CN114219134A - Method and system for predicting ground settlement caused by shield construction - Google Patents
Method and system for predicting ground settlement caused by shield construction Download PDFInfo
- Publication number
- CN114219134A CN114219134A CN202111445389.7A CN202111445389A CN114219134A CN 114219134 A CN114219134 A CN 114219134A CN 202111445389 A CN202111445389 A CN 202111445389A CN 114219134 A CN114219134 A CN 114219134A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction
- shield
- predicting
- parameters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000010801 machine learning Methods 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 238000012360 testing method Methods 0.000 claims abstract description 22
- 238000007781 pre-processing Methods 0.000 claims abstract description 17
- 238000004519 manufacturing process Methods 0.000 claims abstract description 7
- 238000009826 distribution Methods 0.000 claims description 17
- 230000003321 amplification Effects 0.000 claims description 14
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 14
- 238000004062 sedimentation Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 7
- 239000002689 soil Substances 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 5
- 230000002194 synthesizing effect Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 12
- 230000006872 improvement Effects 0.000 description 9
- 238000012544 monitoring process Methods 0.000 description 7
- 238000011160 research Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004836 empirical method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000005641 tunneling Effects 0.000 description 2
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F16/211—Schema design and management
- G06F16/212—Schema design and management with details for data modelling support
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
Abstract
The invention discloses a method and a system for predicting ground settlement caused by shield construction, which comprises the following steps: s1: acquiring local shield case data, establishing a database, and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises amplifying the data by adopting an SMOTE algorithm; s2: inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction; s3: and evaluating the prediction model by using the indexes, and testing the prediction precision of the prediction model. The method improves the prediction capability of the ground settlement deformation caused by shield construction.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a system for predicting ground settlement caused by shield construction.
Background
With the rapid development of economic construction in China, the rapid development of underground rail transit becomes an effective way for solving traffic problems in large cities in China. In dense urban wire network planning, the construction of a newly-built tunnel inevitably disturbs soil layers, so that the ground surface sinks and the building cracks, and therefore, the control of the soil layer sinking caused by the penetration of the tunnel becomes an important research direction for the construction of urban underground engineering at present.
Common research methods for the problem of surface deformation caused by shield construction include an empirical method, a numerical method, an analytical method and the like. The most representative Peck formula in the empirical method reveals that the soil layer settlement is in transverse quasi-normal distribution through a large amount of sorting analysis on related engineering case data. Some researchers have studied factors such as synchronous grouting, secondary grouting, earth cabin pressure, shield tunneling speed and the like by using finite element software, and the results show that the factors cause less surface subsidence when the tunneling speed is higher. The traditional method is largely applied to engineering practice, but in actual engineering, because a plurality of uncertain factors are not considered, a large error still exists between the traditional method and a settlement value in the engineering practice.
In recent years, machine learning has been widely used in the field of underground engineering due to its strong nonlinear fitting capability. In deep learning, the number of samples is generally required to be sufficient, and the more the number of samples is, the better the effect of the trained model is. However, in actual prediction, the number of samples is often insufficient or the quality of the samples is not good enough, so that a large error exists in a model prediction result.
Disclosure of Invention
The invention aims to provide a method and a system for predicting ground settlement caused by shield construction, which improve the prediction capability of ground settlement deformation caused by shield construction.
In order to solve the technical problem, the invention provides a method for predicting ground settlement caused by shield construction, which comprises the following steps:
s1: acquiring local shield case data, establishing a database, and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises amplifying the data by adopting an SMOTE algorithm;
s2: inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction;
s3: and evaluating the prediction model by using the indexes, and testing the prediction precision of the prediction model.
As a further improvement of the present invention, the step S1 specifically includes the following steps:
s1.1: establishing a database, and collecting shield traversing cases of a formulated area in nearly fifteen years, wherein the data types in the database comprise tunnel parameters, stratum parameters and shield parameters;
s1.2: amplifying data by introducing an SMOTE algorithm, analyzing a few types of samples, and artificially synthesizing a new sample according to the few types of samples to be added into a data set;
s1.3: carrying out standardization processing on input data variables before KNN machine learning model training;
s1.4: the amplified data set was divided into 4 subsets, 3 subsets of which were used for training in turn, and the remaining 1 subset was used for testing.
