CN115392615A - Data missing value completion method and system for generating countermeasure network based on information enhancement - Google Patents
Data missing value completion method and system for generating countermeasure network based on information enhancement Download PDFInfo
- Publication number
- CN115392615A CN115392615A CN202210454835.9A CN202210454835A CN115392615A CN 115392615 A CN115392615 A CN 115392615A CN 202210454835 A CN202210454835 A CN 202210454835A CN 115392615 A CN115392615 A CN 115392615A
- Authority
- CN
- China
- Prior art keywords
- data
- information
- representation vector
- vector
- text information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 239000013598 vector Substances 0.000 claims abstract description 94
- 238000000547 structure data Methods 0.000 claims abstract description 33
- 230000005611 electricity Effects 0.000 claims abstract description 13
- 238000004590 computer program Methods 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 10
- 230000006399 behavior Effects 0.000 claims description 9
- 230000007812 deficiency Effects 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 8
- 239000013604 expression vector Substances 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 230000003416 augmentation Effects 0.000 claims description 2
- 238000012549 training Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 13
- 239000011159 matrix material Substances 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000007781 pre-processing Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 230000000295 complement effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005111 flow chemistry technique Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the field of intelligent power utilization, and particularly relates to a data missing value completion method and system for generating a countermeasure network based on information enhancement. The method comprises the steps of acquiring text information in the power data; respectively extracting a global information representation vector and a local information representation vector of the text information; integrating the global information representation vector and the local information representation vector to obtain a representation vector of the text information; constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data; based on the knowledge graph, capturing the mutual correlation among graph structure data to obtain a representation vector of the graph structure data; and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data. The method can effectively improve the accuracy of completing the missing values of the power data.
Description
Technical Field
The invention belongs to the field of intelligent power utilization, and particularly relates to a data missing value completion method and system for generating a countermeasure network based on information enhancement.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Thanks to the development of information technology, the informatization level of the power system is greatly improved in the power industry. In recent years, with the rapid development of deep learning technology, massive power data is an important guarantee for the intellectualization of a power information system. The intelligent analysis and decision system based on the mass power data can greatly improve the operation efficiency of the power system and can greatly reduce the operation cost of the power system. However, due to intrinsic defects of the power system itself and occurrence of a large number of uncontrollable abnormal events such as natural disasters, power system sensor abnormalities, and power transmission line failures, a certain percentage of these huge amounts of power data are missing data.
In the process of constructing an intelligent power system, the missing processing of power data is an urgent problem to be solved, and the missing data can directly influence the analysis and decision performance of the intelligent power system. And missing value completion is an important method for improving the quality of power data. At present, the traditional method mainly depends on manpower such as data experts and business experts to complete missing values. However, as the scale of the power data gradually expands, the conventional method for performing deficiency completion and fade-in on real-time heterogeneous data is insufficient. Meanwhile, manual management also causes the phenomena of error and leakage and the like, so that the service data use requirement cannot be quickly and accurately met. In addition, the cost of missing value completion by manpower is high, the period is long, and the operation and maintenance efficiency of the intelligent power system is seriously influenced.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a data missing value completion method and system for generating a countermeasure network based on information enhancement.
In order to achieve the purpose, the invention adopts the following technical scheme:
a first aspect of the present invention provides a method for generating a data deficiency value completion for a countermeasure network based on information enhancement.
The method for generating the data missing value completion of the confrontation network based on the information enhancement comprises the following steps:
acquiring text information in the power data;
respectively extracting a global information representation vector and a local information representation vector of the text information;
integrating the global information representation vector and the local information representation vector to obtain a representation vector of the text information;
constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data;
based on the knowledge graph, capturing the mutual correlation relationship between graph structure data to obtain a representation vector of the graph structure data;
and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data.
A second aspect of the present invention provides a data deficiency value completion system for generating a countermeasure network based on information enhancement.
The data missing value completion system for generating the countermeasure network based on the information enhancement comprises:
an acquisition module configured to: acquiring text information in the power data;
a first feature extraction module configured to: respectively extracting a global information representation vector and a local information representation vector of the text information;
a fusion module configured to: fusing the global information representation vector and the local information representation vector to obtain a representation vector of the text information;
a knowledge-graph building module configured to: constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data;
a second feature extraction module configured to: based on the knowledge graph, capturing the mutual correlation among graph structure data to obtain a representation vector of the graph structure data;
a completion module configured to: and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data.
