CN115564148A - Chamber safety analysis method based on multi-source heterogeneous data fusion - Google Patents

Chamber safety analysis method based on multi-source heterogeneous data fusion Download PDF

Info

Publication number
CN115564148A
CN115564148A CN202211533494.0A CN202211533494A CN115564148A CN 115564148 A CN115564148 A CN 115564148A CN 202211533494 A CN202211533494 A CN 202211533494A CN 115564148 A CN115564148 A CN 115564148A
Authority
CN
China
Prior art keywords
link
evidence
chamber
data
evidence body
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
Application number
CN202211533494.0A
Other languages
Chinese (zh)
Other versions
CN115564148B (en
Inventor
李鹏
孙亮
王金伟
王宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Science and Technology Beijing USTB
Original Assignee
University of Science and Technology Beijing USTB
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Science and Technology Beijing USTB filed Critical University of Science and Technology Beijing USTB
Priority to CN202211533494.0A priority Critical patent/CN115564148B/en
Publication of CN115564148A publication Critical patent/CN115564148A/en
Application granted granted Critical
Publication of CN115564148B publication Critical patent/CN115564148B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • 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
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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
    • G06Q50/08—Construction
    • 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
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G06Q50/265—Personal security, identity or safety

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Computer Security & Cryptography (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a chamber safety analysis method based on multi-source heterogeneous data fusion, which is characterized in that the initial processing of a chamber index value group is completed through an f link, then chamber index data is predicted from a multi-source angle through a g link-fuzzy processing method, an LSSVM link-least square support vector machine model and an ELM link-ultralimit learning machine, the prediction result is converted into an evidence body, and finally the evidence body is calculated through a DS evidence fusion theory, so that the final chamber safety prediction result is obtained. By adopting the combination method, the data utilization rate is higher, and the prediction result is more accurate.

