CN110705178A - Tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning - Google Patents
Tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning Download PDFInfo
- Publication number
- CN110705178A CN110705178A CN201910933480.XA CN201910933480A CN110705178A CN 110705178 A CN110705178 A CN 110705178A CN 201910933480 A CN201910933480 A CN 201910933480A CN 110705178 A CN110705178 A CN 110705178A
- Authority
- CN
- China
- Prior art keywords
- surrounding rock
- tunnel
- rock
- deformation
- section
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
- G06Q10/04—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
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The invention relates to a dynamic prediction method for surrounding rock deformation in the whole process of tunnel/subway construction based on machine learning, which takes the same design surrounding rock grade section of a tunnel/subway as a research object, and can obtain the surrounding rock deformation convergence value of each selected section through field monitoring measurement. And taking the selected monitoring section as a sample, wherein each sample contains the natural property index of the rock mass of the section and the surrounding rock deformation convergence value. And collecting information of each measuring section of the excavated section to form a sample space. And then the deformation convergence value of the surrounding rock is directly predicted through the natural attribute of the exposed surrounding rock of the current tunnel face. The invention researches a method for predicting surrounding rock deformation response in the whole process of tunnel/subway construction based on machine learning. Compared with the previous research, the method is based on the prior distribution information in the tunnel/subway excavation process, and further a response prediction model of the surrounding rock deformation is constructed. And direct data support is provided for subsequent construction method change and support parameter optimization.
Description
Technical Field
The invention relates to a method for predicting surrounding rock deformation in the whole process of tunnel/subway construction, in particular to a method for rapidly, dynamically and timely predicting the surrounding rock deformation of a tunnel/subway based on machine learning.
Background
Since the introduction of the new Olympic method, the new Olympic method is confirmed by tunnel engineering technical experts all over the world, and is widely popularized and applied to the construction of tunnels and underground projects. As a core meaning, surrounding rock deformation information fed back by field measurement is important, but due to a plurality of influence factors (relating to hydrogeological environment, rock mass structure attributes, excavation supporting construction methods, construction technologies, management levels and the like), a response model between each universal factor and surrounding rock deformation is difficult to construct. However, for a specific tunnel/subway, the same design surrounding rock grade section is taken as a research object, and the engineering influence can be regarded as the same effect level, so that the mapping relation between the natural attribute of the rock mass and the deformation of the surrounding rock is established. As a complete system project, the inherent correlation between the natural property of the rock mass disclosed by the excavated section of a certain tunnel/subway and the surrounding rock deformation is often ignored, and the method has practical guiding significance for the subsequent construction of the tunnel/subway.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, fully utilizes the prior distribution information of the exposed surrounding rock of the excavated segment, and provides a response prediction method of the tunnel face surrounding rock deformation convergence value based on machine learning. The method can provide direct and effective data support for construction method conversion and support parameter optimization in the tunnel/subway excavation process.
In order to achieve the purpose, the invention adopts the following technical scheme:
a dynamic prediction method for the deformation of surrounding rock in the whole process of tunnel/subway construction based on machine learning is characterized by that the same design surrounding rock grade section of tunnel/subway is used as the object of study, the evaluation indexes (geological structure, rock mechanical parameters and structural plane attitude) of the natural properties of rock mass of each excavated tunnel face are obtained from the excavated part of said section, and the deformation convergence value of surrounding rock of each selected section can be obtained by means of on-site monitoring measurement. And taking the selected monitoring section as a sample, wherein each sample contains the natural property index of the rock mass of the section and the surrounding rock deformation convergence value. And collecting information of each measured section of the excavated segment to form a sample space, and constructing a mapping model based on a machine learning algorithm (adopting Gaussian process regression here). And then the deformation convergence value of the surrounding rock is directly predicted through the natural attribute of the exposed surrounding rock of the current tunnel face. The method realizes the advanced prediction of the deformation convergence value of the surrounding rock in the tunnel/subway construction process, thereby carrying out the instant evaluation on the stability condition of the surrounding rock.
The advanced prediction method is based on the prior distribution rule of rock structure information and a machine learning algorithm.
