CN113779665A - Engineering rock mass fracture degradation numerical simulation method based on continuous medium - Google Patents
Engineering rock mass fracture degradation numerical simulation method based on continuous medium Download PDFInfo
- Publication number
- CN113779665A CN113779665A CN202110885245.7A CN202110885245A CN113779665A CN 113779665 A CN113779665 A CN 113779665A CN 202110885245 A CN202110885245 A CN 202110885245A CN 113779665 A CN113779665 A CN 113779665A
- Authority
- CN
- China
- Prior art keywords
- rock mass
- data
- engineering
- numerical simulation
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- 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
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Fuzzy Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of rock mass engineering, and discloses a continuous medium-based engineering rock mass fracture degradation numerical simulation method, which comprises the following steps: collecting data of rock mass related to the consolidation project, and classifying correspondingly according to different conditions; establishing a model by using finite element numerical simulation software; setting numerical parameters of the applied engineering, including the limitation of range conditions in the engineering by utilizing various threshold conditions; and simulating the engineering implementation process by using finite element numerical simulation software, and carrying out iterative calculation of the fracture and degradation of the engineering rock mass in the simulation process to establish a corresponding numerical simulation program. The numerical simulation method is used for carrying out numerical simulation calculation under the fractured rock mass environment, and the dynamic state of the fracture disturbed by excavation in the process of stress redistribution in the surrounding rock is considered, so that the simulation process is more reasonable and rigorous, a more real numerical simulation result is obtained, and scientific reference is provided for engineering simulation.
Description
Technical Field
The invention belongs to the technical field of rock mass engineering, and particularly relates to a continuous medium-based engineering rock mass fracture degradation numerical simulation method.
Background
At present, in the fields of tunnels, mining industry, hydroelectric power, nuclear power and the like of geotechnical engineering, particularly underground engineering, the original stress state of surrounding rocks is broken through by artificial excavation disturbance, and the stress in the surrounding rocks is redistributed. The process enables primary macroscopic and microscopic fractures in the rock to develop, expand and even be communicated with each other, so that the mechanical properties of the surrounding rock are deteriorated and deformation and damage occur to different degrees, and the stability of the surrounding rock is influenced. The stability control of the surrounding rock is an important research subject in the field of geotechnical engineering, and has important significance for ensuring safe and efficient construction and production, but important information is generally ignored for fracture simulation of rock excavation, and the depth analysis of a corresponding big data program is not carried out on a calculation result, so that the problems of unpredictability and wide damage are easily caused after the engineering is really implemented.
Through the analysis, the problems and defects existing in the fracture simulation of rock mass excavation in the prior art are mainly as follows:
(1) the existing rock mass fracture numerical simulation method has no general practicability, and generally, each project needs to be built again by a method, so that the time is very long.
(2) The existing rock mass fracture simulation method usually ignores some important information, and a large data program does not carry out deep analysis on the result.
Disclosure of Invention
Aiming at the problems in the prior art, the invention aims to provide a continuous medium-based engineering rock mass fracture degradation numerical simulation method. The method greatly improves the applicability by arranging the adjustable model grids, carries out numerical simulation calculation under the fractured rock mass environment by using a numerical simulation method, considers the dynamic state of the fracture disturbed by excavation in the process of stress redistribution in the surrounding rock, ensures that the simulation process is more reasonable and rigorous, and further obtains a more real numerical simulation result.
