algorithm and prototype system thereof | |
| CN114595762A (en) | Photovoltaic power station abnormal data sequence extraction method | |
| CN114626433A (en) | Fault prediction and classification method, device and system for intelligent electric energy meter | |
| CN112508254B (en) | Method for determining investment prediction data of transformer substation engineering project | |
| Dong et al. | Fault diagnosis and classification in photovoltaic systems using scada data | |
| CN112418662A (en) | Power distribution network operation reliability analysis method using artificial neural network | |
| CN116879672A (en) | Intelligent identification method for total fault of area based on time convolution network | |
| CN110261693A (en) | A kind of online thermal test method and system based on stable state sampling technique | |
| CN114139408A (en) | Power transformer health state assessment method | |
| CN113537770A (en) | Decision tree configuration life prediction method and system based on cloud computing | |
| CN114168662A (en) | Power distribution network problem combing and analyzing method and system based on multiple data sources |
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 |