CN117386355A - Method for predicting high-strength concrete well wall damage in deep buried soil layer - Google Patents
Method for predicting high-strength concrete well wall damage in deep buried soil layer Download PDFInfo
- Publication number
- CN117386355A CN117386355A CN202311160547.3A CN202311160547A CN117386355A CN 117386355 A CN117386355 A CN 117386355A CN 202311160547 A CN202311160547 A CN 202311160547A CN 117386355 A CN117386355 A CN 117386355A
- Authority
- CN
- China
- Prior art keywords
- concrete
- well wall
- damage
- soil layer
- data
- 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- Withdrawn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
- 230000006378 damage Effects 0.000 title claims abstract description 91
- 239000002689 soil Substances 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 title claims abstract description 43
- 239000011372 high-strength concrete Substances 0.000 title claims abstract description 22
- 239000004567 concrete Substances 0.000 claims abstract description 147
- 239000002245 particle Substances 0.000 claims abstract description 50
- 230000008859 change Effects 0.000 claims abstract description 26
- 238000005516 engineering process Methods 0.000 claims abstract description 18
- 238000012423 maintenance Methods 0.000 claims abstract description 17
- 230000008439 repair process Effects 0.000 claims abstract description 15
- 238000011439 discrete element method Methods 0.000 claims abstract description 14
- 230000007613 environmental effect Effects 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 8
- 238000013135 deep learning Methods 0.000 claims abstract description 8
- 239000000203 mixture Substances 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 29
- 238000012544 monitoring process Methods 0.000 claims description 22
- 238000013527 convolutional neural network Methods 0.000 claims description 14
- 238000012795 verification Methods 0.000 claims description 14
- 238000004088 simulation Methods 0.000 claims description 13
- 238000006073 displacement reaction Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 8
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 239000004568 cement Substances 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000011161 development Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 239000000523 sample Substances 0.000 claims description 6
- 239000011800 void material Substances 0.000 claims description 6
- 230000007774 longterm Effects 0.000 claims description 5
- 239000011435 rock Substances 0.000 claims description 5
- 230000006399 behavior Effects 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 238000001069 Raman spectroscopy Methods 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000007596 consolidation process Methods 0.000 claims description 3
- 238000013016 damping Methods 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 239000003673 groundwater Substances 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012913 prioritisation Methods 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 238000002310 reflectometry Methods 0.000 claims description 3
- 230000002787 reinforcement Effects 0.000 claims description 3
- 238000005204 segregation Methods 0.000 claims description 3
- 230000006641 stabilisation Effects 0.000 claims description 3
- 238000011105 stabilization Methods 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims 1
- 238000002474 experimental method Methods 0.000 description 11
- 238000010200 validation analysis Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 4
- 238000001237 Raman spectrum Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000945 filler Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000011236 particulate material Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
- E21B49/006—Measuring wall stresses in the borehole
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B2200/00—Special features related to earth drilling for obtaining oil, gas or water
- E21B2200/20—Computer models or simulations, e.g. for reservoirs under production, drill bits
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B2200/00—Special features related to earth drilling for obtaining oil, gas or water
- E21B2200/22—Fuzzy logic, artificial intelligence, neural networks or the like
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Geochemistry & Mineralogy (AREA)
- Fluid Mechanics (AREA)
- Environmental & Geological Engineering (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to the technical field of concrete well wall prediction, in particular to a method for predicting high-strength concrete well wall damage in a deep buried soil layer, which comprises the following steps: s1: collecting multidimensional data associated with the concrete borehole wall, the data including temperature, humidity, concrete thickness, concrete composition, stress, strain, and external environmental factors; s2: simulating the damage condition of the concrete well wall by combining the collected data; s3: simulating the microscopic structural change of the concrete by using a particle flow discrete element method so as to predict the damage progress of the concrete well wall under different environmental conditions; s4: the soil layer below the concrete well wall is monitored in real time by combining with a geological radar technology; s5: analyzing data and predicting the damage trend of the concrete well wall through a deep learning algorithm; s6: and providing maintenance and repair suggestions according to the prediction result. The invention can monitor the conditions of the concrete well wall and the soil layer in real time and accurately predict the potential risk of the concrete well wall and the soil layer.
Description
Technical Field
The invention relates to the technical field of concrete well wall prediction, in particular to a method for predicting high-strength concrete well wall damage in a deep buried soil layer.
Background
The well shafts are used as throat channels of mining wells, along with the increasing of the quantity of deep and large well shafts, the well shafts are often under the conditions of high ground pressure and high water pressure, the well walls of deep and thick surface soil layers are designed by adopting the principles of resisting, letting and subtracting, the measure of resisting is most direct, namely, the high-strength concrete well walls are used as core supporting structures, as is well known, the brittleness of the high-strength concrete is a brittle material with lower tensile strength, the brittleness of the brittle material is increased along with the increasing of strength grade, the brittle failure of the high-strength concrete well walls is very easy to occur, the brittle failure of the high-strength concrete well walls is similar to the rock explosion phenomena of tunnels, underground chambers and the like, and once the brittle failure of the high-strength concrete well walls occurs, the sudden cement flooding disaster is caused, and huge personnel and property losses are caused.
