CN112529330B - Tunnel surrounding rock geological grading information prediction method based on Bayesian neural network - Google Patents
Tunnel surrounding rock geological grading information prediction method based on Bayesian neural network Download PDFInfo
- Publication number
- CN112529330B CN112529330B CN202011547694.2A CN202011547694A CN112529330B CN 112529330 B CN112529330 B CN 112529330B CN 202011547694 A CN202011547694 A CN 202011547694A CN 112529330 B CN112529330 B CN 112529330B
- Authority
- CN
- China
- Prior art keywords
- tunnel
- surrounding rock
- geological
- neural network
- grading 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.)
- Active
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
Abstract
The invention relates to a tunnel surrounding rock geological grading information prediction method based on a Bayesian neural network, which comprises the steps of collecting surrounding rock geological grading information of an existing tunnel and finely collected under construction and carrying out normalization processing, determining probability distribution of the tunnel surrounding rock geological grading information through Monte Carlo random analysis, and preliminarily determining the node number of an input layer, a hidden layer and an output layer of the Bayesian neural network model, so that a prediction Bayesian neural network prediction model is established by utilizing existing tunnel engineering data with similar geological information; with the continuous forward progress of the working face, updating the prediction model in real time by using the geological grading information of the tunnel surrounding rock newly acquired in the excavation process, and further gradually improving the prediction precision of the model; the prediction method provided by the application has better universality and higher prediction precision, can effectively judge the geological grading information of the unknown section in front of the tunnel excavation in advance, and is suitable for prediction of the geological grading information of most tunnel surrounding rocks.
Description
Technical Field
The invention relates to a tunnel surrounding rock geological grading information prediction method based on a Bayesian neural network, and belongs to the technical field of rock mass tunnel engineering.
Background
The surrounding rock is an integral part of tunnel engineering, and plays a decisive role in the whole life cycle from engineering design construction to later operation and maintenance. At present, the tunnel engineering design is turning to a more reasonable reliability-based design from the application of the traditional safety factor, and the essence of the reliability design needs the statistical characteristics of the geological grading information of the tunnel surrounding rock. The uncertainty of geological grading information is rarely considered when the grade of the tunnel surrounding rock is evaluated by the conventional theory or practice method; meanwhile, in the tunnel excavation process, the dynamic change condition of the surrounding rock is mastered, and the construction and support means are adjusted according to the geological condition, so that the occurrence of geological disasters in the construction process is prevented, and the method is the key of tunnel construction; rock mass excavation has strong uncertainty, and the traditional method has low prediction accuracy on tunnel geological conditions and is inconvenient to apply.
Therefore, how to reasonably utilize existing engineering data and field data, how to consider uncertainty of geological information and how to adopt which method to effectively and accurately predict geological grading information becomes a problem to be solved urgently in the geological grading information prediction process.
Disclosure of Invention
The invention provides a tunnel surrounding rock geological grading information prediction method based on a Bayesian neural network, which solves the problem of insufficient prediction precision of the conventional tunnel surrounding rock geological grading information prediction method.
