CN110750578A - Networked multi-agent system modeling method based on data mining - Google Patents
Networked multi-agent system modeling method based on data mining Download PDFInfo
- Publication number
- CN110750578A CN110750578A CN201910997342.8A CN201910997342A CN110750578A CN 110750578 A CN110750578 A CN 110750578A CN 201910997342 A CN201910997342 A CN 201910997342A CN 110750578 A CN110750578 A CN 110750578A
- Authority
- CN
- China
- Prior art keywords
- data
- agent
- mining
- agent system
- networked multi
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000007418 data mining Methods 0.000 title claims abstract description 18
- 238000005094 computer simulation Methods 0.000 title claims abstract description 9
- 238000005516 engineering process Methods 0.000 claims abstract description 14
- 238000005065 mining Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000004458 analytical method Methods 0.000 claims abstract description 5
- 238000005457 optimization Methods 0.000 claims abstract description 5
- 238000013178 mathematical model Methods 0.000 claims abstract description 3
- 239000012528 membrane Substances 0.000 claims description 17
- 230000006399 behavior Effects 0.000 claims description 14
- 238000004891 communication Methods 0.000 claims description 3
- 230000008878 coupling Effects 0.000 claims description 3
- 238000010168 coupling process Methods 0.000 claims description 3
- 238000005859 coupling reaction Methods 0.000 claims description 3
- 230000036961 partial effect Effects 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 claims description 2
- 238000013528 artificial neural network Methods 0.000 claims 6
- 238000013075 data extraction Methods 0.000 claims 1
- 238000011160 research Methods 0.000 abstract description 7
- 238000013461 design Methods 0.000 abstract description 5
- 230000002452 interceptive effect Effects 0.000 abstract description 2
- 238000007405 data analysis Methods 0.000 description 3
- 238000005034 decoration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/283—Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
Abstract
The invention discloses a networked multi-agent system modeling method based on data mining, belongs to the field of data information optimization control methods, and aims to research the problem of multi-dimensional dynamic compensation of interactive information loss of networked multi-agent systems. And constructing a multi-agent intima data cube by adopting an online analysis processing technology. And mining based on the intima association rule, compensating the lost data and eliminating the influence of data loss on the control performance. And constructing a mathematical model of the optimization problem of the networked multi-agent system by using the relative state deviation. The invention researches a data mining-based distributed optimal control scheme of a networked multi-agent system, provides a method for solving an optimal control problem by using data information under the condition that a system model is completely unknown, and overcomes the limitation of the design of a control method only based on the model.
Description
Technical Field
The invention discloses a networked multi-agent system modeling method based on data mining, and belongs to the field of data information optimization control methods.
Background
At present, the networked multi-agent modeling research based on data mining is still in a primary stage, a set of complete theoretical system is not provided to support the development of related technologies, and the research of the project is urgently needed to be developed. The problem of networked multi-agent modeling based on data mining is the technical field with clear application background and considerable research difficulty, and the traditional technology and method are difficult to adapt to the requirements of practical problems, so that further exploration work needs to be carried out. The invention adopts a data mining mode, and searches the information hidden in the data from the data through a design algorithm so as to estimate or predict the lost data.
Disclosure of Invention
The invention aims to provide a method for estimating or predicting lost data by searching information hidden in data from the data through a design algorithm in a data mining mode under the condition that a system model is completely unknown. And constructing a multi-agent intima data cube by adopting an online analysis processing technology. And mining based on the intima association rule, compensating the lost data and eliminating the influence of data loss on the control performance. And constructing a mathematical model of the optimization problem of the networked multi-agent system by using the relative state deviation. The method is a statistical method for analyzing the coupling correlation between the relative state deviation and the intelligent agent control behavior, mining the correlation rule and researching the control behavior prediction.
The technical scheme of the invention is as follows:
in order to solve the problem of multi-dimensional dynamic compensation of interactive information loss of a networked multi-agent system, the invention adopts a data loss prediction method based on intimal association rule mining. Firstly, constructing a multi-agent intima data cube model; secondly, knowledge is mined based on the inner membrane association rule, and control behaviors which cannot be calculated due to loss of the index data are predicted. It is noted that the data cube is an image of the multi-dimensional model and is not limited to three-dimensional models.
OLAP technology is a general, fast and effective multidimensional data analysis technology, and intimal association rule mining is also a general method for data mining. How to utilize OLAP technology and inner membrane association rules to solve the problem that a networked multi-agent system cannot estimate a distributed control strategy due to data loss is rarely related to in the existing invention or is a public problem. The following analysis needs to be carried out:
The invention has the advantages and effects that:
the existing research of a networked multi-agent system with data loss rarely adopts an OLAP technology and an inner membrane association rule method to research a data loss compensation strategy and estimate or predict lost data, and the invention has wide application value and profound theoretical value. The invention adopts a data mining mode, and searches the information hidden in the data from the data through a design algorithm so as to estimate or predict the lost data. The problem that data are easily lost in equipment, production and operation containing a large amount of information is solved, and the risk that the system performance is influenced and the system cannot normally operate is avoided. Data-driven based control methods are widely considered to be highly competitive, a new generation of advanced control technology well suited for large-scale complex application systems.