Traditional concrete well wall damage detection and prediction methods mainly depend on regular physical inspection and basic instrument testing, however, considering the special environment of a deep buried soil layer, the methods often cannot provide real-time and accurate data, and the potential damage risk is more difficult to predict, so with the application of advanced technologies such as geological radar technology, sensor technology, spectral analysis and the like, more detailed and accurate data sources are provided, and powerful support is provided for well wall state analysis.
Recently, particle flow discrete element methods have provided a powerful tool for modeling the physical behavior of particulate materials, allowing us to study structural changes of concrete on a microscopic scale, and furthermore, deep learning algorithms, particularly Convolutional Neural Networks (CNNs), have shown superior performance in many application areas, providing the possibility to process large-scale, high-dimensional data.
However, the simple dependence on technology and algorithm may cause over-fitting and instability of the model, so that it is crucial to adopt methods such as early-stop strategy to ensure generalization capability and accuracy of the model, and according to the result of model prediction, a targeted suggestion can be provided for maintenance and repair of the concrete well wall, so that long-term stability and safety of the concrete well wall are ensured.
Disclosure of Invention
Based on the purpose, the invention provides a method for predicting the damage of the high-strength concrete well wall in the deep buried soil layer.
The method for predicting the damage of the high-strength concrete well wall in the deep buried soil layer comprises the following steps:
s1: collecting multidimensional data associated with the concrete borehole wall, the data including temperature, humidity, concrete thickness, concrete composition, stress, strain, and external environmental factors;
s2: simulating the damage condition of the concrete well wall by combining the collected data;
s3: simulating the microscopic structural change of the concrete by using a particle flow discrete element method so as to predict the damage progress of the concrete well wall under different environmental conditions;
s4: the soil layer below the concrete well wall is monitored in real time by combining with a geological radar technology so as to predict the displacement or damage of the concrete well wall caused by geological change;
s5: analyzing data and predicting the damage trend of the concrete well wall through a deep learning algorithm;
s6: and providing maintenance and repair suggestions according to the prediction result.
Further, the specific way of collecting the multidimensional data related to the concrete well wall in the step S1 includes:
s11: collecting temperature, humidity and stress data in real time by using a miniature wireless sensor embedded in concrete;
s12: monitoring the concrete composition in real time by adopting a Raman spectrometer or an infrared spectrometer, so as to identify and track the chemical reaction of the cement gel and the existence of harmful chemical substances in the concrete;
s13: measuring the thickness of the concrete in real time by using a portable digital thickness gauge;
s14: and acquiring the related data of the soil type, the hierarchical structure and the groundwater flow condition around the concrete well wall by using geological detection equipment.
Further, the mode of simulating the damage condition of the concrete well wall in the step S2 specifically includes:
s21: preprocessing the collected data, wherein the preprocessing mode comprises filtering, denoising and normalizing so as to ensure the quality and consistency of the data;
s22: using finite element analysis software to establish a three-dimensional model of the concrete well wall based on the preprocessed data;
s23: inputting temperature, humidity, strain and other environmental parameters into the model, and simulating stress distribution and crack propagation in the concrete under the influence of the variables;
s24: and analyzing the damage progress from a microscale to a macroscale by using a multiscale simulation technology, and predicting the potential damage position and degree.
Further, the three-dimensional model of the concrete well wall adopts a nonlinear elastic plastic theory, and based on the characteristics of the concrete, the model uses the combination of a Drucker-Prager plastic criterion and a damage plastic theory to describe the nonlinear response of the concrete, and a specific model formula is as follows:
σ=D(∈)·∈+α·p
where σ is stress, e is strain, D (e) is a matrix of stiffness of the material in dependence of the damage variable, p is a pressure component of plastic flow, and α is a parameter related to the hardening behaviour of the concrete, in the model the formation and propagation of cracks is described based on the energy release rate and the corresponding crack propagation criterion.
Further, the specific step of simulating the change of the microstructure of the concrete by applying the particle flow discrete element method in the step S3 comprises the following steps:
s31: in the initialization model, the microscopic structure of the concrete walls is discretized into several or millions of particles, where each particle p i Can represent aggregate, cement matrix or void in concrete, with its particle position x i And velocity v i ;
S32: each particle p i When subjected to external environmental influences, the forces therebetween are given by the following formula: f (F) ij =k n δ ij τ ij +k t δ ij τ ij +γ(v j -v i )·n ij Wherein F is ij Is the interaction force between particles i and j, k n And k t Elastic constants in normal and shear directions, respectively, delta ij Is the distance between two overlapped particles, n ij Is the unit normal vector between two particles, τ ij Is the relative displacement in tangential direction, and γ is the damping coefficient;
s33: when the relative movement among the particles causes local damage or crack expansion, acting force among the particles can be properly adjusted to simulate the development process of the microscopic damage;
s34: evaluating changes in the microstructure of the concrete by monitoring movement of the particles and changes in the force between the particles, including crack formation, propagation and particle segregation;
s35: and comparing the model parameters with actual observation data by using a statistical method, and timely adjusting the model parameters to improve the simulation accuracy.