A third aspect of the invention provides a computer-readable storage medium.
A computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the method for generating a data-missing value complement for a countermeasure network based on information enhancement as described in the first aspect above.
A fourth aspect of the invention provides a computer apparatus.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the method for generating a data-missing value complement against a network based on information enhancement as described in the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
the method is based on the power data, and performs representation learning from two aspects of graph structure data constructed based on the power data and text information in the power data, so as to enhance the representation capability of the data. Firstly, constructing a knowledge graph of user electricity consumption behaviors based on historical power data of users, and capturing mutual association relation among graph structure data by adopting a graph attention network to obtain a representation vector of the graph structure data; then, based on text information in the power data, representation learning is carried out on the text information by adopting a method of combining Kalman filtering and a one-dimensional convolutional neural network, so that a representation vector of the text information is obtained; the method takes a Transformer as a generator and an SVM as a discriminator, builds a model based on a generation countermeasure network to realize the completion of the missing value of the electric power data, and improves the completion performance through the mutual countermeasure learning of the generator and the discriminator. The method improves the accuracy of completing the missing values of the power data and reduces the cost.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are included to illustrate an exemplary embodiment of the invention and not to limit the invention.
Fig. 1 is an overall flowchart of a data missing value completing method according to an embodiment of the present invention;
fig. 2 is a flow chart of data flow processing for completing missing data values according to an embodiment of the present invention;
FIG. 3 is a flowchart of a data missing value completion method according to an embodiment of the present invention;
fig. 4 is a schematic diagram illustrating an effect of the data missing value completion method on the data amount according to an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
It is noted that the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present invention. It should be noted that each block in the flowchart or block diagrams may represent a module, a segment, or a portion of code, which may comprise one or more executable instructions for implementing the logical function specified in the respective embodiment. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Example one
As shown in fig. 1, the present embodiment provides a method for generating a data missing value complement for a countermeasure network based on information enhancement, and the present embodiment is illustrated by applying the method to a server, it is understood that the method may also be applied to a terminal, and may also be applied to a system including a terminal and a server, and is implemented by interaction between the terminal and the server. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network server, cloud communication, middleware service, a domain name service, a security service CDN, a big data and artificial intelligence platform, and the like. The terminal may be, but is not limited to, a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and the like. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein. In this embodiment, the method includes the steps of:
the method for generating the data missing value completion of the confrontation network based on the information enhancement comprises the following steps:
acquiring text information in the power data;
respectively extracting a global information representation vector and a local information representation vector of the text information;
fusing the global information representation vector and the local information representation vector to obtain a representation vector of the text information;
constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data;
based on the knowledge graph, capturing the mutual correlation among graph structure data to obtain a representation vector of the graph structure data;
and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data.
As shown in fig. 1, fig. 2, and fig. 3, the above-mentioned scheme of this embodiment can be implemented by the following steps:
A. the method comprises the steps of collecting mass power data, preprocessing the data, marking missing values, standardizing non-missing values, defining and storing the data and the like.
B. For electricity consumption data acquired in a certain provincial power system in China, 24 time point data are acquired once per hour and per day, and the data comprise 334656 pieces of electricity consumption data from 996 users in 2019, month 1 and 2019, month 9 and month 14. It should be noted that the present embodiment takes data taken at each time point as a piece of electricity consumption data.
C. For text descriptions in the power data, such as information of device information description, user addresses, and the like, the present embodiment uses a kalman filter and a one-dimensional convolutional neural network to mine global information and local information in the text sequence, and uses an Attention mechanism (Attention) to fuse the two types of dependency information, thereby obtaining a representation vector of the text information.
C1. Based on text information in the electric power data, a Kalman filter is adopted for data feature extraction, so that the global information expression of the text data is obtained, and the formula is as follows:
x k =Ax k-1 +Bu k-1 +w k-1
wherein ,wk-1 Is noise, x, following a Gaussian distribution k-1 and uk-1 The a-posteriori state estimates at time k-1 and k, respectively, are represented, a-state transition matrix, B is the matrix that converts the input to state.