Description

Chamber safety analysis method based on multi-source heterogeneous data fusion
Technical Field
The invention relates to the technical field of information fusion, in particular to a chamber safety analysis method based on multi-source heterogeneous data fusion.
Background
The chamber is a horizontal tunnel with large cross section and short length which is not directly connected to the surface outlet and is used for installing various devices and machines, storing materials and tools or being used for other special purposes, such as machine room, explosive storehouse, rest room and the like. In the prior art, the safety analysis of the underground chambers is based on the evaluation of risk sources, and an LS method or an LEC method is mostly adopted, and the comparison depends on qualitative description and expert scoring. Relatively speaking, the emphasis is on the macro evaluation of the chamber, and the chamber specific conditions cannot be described from a microscopic perspective. Of course, an expert or a learner collects specific data of the chamber through a sensor and then analyzes the safety of the chamber by using a machine learning model, but the expert or the learner mostly belongs to single data source analysis and is not accurate enough.
Disclosure of Invention
The invention aims to provide a multi-source heterogeneous data analysis method to quantitatively analyze the safety state of a chamber, thereby solving the problems that the existing analysis method cannot reflect the microscopic condition of the chamber and the evaluation is inaccurate.
In order to achieve the purpose, the invention provides the following technical scheme:
the chamber safety analysis method based on multi-source heterogeneous data fusion divides chamber safety states into three types, namely safety, slight accidents and serious accidents, and then carries out the following processing on index data:
a step g, setting three membership functions aiming at three states, then respectively calculating the membership of each index corresponding to the three states, and taking the calculation result as the three-dimensional prediction data of the current link;
the LSSVM link is used for comprehensively calculating all indexes by utilizing a least square support vector machine model to obtain three-dimensional prediction data of the current link corresponding to three states;
an ELM link, which utilizes an overrun learning machine model to carry out comprehensive calculation on all indexes to obtain three-dimensional prediction data corresponding to three states in the current link;
and the DS link takes the three-dimensional prediction data of the g link, the LSSVM link and the ELM link as an evidence body, and then calculates the evidence body by using a DS evidence fusion model so as to obtain a final chamber state prediction result.
Preferably, an f link is arranged before the g link, and the f link fuses a plurality of pieces of data of the same index by adopting a weighted average method or a Bayesian method.
Preferably, the DS step further includes performing zero processing and weight assignment on the evidence body;
zero processing corrects the zero factor using the maximum value in the evidence body; the evidential body is shown as
Figure 561274DEST_PATH_IMAGE001
If, if
Figure 100002_DEST_PATH_IMAGE002
Then the evidence body after zero processing is represented as
Figure 7429DEST_PATH_IMAGE003
Wherein
Figure 100002_DEST_PATH_IMAGE004
Is a correction factor;
weight assignment first calculates a correlation matrix between evidence bodies, the calculation formula is as follows:
Figure 124421DEST_PATH_IMAGE005
，
Figure 100002_DEST_PATH_IMAGE006
in the formula (I), the compound is shown in the specification,
Figure 71649DEST_PATH_IMAGE007
is a correlation matrix;
Figure 100002_DEST_PATH_IMAGE008
total number of evidence bodies;
Figure 121644DEST_PATH_IMAGE009
for the element of the ith row and the jth column in the correlation matrix, the evidence body is represented
Figure 100002_DEST_PATH_IMAGE010
The correlation between them;
Figure 4281DEST_PATH_IMAGE011
respectively represent
Figure 100002_DEST_PATH_IMAGE012
Is first and second
Figure 546252DEST_PATH_IMAGE013
The standard deviation of the individual's body of evidence,
Figure 100002_DEST_PATH_IMAGE014
respectively represent
Figure 969274DEST_PATH_IMAGE015
A first and a second
Figure 100002_DEST_PATH_IMAGE016
The mean value of the individual's body of evidence,
Figure 811459DEST_PATH_IMAGE017
indicates a desire;
Figure 100002_DEST_PATH_IMAGE018
is as follows
Figure DEST_PATH_IMAGE020
The weights are then determined by the correlation matrix, which is calculated as follows:
Figure 831947DEST_PATH_IMAGE021
and finally, updating the evidence body by using the weight, wherein the calculation formula is as follows:
Figure 100002_DEST_PATH_IMAGE022
in the formula (I), the compound is shown in the specification,
Figure 58660DEST_PATH_IMAGE023
is an updated evidence body.