The invention is realized in such a way, the engineering rock mass fracture degradation numerical simulation method based on the continuous medium comprises the following steps:
collecting data of rock masses related to a consolidation project, and classifying the data according to different conditions;
secondly, establishing a model by using finite element numerical simulation software FLAC3D according to the collected and sorted data;
setting numerical parameters of the applied engineering, wherein the numerical parameters comprise the limit of range conditions in the engineering by using threshold conditions in multiple aspects, such as the limit of boundary threshold values by applying gravity and stress;
step four, simulating the engineering implementation process by using finite element numerical simulation software FLAC3D, and carrying out iterative calculation of the fracture and degradation of the engineering rock mass in the simulation process;
step five, establishing a corresponding numerical simulation program, establishing a program algorithm by using a deep convolutional neural network, and inputting the operation result in the step four into the corresponding program for result analysis;
giving the calculated values in the analysis result to three-dimensional model material parameters such as rock strength, density, Poisson's ratio, cohesion, friction angle, volume weight and the like, and then carrying out fracture degradation numerical simulation on the engineering rock;
the fracture degradation numerical simulation of the engineering rock mass comprises the following steps:
performing numerical simulation on the rock mass in an uncracked state to obtain an initial stress field;
carrying out numerical simulation of dynamic cracking of the rock mass to obtain a dynamic stress field of the rock mass in the dynamic cracking process of the rock mass;
the dynamic cracking process of the rock mass refers to a continuous process that the rock mass stops diffusing from the first cracking gap to all the cracking gaps, and a group of numerical simulation results of dynamic rock mass cracking are obtained when the rock mass cracks every time dynamically;
the first step specifically comprises the following steps:
searching related data on a network by using a vertical search engine, and collecting data information of similar projects on site;
and classifying and sorting the collected data, and storing a sorting result by using an SQL database.
Further, the specific process of the data search is as follows:
carrying out sequential scanning from one end of a network data structure linear table; comparing the scanned node key words with a given fixed value, and when the scanned node key words are equal to the given fixed value, indicating that the search is successful;
when the scanned node key is not equal to the given constant value, it indicates that the search is not successful.
Further, the second step specifically includes the following steps:
setting a background grid by using FLAC3D, wherein the density of the set grid is flexibly controlled according to the data volume of an application project;
and the rock mass data related to the engineering numerical model is correspondingly sorted, so that the next operation is facilitated.
Further, the third step specifically includes the following contents:
inputting a rock mass model into the FLAC3D, setting the model in the FLAC3D as a strain softening model, and then initializing parameters according to the strain softening model and the data of the rock mass in engineering, wherein the parameters specifically comprise the following steps: young modulus, Poisson's ratio, volume force, cohesion, internal friction angle, external friction angle, tensile strength, residual Young modulus and plastic strain;
secondly, establishing a quantitative relation between the rock mass fracture degree and the residual Young modulus according to a GSI address strength index system, wherein the corresponding formula is as follows:
in the formula, ErIs the residual Young modulus sigma under the action of crack development after the tensile failure of the rock massmThe compressive strength of the rock mass is obtained by rock physical and mechanical property tests and rock mass strength estimation, GSItThe development degree of the crack generated by the tensile failure of the rock mass.
Further, the fourth step specifically includes the following steps:
assigning the position and the related neighborhood of the implementation project defined by the project as a null model;
performing iterative computation by using an explicit finite difference computation method;
checking whether the maximum unbalance value of the model is lower than a default standard value (le-5) in real time, if so, enabling the model to reach a balance state, and ending the simulation; and if the balance state is not reached, conveniently detecting all rock mass models and marking damaged model units.
Further, the step five specifically includes the following contents:
establishing an initial model, wherein a framework is a convolutional neural network, and initializing parameters of the model;
inputting the collected data into an initial model for training, and continuously optimizing parameters through continuous training of the data until the parameters are not obviously changed and stable;
and selecting engineering data outside the training sample, inputting the engineering data into the model to verify the accuracy of the model, finishing the training if the engineering data is accurate, and repeating the process if the engineering data is not accurate.