As a further improvement of the present invention, the introduction of SMOTE algorithm to amplify data specifically comprises:
wherein x isiFor a minority class of data samples, xijIs xiIs determined by the data samples of the adjacent data,
is a new data sample.
As a further improvement of the invention, the normalization process:
wherein the content of the first and second substances,
is the mean of the initial sample data, σ is the standard deviation of the original data, x*Is a normalized value.
As a further improvement of the present invention, the step S2 specifically includes the following steps:
s2.1: the earth surface transverse subsidence tank caused by the shield is approximately in normal distribution, on the basis that the curve of the subsidence tank conforms to Gaussian distribution, the earth surface subsidence is assumed to occur under the condition of no water drainage, and the transverse earth surface subsidence above the weak stratum tunnel is described by an earth surface subsidence formula:
wherein s is the sedimentation value of any point on the ground, smaxIs the maximum value of the ground subsidence, located on the symmetrical center of the subsidence curve, y is the distance from the center of the subsidence curve to the calculated point, i is the distance from the symmetrical center of the subsidence curve to the inflection point of the curve, V1The stratum volume loss rate is shown, and D is the tunnel depth;
after analyzing a large amount of surface subsidence data and engineering data, obtaining that i is generally related to the depth D of the tunnel and the internal friction angle of the surrounding stratum, and the parameters are mainly related to engineering geological conditions, a tunnel construction method and construction technology level factors, namely selecting the cohesive force and the internal friction angle of the soil layer, the buried depth of the tunnel and the diameter of the shield as over parameters;
s2.2: selecting and adjusting the hyper-parameters, and finding out a group of most suitable parameters by a grid optimization method, wherein the grid optimization method is to combine all possible parameters, then train each group of parameters, and verify whether the result is optimal, namely find out the optimal parameter combination;
s2.3: and combining the optimal parameters with the corresponding KNN machine learning model to serve as a prediction model, and outputting the maximum earth surface settlement value caused by shield construction.
As a further improvement of the present invention, the step S2 further includes the following steps:
and predicting the test set by using the trained prediction model, and predicting the original data set and the data set after SMOTE preprocessing by using a KNN machine learning model by using a Python Scikt-learn library, wherein the test set is used for testing the prediction performance of the optimal hyper-parameter combination of each model.
As a further improvement of the present invention, the step S3 specifically includes the following steps: and (3) evaluating the model by using two indexes of root mean square error RMSE and average absolute error MAE:
RMSE：
MAE：
where n is the total number of samples, riIs the predicted sedimentation value, piIs the measured sedimentation value.
As a further improvement of the invention, the output predictive sedimentation value r is used for any group of training samplesiWith the true measured sedimentation value piCertain errors exist between the prediction models, and after the prediction models are calculated by using the calculation formulas of RMSE and MAE, the smaller the RMSE and the MAE, the better the prediction accuracy of the prediction models is.
A system for predicting ground subsidence caused by shield construction comprises:
the data processing unit is used for acquiring local shield case data, establishing a database and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises the step of amplifying the data by adopting an SMOTE algorithm;
the prediction model unit is used for inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction;
and the evaluation unit is used for evaluating the prediction model by using the indexes and testing the prediction precision of the prediction model.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the above method are implemented when the processor executes the computer program.