Drawings
FIG. 1 is a general scheme of data loss prediction based on inner-membrane association rule mining;
FIG. 2 is a smart internal membrane data cube.
Detailed Description
In order to further illustrate the present invention, the following detailed description of the present invention is given with reference to the accompanying drawings and examples, which should not be construed as limiting the scope of the present invention.
Example (b):
the principle of the method of the invention is described as follows: the networked multi-agent system has large-scale characteristics, and unfavorable phenomena such as data loss and the like exist in the mutual communication of the agents, so that accurate and reliable data analysis and data loss compensation are needed. As shown in FIG. 1, the present invention adopts a data mining method, and searches the information hidden in the data from the data through a design algorithm so as to estimate or predict the lost data. The invention relates to an on-line analytical processing (OLAP) technology, which is a quick and effective multidimensional data analysis technology. In view of the unavoidable nature of the distributed relationship between the agents (the agents can only communicate with the neighbor agents) and the phenomenon of communication information loss (part of information is lost due to transmission delay exceeding a given threshold value or due to interference, collision and the like because of limited bandwidth) and the influence of information loss on the performance of the networked multi-agent system, the invention aims to adopt an online analytical processing (OLAP) technology and an association rule mining method to compensate the lost data.
The scientific challenges to be solved in the present invention at this stage include: a) how to determine the dimension of the data cube and the information of the networked multi-agent system represented by each dimension; b) how to make a quantitative analysis of the information contained in each dimension. First, for each agent，
Is the total number of agents. Determining the dimensionality of the data cube based on the number of neighboring agents and the number of leaders (0 or 1), and designating each dimension to represent an agent
Relative state deviation from neighboring agents (including the leader). The problem a) is solved; secondly, by utilizing the tight set limited coverage principle, each dimension is divided intoA limiting portion; and finally, determining the value of each unit in the data cube according to the observation data, wherein the summary (Total) in each dimension represents an aggregation value. The problem b) is thus solved. Obtaining the intelligent agent by the three steps
An inner membrane data cube as shown in fig. 2.
The scientific problem to be solved in the present invention at this stage is how to determine the high frequency relative state deviation and control action pair. We want to statistically analyze the relationship between the relative state deviation and the NN weight of the control behavior. And specifying a minimum support degree, and determining a high-frequency relative state deviation and control behavior NN weight pair based on the inner membrane data cube data information of the intelligent agent constructed in the first stage according to the minimum confidence degree. Control behaviors that cannot be calculated due to partial information loss of neighbor agents (including the leader) are predicted.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (6)
1. A networked multi-agent system modeling method based on data mining is characterized in that: the method comprises the steps of constructing a multi-agent inner membrane data cube by adopting an online analysis (OLAP) processing technology, mining based on inner membrane association rules, compensating lost data, eliminating the influence of data loss on control performance, constructing a networked multi-agent system optimization problem mathematical model by utilizing relative state deviation, analyzing coupling association between the relative state deviation and agent control behaviors, and mining inner membrane association rules.
2. The networked multi-agent system modeling method based on data mining as claimed in claim 1, wherein: the modeling of the intelligent internal membrane data cube comprises the following steps:
first, for each agent
，
Is the total number of agents; determining the dimensionality of the data cube based on the number of neighboring agents and the number of leaders (0 or 1), and designating each dimension to represent an agent
Relative state deviation from a neighbor agent;
secondly, dividing each dimension into a limited number of parts by using a tight set limited coverage principle;
thirdly, determining the value of each unit in the data cube according to the observation data, wherein the summary in each dimension represents an aggregation value; obtaining the intelligent agent by the three steps
An inner membrane data cube of (1).
3. The networked multi-agent system modeling method based on data mining as claimed in claim 1, wherein: the mining of the inner membrane association rule comprises the following steps:
analyzing the relation between the relative state deviation and the NN weight of the control behavior; specifying a minimum support degree, and determining NN weight pairs of high-frequency relative state deviation and control behaviors based on the data information of the inner membrane data cube of the intelligent agent constructed in the first stage according to the minimum confidence degree; and predicting the control behavior which cannot be calculated due to partial information loss of the neighbor intelligent part.
4. The networked multi-agent system modeling method based on data mining as claimed in claim 1, wherein: organizing data in a database into a multi-dimensional structure by an online analysis processing technology, wherein each dimension comprises multi-level data extraction; in view of the distribution relationship between the agents and the unavailability of the phenomenon of loss of communication information, and the influence of information loss on the performance of the networked multi-agent system.