The technical scheme adopted by the invention for solving the technical problem is as follows:
a tunnel surrounding rock geological grading information prediction method based on a Bayesian neural network specifically comprises the following steps:
the method comprises the steps of firstly, collecting existing tunnels and surrounding rock geological grading information which is finely collected in tunnels under construction, analyzing and sorting the obtained data, establishing a database, and carrying out normalization processing on the data in the database;
secondly, determining the subentry index of the tunnel surrounding rock geological grading information through the data subjected to normalization processing, and performing Monte Carlo random analysis by combining the existing data of the existing tunnel engineering surrounding rock geological grading information to obtain the value ranges and the probability distribution of different subentry indexes, and finally determining the probability distribution of the subentry indexes;
thirdly, determining the number of nodes of an input layer, a hidden layer and an output layer by using the data subjected to normalization processing in the first step, and constructing a Bayesian neural network model;
fourthly, selecting existing tunnel engineering data with similar geological information from the database established in the first step, placing the existing tunnel engineering data into the Bayesian neural network model established in the third step, wherein the similar existing tunnel engineering data are known data of tunnel mark sections selected and matched by comparing survey data on the site of the tunnel to be established, and carrying out value taking according to the uncertainty of geological grading information of surrounding rocks of the tunnel to establish an initial Bayesian neural network prediction model;
fifthly, introducing site survey data of the tunnel under construction based on the Bayesian neural network prediction model established in the fourth step, dividing the introduced survey data set into a training set and a test set, adjusting parameters in the model in real time according to the test result of the Bayesian neural network prediction model, and updating the model to form the tunnel surrounding rock geological grading information prediction model suitable for the engineering under construction;
sixthly, acquiring surrounding rock geological grading information on a tunnel face in real time in the excavation process of the tunnel, and adjusting a tunnel surrounding rock geological grading information prediction model in real time to further improve the Bayesian neural network prediction model;
as a further preferred aspect of the present invention, in the first step, the data in the database is normalized, and the processing formula is:
wherein, X i In order to normalize the processed data values,
in order to normalize the raw data prior to processing,
is the minimum value of the original data set,
is the maximum value of the original data set;
as a further preference of the present invention, in the second step, the different binomial indicators include the strength of the surrounding rock, defined as R 1 (ii) a The number of the discontinuous surfaces of the surrounding rock is defined as R 2 (ii) a The condition score of the discontinuous surface of the surrounding rock is defined as R 3 (ii) a Ground Water score, defined as R 4 ；
The sub-indexes of the tunnel surrounding rock geological grading information comprise uniform distribution, normal distribution, lognormal distribution, negative exponential distribution and Weibull distribution, and the determination of the probability distribution of the sub-indexes means that the data of the tunnel surrounding rock geological grading information are fitted by using a distribution form, and the probability distribution with the minimum fitting variance is selected as the probability distribution form of the sub-indexes;
as a further preferred aspect of the present invention, in the third step, the method for determining the number of nodes of the input layer in the bayesian neural network model is a phase space reconstruction method, and the method for determining the number of nodes of the hidden layer is a trial and error method;
the number of nodes of an output layer in the Bayesian neural network model is one;
as a further preferable mode of the invention, in the fourth step, the value is taken according to the uncertainty of the geological grading information of the tunnel surrounding rock, and the value is based on the variance of the on-site survey data and the selection of the geological measurement equipment;
as a further preferable aspect of the present invention, in the fifth step, the proportion of dividing the training set and the test set is determined according to the number of data in the data set, the evaluation index of the bayesian neural network prediction model is a pure mean square error, and the formula is as follows:
wherein MSPE is pure mean square error, y i For the actual value in the data set,
and (4) predicting the value of the model.
Through the technical scheme, compared with the prior art, the invention has the following beneficial effects:
1. the method determines the possible probability distribution of the tunnel surrounding rock geological grading information through Monte Carlo random analysis, can quickly acquire the statistical characteristics of the tunnel surrounding rock geological grading information, and provides key basic guarantee for the establishment of a prediction model of the tunnel surrounding rock geological grading information;
2. according to the invention, by considering the uncertainty of the geological grading information of the tunnel surrounding rock, the self property of the tunnel surrounding rock information can be completely described, and the geological condition in front of tunnel excavation is deduced in the form of dynamically predicting the geological grading information of unknown sections, so that the established Bayesian neural network prediction model has better universality and higher prediction precision;
3. according to the invention, the newly acquired data in the tunnel construction process is added into the prediction model, so that the Bayesian neural network prediction model is further perfected, the influence of the existing tunnel engineering information and the initial tunnel engineering information on prediction is gradually reduced, the prediction precision is effectively improved, the dynamic change condition of surrounding rocks is further mastered, the construction and support means are adjusted according to the geological condition, and the geological disaster in the construction process is prevented.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is a schematic flow chart of a tunnel surrounding rock geological grading information prediction method based on a Bayesian neural network provided by the invention.
Detailed Description
The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic views illustrating only the basic structure of the present invention in a schematic manner, and thus show only the constitution related to the present invention.
The method comprises the steps of analyzing the current situation that prediction accuracy of a prediction method of tunnel surrounding rock geological grading information is insufficient, when various indexes of the tunnel surrounding rock geological grading information are deduced, similar tunnel surrounding rock geological grading information is not considered, uncertainty of the geological information is ignored, and the precision of a conclusion result is insufficient.