The invention has the beneficial effects that: starting from the machine learning angle, the invention constructs a prediction model aiming at the ground settlement caused by shield construction; by utilizing a machine learning method under the background of fully acquiring regional shield case data, a method for predicting ground settlement caused by shield construction is provided, and the accuracy of prediction of the ground settlement is obviously improved; the invention introduces the technology of synthesizing a few oversampling, can solve the defects of uneven distribution of the database and less samples, can carry out telescopic transformation on the data with different characteristic dimensions, ensures that the characteristics between different measurements have comparability, and simultaneously does not change the distribution of the original data; the KNN model based on data enhancement has the advantages of wide prediction range of settlement, small prediction error and good generalization capability, and provides an effective prediction means for bad results of surface settlement caused by shield construction.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a schematic diagram of the KNN algorithm of the present invention;
FIG. 3 is a schematic view of a tunnel floor layout according to an embodiment of the present invention;
FIG. 4 is a data distribution graph of a data set prior to amplification according to an embodiment of the present invention;
FIG. 5 is a data distribution graph after amplification of a data set according to an embodiment of the present invention;
FIG. 6 is a plot of the boxed data set prior to amplification in an embodiment of the invention;
FIG. 7 is a boxed graph after data set amplification according to an embodiment of the invention;
FIG. 8 is a graph of pre-amplification sedimentation prediction values for a data set according to an embodiment of the present invention;
FIG. 9 is a graph of sedimentation predictions after amplification of a data set according to an embodiment of the present invention.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
As shown in fig. 1, the invention provides a method for predicting ground subsidence caused by shield construction, which comprises the following steps:
s1: acquiring local shield case data, establishing a database, and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises amplifying the data by adopting an SMOTE algorithm;
s2: inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction;
s3: and evaluating the prediction model by using the indexes, and testing the prediction precision of the prediction model.
Specifically, the method comprises the steps of firstly obtaining local shield case data, preprocessing the data, then inputting an obtaining parameter, training the data by using KNN machine learning, predicting a maximum earth surface settlement value caused by the shield, finally evaluating a model by using an index, and testing the prediction precision of the model. Compared with the prior art, the method has the advantages that under the background of obtaining the shield parameters of the regional cases, the machine learning algorithm is utilized, the SMOTE algorithm which is a composite minority class oversampling technology is adopted to amplify the database, and on the basis, a KNN machine learning model is selected to carry out prediction analysis on the settlement. The preprocessed data set has obvious improvement on the prediction capability of the ground subsidence deformation caused by shield construction, has guiding significance on the deformation control of the shield construction, and provides reference for the intellectualization of a construction site.
The KNN model can capture the complex relation between the excavation stability of the underground stope and various input parameters and give more accurate classification accuracy. The KNN algorithm principle is to measure the distance between different samples and then select the nearest K neighbors for classification according to the distance, as shown in fig. 2. If a sample belongs to a certain class in most of the K nearest samples in the feature space, the sample also belongs to this class. The final predicted result is the average of the target property values of all neighboring samples.
Step S1 specifically includes:
s1.1, a database is established, and shield crossing cases in regions formulated in nearly fifteen years are collected. The cases need to be widely distributed and have certain representativeness. The data types in the database comprise geological parameters, shield parameters and the like, and the parameters are mainly selected according to the influence degree of the parameters on the settlement;
s1.2, by introducing a few oversampling technologies, namely a SMOTE data amplification method, the defects of uneven distribution of a database and less samples are overcome. Analyzing the minority samples and artificially synthesizing new samples according to the minority samples to be added into the data set;
s1.3, carrying out standardization processing on input variables before model training, and applying a calculation formula of the standardization processing to solve the problem that the training and the output of a machine learning algorithm are not facilitated due to the fact that data with different characteristics are in different orders of magnitude in a database;
and S1.4, verifying the generalization ability of the prediction model by adopting a 4-fold cross verification method in order to ensure that the prediction model has stronger prediction ability in both a training set and a test set. Before training the model, dividing the amplified data into 4 subsets, and using 3 subsets for training and using the remaining 1 subset for testing. And each group of data can be predicted once as a test set, so that the generalization of the prediction model is relatively objectively reflected.