Further, the specific way of carrying out real-time monitoring on the soil layer below the concrete well wall by combining the geological radar technology in the step S4 is as follows:
s41: depth detection is performed by using a high-resolution geological radar, and a continuous section view of a soil layer below a well wall is generated, wherein the radar can be used for capturing fine structures, cavities, rock stratum fracture or other potential discontinuities of the soil layer;
s42: calculating and analyzing the dielectric constant, density and saturation of the soil layer according to the intensity and time delay of the reflected wave generated by the geological radar, and further judging the stability and possible change trend of the soil layer;
s43: monitoring the change and movement of the soil layer structure in real time by comparing geological radar data in a continuous time period;
s44: and predicting the potential damage of the concrete well wall by combining a particle flow discrete element method model according to the soil layer change predicted by radar data and the calculated stress or displacement condition of the concrete well wall.
Further, the specific way of calculating and analyzing the dielectric constant, density and saturation of the soil layer in the step S42 is as follows:
s421: using time delay and intensity of reflected wave returned by geological radar and using time domain reflectometry technique to estimate dielectric constant epsilon of soil layer r The formula is:wherein c is the speed of light in vacuum, v is the speed of the wave in the earth, calculated as v=d/Δt, where d is the distance between the radar probes;
s422: by analyzing the attenuation degree alpha and the reflection coefficient R of the geological radar signal, the known or estimated dielectric constant of the soil layer is combined, and the formula is expressed as follows: ρ=f (ε) r α, R), where ρ is the density of the soil layer and f is an empirical relationship;
s423: the dielectric constant, density and saturation data of the soil layer are combined, and prediction is carried out by using a one-dimensional consolidation theory of Terzaghi, wherein the specific formula is expressed as follows:
Cc=f(ε r ,ρ)
e=g(ρ,ε r )
τ=h(ε r ,ρ,e)
wherein Cc is the compressibility of the soil layer, e is the void ratio, τ is the shear strength, f, g, and h are empirical or experimental based relationships;
s424: and comparing the soil layer parameter change conditions in the continuous time period, and providing data support for decision making by combining the Terzaghi prediction so as to predict the long-term stability of the soil layer and the potential risk of the concrete well wall.
Further, in the step S5, the specific steps of analyzing the data and predicting the damage trend of the concrete well wall by using the deep learning algorithm include the following steps:
s51: utilizing a deep neural network, wherein the deep neural network is a recurrent neural network to learn a space-time pattern in the data and identify the characteristics possibly causing the damage of the concrete well wall;
s52: constructing a standardized data set by utilizing the multidimensional data collected in the earlier stage;
s53: labeling the data set according to historical concrete well wall damage examples, wherein the examples comprise width and depth of cracks or failure positions and time of concrete;
s54: using convolutional neural networks as the deep learning model of choice, specifically, for each input data point x, its output y is represented by a series of convolutional layers: y=f (W x+b), where W is the convolution kernel weight, b is the bias, x represents the convolution operation, and f is the activation function ReLU;
s55: training a convolutional neural network model by using data with labels through cross verification, and adopting an early-stop strategy to prevent overfitting so as to ensure generalization of the model;
s56: and inputting real-time monitoring data by using the trained convolutional neural network model, and predicting the damage trend of the concrete well wall and the specific position where damage will occur.
Further, the specific way of preventing the overfitting by adopting the early-stop strategy in the step S55 is as follows:
s551: in the model training process, a part of data is divided into a verification set in addition to a main training data set;
s552: training the model by using training data after each training period is finished, and evaluating the performance of the model by using a verification set;
s553: comparing the performance of the verification set of continuous training periods, and considering that the model starts to be over-fitted with training data when the performance is not obviously improved or starts to be reduced;
s554: once the model over-fitting is detected, training is stopped immediately, and the weights of the model are rolled back to the state of the best performance before;
s555: and finally selecting the model with the best performance on the verification set as a final model.
Further, the maintenance and repair advice provided according to the prediction result in the step S6 specifically includes:
s61: analyzing the predicted damage position and trend of the concrete well wall, and determining the severity degree and the development trend of the damage;
s62: providing routine maintenance recommendations for areas predicted to be slightly damaged;
s63: providing special repair suggestions for areas predicted to be moderately or severely damaged;
s64: providing suggestions of integral reinforcement or soil stabilization measures for the concrete well wall by combining the stability of the surrounding soil layer and potential risks;
s65: prioritization is provided for maintenance and repair activities over a period of time in the future based on the output of the predictive model.
The invention has the beneficial effects that:
according to the invention, through the latest progress of wireless sensors, geological radar technology, raman spectrum and infrared spectrometer, the technology can realize high-frequency and high-precision real-time monitoring of the concrete well wall, compared with the traditional physical inspection and simple instrument measurement, the advanced data acquisition mode greatly enhances the detection capability of early damage and small change, thereby not only reducing the potential risk caused by delayed response, but also providing favorable opportunity for taking timely maintenance measures.
According to the invention, through a particle flow discrete element method, structural changes of concrete on a microscopic scale can be observed and simulated, early signs possibly causing macroscopic damage are captured, and the careful simulation provides a precious tool for engineers and decision makers, so that the engineers and decision makers can take corrective measures before the problem is upgraded, and the long-term stability and safety of the concrete well wall are ensured.
