CN109740183A - Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method - Google Patents
Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method Download PDFInfo
- Publication number
- CN109740183A CN109740183A CN201811484191.8A CN201811484191A CN109740183A CN 109740183 A CN109740183 A CN 109740183A CN 201811484191 A CN201811484191 A CN 201811484191A CN 109740183 A CN109740183 A CN 109740183A
- Authority
- CN
- China
- Prior art keywords
- tunnel
- face
- stability
- sample
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The present invention provides a kind of tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method, pass through Orthogonal Experiment and Design, it chooses most representative parameter combination and establishes two-dimensional mathematical model progress strength degradation calculating, and sample is demarcated according to strength reduction factor.The sample for importing the training calibration of support vector machines module using Python again is fitted a plurality of decision boundary by being embedded in different kernel functions.Then pass throughkThe Average Accuracy that cross validation calculates classifier is rolled over, the higher classifier of accuracy rate is chosen and carries out judgement of stability, and utilize collective study mechanism, different classifications device is used while classification is to be mutually authenticated.In the case of judging result is unstable, using 0.01m as step-length, tunnel diameter is gradually reduced through iteration, and calculate the value of decision function in each iterative process, when decision function is greater than or equal to zero, show that face is critical state, calculating tunnel diameter at this time is the maximum step height that benching tunnelling method excavates.
Description
Technical field
The present invention relates to Tunnel Design and risk assessment technical fields, refer specifically to a kind of tunnel tunnel face estimation of stability mould
Type and benching tunnelling method excavation height design method.
Background technique
The tunnel of the tunnel constructed in soft stratum, especially big cross section is easy to that face landslide occurs, and greatly endangers
And the safety of construction personnel and equipment, and the works and building that close on are adversely affected.Therefore, in Tunnel Design rank
Before section, or construction, tunnel face face stability should be assessed.When assessment result is to stablize, tunneling boring can be used and open
Dig into capable construction.But when assessment result is unstable, need to use certain auxiliary construction measure to guarantee face safety,
Certain confining pressure such as shield or face anchor pole such as are applied to face, to the front of tunnel heading soil body carry out grouting and reinforcing with
Its strength and stiffness is improved, or i.e. such as benching tunnelling method reduces disposable excavation height using step excavation.
Existing tunnel tunnel face judgment method is mainly numerical method.By taking displacement finite element method as an example, pass through what is given
Parameter establishes numerical model and carries out numerical simulation, when Convergence of Numerical Calculation, can determine whether face to stablize；When numerical value calculates not
When convergence, it can determine whether that face is unstable.
According to existing research, the stability of tunnel tunnel face is mainly tunnel diameter (D) respectively by six state modulators,
Buried depth (H), distribution on ground load (σ), the severe (γ) of country rock, cohesive strength (c) and internal friction angleGive one group of parameterThe strength reduction factor that can be calculated and obtain by numerical value judges its stability.But it is right
Same tunnel, in the above parameter, mainly buried depth, cohesive strength and internal friction angle are changed greatly along tunnel axis, it is therefore desirable to
Tunnel is divided into many sections, each section requires to establish numerical model progress face stability assessment, assesses generation
Valence is higher.Ideal method is can to calculate judgement by theoretical formula or empirical equation in the case where giving one group of parameter
Area face stability.
Summary of the invention
The object of the present invention is to provide one include tunnel tunnel face stability influence parameter support vector machine classifier,
For any one group of given parameters, so that it may assess area face stability, and provide corresponding suggest to design and construction.
The purpose of the present invention is realized through the following technical scheme:
Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method, comprising the following steps:
(1) it data mining: by Orthogonal Experiment and Design, chooses most representative parameter combination and establishes two-dimensional mathematical model
Strength degradation calculating is carried out, and according to strength reduction factor FsSample is demarcated, as strength reduction factor FsIt is to stablize sample when greater than 1
This, as strength reduction factor FsIt is unstable sample when less than 1；
(2) it establishes disaggregated model: the sample of support vector machines module training calibration is imported using Python, by embedding
Enter different kernel functions and is fitted a plurality of decision boundary to get different classification functions is arrived；
(3) model measurement: the Average Accuracy that cross validation calculates classifier is rolled over by k-, chooses higher point of accuracy rate
Class device carries out face judgement of stability, and utilizes collective study mechanism, uses different classifications device while classification is to be mutually authenticated
To assess the generalization ability of disaggregated model；
(4) face stability assessment: calculating the numerical value of classification function by program, to assess the stability of face,
When the value of classification function is greater than zero, face is stablized, and when the value of classification function is less than zero, face is unstable；
(5) it designs benching tunnelling method highest excavation height: in the case of face judging result is unstable, being with 0.01m
Step-length gradually reduces tunnel diameter by iteration, and calculates the numerical value of decision function in each iterative process, when decision function is rigid
When being greater than or equal to zero well, show that face is critical state, calculating tunnel diameter at this time is the maximum that benching tunnelling method excavates
Step height.