Step S2 specifically includes:
s2.1, in the prediction of ground settlement caused by shield construction, the selection of parameters is directly related to the quality of model training. The earth surface transverse subsidence tank caused by the shield is approximately in normal distribution, and on the basis that the curve of the subsidence tank conforms to Gaussian distribution, the earth surface subsidence is supposed to occur under the condition of no water drainage, and the transverse earth surface subsidence above the weak stratum tunnel can be described by an earth surface subsidence formula. The data types in the database comprise geological parameters, shield parameters and the like, the parameters are mainly selected according to the influence degree of the parameters on the settlement, i is obtained to be generally related to the depth D of the tunnel and the internal friction angle of the surrounding stratum after a large amount of surface settlement data and engineering data are analyzed, and the parameters are mainly related to factors such as engineering geological conditions, tunnel construction methods and construction technical levels. According to the concrete case, the cohesive force and the internal friction angle of a soil layer, the buried depth of a tunnel and the diameter of a shield are selected as research parameters;
s2.2, selection of the hyper-parameters is related to model quality and the ability to infer correct results on new input data, and meanwhile, the hyper-parameters also influence the running time of the algorithm. The purpose of grid optimization is to find a group of most suitable parameters, and grid search is to combine all possible parameters, then train each group of parameters, and verify whether the result is optimal, i.e. find the optimal parameter combination.
Step S3 specifically includes:
and S3.1, predicting the test set by using the trained model. The KNN machine learning model was used to predict the raw data set and SMOTE pre-processed data set using the Python's Scikt-left library, where the test set was used to test the prediction performance of the optimal hyper-parameter combination for each model.
S3.2, outputting a predicted value r for any group of training samplesiAnd true value piThere is a certain error therebetween. Usually, the model is evaluated by two indexes, namely Root Mean Square Error (RMSE) and Mean Absolute Error (MAE), and after calculation by using a calculation formula of the RMSE and the MAE, the smaller the RMSE and the MAE are, the better the prediction accuracy of the model is.
Based on the same inventive concept, as shown in fig. 1, the system for predicting the ground settlement caused by the shield construction is provided, the principle of solving the problem is similar to the method for predicting the ground settlement caused by the shield construction, and repeated parts are not repeated.
A system for predicting ground subsidence caused by shield construction comprises:
the data processing unit is used for acquiring local shield case data, establishing a database and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises the step of amplifying the data by adopting an SMOTE algorithm;
the prediction model unit is used for inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction;
and the evaluation unit is used for evaluating the prediction model by using the indexes and testing the prediction precision of the prediction model.
The invention provides a prediction system for ground settlement caused by shield construction by utilizing a machine learning method under the background of fully acquiring regional shield case data, and the accuracy of the prediction of the ground settlement is obviously improved.
The invention has the following advantages:
(1) by introducing a few synthesis oversampling technologies, the defects of uneven distribution of the database and less samples can be overcome;
(2) the data of different feature dimensions can be subjected to telescopic transformation, so that the features of different measures are comparable, and the distribution of original data is not changed;
(3) the KNN model based on data enhancement has the advantages of wide prediction range of settlement, small prediction error and good generalization capability, and provides an effective prediction means for bad results of surface settlement caused by shield construction.
Examples
As shown in figure 3, an example of a west dam river-ternary bridge section tunnel using a 12 # line of track traffic engineering in Beijing city is taken, a right line shield tunnel is taken as a research object, the section mileage is right CSK116+ 979.16-right CSK118+348.21, a shield is driven to the west dam river direction from the direction of the ternary bridge, the construction is carried out by adopting a shield method, the diameter of a shield cutter head is 6.68m, the outer diameter of a segment is 6.4m, and the width of the segment is 1.2 m. The right shield tunnel mainly penetrates through the stratum and is sequentially divided into three layers, and geological exploration parameters are shown in a table 1. And selecting 40 monitoring points of 60 monitoring sections of the center line of the right tunnel. The ground surface settlement monitoring adopts automatic monitoring, the automatic monitoring frequency is 20-60 minutes/time, the engineering settlement monitoring is mainly carried out by a Trimble Dini03 precision electronic level, and the monitoring precision is +/-0.3 mm.
Table 1:
(1) comparison before and after distribution of SMOTE amplification data:
and selecting K similar sample points closest to the sample points by using a KNN algorithm (the nearest neighbor is taken as 5 in the research), randomly selecting the sample points from the K similar sample points, constructing a new sample point by using a SMOTE algorithm for each randomly selected sample point, and repeating the steps to synthesize the class I, class II and class III samples. A variable scatter diagram is drawn by using the pandas library in Python and a data distribution table is arranged, and FIG. 4 and FIG. 5 are data distribution diagrams before and after amplification respectively.