As shown in fig. 1, the method comprises the following steps:
the method comprises the steps of firstly, collecting existing tunnels and surrounding rock geological grading information which is finely collected in tunnels under construction, analyzing and sorting the obtained data, establishing a database, and carrying out normalization processing on the data in the database;
secondly, determining the subentry index of tunnel surrounding rock geological grading information through the data subjected to normalization processing, combining the existing data of the existing tunnel engineering surrounding rock geological grading information, performing Monte Carlo random analysis to obtain the value range and probability distribution of different subentry indexes, and finally determining the probability distribution of the subentry indexes;
thirdly, determining the number of nodes of an input layer, a hidden layer and an output layer by using the data subjected to normalization processing in the first step, and constructing a Bayesian neural network model;
fourthly, selecting existing tunnel engineering data with similar geological information from the database established in the first step, placing the existing tunnel engineering data into the Bayesian neural network model established in the third step, wherein the similar existing tunnel engineering data are known data of tunnel mark sections selected and matched by comparing on-site survey data of the tunnel under construction, and establishing an initial Bayesian neural network prediction model by taking values according to the uncertainty of geological grading information of surrounding rocks of the tunnel;
fifthly, introducing site survey data of the tunnel under construction based on the Bayesian neural network prediction model established in the fourth step, dividing an introduced survey data set into a training set and a test set, adjusting parameters in the model in real time according to a test result of the Bayesian neural network prediction model, and updating the model to form a tunnel surrounding rock geological grading information prediction model suitable for the project under construction;
and sixthly, acquiring surrounding rock geological grading information on a tunnel face in real time in the excavation process of the tunnel, adjusting a tunnel surrounding rock geological grading information prediction model in real time, and further improving the Bayesian neural network prediction model.
In order to better illustrate the technology of the application and prove the advantages of the technology, a preferred embodiment related to the application is provided, firstly, the strength information prediction of the tunnel surrounding rock of a certain practical engineering is taken as an example, and the other three indexes are the number R of the discontinuous surfaces of the surrounding rock 2 And scoring R of discontinuous surrounding rock noodle pieces 3 And groundwater score R 4 The prediction can be performed by the same procedure as follows. In order to achieve the purpose, point load testing is firstly carried out on an engineering site to obtain tunnel surrounding rock strength values at different standard sections of the tunnel. According to the test result, the geological grading information of the tunnel surrounding rock is predicted, and the method specifically comprises the following steps:
the method comprises the steps of firstly, collecting surrounding rock strength information of an existing tunnel and a finely collected tunnel under construction, adopting a point load test method, repeatedly performing the point load test at least 20 times on each test point at each mark section of the tunnel in order to reduce uncertainty of the surrounding rock strength information, calculating to obtain an average value as a current mark section tunnel surrounding rock strength value, further analyzing and sorting the obtained multiple groups of data, establishing a tunnel surrounding rock strength information database, and performing normalization processing on the data in the database, wherein a formula used for the normalization processing is as follows:
wherein, X i In order to normalize the processed data values,
in order to normalize the raw data prior to processing,
is the minimum value of the original data set,
is the maximum value of the original data set.
Secondly, determining the tunnel surrounding rock geological condition subentry index predicted by the project as tunnel surrounding rock strength R 1 Number of discontinuity of surrounding rock R 2 And scoring R of discontinuous surrounding rock noodle pieces 3 Groundwater score R 4 (ii) a Carrying out Monte Carlo random analysis based on the existing tunnel engineering and the database of the surrounding rock strength information of the tunnel under construction established in the first step, and adopting a mode of fitting surrounding rock strength information data, wherein the probability distribution with the minimum fitting variance is a selected probability distribution form; therefore, the strength distribution form of the tunnel surrounding rock is determined to be normal distribution, and the value range of the tunnel surrounding rock strength information is 0-500MPa.