Preferably, the g link, the LSSVM link and the ELM link are used for completing index calculation in a parallel mode.
In conclusion, by adopting the chamber safety analysis method, the invention acquires the microscopic data of the chamber through the sensor, and respectively carries out quantitative analysis on the acquired data by combining the fuzzy processing method, the least square support vector machine model and the overrun learning machine to obtain the evidence body, and then carries out comprehensive calculation on the evidence body through the improved DS evidence fusion theory, thereby realizing the accurate evaluation of the safety state of the chamber at the multi-source angle.
Drawings
FIG. 1 is an overall flow diagram of an embodiment of the present invention;
FIG. 2 is a flow chart of a procedure f according to an embodiment of the present invention;
FIG. 3 is a flowchart of a link g in an embodiment of the present invention;
FIG. 4 is a flow chart of an LSSVM link in an embodiment of the present invention;
FIG. 5 is a flow chart of an ELM link according to an embodiment of the present invention;
FIG. 6 is a flowchart of a DS link according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained by combining the drawings and the embodiment.
Stress, a B value, heat radiation and deformation are four indexes reflecting the safety state of the chamber, and the safety state of the chamber is judged according to the indexes through a data fusion technology. As shown in fig. 1, in the chamber safety analysis method based on multi-source heterogeneous data fusion, first, respective index value sets are respectively measured through a plurality of groups of sensors. The values of each index value group are output through the f link
Figure 100002_DEST_PATH_IMAGE024
。
Figure 286510DEST_PATH_IMAGE025
The method has three purposes as input, namely obtaining an evidence body through a g link
Figure DEST_PATH_IMAGE026
Secondly, obtaining an evidence body through an LSSVM model
Figure 838845DEST_PATH_IMAGE027
Thirdly, obtaining an evidence body through an ELM model
Figure DEST_PATH_IMAGE028
. The evidence body obtained in the three links obtains final three-dimensional output through a DS evidence fusion algorithm, and each dimension of the three-dimensional output represents the probability of three states (safety, light accidents and serious accidents).
And f, link: as shown in fig. 2, a weighted average method or a bayesian method is used to perform a preliminary data level fusion process to improve the accuracy of the final prediction result, and the partial data processing results in this link are shown in table 1.
Table 1 f four kinds of index values after link processing
Figure 948884DEST_PATH_IMAGE029
And g, link: as shown in fig. 3, a fuzzy processing method is used to process the data output by the f-element. Three safety states, namely safety, light accidents and serious accidents, are involved, and corresponding outputs also need to respectively give predicted values of the three states, namely three-dimensional predicted data is output. Therefore, it is necessary to set a membership function for each of the three safety states, and the expression thereof is as follows.
Figure DEST_PATH_IMAGE030
And (3) respectively carrying out g links on the four indexes of stress, deformation, B value and heat radiation, and finally obtaining 4 three-dimensional prediction data, namely four evidence bodies. Taking the third data in table 1 as an example, the probability distributions of stress, deformation, B value and thermal radiation obtained by calculation in the g link are respectively
Figure 979288DEST_PATH_IMAGE031
，
Figure DEST_PATH_IMAGE032
，
Figure 858382DEST_PATH_IMAGE033
，
Figure DEST_PATH_IMAGE034
。
LSSVM link: as shown in fig. 4, the input index is calculated by using a least squares support vector machine model (LSSVM model), and three-dimensional prediction data is obtained. And the LSSVM model carries out binary classification on the data according to a supervised learning mode. Of course, the LSSVM model is one of the machine learning models, and its use requires training the initial model through a historical data set to find the best parameters representing the three-state classifier. Taking the third data in Table 1 as an example, the data measurement output of the LSSVM model is
Figure 847198DEST_PATH_IMAGE035
。
An ELM link: as shown in fig. 5, the input index is calculated by using an overrun learning machine model (ELM model), and three-dimensional prediction data is obtained. The ELM is a machine learning model constructed based on a feedforward neural network, and the initial model also needs to be trained through a historical data set before use so as to find the optimal input weight, the number of hidden nodes and the like, and finally predicted values representing three states are respectively output through three output layers. Taking the third data of Table 1 as an example, the measured output of the ELM model is