Fig. 6 and 7 are box diagrams of the machine learning algorithm before and after amplification, respectively, and prediction ranges of the preprocessed KNN algorithm are improved, so that prediction capability is remarkably enhanced.
The engineering is predicted by using the KNN models before and after amplification, and the predicted value and the measured value of the western-three interval settlement are compared, as shown in FIGS. 8 and 9: the MAE and the RMSE are respectively reduced from 6.21mm and 8.05mm before amplification to 3.75mm and 5.46mm after amplification, which shows that the KNN model based on the SMOTE data enhancement has good effect in practical engineering application.
Compared with the prior art, the method provided by the invention has the advantages that under the background of obtaining the shield parameters of the regional cases, the machine learning algorithm is utilized, the SMOTE algorithm which is a composite minority oversampling technology is adopted to amplify the database, and on the basis, a KNN machine learning model is selected to carry out prediction analysis on settlement. The preprocessed data set has obvious improvement on the prediction capability of the ground subsidence deformation caused by shield construction, has guiding significance on the deformation control of the shield construction, and provides reference for the intellectualization of a construction site.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (10)
1. A method for predicting ground settlement caused by shield construction is characterized by comprising the following steps: the method comprises the following steps:
s1: acquiring local shield case data, establishing a database, and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises amplifying the data by adopting an SMOTE algorithm;
s2: inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction;
s3: and evaluating the prediction model by using the indexes, and testing the prediction precision of the prediction model.
2. The method for predicting the ground subsidence caused by shield construction according to claim 1, wherein: the step S1 specifically includes the following steps:
s1.1: establishing a database, and collecting shield traversing cases of a formulated area in nearly fifteen years, wherein the data types in the database comprise tunnel parameters, stratum parameters and shield parameters;
s1.2: amplifying data by introducing an SMOTE algorithm, analyzing a few types of samples, and artificially synthesizing a new sample according to the few types of samples to be added into a data set;
s1.3: carrying out standardization processing on input data variables before KNN machine learning model training;
s1.4: the amplified data set was divided into 4 subsets, 3 subsets of which were used for training in turn, and the remaining 1 subset was used for testing.
3. The method for predicting the ground subsidence caused by shield construction according to claim 2, wherein: the introduction of the SMOTE algorithm to data amplification specifically comprises the following steps:
wherein x isiFor a minority class of data samples, xijIs xiIs determined by the data samples of the adjacent data,
is a new data sample.
4. The method for predicting the ground subsidence caused by shield construction according to claim 2, wherein: the normalization processing comprises the following steps:
wherein the content of the first and second substances,
is the mean of the initial sample data, σ is the standard deviation of the original data, x*Is a normalized value.
5. The method for predicting the ground subsidence caused by shield construction according to claim 2, wherein: the step S2 specifically includes the following steps:
s2.1: the earth surface transverse subsidence tank caused by the shield is approximately in normal distribution, on the basis that the curve of the subsidence tank conforms to Gaussian distribution, the earth surface subsidence is assumed to occur under the condition of no water drainage, and the transverse earth surface subsidence above the weak stratum tunnel is described by an earth surface subsidence formula:
wherein s is the sedimentation value of any point on the ground, smaxIs the maximum value of the ground subsidence, located on the symmetrical center of the subsidence curve, y is the distance from the center of the subsidence curve to the calculated point, i is the distance from the symmetrical center of the subsidence curve to the inflection point of the curve, V1The stratum volume loss rate is shown, and D is the tunnel depth;
after analyzing a large amount of surface subsidence data and engineering data, obtaining that i is generally related to the depth D of the tunnel and the internal friction angle of the surrounding stratum, and the parameters are mainly related to engineering geological conditions, a tunnel construction method and construction technology level factors, namely selecting the cohesive force and the internal friction angle of the soil layer, the buried depth of the tunnel and the diameter of the shield as over parameters;
s2.2: selecting and adjusting the hyper-parameters, and finding out a group of most suitable parameters by a grid optimization method, wherein the grid optimization method is to combine all possible parameters, then train each group of parameters, and verify whether the result is optimal, namely find out the optimal parameter combination;
s2.3: and combining the optimal parameters with the corresponding KNN machine learning model to serve as a prediction model, and outputting the maximum earth surface settlement value caused by shield construction.