Further, the specific method for initializing the parameters includes:
establishing a model based on a data set of the model and describing data characteristics, and determining the number of singular value elements reserved in a singular value matrix by using the descending trend of the singular value matrix elements and the sum of the first N elements representing most information of the singular value matrix and the dimensionality of the data matrix, wherein the number of the singular value elements is used as an initial clustering number of an original data matrix to obtain an initial value of a model component number;
obtaining an initial clustering subset of the network traffic data set X according to the left singular matrix, thereby realizing the preliminary division of the data set into a plurality of categories;
and taking the ratio of the length of the subset in the initial cluster described by each model component to the length of the original data set as the parameter initial value of the mixing coefficient of each model component.
Further, in the first step, the classification method adopted when performing corresponding classification according to different situations includes:
obtaining a plurality of test samples from the data of the rock mass by using the weighted Euclidean distance of the characteristics to respectively form a first neighborhood and a second neighborhood of the test samples;
acquiring a plurality of different possible condition categories, and calculating the probability that any sample in a first neighborhood and a second neighborhood of a test sample belongs to each condition category;
respectively calculating the average value of the classification prediction evaluation capability indexes of each condition class to the nearest samples, and taking the average value as the local classification prediction capability of each condition class to rock mass data;
and taking the classification result and the weighted probability of the classifier corresponding to each condition category as the final classification result of the rock mass data.
Further, the probability that any sample in the first neighborhood and the second neighborhood of the test sample belongs to each case category is calculated by the following formula:
wherein, ajRepresents the probability that any sample in the first neighborhood of test samples belongs to the jth case class, J ∈ {1, 2.., J }, J being the total number of case classes, NjThe number of samples belonging to the jth case category in the first neighborhood of the test samples;
the calculation formula of the classification prediction evaluation capability index of the nearest sample is as follows:
in the formula, riAnd representing the classification prediction evaluation capability index of the ith image characteristic on the test sample, i belongs to {1, 2.
Further, collecting data of rock mass related to the consolidation project further comprises:
data cleaning is carried out on filling missing values in data of rock masses relevant to collected and sorted engineering, smooth denoising is carried out, and outliers are identified or deleted;
the data integration combines data in data sources for collecting engineering-related rock masses;
the data transformation converts collected data of engineering-related rock masses into a form suitable for data mining in a mode of smooth aggregation, data generalization, normalization and the like.
By combining all the technical schemes, the invention has the advantages and positive effects that:
(1) the method greatly increases the applicability by arranging the adjustable model grids, so that the rock mass model in each project has higher universality, and the numerical simulation calculation under the fractured rock mass environment is carried out by using the numerical simulation method, so that the problems of dynamic development of the fracture disturbed by excavation and surrounding rock Young modulus degradation caused in the stress redistribution process in the surrounding rock are considered, the numerical simulation process based on the continuous medium is more reasonable and rigorous, and a more real numerical simulation result is obtained.
(2) According to the invention, the result analysis model is established, so that the calculation result can be deeply analyzed, and the engineering efficiency and the implementation reliability are improved.
Drawings
FIG. 1 is a flow chart of a continuous medium-based engineering rock mass fracture degradation numerical simulation method provided by an embodiment of the invention.
FIG. 2 is a flow chart of a method for collecting and collating rock mass data related to engineering provided by the embodiment of the invention.
Fig. 3 is a flowchart of a model building method according to an embodiment of the present invention.
FIG. 4 is a flow chart of a calculation method for iteration of fracture and degradation of the engineered rock mass in the simulation process provided by the embodiment of the invention.
Fig. 5 is a flowchart of a result analysis method according to an embodiment of the present invention.
Fig. 6 is a schematic diagram of a deformation law of surrounding rock during working recovery according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Aiming at the problems in the prior art, the invention provides a continuous medium-based engineering rock mass fracture degradation numerical simulation method, which is described in detail below with reference to the accompanying drawings.