A dynamic prediction method for surrounding rock deformation in the whole process of tunnel/subway construction based on machine learning is characterized by comprising the following steps:
A. the method comprises the following steps of collecting rock mass natural attribute evaluation indexes and surrounding rock convergence deformation values of monitoring sections of an excavated tunnel/subway section as training samples, wherein the input sample pairs of a GPR model are as follows: (x i ，y i )，i=1，2，…，kWherein, in the step (A),x i to represent several quantitative control factors (here only natural property influencing factors such as rock strength, rock integrity, self-characteristics of rock structure, volume fraction, grade of surrounding rock, etc.),y i the convergence value of the deformation of the surrounding rock is taken as the value;
B. the construction of a mapping model between the natural attribute of the rock mass and the deformation convergence value of the surrounding rock requires the provision of a training sample of a functional function Z, and the selected sample point is information of each section of an excavated part of the same tunnel/subway in a surrounding rock grade section of the same design. In the practical application process, along with the continuous excavation of the tunnel/subway, the sample set is updated in a rolling mode, the selected n samples need to be guaranteed to be the sample information of n sections nearest to the current tunnel face, namely, the sample points farthest from the tunnel face are abandoned while the next section is added as the sample point;
C. when the magnitude order of various control factors influencing the deformation of the surrounding rock, such as rock strength, RQD value, rock integrity degree, tunnel/subway burial depth and the like, has large difference, the standard processing needs to be carried out on the sample data
In the formula:
x i is as followsiAn index;P i is a normalized value;
D. based on the constructed sample space, performing model training by using the existing machine learning algorithm to establish control influence factorsx i And output responsey i Mapping relation between them, obtaining the influence of given engineeringXCorresponding surrounding rock deformation convergence predicted valuey*. And providing data support for the current tunnel/subway excavation construction method and support parameter optimization.
The machine learning algorithm and rock mass structure information acquisition equipment (including non-contact measurement technologies such as binocular photogrammetry and three-dimensional laser scanning) in the invention are all existing things, and are not described herein again.
It should be noted that the use of machine learning algorithm herein does not mean a single algorithm, including currently mainstream Artificial Neural Networks (ANN), Support Vector Machines (SVM), gaussian process regression theory (GPR), and the like.
It should be noted that the above description is intended to provide further explanation of the present application. The preferred embodiments of the present invention are merely examples, which are not intended to limit the present invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
The method takes the same design surrounding rock grade section of the tunnel/subway as a research object, obtains rock natural attribute evaluation indexes (geological structure, rock mechanical parameters, structural plane attitude and the like) of each excavation working face from the excavated part, and can obtain the surrounding rock deformation convergence value of each selected section through field monitoring measurement. And taking the selected monitoring section as a sample, wherein each sample contains the natural property index of the rock mass of the section and the surrounding rock deformation convergence value. And collecting information of each measured section of the excavated section to form a sample space, and constructing a mapping model based on a machine learning algorithm. And then the deformation convergence value of the surrounding rock is directly predicted through the natural attribute of the exposed surrounding rock of the current tunnel face.
The invention researches a tunnel/subway surrounding rock deformation response prediction method based on machine learning, realizes advanced prediction of a surrounding rock deformation convergence value in the tunnel/subway construction process, and thus immediately evaluates the surrounding rock stability condition. Compared with the previous research, the method is different from the time sequence prediction of displacement, fully considers the natural attributes of all rock masses influencing the deformation of the surrounding rock, and constructs a response prediction model of the deformation of the surrounding rock based on the prior distribution information in the tunnel/subway excavation process. And direct data support is provided for subsequent construction method change and support parameter optimization.
The method solves the problem of immediate prediction of the deformation convergence value of the surrounding rock in the tunnel/subway construction process, and has the following advantages:
1. predicting the deformation of the surrounding rock based on a machine learning algorithm without manually analyzing and processing data; the manpower and financial resources are saved; with the continuous construction of the sample library, the prediction precision of the sample library can be continuously improved;
2. the deformation convergence of the surrounding rock is not required to be waited, the defects of long-term and accurate acquisition of displacement information and limited application section in displacement time sequence prediction are avoided, and the advanced prediction of the deformation value of the surrounding rock in the tunnel/subway construction process is realized;
3. in the conventional evaluation analysis model or method, because the object-oriented objects are different tunnel/subway projects across the country, all influence factors and project feasibility need to be considered comprehensively and comprehensively, so that the evaluation result of the established mathematical model is mostly on the level of a grade interval. The sample library constructed by the method is based on the prior distribution information of the excavated exposed rock mass of the researched tunnel/subway, so that specific problems are fully considered, and the specific guidance of specific engineering is realized;
4. for a specific tunnel/subway, under the condition of the same design surrounding rock level section, a certain mapping relation exists between the natural attribute and the deformation of the surrounding rock, and corresponding response analysis is carried out through a machine learning algorithm. Although the predicted displacement has errors, the variation trend of the predicted displacement can basically reflect the actual engineering, so that a quantitative basis is provided for quick change decision, and the condition that the construction period is delayed due to change can be reduced.
FIG. 1 is a flow chart of a prediction method of the present invention.