Step (1) the strength reduction factor FsCalculating formula are as follows:
In formula c andFor the cohesive strength and internal friction angle of country rock, ccrWithTo keep facing required for area face stability
Tunnel tunnel face analysis of stability can be two classes, F by boundary's intensive parameter, the strength reduction factorsWhen >=1, face is to stablize shape
State；FsIt is unstable state when < 1.
Average Accuracy described in step (3) is defined as follows:
N in formulaFAnd nTRespectively represent the quantity of error prediction result and correctly predicted result.
Compared with prior art, the present invention has following effect:
1. using estimation of stability model and design method of the invention, calculated without establishing numerical model, as long as
Tunnel is divided into multiple sections, determines six relevant parameters in each section, face can be stablized by decision function
Property assessed, and to design and construction provide it is corresponding suggest, improve the efficiency of constructing tunnel design.
Detailed description of the invention
The step flow chart of Fig. 1 the method for the invention；
Fig. 2 is the influence factor and level of tunnel face face stability；
The combination of Fig. 3, Fig. 4 orthogonal test and sample calibration；
Fig. 5 is strength reduction factor result figure；
Fig. 6 is SVM principle of classification figure；
Fig. 7 is the accuracy rate and corresponding supporting vector quantity of the disaggregated model established according to different kernel functions；
Fig. 8 is using the correlativity between the classification function numerical value and respective intensities reduction coefficient of linear kernel；
Fig. 9 is to roll over using the sample classification function value of quadratic polynomial kernel function and cubic polynomial kernel function and intensity
Subtract the relationship between coefficient；
Relationship of the Figure 10 between gaussian kernel function classifier and strength reduction factor；
Figure 11 is the Average Accuracy that k- rolls over the classification function based on different kernel functions that cross validation method obtains；
Figure 12 is that benching tunnelling method excavates schematic diagram；
Figure 13 is the process table that maximum bench excavation height is determined using quadratic polynomial classification function
Specific embodiment
The present invention is made further instructions below by embodiment.
Embodiment 1
Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method, comprising the following steps:
(1) it data mining: by Orthogonal Experiment and Design, chooses most representative parameter combination and establishes two-dimensional mathematical model
Strength degradation calculating is carried out, and according to strength reduction factor FsSample is demarcated, as strength reduction factor FsIt is to stablize sample when greater than 1
This, as strength reduction factor FsIt is unstable sample when less than 1；
(2) it establishes disaggregated model: the sample of support vector machines module training calibration is imported using Python, by embedding
Enter different kernel functions and is fitted a plurality of decision boundary to get different classification functions is arrived；
(3) model measurement: the Average Accuracy that cross validation calculates classifier is rolled over by k-, chooses higher point of accuracy rate
Class device carries out face judgement of stability, and utilizes collective study mechanism, uses different classifications device while classification is to be mutually authenticated
To assess the generalization ability of disaggregated model；
(4) face stability assessment: calculating the numerical value of classification function by program, to assess the stability of face,
When the value of classification function is greater than zero, face is stablized, and when the value of classification function is less than zero, face is unstable；
(5) it designs benching tunnelling method highest excavation height: in the case of face judging result is unstable, being with 0.01m
Step-length gradually reduces tunnel diameter by iteration, and calculates the numerical value of decision function in each iterative process, when decision function is rigid
When being greater than or equal to zero well, show that face is critical state, calculating tunnel diameter at this time is the maximum that benching tunnelling method excavates
Step height.