According to the invention, a method for processing large-scale and high-dimensional data and extracting key insight from the data is provided through the convolutional neural network model, the predicted output of the model not only can indicate the possible damage trend and position of the concrete well wall, but also can provide targeted suggestions for maintenance and repair, and the method ensures the high generalization capability of the model by combining with skills such as early-stop strategies and the like, thereby providing firm and scientific support for the management and maintenance decision of the concrete well wall.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only of the invention and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a prediction method according to an embodiment of the present invention.
Detailed Description
The present invention will be further described in detail with reference to specific embodiments in order to make the objects, technical solutions and advantages of the present invention more apparent.
It is to be noted that unless otherwise defined, technical or scientific terms used herein should be taken in a general sense as understood by one of ordinary skill in the art to which the present invention belongs. The terms "first," "second," and the like, as used herein, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
As shown in FIG. 1, the method for predicting the damage of the high-strength concrete well wall in the deep buried soil layer comprises the following steps:
s1: collecting multidimensional data associated with the concrete borehole wall, the data including temperature, humidity, concrete thickness, concrete composition, stress, strain, and external environmental factors;
s2: simulating the damage condition of the concrete well wall by combining the collected data;
s3: simulating the microscopic structural change of the concrete by using a particle flow discrete element method so as to predict the damage progress of the concrete well wall under different environmental conditions;
s4: the soil layer below the concrete well wall is monitored in real time by combining with a geological radar technology so as to predict the displacement or damage of the concrete well wall caused by geological change;
s5: analyzing data and predicting the damage trend of the concrete well wall through a deep learning algorithm;
s6: and providing maintenance and repair suggestions according to the prediction result.
The specific mode for collecting multidimensional data related to the concrete well wall in the step S1 comprises the following steps:
s11: collecting temperature, humidity and stress data in real time by using a miniature wireless sensor embedded in concrete;
s12: monitoring the concrete composition in real time by adopting a Raman spectrometer or an infrared spectrometer, so as to identify and track the chemical reaction of the cement gel and the existence of harmful chemical substances in the concrete;
s13: the thickness of the concrete is measured in real time by using a portable digital thickness gauge, and the thickness gauge has high precision and speed, and can provide accurate thickness readings at different concrete well wall positions;
s14: and using geological detection equipment, such as acoustic wave or electromagnetic wave geological radar, to obtain the related data of soil type, hierarchical structure and groundwater flow condition around the concrete well wall so as to know the potential influence of the soil on the concrete well wall.
The mode for simulating the damage condition of the concrete well wall in the S2 step specifically comprises the following steps:
s21: preprocessing the collected data, wherein the preprocessing mode comprises filtering, denoising and normalizing so as to ensure the quality and consistency of the data;
s22: using finite element analysis software, and based on the preprocessed data, establishing a three-dimensional model of the concrete well wall, wherein the model considers anisotropy, crack propagation mechanism and fine structure inside the concrete;
s23: inputting temperature, humidity, strain and other environmental parameters into the model, and simulating stress distribution and crack propagation in the concrete under the influence of the variables;
s24: the multi-scale simulation technique is used to analyze the damage progress from microscopic scale (such as the microscopic structure of concrete) to macroscopic scale (such as the structure of the whole well wall) and predict the potential damage location and extent.
The three-dimensional model of the concrete well wall adopts a nonlinear elastic plastic theory, wherein damage, crack formation and crack expansion of the concrete are considered, and based on the characteristics of the concrete, the model uses the combination of a Drucker-Prager plastic criterion and the damage plastic theory to describe the nonlinear response of the concrete, and a specific model formula is as follows:
σ=D(∈)·∈+α·p
where σ is stress, e is strain, D (e) is a matrix of stiffness of the material in dependence of the damage variable, p is a pressure component of plastic flow, and α is a parameter related to the hardening behaviour of the concrete, in the model the formation and propagation of cracks is described based on the energy release rate and the corresponding crack propagation criterion.
The concrete microstructure change simulation method by applying the particle flow discrete element method in the S3 comprises the following specific steps of:
s31: in the initialization model, the microscopic structure of the concrete walls is discretized into several or millions of particles, where each particle p i Can represent aggregate, cement matrix or void in concrete, with its particle position x i And velocity v i ;
S32: each particle p i When subjected to external environmental influences, the forces therebetween are given by the following formula: f (F) ij =k n δ ij τ ij +k t δ ij τ ij +γ(v j -v i )·n ij Wherein F is ij Is the interaction force between particles i and j, k n And k t Elastic constants in normal and shear directions, respectively, delta ij Is the distance between two overlapped particles, n ij Is the unit normal vector between two particles, τi j is Tangential relative displacement, while γ is the damping coefficient;
s33: when the relative movement among the particles causes local damage or crack expansion, acting force among the particles can be properly adjusted to simulate the development process of the microscopic damage;
s34: evaluating changes in the microstructure of the concrete by monitoring movement of the particles and changes in the force between the particles, including crack formation, propagation and particle segregation;
s35: and comparing the model parameters with actual observation data by using a statistical method, and timely adjusting the model parameters to improve the simulation accuracy.