Claims (3)
1. A tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning fully utilizes prior distribution information of exposed surrounding rocks of an excavated section, collects rock natural attribute evaluation indexes (geological structure, rock mechanical parameters, structural plane attitude and the like) of each excavated tunnel face obtained from an excavated part, and simultaneously obtains surrounding rock deformation convergence values of each selected section through field monitoring measurement, and each sample contains the natural attribute indexes and the surrounding rock deformation convergence values of the rock mass of the section by taking the selected monitored section as a sample; collecting information of each measured section of an excavated section to form a sample space, taking natural attribute indexes of a revealed rock mass as input samples and surrounding rock deformation convergence values as output samples, and constructing a mapping model between the natural attribute and the deformation of the rock mass in the same design surrounding rock grade section based on a machine learning algorithm to realize response prediction of the surrounding rock deformation convergence values of the current face and instant evaluation of the stability condition of the surrounding rock; therefore, data support is provided for tunnel/subway excavation construction method and support parameter optimization.
2. The machine learning-based dynamic prediction method for the deformation of the surrounding rock in the whole process of tunnel/subway construction according to claim 1, characterized in that: the prediction method is based on the prior distribution rule of rock mass structure information and a machine learning algorithm.
3. A dynamic prediction method for surrounding rock deformation in the whole process of tunnel/subway construction based on machine learning is characterized by comprising the following steps:
A. the method comprises the following steps of collecting rock mass natural attribute evaluation indexes and surrounding rock convergence deformation values of monitoring sections of an excavated tunnel/subway section as training samples, wherein the input sample pairs of a GPR model are as follows: (x i ，y i )，i=1，2，…，kWherein, in the step (A),x i to represent several quantitative control factors (here only natural property influencing factors such as rock strength, rock integrity, self-characteristics of rock structure, volume fraction, grade of surrounding rock, etc.),y i the convergence value of the deformation of the surrounding rock is taken as the value;
B. constructing a mapping model between the natural attribute of the rock mass and the deformation convergence value of the surrounding rock, wherein a training sample of a functional function Z needs to be provided, and the selected sample point is information of each section of an excavated part of the same tunnel/subway in a surrounding rock grade section with the same design; in the practical application process, along with the continuous excavation of the tunnel/subway, the sample set is updated in a rolling mode, the selected n samples need to be guaranteed to be the sample information of n sections nearest to the current tunnel face, namely, the sample points farthest from the tunnel face are abandoned while the next section is added as the sample point;
C. when the magnitude order of various control factors influencing the deformation of the surrounding rock, such as rock strength, RQD value, rock integrity degree, tunnel/subway burial depth and the like, has large difference, the standard processing needs to be carried out on the sample data
In the formula:
x i is as followsiAn index;P i is a normalized value;
D. based on the constructed sample space, performing model training by using the existing machine learning algorithm to establish control influence factorsx i And output responsey i Mapping relation between them, obtaining the influence of given engineeringXCorresponding surrounding rock deformation convergence predicted valueyA first step of; and providing data support for the current tunnel/subway excavation construction method and support parameter optimization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910933480.XA CN110705178A (en) | 2019-09-29 | 2019-09-29 | Tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910933480.XA CN110705178A (en) | 2019-09-29 | 2019-09-29 | Tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110705178A true CN110705178A (en) | 2020-01-17 |
Family
ID=69197263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910933480.XA Pending CN110705178A (en) | 2019-09-29 | 2019-09-29 | Tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110705178A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340275A (en) * | 2020-02-18 | 2020-06-26 | 山东科技大学 | Tunnel support mode selection real-time prediction method based on detection while drilling technology |
| CN112614021A (en) * | 2020-12-24 | 2021-04-06 | 东南大学 | Tunnel surrounding rock geological information prediction method based on built tunnel information intelligent identification |
| CN112664174A (en) * | 2020-12-21 | 2021-04-16 | 中铁四局集团第五工程有限公司 | Tunnel surrounding rock grade determination method and system based on multiple drill holes |
| CN113268808A (en) * | 2021-07-21 | 2021-08-17 | 中铁大桥科学研究院有限公司 | Digital detection method for top-lifting construction of extremely-soft and weak broken surrounding rock door type system |
| CN113360998A (en) * | 2021-07-14 | 2021-09-07 | 四川绵九高速公路有限责任公司 | Large deformation trend dynamic judgment and construction decision method for large deformation tunnel |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103870677A (en) * | 2014-02-07 | 2014-06-18 | 上海交通大学 | Setting method for tunneling parameters of tunneling machine |