C2. Based on text information in the electric power data, a one-dimensional convolution neural network is adopted to extract local features of the data, so that local information expression of the text data is obtained, and the formula is as follows:
and (3) convolution operation:
c={c 1 ,c 2 ,...,c t-h+1 }
where ReLU denotes the activation function, v i Representing an embedded representation of the text information, W c Representing weight vectors, b c Representing the basis vectors of the feature map, c i The vectors of the previous layer are shown.
And (3) pooling operation:
h=max{c}
wherein, the above formula uses maximal pooling operation to perform feature dimension reduction processing; h represents a finally obtained text data local information representation vector.
C3. The Attention-drawing mechanism (Attention) fuses the local information representation vector and the global information representation vector to obtain a representation vector x' of the text information, the formula is as follows:
x′=Attention(Concat(x k ,h))
D. the method comprises the steps of constructing a knowledge graph of electricity utilization behaviors of users based on historical electricity data, and then capturing the mutual association relation among graph structure data by using a graph attention network to obtain a representation vector of the graph structure data.
D1. Based on historical power data, the implementation firstly constructs a knowledge graph of the power utilization behavior of the user, and uses G = (V, E, A, X) to represent, wherein V ∈ R n Representing the nodes of the graph G, and n represents the number of nodes; e is the edge of FIG. G; a is an element of R n ×n An adjacency matrix of G; for theIf e ij E is E, then A ij =1, otherwise A ij =0;X∈R n×m The attribute information of the node is shown, and m is the dimension of the attribute information.
D2. A Graph neural network (GAT) based Attention mechanism is adopted to learn the low-dimensional embedded vector of each node. Suppose the graph comprises N nodes, and the feature vector of each node is H i Dimension F, as follows:
H={H 1 ,H 2 ,...,H N }
the node feature vector H is subjected to linear transformation to obtain a new feature vector H' i As follows:
H′ i =WH i
H′={H′ 1 ,H′ 2 ,...,H′ N },H′ i ∈R F′
wherein ,W∈RF′×F For a linear transformed matrix, F' denotes the dimension of the transformed matrix.
Drawing attention network is introduced, and feature vectors H 'of nodes i, j are' i ,H′ j Spliced together and then the inner product is calculated with a 2F' dimensional vector a. The activating function adopts ReLU function and formulaThe following:
where, σ denotes the activation function,the representation graph structure data represents a vector.
E. Based on the text information expression vector x' obtained in step C and the graph structure data expression vector obtained in step DIn the embodiment, a Transformer is used as a generator, an SVM is used as a discriminator, a generation confrontation network model is constructed, and data missing value completion is realized through the mutual confrontation learning between the generator and the discriminator.
E1. In generating the confrontation network model, the Transformer generator is represented as follows:
Multihead(Q,K,V)=concat(head 1 ,...,head 2 ,...,head n )W O
wherein ,W,W O a parameter vector is represented. The remaining portion of the Transformer is represented as follows:
H 1 =LN(Multihead(Q,K,V))
H 2 =ReLU(H 1 )W f +b f
H 3 =LN(H 1 +H 2 )
where LN denotes the layer normalization method, reLU denotes the activation function, W f and bf Weight matrix of the table, H 1 and H2 Hidden vector in transform, H, representation 3 Represented is a comprehensive representation vector of the power data.
E2. Finally, the H obtained by the Transformer is used 3 Inputting a softmax layer for data missing completion, and completing the result:
F. and E, based on the electric power data completion result obtained in the step E, combining the real data, and constructing a discriminator by using the SVM, wherein whether the completion result of the discriminator belongs to the real data or not is judged. The optimization problem of SVM is as follows:
where ω and b represent weight vectors.
The lagrange multiplier method is used for the above equation to convert to the optimized form:
where α is the lagrange multiplier.
G. And F, feeding back and updating data information in the generator based on the judgment result obtained in the step F, continuously optimizing the data weight value in the model through mutual confrontation, continuously improving the missing data completion effect until the loss of the model is not obviously reduced any more and the model integrally reaches a convergence state, and terminating the model training by using an early-stop strategy. Using the cross entropy as a loss function, if y is a real category distribution, the loss function is defined as follows:
wherein ,Spositive and Snegative Positive and negative sample data are shown separately.Representing a sampleProbability of belonging to the real data.