Figure DEST_PATH_IMAGE036
。
The LSSVM link and the ELM link are models which take four indexes of stress, deformation, a B value and heat radiation as input at the same time and carry out comprehensive prediction, and only one evidence body is obtained in each model.
And (3) a DS link: and calculating six evidence bodies obtained in the g link, the ELM link and the LSSVM link based on a DS evidence fusion model so as to obtain a final prediction result. However, the traditional DS evidence theory cannot be used in practice, and because of the problem of confidence conflict, the DS evidence theory model needs to be improved. As shown in fig. 6, the solution first performs zero processing and weight assignment on the evidence body before the fusion calculation.
Zero processing is to correct the zero factor, for example, there is an evidence body of [0,0.99,0.01 ]]If there is a zero factor in the evidential volume, then the highest value of 0.99 uniform point (i.e., correction factor q = 0.01) in the evidential volume is given the zero factor, i.e., [0.01,0.98,0.01]. Taking the third row of data in Table 1 as an example, the six evidence bodies are changed into zero after being processed
Figure 788740DEST_PATH_IMAGE037
，
Figure DEST_PATH_IMAGE038
，
Figure 787218DEST_PATH_IMAGE039
，
Figure DEST_PATH_IMAGE040
，
Figure 724081DEST_PATH_IMAGE041
，
Figure DEST_PATH_IMAGE042
。
The weight distribution firstly calculates a correlation matrix between the evidence bodies, and then finds the weight coefficient of each evidence body by using the correlation matrix. The correlation matrix calculation formula is as follows:
Figure 618219DEST_PATH_IMAGE043
，
Figure DEST_PATH_IMAGE044
in the formula (I), the compound is shown in the specification,
Figure 702850DEST_PATH_IMAGE045
is a correlation matrix;ntotal number of evidence bodies;
Figure DEST_PATH_IMAGE046
representing the evidence body for the element of the ith row and the jth column in the correlation matrix
Figure 137373DEST_PATH_IMAGE047
The correlation between them;
Figure DEST_PATH_IMAGE048
respectively represent
Figure DEST_PATH_IMAGE049
A first and a second
Figure DEST_PATH_IMAGE050
The standard deviation of the individual body of evidence,
Figure 69688DEST_PATH_IMAGE051
respectively represent the first
Figure DEST_PATH_IMAGE052
A first and a second
Figure 603569DEST_PATH_IMAGE053
The mean value of the individual evidence bodies,
Figure DEST_PATH_IMAGE054
indicating a desire.
Figure 113179DEST_PATH_IMAGE055
Is as follows
Figure DEST_PATH_IMAGE056
After the correlation matrix is obtained, the weight coefficient is calculated by using the following formula.
Figure 85814DEST_PATH_IMAGE057
And processing items which are less than zero, namely assigning values to be 0 if the items are less than zero, and finally updating the evidence body through the weight coefficient, namely multiplying the evidence body by the weight coefficient respectively.
Figure DEST_PATH_IMAGE058
Taking the third data in Table 1 as an example, first, the correlation matrix of Pearson is obtained according to the above correlation calculation formula
Figure 794007DEST_PATH_IMAGE059
Assigning a value of zero to a term less than zero and calculating the weight of the six evidence volumes as
Figure DEST_PATH_IMAGE060
Multiplying the weight with the corresponding evidence body to obtain
Figure 561106DEST_PATH_IMAGE061
Then to the obtained
Figure DEST_PATH_IMAGE062
And carrying out subsequent treatment.
And after obtaining the new six evidence bodies, calculating by using a formula of the traditional DS evidence fusion theory to obtain a final fusion result. The DS evidence fusion calculation method is as follows:
Figure 354750DEST_PATH_IMAGE063
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE064
being a collection of events, secure in this context
Figure 927813DEST_PATH_IMAGE065
Minor accident
Figure DEST_PATH_IMAGE066
Serious accident
Figure 428196DEST_PATH_IMAGE067
A set of these three events;
Figure DEST_PATH_IMAGE068
is a collision coefficient;
Figure DEST_PATH_IMAGE069
showing the evidence body generated by each link.
Taking the third row of data in table 1 as an example, the final data fusion result is obtained
Figure DEST_PATH_IMAGE070
Therefore, the fusion result judges the chamber state as a slight accident.
In addition, the scheme adopts a parallel computing mode in a g link, an ELM link and an LSSVM link so as to accelerate the data processing speed.
The above is a specific embodiment of the present invention, but the scope of the present invention should not be limited to the analysis of chamber safety. Any changes or substitutions that can be easily made by those skilled in the art within the technical scope of the present invention are included in the protection scope of the present invention, and therefore, the protection scope of the present invention is subject to the protection scope defined by the appended claims.