Step (1) the strength reduction factor FsCalculating formula are as follows:
In formula c andFor the cohesive strength and internal friction angle of country rock, ccrWithTo keep facing required for area face stability
Tunnel tunnel face analysis of stability can be two classes, F by boundary's intensive parameter, the strength reduction factorsWhen >=1, face is to stablize shape
State；FsIt is unstable state when < 1.
Average Accuracy described in step (3) is defined as follows:
N in formulaFAnd nTRespectively represent the quantity of error prediction result and correctly predicted result.
The stability of tunnel tunnel face is mainly tunnel diameter (D), buried depth (H), ground distributor respectively by six state modulators
Cloth load (σ), the severe (γ) of country rock, cohesive strength (c) and internal friction angle
(1) for data mining
In order to enable the classifier established has preferable generalization ability, the sample of acquisition needs to cover above 6 parameters
Main distribution.To each parameter, 12 levels are considered, as shown in Figure 2.Tunnel minimum diameter is Dmin=4m, it is maximum
Diameter is Dmax=15m, step-length Di+1-Di=1m.The diameter in most of tunnels is in scope of design, when diameter is more than 15m
When, it is large section tunnel, need to be specialized in.Buried depth minimum value is Hmin=5m, maximum value Hmax=60m, step-length Hi+1-
Hi=5m.When buried depth is greater than 60m, buried depth diameter ratio is greater than 4, is deep tunnel, will form arching above tunnel tunnel face,
Buried depth is little to face stability influence, can assume that edpth of tunnel is 60m at this time.
Earth's surface evenly load minimum value is 0kPa, maximum value σmax=110kPa, step-length 10kPa；Severe minimum value is
13kN/m3, maximum value 24kN/m3, step-length 1kN/m3, the severe of all soil bodys, including buoyant weight degree can be covered.The soil body is viscous
Poly- power minimum value σmin=0kPa, maximum cmax=44kPa, step-length ci+1-ci=4kPa；Soil body angle of friction minimum valueMaximum valueStep-length
Comprehensive Designing method is such as used, all groups of 6 factor, 12 level are combined into n=126=2985984.Data splitting amount
It is too big, it can not acquire, be also not suitable for being trained with support vector machines.More feasible method is to use Orthogonal Experiment and Design,
Choose its most representative factor combined sample.By orthogonal design, available 144 groups of Orthogonal Parameters combination, to every group
Parameter combination establishes two-dimensional mathematical model using OptumG2 software, and carries out strength degradation calculating, obtains corresponding strength degradation
Coefficient.First 72 groups and rear 72 groups of test combinations and sample calibration situation difference are as shown in Figure 3 and Figure 4.The intensity of all tests
Reduction coefficient is according to test number distribution results as shown in figure 5, wherein stabilization sample of the intensity more than or equal to 1.0 has 49 groups；It is small
Unstable sample in 1.0 has 95 groups.
(2) about establishing disaggregated model
1. support vector machine classifier
According to calibration sample achieved above, a classification boundaries can be fitted by support vector machines, this boundary can
Classify to the i.e. given parameters value of the sample newly inputted, principle is as shown in Figure 6.Its orbicular spot represents stable sample, side
Block represents unstable sample.These samples appear in 6 dimension parameter spaces being made of 6 influence factors in vector form
In.The realization of classification is to establish the hyperplane of one 5 dimension in this space, and 6 dimension spaces are divided into two sub-spaces, a sub-spaces
Only stable sample, another subspace only have unstable sample.The classification of new samples can be by detecting its parameter vector
Which realized in subspace.
WhereinOne group of tunnel operating condition is represented,Represent Optimal Separating Hyperplane
Function, that is, decision boundary expression formula.To determine this boundary, for all stable samples nearest from classification boundaries, enable it full
Foot:
Its satisfaction is enabled for all unstable samples nearest from decision boundary:
WhereinWithRepresent be parallel to two of classification boundaries it is parallel super flat
Face,The sample being located on the two parallel hyperplane is represented,The normal vector of these hyperplane is represented,It represents
Hyperplane is from origin along the offset of normal direction.It is divided between two parallel hyperplaneThe principle of support vector machines is most
Change the interval greatly, and enabling the middle position of two hyperplane is classification boundaries.