As shown in FIG. 1, the method for simulating the engineering rock mass fracture degradation based on the continuous medium provided by the embodiment of the invention comprises the following steps:
s101, collecting data of rock masses related to the consolidation project, and classifying the data according to different conditions;
s102, establishing a model by using finite element numerical simulation software FLAC3D according to the collected and sorted data;
s103, setting numerical parameters of the applied engineering, wherein the numerical parameters comprise the limit of range conditions in the engineering by using multi-aspect threshold conditions, such as the limit of threshold values of applied gravity and stress;
s104, simulating an engineering implementation process by using finite element numerical simulation software FLAC3D, and carrying out iterative calculation of fracture and degradation of the engineering rock mass in the simulation process;
and S105, establishing a corresponding numerical simulation program, establishing a program algorithm by using the deep convolutional neural network, and inputting the operation result in the step S104 into the corresponding program to analyze the result.
And S106, giving the calculated values in the analysis result to three-dimensional model material parameters including rock mass strength, density, Poisson' S ratio, cohesion, friction angle, volume weight and the like, and then carrying out fracture degradation numerical simulation on the engineering rock mass.
In S101 in the embodiment of the present invention, a classification method adopted when performing corresponding classification according to different situations includes:
obtaining a plurality of test samples from the data of the rock mass by using the weighted Euclidean distance of the characteristics to respectively form a first neighborhood and a second neighborhood of the test samples;
acquiring a plurality of different possible condition categories, and calculating the probability that any sample in a first neighborhood and a second neighborhood of a test sample belongs to each condition category;
respectively calculating the average value of the classification prediction evaluation capability indexes of each condition class to the nearest samples, and taking the average value as the local classification prediction capability of each condition class to rock mass data;
and taking the classification result and the weighted probability of the classifier corresponding to each condition category as the final classification result of the rock mass data.
Calculating the probability that any sample in the first neighborhood and the second neighborhood of the test sample belongs to each case category by using the following formula:
wherein, ajRepresents the probability that any sample in the first neighborhood of test samples belongs to the jth case class, J ∈ {1, 2.., J }, J being the total number of case classes, NjBelongs to the jth situation in the first neighborhood of the test sampleThe number of samples of condition categories;
the calculation formula of the classification prediction evaluation capability index of the nearest sample is as follows:
in the formula, riAnd representing the classification prediction evaluation capability index of the ith image characteristic on the test sample, i belongs to {1, 2.
As shown in fig. 2, S101 provided in the embodiment of the present invention specifically includes the following steps:
s201, searching related data on a network by using a vertical search engine, and collecting related similar engineering data information on the spot;
s202, the collected data are classified and sorted, and the sorted result is stored by using an SQL database.
The specific process of data search provided by the embodiment of the invention is as follows:
carrying out sequential scanning from one end of a network data structure linear table; comparing the scanned node key words with a given fixed value, and when the scanned node key words are equal to the given fixed value, indicating that the search is successful;
when the scanned node key is not equal to the given constant value, it indicates that the search is not successful.
The data of collecting and collating engineering related rock masses provided by the embodiment of the invention further comprises the following steps:
data cleaning is carried out on filling missing values in data of rock masses relevant to collected and sorted engineering, smooth denoising is carried out, and outliers are identified or deleted;
the data integration combines data in data sources for collecting engineering-related rock masses;
the data transformation converts collected data of engineering-related rock masses into a form suitable for data mining in a mode of smooth aggregation, data generalization, normalization and the like.
As shown in fig. 3, S102 provided in the embodiment of the present invention specifically includes the following steps:
s301, setting a background grid by using FLAC3D, wherein the density of the set grid is flexibly controlled according to the data volume of an application project;
and S302, correspondingly sorting rock mass data related to the engineering numerical model, and facilitating the next operation.
The step S103 provided in the embodiment of the present invention specifically includes the following steps:
inputting a rock mass model into the FLAC3D, setting the model in the FLAC3D as a strain softening model, and then initializing parameters according to the strain softening model and the data of the rock mass in engineering, wherein the parameters specifically comprise the following steps: young modulus, Poisson's ratio, volume force, cohesion, internal friction angle, external friction angle, tensile strength, residual Young modulus and plastic strain;
secondly, establishing a quantitative relation between the rock mass fracture degree and the residual Young modulus according to a GSI address strength index system, wherein the corresponding formula is as follows:
in the formula, ErIs the residual Young modulus sigma under the action of crack development after the tensile failure of the rock massmThe compressive strength of the rock mass is obtained by rock physical and mechanical property tests and rock mass strength estimation, GSItThe development degree of the crack generated by the tensile failure of the rock mass.