5. The networked multi-agent system modeling method based on data mining as claimed in claim 1, wherein: estimating a control protocol by using a Neural Network (NN), thereby statistically analyzing the relation between the relative state deviation and NN weight of a control behavior; specifying a minimum support degree, and determining NN weight pairs of high-frequency relative state deviation and control behaviors based on the data information of the inner membrane data cube of the intelligent agent constructed in the first stage according to the minimum confidence degree; and predicting the control behavior which cannot be calculated due to partial information loss of the neighbor intelligent part.
6. The method of claim 1, wherein the modeling of the networked multi-agent system based on data mining is performed by a computer
The method comprises the following steps: the data cube is not limited to three-dimensional models, but multi-dimensional models.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910997342.8A CN110750578A (en) | 2019-10-21 | 2019-10-21 | Networked multi-agent system modeling method based on data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910997342.8A CN110750578A (en) | 2019-10-21 | 2019-10-21 | Networked multi-agent system modeling method based on data mining |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110750578A true CN110750578A (en) | 2020-02-04 |
Family
ID=69278960
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910997342.8A Pending CN110750578A (en) | 2019-10-21 | 2019-10-21 | Networked multi-agent system modeling method based on data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110750578A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040181376A1 (en) * | 2003-01-29 | 2004-09-16 | Wylci Fables | Cultural simulation model for modeling of agent behavioral expression and simulation data visualization methods |
| CN109541944A (en) * | 2018-12-20 | 2019-03-29 | 哈尔滨理工大学 | Discrete networks multi-agent system finite-time control method containing communication delay |
-
2019
- 2019-10-21 CN CN201910997342.8A patent/CN110750578A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040181376A1 (en) * | 2003-01-29 | 2004-09-16 | Wylci Fables | Cultural simulation model for modeling of agent behavioral expression and simulation data visualization methods |
| CN109541944A (en) * | 2018-12-20 | 2019-03-29 | 哈尔滨理工大学 | Discrete networks multi-agent system finite-time control method containing communication delay |
Non-Patent Citations (3)
| Title |
|---|
| MEHMET KAYA 等: "A Novel Approach to Multiagent Reinforcement Learning: Utilizing OLAP Mining in the Learning Process" * |
| 丁钰;许红艳;: "数据仓库在企业决策支持系统中的应用研究", 河南工程学院学报(自然科学版) * |
| 郭起霖;钟伟锋;张浩川;张涵;余荣;: "智能电网中异构通信网络的自适应速率控制" * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220413182A1 (en) | Recommendation engine for a cognitive reservoir system | |
| CN112925989B (en) | Group discovery method and system of attribute network | |
| CN107391542B (en) | Open source software community expert recommendation method based on file knowledge graph | |
| US9069916B2 (en) | Model selection from a large ensemble of models | |
| Graening et al. | Shape mining: A holistic data mining approach for engineering design | |
| CN108804473A (en) | The method, apparatus and Database Systems of data query | |
| Aydin et al. | SKATER-CON: Unsupervised regionalization via stochastic tree partitioning within a consensus framework using random spanning trees | |
| CN117454492A (en) | Electromechanical construction method and system based on BIM technology | |
| CN110750578A (en) | Networked multi-agent system modeling method based on data mining | |
| Setiyoko et al. | Minimax approach for semivariogram fitting in ordinary kriging | |
| Mishra et al. | Fuzzy frequent pattern mining from gene expression data using dynamic multi-swarm particle swarm optimization | |
| US20150285951A1 (en) | Attribute Importance Determination | |
| CN106815653B (en) | Distance game-based social network relationship prediction method and system | |
| US20220261696A1 (en) | Recommmender system for adaptive computation pipelines in cyber-manufacturing computational services | |
| CN115600268A (en) | High-dimensional data set visualization method and device, electronic equipment and storage medium | |
| Jarry et al. | Interactive trajectory modification and generation with FPCA | |
| Peng et al. | Uck-means: A customized k-means for clustering uncertain measurement data | |
| CN113095466A (en) | Algorithm of satisfiability model theoretical solver based on meta-learning model | |
| Miyata et al. | Concept drift detection on data stream for revising DBSCAN cluster | |
| CN111859164A (en) | Local structure based microblog network important node discovery method, device and medium | |
| Sadi-Nezhad et al. | A new fuzzy clustering algorithm based on multi-objective mathematical programming | |
| CN115964640B (en) | Improved template matching-based secondary target grouping method | |
| Zakharova et al. | Methods for Construction and Research Stereometric Visual Images of Multi-criteria Alternatives | |
| CN117151934B (en) | Multi-dimensional cluster analysis method and device for uninterrupted operation project of power distribution network | |
| CN113326584B (en) | Electrical equipment optimization design method taking robustness and reliability into consideration |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200204 |
|
| WD01 | Invention patent application deemed withdrawn after publication |