And S4, the concrete mode of carrying out real-time monitoring on the soil layer below the concrete well wall by combining a geological radar technology is as follows:
s41: depth detection is performed by using a high-resolution geological radar, and a continuous section view of a soil layer below a well wall is generated, wherein the radar can be used for capturing fine structures, cavities, rock stratum fracture or other potential discontinuities of the soil layer;
s42: calculating and analyzing the dielectric constant, density and saturation of the soil layer according to the intensity and time delay of the reflected wave generated by the geological radar, and further judging the stability and possible change trend of the soil layer;
s43: by comparing geological radar data in a continuous time period, the change and movement of a soil layer structure, such as soil layer sinking, soil saturation change or rock stratum displacement, are monitored in real time;
s44: and predicting the potential damage of the concrete well wall by combining a particle flow discrete element method model according to the soil layer change predicted by radar data and the calculated stress or displacement condition of the concrete well wall.
The specific way of calculating and analyzing the dielectric constant, density and saturation of the soil layer in the step S42 is as follows:
s421: using time delay and intensity of reflected wave returned by geological radar and using time domain reflectometry technique to estimate dielectric constant epsilon of soil layer r The formula is:wherein c is the speed of light in vacuum, v is the speed of the wave in the earth, calculated as v=d/Δt, where d is the distance between the radar probes;
s422: by analyzing the attenuation degree alpha and the reflection coefficient R of the geological radar signal, the known or estimated dielectric constant of the soil layer is combined, and the formula is expressed as follows: ρ=f (ε) r α, R), where ρ is the density of the soil layer and f is an empirical relationship;
s423: the dielectric constant, density and saturation data of the soil layer are combined, and prediction is carried out by using a one-dimensional consolidation theory of Terzaghi, wherein the specific formula is expressed as follows:
Cc=f(ε r ,ρ)
e=g(ρ,ε r )
τ=h(ε r ,ρ,e)
wherein Cc is the compressibility of the soil layer, e is the void ratio, τ is the shear strength, f, g, and h are empirical or experimental based relationships;
s424: and comparing the soil layer parameter change conditions in the continuous time period, and providing data support for decision making by combining the Terzaghi prediction so as to predict the long-term stability of the soil layer and the potential risk of the concrete well wall.
And S5, analyzing the data and predicting the damage trend of the concrete well wall by using a deep learning algorithm, wherein the concrete well wall damage trend prediction method specifically comprises the following steps of:
s51: utilizing a deep neural network, which is a Recurrent Neural Network (RNN), to learn a space-time pattern in the data and identify features that may cause damage to the concrete borehole wall;
s52: constructing a standardized data set by utilizing multidimensional data collected in the earlier stage, such as physical and chemical properties of concrete, and dielectric constants, densities and saturation of soil layers;
s53: labeling the data set according to historical concrete well wall damage examples, wherein the examples comprise width and depth of cracks or failure positions and time of concrete;
s54: using convolutional neural networks as the deep learning model of choice, specifically, for each input data point x, its output y is represented by a series of convolutional layers: y=f (W x+b), where W is the convolution kernel weight, b is the bias, x represents the convolution operation, and f is the activation function ReLU;
s55: training a convolutional neural network model by using data with labels through cross verification, and adopting an early-stop strategy to prevent overfitting so as to ensure generalization of the model;
s56: and inputting real-time monitoring data by using the trained convolutional neural network model, and predicting the damage trend of the concrete well wall and the specific position where damage will occur.
The specific mode of preventing overfitting by adopting the early-stop strategy in the step S55 is as follows:
s551: in the model training process, a part of data is divided into a verification set in addition to a main training data set;
s552: after each training period is finished, training the model by using training data, and evaluating the performance of the model by using a verification set, wherein the main monitored index can be a loss function value or other related indexes on the verification set;
s553: comparing the performance of the validation set for consecutive training periods (e.g., 5 or 10), when there is no significant improvement in performance or a start to decline, the model is considered to start to over-fit the training data;
s554: once the model over-fitting is detected, training is stopped immediately, and the weights of the model are rolled back to the state of the best performance before;
s555: the model that performs best on the validation set is ultimately chosen as the final model, rather than simply selecting the model that has undergone the most training periods.
The maintenance and repair suggestions provided according to the prediction result in the step S6 specifically comprise:
s61: analyzing the predicted damage position and trend of the concrete well wall, and determining the severity degree and the development trend of the damage;
s62: for areas predicted to be slightly damaged, conventional maintenance recommendations are provided, such as periodic inspection, cleaning or coating with protective materials, etc.;
s63: for areas predicted to be moderately or severely damaged, specialized repair suggestions are provided, such as using specialized fillers, reinforcing or replacing damaged portions, etc.;
s64: providing suggestions of integral reinforcement or soil stabilization measures for the concrete well wall by combining the stability of the surrounding soil layer and potential risks;
s65: prioritization is provided for maintenance and repair activities over a period of time in the future based on the output of the predictive model to ensure that the most urgent and critical problems are handled first.