Claims (4)

1. The method for analyzing the safety of the underground chamber based on the multi-source heterogeneous data fusion is characterized in that the safety state of the underground chamber is divided into three types, namely safety, slight accidents and serious accidents, and then index data are processed as follows:
a step g, setting three membership functions aiming at three states, then respectively calculating the membership of each index corresponding to the three states, and taking the calculation result as the three-dimensional prediction data of the current link;
the LSSVM link is used for comprehensively calculating all indexes by using a least square support vector machine model to obtain three-dimensional prediction data corresponding to three states in the current link;
an ELM link, which utilizes an overrun learning machine model to carry out comprehensive calculation on all indexes to obtain three-dimensional prediction data corresponding to three states in the current link;
and the DS link is used for taking the three-dimensional prediction data of the g link, the LSSVM link and the ELM link as an evidence body, and then calculating the evidence body by using a DS evidence fusion model so as to obtain a final chamber state prediction result.
2. The method as claimed in claim 1, wherein the g link is preceded by an f link, and the f link fuses a plurality of data of the same index by using a weighted average method or a bayesian method.
3. The method for chamber security analysis of claim 1, wherein the DS element further comprises zero processing and weight assignment of the evidence body;
zero processing corrects the zero factor using the maximum value in the evidence body; the evidence body is expressed as
Figure 124783DEST_PATH_IMAGE001
If, if
Figure DEST_PATH_IMAGE002
Then the evidence body after zero processing is represented as
Figure 386131DEST_PATH_IMAGE003
Wherein
Figure DEST_PATH_IMAGE004
Is a correction factor;
weight assignment first calculates a correlation matrix between evidence bodies, the calculation formula is as follows:
Figure 572393DEST_PATH_IMAGE005
，
Figure DEST_PATH_IMAGE006
in the formula (I), the compound is shown in the specification,
Figure 577389DEST_PATH_IMAGE007
is a correlation matrix; n is the total number of evidence bodies;
Figure DEST_PATH_IMAGE008
is the first in the correlation matrix
Figure 798286DEST_PATH_IMAGE009
Go to the first
Figure DEST_PATH_IMAGE010
Elements of the column, representing evidence bodies
Figure 292852DEST_PATH_IMAGE011
The correlation between them;
Figure DEST_PATH_IMAGE012
respectively represent the first
Figure 966410DEST_PATH_IMAGE013
A first and a second
Figure DEST_PATH_IMAGE014
The standard deviation of the individual's body of evidence,
Figure 571835DEST_PATH_IMAGE015
respectively represent
Figure DEST_PATH_IMAGE016
Is first and second
Figure 850501DEST_PATH_IMAGE017
The mean value of the individual evidence bodies,
Figure DEST_PATH_IMAGE018
indicating a desire;
Figure 656914DEST_PATH_IMAGE019
is calculated as follows
Figure 286609DEST_PATH_IMAGE021
The weights are then determined by a correlation matrix, the calculation formula is as follows:
Figure DEST_PATH_IMAGE022
and finally, updating the evidence body by using the weight, wherein the calculation formula is as follows:
Figure 430146DEST_PATH_IMAGE023
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE024
is an updated evidence body.
4. The chamber safety analysis method of claim 1, wherein the g-link, the LSSVM-link, and the ELM-link perform index calculations in a parallel manner.
CN202211533494.0A 2022-12-01 2022-12-01 Chamber safety analysis method based on multi-source heterogeneous data fusion Active CN115564148B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211533494.0A CN115564148B (en) 2022-12-01 2022-12-01 Chamber safety analysis method based on multi-source heterogeneous data fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211533494.0A CN115564148B (en) 2022-12-01 2022-12-01 Chamber safety analysis method based on multi-source heterogeneous data fusion

Publications (2)

Publication Number Publication Date
CN115564148A true CN115564148A (en) 2023-01-03
CN115564148B CN115564148B (en) 2023-04-07

Family

ID=84769733

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211533494.0A Active CN115564148B (en) 2022-12-01 2022-12-01 Chamber safety analysis method based on multi-source heterogeneous data fusion

Country Status (1)

Country Link
CN (1) CN115564148B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5787235A (en) * 1995-05-09 1998-07-28 Gte Government Systems Corporation Fuzzy logic-based evidence fusion tool for network analysis
CN107842394A (en) * 2017-10-23 2018-03-27 青岛理工大学 Large Span Underground chamber exploits the Dynamic Elastic Module detection method of roof stability
CN110533091A (en) * 2019-08-22 2019-12-03 贵州大学 A kind of more evident information fusion methods for improving DS evidence theory
CN111667193A (en) * 2020-06-12 2020-09-15 中国矿业大学(北京) Coal mine gas safety evaluation method based on D-S evidence theory
CN111950627A (en) * 2020-08-11 2020-11-17 重庆大学 Multi-source information fusion method and application thereof
US20220112806A1 (en) * 2020-10-13 2022-04-14 Institute Of Rock And Soil Mechanics, Chinese Academy Of Sciences Safety early warning method and device for full-section tunneling of tunnel featuring dynamic water and weak surrounding rock