Problems described above belongs to convex optimization problem, can be solved by solving the approach of Lagrangian.Scikit-
Learn provides SVM module, uses the available required classification function solution of the module.Input one group of parameterClassification problem is just changed into the value for calculating following sign function.
When function result is timing, showing to input is positive sample, and tunnel tunnel face is predicted as stable state, works as classification
When function result is negative value, show that input is negative sample, tunnel tunnel face is predicted as unstable state.Using superior function as line
Property classification function, when data be linearly inseparable when, need to using kernel function to classification function carry out Nonlinear Mapping, so that it is had
Nonlinear Classification function.The Polynomial kernel function mainly used, gaussian radial basis function and the following institute of sigmoid kernel function
Show:
For Nonlinear Classification model, stability prediction can be carried out by calculating with the symbol of minor function by inputting new samples.
WhereinAll supporting vectors are represented,Represent the new samples vector of input.
2. establishing disaggregated model by different kernel functions
Python program is worked out, and imports SVM module and different kernel functions from scikit-learn, to the data of calibration
Sample is trained, and can establish different classifiers.But before application model, need to assess its accuracy,
In Average Accuracy be defined as follows shown in:
N in formulaFAnd nTRespectively represent the quantity of error prediction result and correctly predicted result.Fig. 7 gives according to different IPs
The accuracy rate for the disaggregated model that function is established and related supporting vector quantity.The results show that using sigmoid core letter
Several disaggregated model accuracys rate is lower, belongs to poor fitting, it is impossible to be used in the prediction of new samples.
For the disaggregated model of application linear kernel function, the coefficient and intercept of classification function are as follows:
Give one group of parameterThe numerical value of classification function can be calculated by following equation, and
Judge positive and negative.
Fig. 8 gives the correlativity between classification function numerical value and respective intensities reduction coefficient using linear kernel.Wherein
TN and FN respectively represents correct and wrong unstable sample, and TP and FP respectively represent correct and wrong stabilization sample.Fig. 8
(b) show the sample of a total of 6 mistakes classification, wherein 3 wrong positive samples stablize sample, 3 wrong negative samples
I.e. unstable sample.
Fig. 9 gives using the sample classification function value of quadratic polynomial kernel function and cubic polynomial kernel function and strong
The relationship between reduction coefficient is spent, Figure 10 gives the relationship between gaussian kernel function classifier and strength reduction factor.As can be seen that
These three disaggregated models have carried out correct classification to all models, do not occur wrong classification samples, accuracy rate 100%.
But what the detection of above accuracy was directed to is all training sample, can only judge whether disaggregated model is poor fitting, can not be judged point
Whether class model is over-fitting, i.e., can not assessment models to the predictablity rates of new samples.Therefore, it also needs to survey the model
Examination, to assess its predictablity rate to new samples.
(3) about test model
For the generalization ability for assessing disaggregated model, it need to judge whether disaggregated model has higher accuracy rate to new data.?
In the case where lacking enough verify datas, the generalization ability of k- folding cross validation assessment models usually can be used.This method will instruct
Practice data acquisition system and is equally divided into k subset.Reserved a subset forms training set by residuary subset as verifying collection, assessment every time
The accuracy rate of the classification function of fitting.As k=5, verification process is carried out 5 times.Figure 11 is obtained using this verification method
The Average Accuracy of classification function based on different kernel functions, wherein μ and σ respectively indicates the average value and mark of 5 Evaluation accuracies
It is quasi- poor.It can be seen that Gaussian kernel is lower to the prediction Average Accuracy of new samples, illustrate that the classification function is over-fitting.Linearly
The Average Accuracy of core, quadratic polynomial core and cubic polynomial verification verify data is higher, can apply these classification results
Carry out face stability prediction.
Compared to single disaggregated model, more reliable method is simultaneously using based on linear kernel, quadratic polynomial core and three
The disaggregated model of order polynomial core is predicted simultaneously, is adopted when prediction result is consistent, when prediction result conflict, with
Subject to unfavorable prediction result, that is, it is predicted as unstable.