As shown in fig. 4, S104 provided in the embodiment of the present invention specifically includes the following steps:
s401, assigning the position and the related neighborhood of the implementation project defined by the project as a null model;
s402, performing iterative computation by using an explicit finite difference computation method;
s403, checking whether the maximum unbalance value of the model is lower than a default standard value (le-5) in real time, if so, enabling the model to reach a balance state, and ending the simulation; and if the balance state is not reached, conveniently detecting all rock mass models and marking damaged model units.
As shown in fig. 5, S105 provided in the embodiment of the present invention specifically includes the following steps:
s501: establishing an initial model, wherein a framework is a convolutional neural network, and initializing parameters of the model;
s502: inputting the collected data into an initial model for training, and continuously optimizing parameters through continuous training of the data until the parameters are not obviously changed and stable;
s503: and (5) selecting engineering data except the training sample, inputting the engineering data into the model to verify the accuracy of the model, finishing the training if the engineering data is accurate, and repeating the step (S502) if the engineering data is not accurate.
In step S501, the specific method for initializing the parameters includes: establishing a model based on a data set of the model and describing data characteristics, and utilizing the descending trend of elements of a singular value matrix, the sum of the first N elements to represent most information of the singular value matrix and the dimensionality of the data matrix to further determine the number of the singular value elements reserved in the singular value matrix, and using the number as an initial clustering number of an original data matrix to obtain an initial value of a model component number;
obtaining an initial clustering subset of the network traffic data set X according to the left singular matrix, thereby realizing the preliminary division of the data set into a plurality of categories;
and taking the ratio of the length of the subset in the initial cluster described by each model component to the length of the original data set as the parameter initial value of the mixing coefficient of each model component.
In S106 provided by the embodiment of the present invention, the fracture degradation numerical simulation of the engineering rock mass includes:
performing numerical simulation on the rock mass in an uncracked state to obtain an initial stress field;
carrying out numerical simulation of dynamic cracking of the rock mass to obtain a dynamic stress field of the rock mass in the dynamic cracking process of the rock mass;
the dynamic cracking process of the rock mass refers to a continuous process that the rock mass stops diffusing from the first cracking gap to all the cracking gaps, and a group of numerical simulation results of dynamic rock mass cracking are obtained when the rock mass cracks every time dynamically.
The technical solution of the present invention is further described below with reference to simulation experiments.