Thirdly, determining the number of nodes of an input layer, a hidden layer and an output layer by using the data subjected to normalization processing in the first step, and constructing a Bayesian neural network model; the method for determining the number of the input layer nodes of the Bayesian neural network model is a phase space reconstruction method, and the specific flow is as follows:
and 31, assuming that the tunnel surrounding rock strength information data sequence is as follows: [ x ] of 1 ,x 2 ,x 3 ,...,x N ]Where N is the length of the sequence, from x 1 Starting to select a time lag tau to be inserted into an n-dimensional Euclidean space, thereby obtaining a point y of the n-dimensional space 1 ，
y 1 ＝[x 1 ,x 1+τ ,x 1+2τ ,...,x 1+(n-1)τ ] T (3)
Step 32, from x 1 Starting to select a time lag tau to be inserted into an n-dimensional Euclidean space, and obtaining a point y of the n-dimensional space by the same method 2 ，
y 2 ＝[x 2 ,x 2+τ ,x 2+2τ ,...,x 2+(n-1)τ ] T (4)
In step 33, according to the same method, the sequence with the length of N can finally obtain N- (N-1) tau N-dimensional space points:
through conversion, the sequence with the length of N is converted into N- (N-1) tau points in an N-dimensional space, and the N-dimensional phase space can be reconstructed by selecting proper tau and N:
Y＝[y 1 ,y 2 ,y 3 ,...,y N-(n-1)τ ] (6)
and then the optimal value of n is calculated by a mutual information method to obtain 6, the number of nodes of an input layer of the Bayesian neural network model is (n-1), namely the number of nodes of the input layer is 5, the method for determining the number of nodes of a hidden layer of the model is a trial-and-error method, the number of the nodes of the hidden layer of the model is finally obtained to be 3, and the number of the nodes of an output layer of the model is 1 because the result of model prediction is the surrounding rock geological grading information on the working sub-surface.
And fourthly, selecting known data of corresponding tunnel mark sections according to comparison of survey data on the site of the tunnel under construction in the database established in the first step, wherein the selected tunnel surrounding rock strength data are shown in the table 1 and have 60 groups of data. On the basis of considering uncertainty of geological grading information of tunnel surrounding rocks, based on the related design of the Bayesian neural network framework in the third step, an initial neural network prediction model is established according to 5, 3 and 1 model input layers, hidden layers and output layers respectively.
TABLE 1 uniaxial compressive strength data of tunnel surrounding rock rocks with similar geological conditions
Fifthly, introducing site survey data of the excavated standard segments of the construction project according to the initial neural network prediction model established in the fourth step, wherein 36 groups of data of the first 9 standard segments in the table 2 are selected for the survey data, and 96 groups of data are combined in the data set by combining 60 groups of data in the table 1;
TABLE 2 uniaxial compressive strength data of surrounding rock rocks in tunnel construction
Dividing a data set into a training set and a testing set, wherein the proportion of the training set to the testing set is 7:1; adjusting the parameters of the model according to the test result of the prediction model, wherein the evaluation index of the test result of the model is a pure mean square error (MSPE), and the formula is as follows:
wherein, y i For the actual value in the data set,
is a model prediction value. The smaller the pure mean square error is, the higher the accuracy of the established prediction model is, and the model is updated on the basis, so that the tunnel surrounding rock geological grading information prediction model suitable for the current engineering is formed.
Sixthly, applying the established Bayes neural network model to the prediction of the current prediction surface (SM 61+ 320), wherein the prediction value of the model of the section is mean value mu =155MPa, standard deviation sigma =11MPa, and compared with the actual mean value 161MPa, the error is 3.7%, the error is small, and the prediction precision requirement is met;
secondly, continuously adjusting parameters of the prediction model by using geological grading information acquired by the excavated face, adding Data newly observed in the construction process into the prediction model in real time along with the forward advance of the working face, and sequentially adding two groups of Data (SM +320Data 1-Data 4 and SM +370Data 1-Data 4) into the previous prediction model to further perfect the prediction model, wherein the predicted value of the model at the position of the obtained prediction face (SM 61+ 370) is 110MPa, the standard deviation sigma =7MPa, and the error is 3.3% compared with the actual average value 113.8 MPa; the predicted value of the model at the prediction plane (SM 61+ 390) is 102MPa, and compared with the actual average value of 103.5MPa, the standard deviation σ =6MPa, and the error is 1.4%.
Therefore, with the continuous addition of new data, the influence of the existing tunnel engineering information and the initial information of the tunnel engineering under construction on prediction is gradually reduced, the model prediction precision can be effectively improved, and the dynamic change condition of the surrounding rock can be further mastered.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The meaning of "and/or" as used herein is intended to include both the individual components or both.