(4) about face stability assessment
According to obtained disaggregated model, one group of parameter is givenClassification function number can be calculated
Value, and provide prediction result.It is assumed that the parameter provided is x=[9,26,110,20,12,41], the calculating knot of linear classification model
Fruit are as follows:
Calculated result is negative, and shows that the corresponding tunnel tunnel face of this group of parameter is unstable state.Meanwhile using secondary more
The numerical value for the classification function that item formula kernel function and cubic polynomial kernel function are calculated is respectively -2.067 and -2.787, equal table
The bright tunnel tunnel face is unstable state.It is emphasized that the kernel function due to use is different, obtained different classifications device
It only needs symbol identical the functional value that same group of parameter obtains, does not need numerically to coincide.
Linear classification model can directly be calculated using formula, therefore can be conveniently used with the preliminary of stability
Assessment.Rather than linear classification model includes all supporting vectors in calculating due to needing, calculating process is complicated, Ying Caiyong journey
Sequence calculates, to avoid error.
(5) design of benching tunnelling method highest excavation height
When face is unstable, needs to adopt various measures and improve its stability, to guarantee the safety of work progress.
One of method is excavated using benching tunnelling method, and the height once excavated is reduced.Figure 12 gives the schematic diagram of benching tunnelling method excavation.
Such as being given above data x=[9,26,110,20,12,41], by prediction, face is unstable state.Benching tunnelling method
Tunnel excavation height is exactly dropped to D from D by the essence of excavation1.In order to determine maximum D1, can by way of iteration, with
Step-length 0.01m one small constantly reduces the numerical value of initial diameter D, iterates to calculate its classification function numerical value every time, works as classification function
When greater than zero, the diameter used at this time is D1Maximum value.
Figure 13, which gives, determines maximum bench excavation high process using quadratic polynomial classification function, wherein first group of number
According to for primary data, after iteration, the diameter D of (DF=0.003) when classification function is greater than zero1As step maximum is opened
Dig height.The strength reduction factor for using this group of parameter progress numerical value to be calculated is Fs=0.98, show area at this time
Face is approximately critical state (i.e. Fs=1.000).
Safer solution in order to obtain, it is proposed that using DF=1.000 as the target value of classification function, because the value is all
The constraint condition of stable training sample.Determining D at this time1=5.97m is the maximum excavation height (F of steps=1.06).When
The bench excavation height of design is less than this critical value, is just avoided that face caves in.
Using estimation of stability model of the invention and design method, area is carried out it is not necessary that tunnel is divided into multiple sections
Face stability assessment, for any one group of given parameters, so that it may area face stability be assessed, and to design and construction
Corresponding suggestion is provided, the efficiency of constructing tunnel design is improved.
The aforementioned description to specific exemplary embodiment of the invention is in order to illustrate and illustration purpose.These descriptions
It is not wishing to limit the invention to disclosed precise forms, and it will be apparent that according to the above instruction, can much be changed
And variation.The purpose of selecting and describing the exemplary embodiment is that explaining specific principle of the invention and its actually answering
With so that those skilled in the art can be realized and utilize a variety of different exemplary implementation schemes of the invention and
Various chooses and changes.The scope of the present invention is intended to be limited by claims and its equivalents.
Claims (3)
1. tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method, which is characterized in that including following step
It is rapid:
(1) it data mining: by Orthogonal Experiment and Design, chooses most representative parameter combination and establishes two-dimensional mathematical model progress
Strength degradation calculates, and according to strength reduction factor FsSample is demarcated, as strength reduction factor FsTo stablize sample when greater than 1, when
Strength reduction factor FsIt is unstable sample when less than 1；
(2) it establishes disaggregated model: the sample of support vector machines module training calibration is imported using Python, by being embedded in not
Same kernel function is fitted a plurality of decision boundary to get different classification functions is arrived；
(3) model measurement: the Average Accuracy that cross validation calculates classifier is rolled over by k-, chooses the higher classifier of accuracy rate
Face judgement of stability is carried out, and utilizes collective study mechanism, different classifications device is used while classification is commented with being mutually authenticated
Estimate the generalization ability of disaggregated model；
(4) face stability assessment: by program calculate classification function numerical value, to assess the stability of face, when point
Face is stablized when the value of class function is greater than zero, and when the value of classification function is less than zero, face is unstable；
(5) benching tunnelling method highest excavation height is designed: in the case of face judging result is unstable, using 0.01m as step-length,
Tunnel diameter is gradually reduced by iteration, and calculates the numerical value of decision function in each iterative process, when decision function is just big
When zero, show that face is critical state, calculating tunnel diameter at this time is the maximum step height that benching tunnelling method excavates
Degree.