To verify the effect of the present invention, the following experiments were performed
Experiment 1: real-time monitoring effect test
The purpose of the experiment is as follows: and verifying the real-time performance and accuracy of the wireless sensor, the Raman spectrum and the infrared spectrometer in concrete well wall monitoring.
The specific method comprises the following steps:
setting 5 known damage points in a known concrete well wall structure;
monitoring by a wireless sensor, a Raman spectrum and an infrared spectrometer;
comparing the known damage points with the monitoring results, and evaluating the accuracy;
data results:
knowing the location of the damage point: [5m,10m,15m,20m,25m ];
monitoring the result position in real time: [5.1m,9.9m,15.3m,20.2m,24.8m ];
error rate: less than or equal to 2 percent.
Experiment 2: mesoscopic structural change simulation test
The purpose of the experiment is as follows: and verifying the simulation effect of the particle flow discrete element method on the change of the concrete microstructure.
The specific method comprises the following steps:
selecting a concrete sample with the size of 10cm x10cm, and performing a controlled destruction experiment;
simulating the sample by using a particle flow discrete element method;
comparing the simulation result with the experimental result, and evaluating the similarity and accuracy of the simulation result;
data results:
control of fracture depth: 3cm;
simulation of crack depth: 2.9cm;
error rate: less than or equal to 3.5 percent.
Experiment 3: deep learning predictive model validation
The purpose of the experiment is as follows: and verifying the accuracy of the convolutional neural network in predicting the damage trend of the concrete well wall.
The specific method comprises the following steps:
collecting historical damage data and corresponding monitoring data of a large number of concrete well walls;
a convolutional neural network model was trained using 70% of the data, 30% for validation;
comparing the damage trend predicted by the model with the damage condition actually occurring, and calculating the prediction accuracy;
data results:
training data sample number: 7000;
verifying the number of data samples: 3000;
prediction accuracy: 91%.
Experiment 4: soil layer stability prediction for geological radar technology
The purpose of the experiment is as follows: verifying the accuracy of geological radar technology on soil layer stability and potential risk prediction of a concrete well wall;
the specific method comprises the following steps:
soil disturbance controlled at a specific test site;
monitoring by using a geological radar technology, and predicting the stability of a soil layer and the potential risk of a concrete well wall;
comparing the prediction result with actual observation data, and evaluating the accuracy and reliability of prediction;
data results:
soil layer disturbance depth: 2m;
predicting disturbance depth: 1.9m;
error rate: less than or equal to 5 percent,
the above experiments demonstrate the validation process of the present invention through detailed data and methods, thereby providing powerful support for its practical application.
The present invention is intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent replacement, improvement, etc. of the present invention should be included in the scope of the present invention.
Claims (10)
1. The method for predicting the damage of the high-strength concrete well wall in the deep buried soil layer is characterized by comprising the following steps of:
s1: collecting multidimensional data associated with the concrete borehole wall, the data including temperature, humidity, concrete thickness, concrete composition, stress, strain, and external environmental factors;
s2: simulating the damage condition of the concrete well wall by combining the collected data;
s3: simulating the microscopic structural change of the concrete by using a particle flow discrete element method so as to predict the damage progress of the concrete well wall under different environmental conditions;
s4: the soil layer below the concrete well wall is monitored in real time by combining with a geological radar technology so as to predict the displacement or damage of the concrete well wall caused by geological change;
s5: analyzing data and predicting the damage trend of the concrete well wall through a deep learning algorithm;
s6: and providing maintenance and repair suggestions according to the prediction result.
2. The method for predicting the damage of a high-strength concrete well wall in a deep buried soil layer according to claim 1, wherein the specific manner of collecting multidimensional data related to the concrete well wall in step S1 comprises:
s11: collecting temperature, humidity and stress data in real time by using a miniature wireless sensor embedded in concrete;
s12: monitoring the concrete composition in real time by adopting a Raman spectrometer or an infrared spectrometer, so as to identify and track the chemical reaction of the cement gel and the existence of harmful chemical substances in the concrete;
s13: measuring the thickness of the concrete in real time by using a portable digital thickness gauge;
s14: and acquiring the related data of the soil type, the hierarchical structure and the groundwater flow condition around the concrete well wall by using geological detection equipment.
3. The method for predicting the damage of a high-strength concrete well wall in a deep buried soil layer according to claim 1, wherein the method for simulating the damage condition of the concrete well wall in the step S2 is specifically as follows:
s21: preprocessing the collected data, wherein the preprocessing mode comprises filtering, denoising and normalizing so as to ensure the quality and consistency of the data;
s22: using finite element analysis software to establish a three-dimensional model of the concrete well wall based on the preprocessed data;
s23: inputting temperature, humidity, strain and other environmental parameters into the model, and simulating stress distribution and crack propagation in the concrete under the influence of the variables;
s24: and analyzing the damage progress from a microscale to a macroscale by using a multiscale simulation technology, and predicting the potential damage position and degree.