The method comprises the steps of completing missing values of samples to be tested, pushing detection results, and comparing the results with actual power data, wherein Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) are used as evaluation indexes.
TABLE 1 comparison of different deficiency completion methods on RMSE
10% | 20% | 30% | 40% | 50% | |
LI | 4.8863 | 5.9194 | 6.7094 | 8.5074 | 8.6406 |
LSTM | 4.5091 | 5.4308 | 6.3513 | 8.0623 | 8.1192 |
GAN | 4.4793 | 5.2251 | 6.1371 | 7.7657 | 7.8047 |
ConvNN-DAE | 4.6868 | 6.3127 | 6.6556 | 7.8151 | 7.8978 |
ST-LRAGAN | 3.5078 | 4.3629 | 5.1809 | 4.8096 | 5.4563 |
This example | 3.0162 | 3.1833 | 3.3974 | 3.5465 | 3.7236 |
TABLE 2 comparison of different deficiency completion methods on MAE
10% | 20% | 30% | 40% | 50% | |
LI | 0.4017 | 0.4841 | 0.5479 | 0.6876 | 0.8011 |
LSTM | 0.3952 | 0.4716 | 0.5253 | 0.6604 | 0.7803 |
GAN | 0.3781 | 0.4558 | 0.5104 | 0.6339 | 0.7554 |
ConvNN-DAE | 0.3526 | 0.4335 | 0.4962 | 0.5725 | 0.6386 |
ST-LRAGAN | 0.3703 | 0.4607 | 0.5182 | 0.6377 | 0.6052 |
This example | 0.3649 | 0.4319 | 0.4617 | 0.5433 | 0.5704 |
Based on the results in tables 1 and 2, the missing value completion model proposed in this example outperformed other methods in the course of the data missing rate from 10% to 50%. From table 2, the missing value completion performance is improved by about 17.39% on average compared with the conventional baseline method such as LSTM. In addition, fig. 4 shows the change of the data amount after completion in the process of the data missing rate from 10% to 50%. As can be seen from fig. 4, when the data missing rate is 10% to 50%, the missing data amount is not more than about 5% of the total data amount after the missing data is completed by the completion method of the present embodiment.
Example two
The embodiment provides a data missing value completion system for generating a countermeasure network based on information enhancement.
A missing-data-value-complementing system for generating a countermeasure network based on information augmentation, comprising:
an acquisition module configured to: acquiring text information in the power data;
a first feature extraction module configured to: respectively extracting a global information representation vector and a local information representation vector of the text information;
a fusion module configured to: integrating the global information representation vector and the local information representation vector to obtain a representation vector of the text information;
a knowledge-graph building module configured to: constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data;
a second feature extraction module configured to: based on the knowledge graph, capturing the mutual correlation among graph structure data to obtain a representation vector of the graph structure data;
a completion module configured to: and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data.
As one or more embodiments, the system further comprises:
and the preprocessing module is used for preprocessing the power data before representation learning of the power data, and comprises non-missing value feature standardization, missing value feature labeling and the like.
It should be noted here that the acquiring module, the first feature extracting module, the fusing module, the knowledge graph constructing module, the second feature extracting module, the complementing module and the preprocessing module are the same as the example and the application scenario realized by the steps in the first embodiment, but are not limited to the disclosure in the first embodiment. It should be noted that the modules described above as part of a system may be implemented in a computer system such as a set of computer-executable instructions.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the data missing value completion method for generating a countermeasure network based on information enhancement as described in the first embodiment above.
Example four
The embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the processor implements the steps in the data missing value completion method for generating an anti-network based on information enhancement as described in the first embodiment.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention 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, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by a computer program, which may be stored in a computer readable storage medium and executed by a computer to implement the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The method for generating the data missing value completion of the countermeasure network based on the information enhancement is characterized by comprising the following steps:
acquiring text information in the power data;
respectively extracting a global information representation vector and a local information representation vector of the text information;
integrating the global information representation vector and the local information representation vector to obtain a representation vector of the text information;
constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data;
based on the knowledge graph, capturing the mutual correlation relationship between graph structure data to obtain a representation vector of the graph structure data;
and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data.