As shown in fig. 6, according to the above data, the problems of the roof sinking, the two sides moving in, the bottom plate bulging and the like in the complete service period of the mining roadway can be evaluated and analyzed, and secondary reinforcing support and the like can be guided. In actual engineering, roadway deformation needs to be observed in time, secondary reinforcement support is implemented at positions where surrounding rocks are loosened, broken, poor in stability and large in deformation, and reinforcement support such as single hydraulic props are assumed on a working face ahead during working face extraction, so that normal use of the extraction roadway within a service period is maintained. The numerical simulation method is used for carrying out numerical simulation calculation under the fractured rock mass environment, the problems of dynamic development of the fractures due to excavation disturbance and surrounding rock Young modulus deterioration caused in the stress redistribution process in the surrounding rock are considered, the optimal balance point is calculated, and the result is evaluated by using the deep convolution neural network.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A continuous medium-based engineering rock mass fracture degradation numerical simulation method is characterized by comprising the following steps:
collecting data of rock masses related to a consolidation project, and classifying the data according to different conditions;
secondly, establishing a model by using finite element numerical simulation software FLAC3D according to the collected and sorted data;
setting numerical parameters of the applied engineering, wherein the numerical parameters comprise the limit of range conditions in the engineering by using threshold conditions in multiple aspects, such as the limit of boundary threshold values by applying gravity and stress;
step four, simulating the engineering implementation process by using finite element numerical simulation software FLAC3D, and carrying out iterative calculation of the fracture and degradation of the engineering rock mass in the simulation process;
step five, establishing a corresponding numerical simulation program, establishing a program algorithm by using a deep convolutional neural network, and inputting the operation result in the step four into the corresponding program for result analysis;
giving the calculated numerical value of the analysis result to the three-dimensional model material parameters, such as rock strength, density, Poisson's ratio, cohesion, friction angle, volume weight and the like, and then carrying out fracture degradation numerical simulation on the engineering rock;
the fracture degradation numerical simulation of the engineering rock mass comprises the following steps:
performing numerical simulation on the rock mass in an uncracked state to obtain an initial stress field;
carrying out numerical simulation of dynamic cracking of the rock mass to obtain a dynamic stress field of the rock mass in the dynamic cracking process of the rock mass;
the dynamic cracking process of the rock mass refers to a continuous process that the rock mass stops diffusing from the first cracking gap to all the cracking gaps, and a group of numerical simulation results of dynamic rock mass cracking are obtained when the rock mass cracks every time dynamically;
the first step specifically comprises the following steps:
searching related data on a network by using a vertical search engine, and collecting data information of similar projects on site;
and classifying and sorting the collected data, and storing a sorting result by using an SQL database.
2. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 1, wherein the specific process of data search is as follows:
carrying out sequential scanning from one end of a network data structure linear table; comparing the scanned node key words with a given fixed value, and when the scanned node key words are equal to the given fixed value, indicating that the search is successful;
when the node key is scanned to be not equal to the given constant value, the search is not successful.
3. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 1, wherein the second step specifically comprises the following steps:
setting a background grid by using FLAC3D, wherein the density of the set grid is flexibly controlled according to the data volume of an application project;
and the rock mass data related to the engineering numerical model is correspondingly sorted, so that the next operation is facilitated.
4. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 1, wherein the third step specifically comprises the following steps:
inputting a rock mass model into the FLAC3D, setting the model in the FLAC3D as a strain softening model, and then initializing parameters according to the strain softening model and the data of the rock mass in engineering, wherein the parameters specifically comprise the following steps: young modulus, Poisson's ratio, volume force, cohesion, internal friction angle, external friction angle, tensile strength, residual Young modulus and plastic strain;
secondly, establishing a quantitative relation between the rock mass fracture degree and the residual Young modulus according to a GSI address strength index system, wherein the corresponding formula is as follows:
in the formula, ErIs the residual Young modulus sigma under the action of crack development after the tensile failure of the rock massmThe compressive strength of rock mass is obtained by rock physical and mechanical property test and rock mass strength estimation, GSItThe development degree of the crack generated by the tensile failure of the rock mass.
5. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 1, wherein the fourth step specifically comprises the following steps:
assigning the position and the related neighborhood of the implementation project defined by the project as a null model;
performing iterative computation by using an explicit finite difference computation method;
checking whether the maximum unbalance value of the model is lower than a default standard value (le-5) in real time, if so, enabling the model to reach a balance state, and ending the simulation; and if the balance state is not reached, conveniently detecting all rock mass models and marking damaged model units.
6. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 1, wherein the fifth step specifically comprises the following steps:
establishing an initial model, wherein a framework is a convolutional neural network, and initializing parameters of the model;
inputting the collected data into an initial model for training, and continuously optimizing parameters through continuous training of the data until the parameters are not obviously changed and stable;
and selecting engineering data outside the training sample and inputting the data into the model to verify the accuracy of the model, finishing the training if the data is accurate, and repeating the process if the data is not accurate.
7. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 6, wherein the specific parameter initialization method comprises the following steps:
establishing a model based on a data set of the model and describing data characteristics, and determining the number of singular value elements reserved in a singular value matrix by using the descending trend of the singular value matrix elements and the sum of the first N elements representing most information of the singular value matrix and the dimensionality of the data matrix, wherein the number of the singular value elements is used as an initial clustering number of an original data matrix to obtain an initial value of a model component number;
obtaining an initial clustering subset of the network traffic data set X according to the left singular matrix, thereby realizing the preliminary division of the data set into a plurality of categories;
and taking the ratio of the length of the subset in the initial cluster described by each model component to the length of the original data set as the parameter initial value of the mixing coefficient of each model component.
8. The continuous medium-based engineering rock mass fracture degradation numerical simulation method of claim 1, wherein in the step one, the classification method adopted when performing corresponding classification according to different situations comprises the following steps:
obtaining a plurality of test samples from the data of the rock mass by using the weighted Euclidean distance of the characteristics to respectively form a first neighborhood and a second neighborhood of the test samples;
acquiring a plurality of different possible condition categories, and calculating the probability that any sample in a first neighborhood and a second neighborhood of a test sample belongs to each condition category;
respectively calculating the average value of the classification prediction evaluation capability indexes of each condition class to the nearest samples, and taking the average value as the local classification prediction capability of each condition class to rock mass data;
and taking the classification result and the weighted probability after weighting according to the classifiers corresponding to the categories of all the conditions as the final classification result of the rock mass data.
9. The continuous medium-based numerical simulation method for fracture and degradation of an engineered rock mass according to claim 8, wherein the probability that any one of the first neighborhood and the second neighborhood of the test sample belongs to each case category is calculated by using the following formula:
wherein, ajRepresents the probability that any sample in the first neighborhood of test samples belongs to the jth case class, J ∈ {1, 2.., J }, J being the total number of case classes, NjThe number of samples belonging to the jth case category in the first neighborhood of the test samples;
the calculation formula of the classification prediction evaluation capability index of the nearest sample is as follows:
in the formula, riAnd representing the classification prediction evaluation capability index of the ith image characteristic on the test sample, i belongs to {1, 2.
10. The continuous medium-based numerical simulation method for fracture and degradation of an engineered rock mass according to claim 1, wherein the collecting data of the associated rock mass for consolidation further comprises:
data cleaning is carried out on filling missing values in data of rock masses relevant to collected and sorted engineering, smooth denoising is carried out, and outliers are identified or deleted;
the data integration combines data in data sources for collecting engineering-related rock masses;
the data transformation converts collected data of engineering-related rock masses into a form suitable for data mining in a mode of smooth aggregation, data generalization, normalization and the like.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110885245.7A CN113779665A (en) | 2021-08-03 | 2021-08-03 | Engineering rock mass fracture degradation numerical simulation method based on continuous medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110885245.7A CN113779665A (en) | 2021-08-03 | 2021-08-03 | Engineering rock mass fracture degradation numerical simulation method based on continuous medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113779665A true CN113779665A (en) | 2021-12-10 |
Family
ID=78836645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110885245.7A Withdrawn CN113779665A (en) | 2021-08-03 | 2021-08-03 | Engineering rock mass fracture degradation numerical simulation method based on continuous medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113779665A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115048884A (en) * | 2022-08-16 | 2022-09-13 | 武九铁路客运专线湖北有限责任公司 | Rock damage fracture datamation recording method under seepage stress coupling effect |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550441A (en) * | 2015-12-12 | 2016-05-04 | 山东科技大学 | Engineering rock mass fracture degradation numerical simulation method based on continuous medium |