2. tunnel tunnel face estimation of stability model as described in claim 1 and benching tunnelling method excavation height design method, special
Sign is: step (1) the strength reduction factor FsCalculating formula are as follows:
In formula c andFor the cohesive strength and internal friction angle of country rock, ccrWithTo keep critical strong required for area face stability
Parameter is spent, which can be two classes, F by tunnel tunnel face analysis of stabilitysWhen >=1, face is stable state；Fs<
It is unstable state when 1.
3. tunnel tunnel face estimation of stability model as described in claim 1 and benching tunnelling method excavation height design method, special
Sign is: Average Accuracy described in step (3) is defined as follows:
N in formulaFAnd nTRespectively represent the quantity of error prediction result and correctly predicted result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811484191.8A CN109740183B (en) | 2018-12-06 | 2018-12-06 | Tunnel face stability evaluation model and step method excavation height design method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811484191.8A CN109740183B (en) | 2018-12-06 | 2018-12-06 | Tunnel face stability evaluation model and step method excavation height design method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109740183A true CN109740183A (en) | 2019-05-10 |
| CN109740183B CN109740183B (en) | 2021-05-04 |
Family
ID=66358566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811484191.8A Active CN109740183B (en) | 2018-12-06 | 2018-12-06 | Tunnel face stability evaluation model and step method excavation height design method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109740183B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111475765A (en) * | 2020-04-14 | 2020-07-31 | 交通运输部公路科学研究所 | Method for discriminating tunnel step method |
| CN112098120A (en) * | 2020-08-04 | 2020-12-18 | 中国铁建股份有限公司 | Experimental method and device for measuring influence of split grouting on tunnel face stability |
| CN112182729A (en) * | 2020-10-26 | 2021-01-05 | 武汉理工大学 | Tunnel face stability rapid determination method based on naive Bayes |
| CN112307547A (en) * | 2020-11-02 | 2021-02-02 | 武汉理工大学 | Design method for tunnel face support pressure |
| CN112365044A (en) * | 2020-11-09 | 2021-02-12 | 武汉理工大学 | Tunnel face failure probability prediction method based on k nearest neighbor algorithm and support vector machine |
| CN113515795A (en) * | 2021-06-29 | 2021-10-19 | 中国建筑第八工程局有限公司 | Tunnel safety factor calculation method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102155231A (en) * | 2011-03-18 | 2011-08-17 | 大连海事大学 | Quick feedback analyzing system in tunnel constructing process |
| JP2012112111A (en) * | 2010-11-22 | 2012-06-14 | Shimizu Corp | Stability evaluation method and stability evaluation device |
| CN106570292A (en) * | 2016-11-14 | 2017-04-19 | 中建山东投资有限公司 | Super-large-cross-section tunnel-based surrounding rock classification method |
| CN107357966A (en) * | 2017-06-21 | 2017-11-17 | 山东科技大学 | A kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure |
| CN108875152A (en) * | 2018-05-28 | 2018-11-23 | 湖南科技大学 | A kind of tunnel tunnel face calculating method for stability considering penetration |
-
2018
- 2018-12-06 CN CN201811484191.8A patent/CN109740183B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012112111A (en) * | 2010-11-22 | 2012-06-14 | Shimizu Corp | Stability evaluation method and stability evaluation device |
| CN102155231A (en) * | 2011-03-18 | 2011-08-17 | 大连海事大学 | Quick feedback analyzing system in tunnel constructing process |
| CN106570292A (en) * | 2016-11-14 | 2017-04-19 | 中建山东投资有限公司 | Super-large-cross-section tunnel-based surrounding rock classification method |
| CN107357966A (en) * | 2017-06-21 | 2017-11-17 | 山东科技大学 | A kind of surrounding rock of actual mining roadway stability prediction and appraisal procedure |
| CN108875152A (en) * | 2018-05-28 | 2018-11-23 | 湖南科技大学 | A kind of tunnel tunnel face calculating method for stability considering penetration |
Non-Patent Citations (6)
| Title |
|---|