2. The method for complementing data missing values based on information enhancement generation countermeasure network of claim 1, wherein if the countermeasure network model is a model to be optimized, the method is characterized in that whether the complementing result belongs to real data or not is judged by using a discriminator based on the electric power data complementing result in combination with the real data to obtain a discrimination result; and feeding back and updating parameters of the confrontation network model based on the discrimination result, continuously optimizing the data weight value in the model through mutual confrontation, and terminating the model training until the model reaches a convergence state to obtain the optimized confrontation network model.
3. The method for generating a data deficiency value completion for an anti-network based on information enhancement as claimed in claim 1, wherein the extracting the global information representation vector and the local information representation vector of the text information respectively specifically comprises:
based on text information in the electric power data, performing data feature extraction by adopting a Kalman filter to obtain a global information expression vector of the text information;
and based on the text information in the electric power data, extracting the local features of the data by adopting a one-dimensional convolution neural network to obtain a local information expression vector of the text information.
4. The method for generating a completion of data missing value for an anti-network based on information enhancement as claimed in claim 1, wherein the fusing the global information representation vector and the local information representation vector to obtain the representation vector of the text information specifically comprises: and (4) introducing an attention mechanism to fuse the local information representation vector and the global information representation vector to obtain a representation vector of the text information.
5. The method for generating data missing value completion for countermeasure network based on information enhancement as claimed in claim 1, wherein the capturing the correlation between graph structure data based on the knowledge graph to obtain the representation vector of the graph structure data specifically comprises:
learning a low-dimensional embedded vector of each node in the knowledge graph by adopting a graph neural network based on an attention mechanism, and preliminarily obtaining a characteristic vector of each node; and then, for each node, using an attention mechanism to aggregate information of neighbor nodes and generate node representation vectors, and finally, using average pooling operation to aggregate the node representation vectors of all nodes in the knowledge graph to obtain the representation vectors of the graph structure data.
6. The method for generating the data missing value completion of the countermeasure network based on the information enhancement as claimed in claim 1, wherein the countermeasure network model comprises a generator and a discriminator, and the data missing value completion is realized through the mutual countermeasure learning between the generator and the discriminator.
7. The method for generating the data missing value completion of the countermeasure network based on the information enhancement as claimed in claim 6, wherein the process of obtaining the power data completion result by using the countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data comprises: and obtaining a comprehensive expression vector of the electric power data by adopting a generator based on the expression vector of the text information and the expression vector of the graph structure data.
8. The data missing value completion system for generating the countermeasure network based on the information enhancement is characterized by comprising the following steps:
an acquisition module configured to: acquiring text information in the power data;
a first feature extraction module configured to: respectively extracting a global information representation vector and a local information representation vector of the text information;
a fusion module configured to: integrating the global information representation vector and the local information representation vector to obtain a representation vector of the text information;
a knowledge-graph building module configured to: constructing a knowledge graph of the electricity utilization behavior of the user based on the electric power data;
a second feature extraction module configured to: based on the knowledge graph, capturing the mutual correlation among graph structure data to obtain a representation vector of the graph structure data;
a completion module configured to: and obtaining a power data completion result by adopting a countermeasure network model based on the representation vector of the text information and the representation vector of the graph structure data.