| JP2016160161A (en) * | 2015-03-04 | 2016-09-05 | 太平洋セメント株式会社 | Hydraulic composition and heat resistant structure |
| CN106844845A (en) * | 2016-12-15 | 2017-06-13 | 长江水利委员会长江科学院 | Consider that the three-dimensional season cracking model building method of effect is contributed in bending |
| CN106989968A (en) * | 2017-03-27 | 2017-07-28 | 山东大学 | The producing device and method of a kind of Seepage-stress coupling experiment three-dimensional fracture rocks test specimen |
-
2021
- 2021-08-03 CN CN202110885245.7A patent/CN113779665A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016160161A (en) * | 2015-03-04 | 2016-09-05 | 太平洋セメント株式会社 | Hydraulic composition and heat resistant structure |
| CN105550441A (en) * | 2015-12-12 | 2016-05-04 | 山东科技大学 | Engineering rock mass fracture degradation numerical simulation method based on continuous medium |
| CN106844845A (en) * | 2016-12-15 | 2017-06-13 | 长江水利委员会长江科学院 | Consider that the three-dimensional season cracking model building method of effect is contributed in bending |
| CN106989968A (en) * | 2017-03-27 | 2017-07-28 | 山东大学 | The producing device and method of a kind of Seepage-stress coupling experiment three-dimensional fracture rocks test specimen |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115048884A (en) * | 2022-08-16 | 2022-09-13 | 武九铁路客运专线湖北有限责任公司 | Rock damage fracture datamation recording method under seepage stress coupling effect |
| CN115048884B (en) * | 2022-08-16 | 2023-02-28 | 武九铁路客运专线湖北有限责任公司 | Rock damage fracture datamation recording method under seepage stress coupling effect |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107153843B (en) | Ground settlement prediction system and method based on support vector machine | |
| CN108287808A (en) | A kind of general dynamic tracing sequential sampling method of structure-oriented fail-safe analysis | |
| US20210390231A1 (en) | Method for Rapidly Acquiring Multi-Field Response of Mining-induced Coal Rock | |
| Javadi et al. | Neural network for constitutive modelling in finite element analysis | |
| CN109241627B (en) | Probability grading dynamic support method and device for automatically designing support scheme | |
| Kirts et al. | Soil-compressibility prediction models using machine learning | |
| CN104572449A (en) | Automatic test method based on case library | |
| García-Macías et al. | Metamodel-based pattern recognition approach for real-time identification of earthquake-induced damage in historic masonry structures | |
| CN105183796A (en) | Distributed link prediction method based on clustering | |
| CN103617147A (en) | Method for identifying mine water-inrush source | |
| CN103345566A (en) | Geochemical exploration anomaly recognizing and evaluating method based on contents contained in geology | |
| CN112668802A (en) | Construction carbon emission prediction method based on design parameters | |
| CN111914943B (en) | Information vector machine method and device for comprehensively judging stability of dumping type karst dangerous rock | |
| CN116384837B (en) | Information processing method, system and inspector terminal for engineering quality inspection | |
| CN116090825A (en) | Dangerous waste outlet risk early warning method based on K-means clustering and neural network | |
| CN113779665A (en) | Engineering rock mass fracture degradation numerical simulation method based on continuous medium | |
| WO2019091134A1 (en) | Method for predicting water-rich levels of sandstone in coal seam roof | |
| CN112365054A (en) | Comprehensive grading prediction method for deep well roadway surrounding rock | |
| Zetterlund et al. | Framework for value of information analysis in rock mass characterization for grouting purposes | |
| CN111767546B (en) | Deep learning-based input structure inference method and device | |
| CN103898890A (en) | Soil layer quantization layering method based on double-bridge static sounding data of BP neural network | |
| CN115324549A (en) | Comprehensive evaluation method of marlite fractured dessert area based on logging information | |
| CN115855686A (en) | Method and device for measuring and calculating shear strength of cement mixing pile composite foundation | |
| CN115018161A (en) | Intelligent rock burst prediction method based on African bald eagle optimization random forest model | |
| CN114462323A (en) | Oil reservoir flow field characterization method based on multi-attribute field fusion |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211210 |