The term "connected" as used herein may mean either a direct connection between components or an indirect connection between components via other components.
In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.
Claims (6)
1. A tunnel surrounding rock geological grading information prediction method based on a Bayesian neural network is characterized by comprising the following steps: the method specifically comprises the following steps:
collecting the existing tunnel and finely collecting surrounding rock geological grading information of the tunnel under construction, analyzing and sorting the obtained data, establishing a database, and performing normalization processing on the data in the database;
secondly, determining the subentry index of the tunnel surrounding rock geological grading information through the data subjected to normalization processing, and performing Monte Carlo random analysis by combining the existing data of the existing tunnel engineering surrounding rock geological grading information to obtain the value ranges and the probability distribution of different subentry indexes, and finally determining the probability distribution of the subentry indexes;
thirdly, determining the number of nodes of an input layer, a hidden layer and an output layer by using the data subjected to normalization processing in the first step, and constructing a Bayesian neural network model;
fourthly, selecting existing tunnel engineering data with similar geological information from the database established in the first step, placing the existing tunnel engineering data into the Bayesian neural network model established in the third step, wherein the similar existing tunnel engineering data are known data of tunnel mark sections selected and matched by comparing survey data on the site of the tunnel to be established, and carrying out value taking according to the uncertainty of geological grading information of surrounding rocks of the tunnel to establish an initial Bayesian neural network prediction model;
fifthly, introducing site survey data of the tunnel under construction based on the Bayesian neural network prediction model established in the fourth step, dividing the introduced survey data set into a training set and a test set, adjusting parameters in the model in real time according to the test result of the Bayesian neural network prediction model, and updating the model to form the tunnel surrounding rock geological grading information prediction model suitable for the engineering under construction;
and sixthly, acquiring the surrounding rock geological grading information on the tunnel face in real time in the excavation process of the tunnel, and adjusting the tunnel surrounding rock geological grading information prediction model in real time to further improve the Bayesian neural network prediction model.
2. The Bayesian neural network-based tunnel surrounding rock geological grading information prediction method as recited in claim 1, wherein: in the first step, the data in the database is normalized, and the processing formula is as follows:
wherein, X i In order to normalize the processed data values,
in order to normalize the raw data prior to processing,
is the minimum value of the original data set,
is the maximum value of the original data set.
3. The Bayesian neural network-based tunnel surrounding rock geological grading information prediction method as recited in claim 1, wherein: in the second step, the different sub-indicators include the wall rock strength, defined as R 1 (ii) a The number of the discontinuous surfaces of the surrounding rock is defined as R 2 (ii) a The condition score of the discontinuous surface of the surrounding rock is defined as R 3 (ii) a Groundwater score, defined as R 4 ；
The sub-indexes of the tunnel surrounding rock geological grading information comprise uniform distribution, normal distribution, lognormal distribution, negative exponential distribution and Weibull distribution, the step of determining the probability distribution of the sub-indexes is to fit the data of the tunnel surrounding rock geological grading information by using a distribution form, and the probability distribution with the minimum fitting variance is selected as the probability distribution form of the sub-indexes.
4. The Bayesian neural network-based tunnel surrounding rock geological grading information prediction method according to claim 1, characterized in that: in the third step, the method for determining the number of the nodes of the input layer in the Bayes neural network model is a phase space reconstruction method, and the method for determining the number of the nodes of the hidden layer is a trial and error method;
the number of the nodes of the output layer in the Bayesian neural network model is one.
5. The Bayesian neural network-based tunnel surrounding rock geological grading information prediction method as recited in claim 1, wherein: and in the fourth step, values are taken according to the uncertainty of the geological grading information of the tunnel surrounding rock, and the values are selected based on the variance of the on-site survey data and geological measurement equipment.