9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the method for generating a data deficiency value completion for a countermeasure network based on information augmentation according to any one of claims 1-7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the method for generating a data deficiency value completion for a countermeasure network based on information enhancement according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210454835.9A CN115392615B (en) | 2022-04-24 | 2022-04-24 | Data missing value completion method and system for generating countermeasure network based on information enhancement |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210454835.9A CN115392615B (en) | 2022-04-24 | 2022-04-24 | Data missing value completion method and system for generating countermeasure network based on information enhancement |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115392615A true CN115392615A (en) | 2022-11-25 |
CN115392615B CN115392615B (en) | 2023-09-29 |
Family
ID=84115753
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210454835.9A Active CN115392615B (en) | 2022-04-24 | 2022-04-24 | Data missing value completion method and system for generating countermeasure network based on information enhancement |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115392615B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210201169A1 (en) * | 2019-10-31 | 2021-07-01 | American Family Mutual Insurance Company, S.I. | Systems and methods for collecting and processing data for insurance-related tasks |
CN113934847A (en) * | 2021-10-21 | 2022-01-14 | 山大地纬软件股份有限公司 | Knowledge graph completion method and system based on unstructured information |
CN114022311A (en) * | 2021-11-16 | 2022-02-08 | 东北大学 | Comprehensive energy system data compensation method for generating countermeasure network based on time sequence condition |
CN114091615A (en) * | 2021-11-26 | 2022-02-25 | 广东工业大学 | Electric energy metering data completion method and system based on generation countermeasure network |
-
2022
- 2022-04-24 CN CN202210454835.9A patent/CN115392615B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210201169A1 (en) * | 2019-10-31 | 2021-07-01 | American Family Mutual Insurance Company, S.I. | Systems and methods for collecting and processing data for insurance-related tasks |
CN113934847A (en) * | 2021-10-21 | 2022-01-14 | 山大地纬软件股份有限公司 | Knowledge graph completion method and system based on unstructured information |
CN114022311A (en) * | 2021-11-16 | 2022-02-08 | 东北大学 | Comprehensive energy system data compensation method for generating countermeasure network based on time sequence condition |
CN114091615A (en) * | 2021-11-26 | 2022-02-25 | 广东工业大学 | Electric energy metering data completion method and system based on generation countermeasure network |
Non-Patent Citations (4)
Title |
---|
"A deep learning approach for anomaly detection and prediction in power consumption data", ENERGY EFFICIENCY * |
冀俭俭;杨刚;: "基于生成对抗网络的分级联合图像补全方法", 图学学报, no. 06 * |
张波等: "SVM在中压配网停电事件补全中的应用研究", 《电力工程技术》, pages 34 - 40 * |
王旭磊: "基于GAN和GRU的时间序列预测和填补方法研究", 基础科学, pages 1 - 69 * |
Also Published As
Publication number | Publication date |
---|---|
CN115392615B (en) | 2023-09-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021000556A1 (en) | Method and system for predicting remaining useful life of industrial equipment, and electronic device | |
CN115249332B (en) | Hyperspectral image classification method and device based on space spectrum double-branch convolution network | |
CN114221790A (en) | BGP (Border gateway protocol) anomaly detection method and system based on graph attention network | |
CN112884758B (en) | Defect insulator sample generation method and system based on style migration method | |
CN115272776B (en) | Hyperspectral image classification method based on double-path convolution and double attention and storage medium | |
CN113268370B (en) | Root cause alarm analysis method, system, equipment and storage medium | |
CN112217674A (en) | Alarm root cause identification method based on causal network mining and graph attention network | |
CN109412161A (en) | A kind of Probabilistic Load calculation method and system | |
CN113010683A (en) | Entity relationship identification method and system based on improved graph attention network | |
CN111783688B (en) | Remote sensing image scene classification method based on convolutional neural network | |
CN113536508A (en) | Method and system for classifying manufacturing network nodes | |
CN115438190B (en) | Power distribution network fault auxiliary decision knowledge extraction method and system | |
CN115392615B (en) | Data missing value completion method and system for generating countermeasure network based on information enhancement | |
CN116527346A (en) | Threat node perception method based on deep learning graph neural network theory | |
CN114357160B (en) | Early rumor detection method and device based on generated propagation structural features | |
CN113886547B (en) | Client real-time dialogue switching method and device based on artificial intelligence and electronic equipment | |
CN112199287B (en) | Cross-project software defect prediction method based on enhanced hybrid expert model | |
CN115209441A (en) | Method, device, equipment and storage medium for predicting base station out-of-service alarm | |
CN111967577A (en) | Energy internet scene generation method based on variational self-encoder | |
CN110727532A (en) | Data restoration method, electronic device and storage medium | |
CN116842073B (en) | Graph data mining method and device and electronic equipment | |
CN114611712B (en) | Prediction method based on heterogeneous federated learning, model generation method and device | |
CN115563506A (en) | Cross-system heterogeneous data fusion characterization method | |
CN116662903A (en) | Power grid terminal equipment identification method and system based on graphic neural network | |
CN118427720A (en) | Power system transient stability judging method and system based on interpretable graph neural tree |
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 |