| CN107702638A (en) * | 2017-11-08 | 2018-02-16 | 山东科技大学 | Country rock excavation deformation overall process monitoring system and application method |
| CN109241627A (en) * | 2018-09-07 | 2019-01-18 | 大连海事大学 | The dynamic shoring method of probability hierarchical and the device of Automated Design supporting scheme |
-
2019
- 2019-09-29 CN CN201910933480.XA patent/CN110705178A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103870677A (en) * | 2014-02-07 | 2014-06-18 | 上海交通大学 | Setting method for tunneling parameters of tunneling machine |
| CN107702638A (en) * | 2017-11-08 | 2018-02-16 | 山东科技大学 | Country rock excavation deformation overall process monitoring system and application method |
| CN109241627A (en) * | 2018-09-07 | 2019-01-18 | 大连海事大学 | The dynamic shoring method of probability hierarchical and the device of Automated Design supporting scheme |
Non-Patent Citations (1)
| Title |
|---|
| 贺鹏等: "基于数据挖掘的隧道围岩变形响应预测与动态变更许可机制", 《岩石力学与工程学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340275A (en) * | 2020-02-18 | 2020-06-26 | 山东科技大学 | Tunnel support mode selection real-time prediction method based on detection while drilling technology |
| CN112664174A (en) * | 2020-12-21 | 2021-04-16 | 中铁四局集团第五工程有限公司 | Tunnel surrounding rock grade determination method and system based on multiple drill holes |
| CN112614021A (en) * | 2020-12-24 | 2021-04-06 | 东南大学 | Tunnel surrounding rock geological information prediction method based on built tunnel information intelligent identification |
| CN113360998A (en) * | 2021-07-14 | 2021-09-07 | 四川绵九高速公路有限责任公司 | Large deformation trend dynamic judgment and construction decision method for large deformation tunnel |
| CN113268808A (en) * | 2021-07-21 | 2021-08-17 | 中铁大桥科学研究院有限公司 | Digital detection method for top-lifting construction of extremely-soft and weak broken surrounding rock door type system |
| CN113268808B (en) * | 2021-07-21 | 2021-10-26 | 中铁大桥科学研究院有限公司 | Digital detection method for top-lifting construction of extremely-soft and weak broken surrounding rock door type system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110705178A (en) | Tunnel/subway construction overall process surrounding rock deformation dynamic prediction method based on machine learning | |
| Qi et al. | Prediction of open stope hangingwall stability using random forests | |
| Shen et al. | Real-time prediction of shield moving trajectory during tunnelling | |
| Ninić et al. | A hybrid finite element and surrogate modelling approach for simulation and monitoring supported TBM steering | |
| CN109165664A (en) | A kind of attribute missing data collection completion and prediction technique based on generation confrontation network | |
| CN109740119B (en) | Rapid estimation method for uniaxial compressive strength of surrounding rock of TBM tunneling tunnel | |
| Harandizadeh et al. | TBM performance prediction developing a hybrid ANFIS-PNN predictive model optimized by imperialism competitive algorithm | |
| Wang et al. | A novel combination of PCA and machine learning techniques to select the most important factors for predicting tunnel construction performance | |
| Yao et al. | Improved support vector machine regression in multi-step-ahead prediction for rock displacement surrounding a tunnel | |
| Ghorbani et al. | Determination of initial stress state and rock mass deformation modulus at Lavarak HEPP by back analysis using ant colony optimization and multivariable regression analysis | |
| Liu et al. | Intelligent information-based construction in tunnel engineering based on the GA and CCGPR coupled algorithm | |
| Yang et al. | A no-tension elastic–plastic model and optimized back-analysis technique for modeling nonlinear mechanical behavior of rock mass in tunneling | |
| Gao et al. | Back analysis of rock mass parameters and initial stress for the Longtan tunnel in China | |
| Yagiz et al. | Application of differential evolution algorithm and comparing its performance with literature to predict rock brittleness for excavatability | |
| CN103885867B (en) | Online evaluation method of performance of analog circuit | |
| Wang et al. | A novel model of the ideal point method coupled with objective and subjective weighting method for evaluation of surrounding rock stability | |
| CN110378574A (en) | Submerged tunnel Pressure Shield Tunnel face stability evaluation method, system and equipment | |
| CN108733952B (en) | Three-dimensional characterization method for spatial variability of soil water content based on sequential simulation | |
| Wang et al. | An extreme gradient boosting technique to estimate TBM penetration rate and prediction platform | |
| Shang et al. | Intelligent back analysis of displacements using precedent type analysis for tunneling | |
| CN111305899A (en) | Method for determining removal length of temporary support for construction of subway station arch cover method | |
| Sun et al. | Inversion of Surrounding Rock Mechanical Parameters in a Soft Rock Tunnel Based on a Hybrid Model EO-LightGBM | |
| Wang et al. | Discussion on the prediction of engineering cost based on improved BP neural network algorithm | |
| Cao et al. | Predication of displacement of tunnel rock mass based on the back-analysis method-BP neural network | |
| CN106980874A (en) | A kind of multi-time Scales dimension data fusion method towards distribution big data |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200814 Address after: 266590 No. 579, Bay Road, Huangdao District, Shandong, Qingdao Applicant after: Shandong University of Science and Technology Applicant after: Sichuan Jingyu Biotechnology Co., Ltd Address before: 266590 No. 579, Bay Road, Huangdao District, Shandong, Qingdao Applicant before: Shandong University of Science and Technology |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200117 |