| GUANG HUANG 等: "《Modelling seismic fragility of a rock mountain tunnelbased on support vector machine》", 《SOIL DYNAMICS AND EARTHQUAKE ENGINEERING》 * |
| WANG KUNPENG等: "《The Parameter Back Analysis of Rock Mass Based on Orthogonal Design》", 《2014 7TH INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTATION TECHNOLOGY AND AUTOMATION》 * |
| 安永林等: "《基于强度折减法的隧道掌子面稳定性及破坏形态》", 《矿业工程研究》 * |
| 李红: "《有限元强度折减法对逐级开挖高边坡的稳定性分析》", 《南京理工大学学报》 * |
| 章志刚: "《基于强度折减法的掌子面稳定性上限分析》", 《路基工程》 * |
| 胡军 等: "《隧道洞室地基稳定性进化支持向量机识别》", 《辽宁工程科技大学学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111475765A (en) * | 2020-04-14 | 2020-07-31 | 交通运输部公路科学研究所 | Method for discriminating tunnel step method |
| CN111475765B (en) * | 2020-04-14 | 2023-06-20 | 交通运输部公路科学研究所 | Discrimination method of tunnel step method |
| CN112098120A (en) * | 2020-08-04 | 2020-12-18 | 中国铁建股份有限公司 | Experimental method and device for measuring influence of split grouting on tunnel face stability |
| CN112182729A (en) * | 2020-10-26 | 2021-01-05 | 武汉理工大学 | Tunnel face stability rapid determination method based on naive Bayes |
| CN112307547A (en) * | 2020-11-02 | 2021-02-02 | 武汉理工大学 | Design method for tunnel face support pressure |
| CN112365044A (en) * | 2020-11-09 | 2021-02-12 | 武汉理工大学 | Tunnel face failure probability prediction method based on k nearest neighbor algorithm and support vector machine |
| CN113515795A (en) * | 2021-06-29 | 2021-10-19 | 中国建筑第八工程局有限公司 | Tunnel safety factor calculation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109740183B (en) | 2021-05-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109740183A (en) | Tunnel tunnel face estimation of stability model and benching tunnelling method excavation height design method | |
| CN109034582A (en) | Tunnel Passing inrush through faults based on cloud model and combination weighting are dashed forward mud risk evaluating method | |
| CN105488248B (en) | Judgment method for deep ore stratum and rock stratum | |
| CN108614944A (en) | A kind of shield track axis correction parameter prediction technique | |
| CN103827698A (en) | Cluster 3d petrophysical uncertainty modeling | |
| CN106529150B (en) | Compound stratum shield tunnel vault load calculation method | |
| CN105493100A (en) | Static earth model calibration methods and systems | |
| AU2013297036B2 (en) | Methods and systems related to hydrocarbon recovery strategy development | |
| US10233727B2 (en) | Induced control excitation for enhanced reservoir flow characterization | |
| CN109826174A (en) | A kind of slope reinforcement deep regional range determining method | |
| CN104471185A (en) | System and method for reservoir pressure data analysis | |
| EP2795527A1 (en) | Systems and methods for selecting facies model realizations | |
| KR102155101B1 (en) | System and method for predicting ground layer information, and a recording medium having computer readable program for executing the method | |
| CN104265279A (en) | Logging while drilling curve predicating method under geological fault condition | |
| CN109740119A (en) | A kind of TBM driving Wall Rock of Tunnel uniaxial compressive strength Method of fast estimating | |
| CN106096081B (en) | The estimation method of reserve of fracture hole type bottom water reservoir | |
| CN104615873A (en) | Disaster-pregnant performance evaluation method for karst region inrush water disaster source | |
| Jung et al. | Probabilistic soil identification based on cone penetration tests | |
| US20160177706A1 (en) | Formation fracturing potential using surrounding pore pressures | |
| NO20180082A1 (en) | Time-to-finish simulation forecaster | |
| CN104459800A (en) | Sand body pinch-out prediction method and device | |
| CN117786794A (en) | Shield tunneling existing tunnel deformation optimization control method and system | |
| CN107780926A (en) | Well cross sectional shape measuring method, device and terminal | |
| CN114488297A (en) | Fault identification method and device | |
| US11543549B2 (en) | Method and system for analyzing a reservoir grid of a reservoir geological formation based on 4D seismic images |
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 |