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5787235A (en) * 1995-05-09 1998-07-28 Gte Government Systems Corporation Fuzzy logic-based evidence fusion tool for network analysis
CN107842394A (en) * 2017-10-23 2018-03-27 青岛理工大学 Large Span Underground chamber exploits the Dynamic Elastic Module detection method of roof stability
CN110533091A (en) * 2019-08-22 2019-12-03 贵州大学 A kind of more evident information fusion methods for improving DS evidence theory
CN111667193A (en) * 2020-06-12 2020-09-15 中国矿业大学(北京) Coal mine gas safety evaluation method based on D-S evidence theory
CN111950627A (en) * 2020-08-11 2020-11-17 重庆大学 Multi-source information fusion method and application thereof
US20220112806A1 (en) * 2020-10-13 2022-04-14 Institute Of Rock And Soil Mechanics, Chinese Academy Of Sciences Safety early warning method and device for full-section tunneling of tunnel featuring dynamic water and weak surrounding rock

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
刘湘平;古德生;罗一忠;谢学斌;: "深井采场凿岩硐室稳定性模糊综合评价" *
岳乾等: "基于证据理论的煤矿避难硐室可靠性等级划分", 《煤炭技术》 *
张炎等: "龙滩水电站地下洞室围岩变形的智能化预测方法", 《水力发电学报》 *
李冬静: "基于LSSVM混淆矩阵和改进DS合成的多源传感器网络安全态势预测", 《计算机测量与控制》 *
田紫圆等: "基于机器学习的重力坝变形监测统计模型及应用", 《水利规划与设计》 *

Also Published As

Publication number Publication date
CN115564148B (en) 2023-04-07

Similar Documents

Publication Publication Date Title
CN108596327B (en) Seismic velocity spectrum artificial intelligence picking method based on deep learning
CN109472088B (en) Shale gas-conditioned production well production pressure dynamic prediction method
CN113762486B (en) Method and device for constructing fault diagnosis model of converter valve and computer equipment
CN110097123B (en) Express mail logistics process state detection multi-classification system
CN112464996B (en) LSTM-XGBoost-based intelligent power grid intrusion detection method
CN106778838A (en) A kind of method for predicting air quality
CN106296434B (en) Grain yield prediction method based on PSO-LSSVM algorithm
CN113156263A (en) Cable state monitoring method and storage medium
CN111898867A (en) Airplane final assembly production line productivity prediction method based on deep neural network
CN111638034A (en) Strain balance temperature gradient error compensation method and system based on deep learning
KR20210039921A (en) Operation method of deep neural network system configured to optimize neural network model
CN115564148B (en) Chamber safety analysis method based on multi-source heterogeneous data fusion
CN112464567B (en) Intelligent data assimilation method based on variational and assimilative framework
CN106682773A (en) Human-machine welding time prediction method based on simulation annealing algorithm
CN114004008B (en) Airplane assembly line resource configuration optimization method based on neural network and genetic algorithm
CN113053536A (en) Infectious disease prediction method, system and medium based on hidden Markov model
CN116706907A (en) Photovoltaic power generation prediction method based on fuzzy reasoning and related equipment
CN115618743B (en) State evaluation method and state evaluation system of sighting telescope system
CN116415501A (en) MGU-A thermal error prediction model creation method and thermal error control system based on digital twin
CN115091725B (en) Intelligent blow molding machine for producing pesticide packing barrel and control method thereof
CN111260056A (en) Network model distillation method and device
CN113688770B (en) Method and device for supplementing long-term wind pressure missing data of high-rise building
CN113722308B (en) Acceleration response data complement method and device based on EEMD-MultiCNN-LSTM
CN115293639A (en) Battlefield situation studying and judging method based on hidden Markov model
CN114707399A (en) Decoupling method of six-dimensional force sensor

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