6. The Bayesian neural network-based tunnel surrounding rock geological grading information prediction method as recited in claim 1, wherein: in the fifth step, the proportion of dividing the training set and the test set is determined according to the data number of the data set, the evaluation index of the Bayes neural network prediction model is a pure mean square error, and the formula is as follows:
wherein MSPE is pure mean square error, y i For the actual value in the data set,
and (4) predicting the value of the model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011547694.2A CN112529330B (en) | 2020-12-24 | 2020-12-24 | Tunnel surrounding rock geological grading information prediction method based on Bayesian neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011547694.2A CN112529330B (en) | 2020-12-24 | 2020-12-24 | Tunnel surrounding rock geological grading information prediction method based on Bayesian neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112529330A CN112529330A (en) | 2021-03-19 |
| CN112529330B true CN112529330B (en) | 2022-11-18 |
Family
ID=74976533
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011547694.2A Active CN112529330B (en) | 2020-12-24 | 2020-12-24 | Tunnel surrounding rock geological grading information prediction method based on Bayesian neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112529330B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113553763A (en) * | 2021-07-12 | 2021-10-26 | 中铁二十局集团有限公司 | Tunnel surrounding rock rapid grading method and system |
| CN115239108B (en) * | 2022-07-13 | 2023-04-18 | 北京交通大学 | Weak broken surrounding rock sensing method based on TBM real-time broken rock data |
| CN115929406B (en) * | 2022-12-07 | 2023-08-18 | 河海大学 | Surrounding rock stability prediction method, system and storage medium based on neural network |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109934398A (en) * | 2019-03-05 | 2019-06-25 | 山东大学 | A kind of drill bursting construction tunnel gas danger classes prediction technique and device |
| CN110210121A (en) * | 2019-05-31 | 2019-09-06 | 成都理工大学 | A kind of dynamic risk acquisition methods of serious deformation disaster |
| CN111291934B (en) * | 2020-02-18 | 2023-04-07 | 山东科技大学 | Surrounding rock real-time grading prediction and self-checking method in tunnel construction process |
-
2020
- 2020-12-24 CN CN202011547694.2A patent/CN112529330B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112529330A (en) | 2021-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112529330B (en) | Tunnel surrounding rock geological grading information prediction method based on Bayesian neural network | |
| US6810332B2 (en) | Method for computing complexity, confidence and technical maturity indices for reservoir evaluations | |
| Kottke et al. | A semi-automated procedure for selecting and scaling recorded earthquake motions for dynamic analysis | |
| Zhang et al. | A fractal discrete fracture network model for history matching of naturally fractured reservoirs | |
| CN103775074B (en) | Method for correcting expanding influence of coalbed methane reservoir density logging | |
| CN110705795A (en) | Tunnel gush risk grade fuzzy prediction method based on comprehensive advanced geological prediction technology | |
| US10890688B2 (en) | Method for generating secondary data in geostatistics using observed data | |
| CN104899411A (en) | Method and system for establishing reservoir capacity prediction model | |
| CN110889440A (en) | Rockburst grade prediction method and system based on principal component analysis and BP neural network | |
| CN115049124A (en) | Deep and long tunnel water inrush prediction method based on Bayesian network | |
| Wei et al. | Progress on the different methods of reserves calculation in the whole life cycle of gas reservoir development | |
| CN109268003A (en) | A kind of densification oil volume increase Mathematical Modeling Methods under Geological Engineering integrated frame | |
| Biçer et al. | Inference on parameters of a geometric process with scaled Muth distribution | |
| Yang et al. | Intelligent rating method of tunnel surrounding rock based on one-dimensional convolutional neural network | |
| Michanos et al. | Short-term load forecasting using a chaotic time series | |
| Niño | Wellbore stability analysis based on sensitivity and uncertainty analysis | |
| CN111929424B (en) | Large-span underground cavern hard surrounding rock sub-classification method based on size effect | |
| CN115310319A (en) | Simulation method for perforation completion under simulated formation condition | |
| Zhang et al. | BPNN-GA-based intelligence information approach to characterize deep fractured strata | |
| CN110029984B (en) | Method for calculating high-pressure physical properties of crude oil based on formation parameters and degassed oil analysis data | |
| Brouwer et al. | A direct inverse model to determine permeability fields from pressure and flow rate measurements | |
| Elshafei et al. | Neural network aided design of oil production units | |
| Zanon et al. | Implementation aspects of sequential simulation | |
| LIN | An introduction of the Chinese standard for engineering classification of rock masses (GB50218-94) | |
| Lødøen et al. | Assessment of uncertainty in reservoir production forecasts using upscaled flow models |
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 |