4. The method for predicting the damage of a high-strength concrete well wall in a deep buried soil layer according to claim 3, wherein the three-dimensional model of the concrete well wall adopts a nonlinear elastic plastic theory, and based on the characteristics of the concrete, the model uses a combination of a Drucker-Prager plastic criterion and a damage plastic theory to describe the nonlinear response of the concrete, and a specific model formula is as follows:
σ=D(∈)·∈+α·p
where σ is stress, e is strain, D (e) is a matrix of stiffness of the material in dependence of the damage variable, p is a pressure component of plastic flow, and α is a parameter related to the hardening behaviour of the concrete, in the model the formation and propagation of cracks is described based on the energy release rate and the corresponding crack propagation criterion.
5. The method for predicting the wall damage of high-strength concrete in a deep buried soil layer according to claim 1, wherein the specific step of simulating the change of a concrete microstructure by applying a particle flow discrete element method in the step S3 comprises the following steps:
s31: in the initialization model, the microscopic structure of the concrete walls is discretized into several or millions of particles, where each particle p i Can represent aggregate, cement matrix or void in concrete, with its particle position x i And velocity v i ;
S32: each particle p i When subjected to external environmental influences, the forces therebetween are given by the following formula: f (F) ij =k n δ ij n ij +k t δ ij τ ij +γ(v j -v i )·n ij Wherein F is ij Is the interaction force between particles i and j, k n And k t Elastic constants in normal and shear directions, respectively, delta ij Is the distance between two overlapped particles, n ij Is the unit normal vector between two particles, τ ij Is the relative displacement in tangential direction, and γ is the damping coefficient;
s33: when the relative movement among the particles causes local damage or crack expansion, acting force among the particles can be properly adjusted to simulate the development process of the microscopic damage;
s34: evaluating changes in the microstructure of the concrete by monitoring movement of the particles and changes in the force between the particles, including crack formation, propagation and particle segregation;
s35: and comparing the model parameters with actual observation data by using a statistical method, and timely adjusting the model parameters to improve the simulation accuracy.
6. The method for predicting the damage of the high-strength concrete well wall in the deep buried soil layer according to claim 1, wherein the specific mode of carrying out real-time monitoring on the soil layer below the concrete well wall by combining a geological radar technology in the step S4 is as follows:
s41: depth detection is performed by using a high-resolution geological radar, and a continuous section view of a soil layer below a well wall is generated, wherein the radar can be used for capturing fine structures, cavities, rock stratum fracture or other potential discontinuities of the soil layer;
s42: calculating and analyzing the dielectric constant, density and saturation of the soil layer according to the intensity and time delay of the reflected wave generated by the geological radar, and further judging the stability and possible change trend of the soil layer;
s43: monitoring the change and movement of the soil layer structure in real time by comparing geological radar data in a continuous time period;
s44: and predicting the potential damage of the concrete well wall by combining a particle flow discrete element method model according to the soil layer change predicted by radar data and the calculated stress or displacement condition of the concrete well wall.
7. The method for predicting the wall damage of high-strength concrete in deep buried soil according to claim 6, wherein the specific way of calculating and analyzing the dielectric constant, density and saturation of the soil in step S42 is as follows:
s421: using time delay and intensity of reflected wave returned by geological radar and using time domain reflectometry technique to estimate dielectric constant epsilon of soil layer r The formula is:wherein c is the speed of light in vacuum, v is the speed of wave in soil, and the calculation methodThe formula v=d/Δt, where d is the distance between the radar probes;
s422: by analyzing the attenuation degree alpha and the reflection coefficient R of the geological radar signal, the known or estimated dielectric constant of the soil layer is combined, and the formula is expressed as follows: ρ=f (ε) r α, R), where ρ is the density of the soil layer and f is an empirical relationship;
s423: the dielectric constant, density and saturation data of the soil layer are combined, and prediction is carried out by using a one-dimensional consolidation theory of Terzaghi, wherein the specific formula is expressed as follows:
Cc=f(ε r ,ρ)
e=g(ρ,ε r )
τ=h(ε r ,ρ,e)
wherein Cc is the compressibility of the soil layer, e is the void ratio, τ is the shear strength, f, g, and h are empirical or experimental based relationships;
s424: and comparing the soil layer parameter change conditions in the continuous time period, and providing data support for decision making by combining the Terzaghi prediction so as to predict the long-term stability of the soil layer and the potential risk of the concrete well wall.
8. The method for predicting the damage of the high-strength concrete well wall in the deep buried soil layer according to claim 1, wherein the step S5 is characterized by comprising the following steps of:
s51: utilizing a deep neural network, wherein the deep neural network is a recurrent neural network to learn a space-time pattern in the data and identify the characteristics possibly causing the damage of the concrete well wall;
s52: constructing a standardized data set by utilizing the multidimensional data collected in the earlier stage;
s53: labeling the data set according to historical concrete well wall damage examples, wherein the examples comprise width and depth of cracks or failure positions and time of concrete;
s54: using convolutional neural networks as the deep learning model of choice, specifically, for each input data point x, its output y is represented by a series of convolutional layers: y=f (W x+b), where W is the convolution kernel weight, b is the bias, x represents the convolution operation, and f is the activation function ReLU;
s55: training a convolutional neural network model by using data with labels through cross verification, and adopting an early-stop strategy to prevent overfitting so as to ensure generalization of the model;
s56: and inputting real-time monitoring data by using the trained convolutional neural network model, and predicting the damage trend of the concrete well wall and the specific position where damage will occur.
9. The method for predicting the damage of a high-strength concrete well wall in a deep buried soil layer according to claim 8, wherein the specific method for preventing the overfitting by adopting the early-stop strategy in the step S55 is as follows:
s551: in the model training process, a part of data is divided into a verification set in addition to a main training data set;
s552: training the model by using training data after each training period is finished, and evaluating the performance of the model by using a verification set;
s553: comparing the performance of the verification set of continuous training periods, and considering that the model starts to be over-fitted with training data when the performance is not obviously improved or starts to be reduced;
s554: once the model over-fitting is detected, training is stopped immediately, and the weights of the model are rolled back to the state of the best performance before;
s555: and finally selecting the model with the best performance on the verification set as a final model.
10. The method for predicting the damage of a high-strength concrete well wall in a deep buried soil layer according to claim 1, wherein the maintenance and repair advice provided according to the prediction result in step S6 specifically comprises:
s61: analyzing the predicted damage position and trend of the concrete well wall, and determining the severity degree and the development trend of the damage;
s62: providing routine maintenance recommendations for areas predicted to be slightly damaged;
s63: providing special repair suggestions for areas predicted to be moderately or severely damaged;
s64: providing suggestions of integral reinforcement or soil stabilization measures for the concrete well wall by combining the stability of the surrounding soil layer and potential risks;
s65: prioritization is provided for maintenance and repair activities over a period of time in the future based on the output of the predictive model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311160547.3A CN117386355A (en) | 2023-09-08 | 2023-09-08 | Method for predicting high-strength concrete well wall damage in deep buried soil layer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311160547.3A CN117386355A (en) | 2023-09-08 | 2023-09-08 | Method for predicting high-strength concrete well wall damage in deep buried soil layer |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117386355A true CN117386355A (en) | 2024-01-12 |
Family
ID=89438064
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311160547.3A Withdrawn CN117386355A (en) | 2023-09-08 | 2023-09-08 | Method for predicting high-strength concrete well wall damage in deep buried soil layer |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117386355A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117627077A (en) * | 2024-01-15 | 2024-03-01 | 广东大禹水利建设有限公司 | Intelligent monitoring method for permeable layer open caisson construction |
-
2023
- 2023-09-08 CN CN202311160547.3A patent/CN117386355A/en not_active Withdrawn
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117627077A (en) * | 2024-01-15 | 2024-03-01 | 广东大禹水利建设有限公司 | Intelligent monitoring method for permeable layer open caisson construction |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jayawickrema et al. | Fibre-optic sensor and deep learning-based structural health monitoring systems for civil structures: A review | |
Webb et al. | Categories of SHM deployments: Technologies and capabilities | |
Jin et al. | Study on the failure mechanism of basalts with columnar joints in the unloading process on the basis of an experimental cavity | |
Mahdiabadi et al. | Prediction of uniaxial compressive strength and modulus of elasticity in calcareous mudstones using neural networks, fuzzy systems, and regression analysis | |
Prakash et al. | Recent advancement of concrete dam health monitoring technology: A systematic literature review | |
Rashidi et al. | Shear modulus prediction of embedded pressurized salt layers and pinpointing zones at risk of casing collapse in oil and gas wells | |
CN117386355A (en) | Method for predicting high-strength concrete well wall damage in deep buried soil layer | |
Ma et al. | Calibration of the microparameters of the discrete element method using a relevance vector machine and its application to rockfill materials | |
Vasanelli et al. | Combining non-invasive techniques for reliable prediction of soft stone strength in historic masonries | |
Fedele et al. | The prediction of road cracks through acoustic signature: Extended finite element modeling and experiments | |
CN115423167A (en) | Subway deep foundation pit construction safety early warning and decision-making assisting method and system | |
Lu et al. | Hybrid machine learning for pullback force forecasting during horizontal directional drilling | |
Yan et al. | Design and implementation of a structural health monitoring system for a large sea‐crossing project with bridges and tunnel | |
Le et al. | Subsurface ground movements due to circular shaft construction | |
Liu et al. | Cracking risk analysis of face slabs in concrete face rockfill dams during the operation period | |
Wang et al. | Monitoring structural health status of asphalt pavement using intelligent sensing technology | |
Afshari et al. | Time-varying structural reliability assessment method: Application to fiber reinforced composites under repeated impact loading | |
Zhou et al. | Prospective forecast of sliding instability time using a precursory AE time series | |
He et al. | Brittle failure modes of underground powerhouses: an insight based on true triaxial compression tests | |
Dai et al. | Crack extension identification based on distributed fiber sensing measurement and optimized support vector regression | |
Liu et al. | Developing a prediction model for segment joint opening in an underwater shield tunnel | |
Guo et al. | Autonomous prediction of rock deformation in fault zones of coal roadways using supervised machine learning | |
Fattahi et al. | Application of monte carlo markov chain and GMDH neural network for estimating the behavior of suction caissons in clay | |
Rezaiee-Pajand et al. | Crack detection in concrete gravity dams using a genetic algorithm | |
Kazemian et al. | A review of bridge scour monitoring techniques and developments in vibration based scour monitoring for bridge foundations |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20240112 |