6. The method for predicting the ground subsidence caused by shield construction according to claim 5, wherein: the step S2 specifically includes the following steps:
and predicting the test set by using the trained prediction model, and predicting the original data set and the data set after SMOTE preprocessing by using a KNN machine learning model by using a Python Scikt-learn library, wherein the test set is used for testing the prediction performance of the optimal hyper-parameter combination of each model.
7. The method for predicting the ground subsidence caused by shield construction according to claim 1, wherein: the step S3 specifically includes the following steps: and (3) evaluating the model by using two indexes of root mean square error RMSE and average absolute error MAE:
RMSE：
MAE：
where n is the total number of samples, riIs the predicted sedimentation value, piIs the measured sedimentation value.
8. The method for predicting the ground subsidence caused by shield construction according to claim 7, wherein: for any set of training samples, the output predicted sedimentation value riWith the true measured sedimentation value piCertain errors exist between the prediction models, and after the prediction models are calculated by using the calculation formulas of RMSE and MAE, the smaller the RMSE and the MAE, the better the prediction accuracy of the prediction models is.
9. The utility model provides a shield constructs prediction system that construction arouses ground subsides which characterized in that: the method comprises the following steps:
the data processing unit is used for acquiring local shield case data, establishing a database and preprocessing the data in the database to manufacture a data set, wherein the preprocessing comprises the step of amplifying the data by adopting an SMOTE algorithm;
the prediction model unit is used for inputting the data set into a KNN machine learning model for training, selecting the KNN machine learning model with the optimal training parameters as a prediction model, and predicting the maximum earth surface settlement value caused by shield construction;
and the evaluation unit is used for evaluating the prediction model by using the indexes and testing the prediction precision of the prediction model.
10. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1-8 are implemented by the processor when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111445389.7A CN114219134A (en) | 2021-11-30 | 2021-11-30 | Method and system for predicting ground settlement caused by shield construction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111445389.7A CN114219134A (en) | 2021-11-30 | 2021-11-30 | Method and system for predicting ground settlement caused by shield construction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114219134A true CN114219134A (en) | 2022-03-22 |
Family
ID=80699056
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111445389.7A Pending CN114219134A (en) | 2021-11-30 | 2021-11-30 | Method and system for predicting ground settlement caused by shield construction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114219134A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115018164A (en) * | 2022-06-09 | 2022-09-06 | 广州市第三市政工程有限公司 | Method, system, equipment and storage medium for predicting sedimentation of slurry balance pipe jacking construction |
| CN117592163A (en) * | 2023-12-04 | 2024-02-23 | 南宁轨道交通建设有限公司 | Auxiliary decision method for treating longitudinal differential settlement of shield tunnel |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144651A (en) * | 2019-12-26 | 2020-05-12 | 杭州鲁尔物联科技有限公司 | Geological disaster prediction method, device and equipment |
| CN111259219A (en) * | 2020-01-10 | 2020-06-09 | 北京金睛云华科技有限公司 | Malicious webpage identification model, identification model establishing method, identification method and identification system |
| CN113128106A (en) * | 2021-04-06 | 2021-07-16 | 汕头大学 | Method for determining surface subsidence caused by shield construction of karst stratum |
| CN113239439A (en) * | 2021-05-21 | 2021-08-10 | 上海大学 | Shield construction ground surface settlement prediction system and method |
| CN113435056A (en) * | 2021-07-08 | 2021-09-24 | 上海交通大学 | Shield utilization rate prediction and operation parameter optimization method and system based on SVR and PSO |
-
2021
- 2021-11-30 CN CN202111445389.7A patent/CN114219134A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144651A (en) * | 2019-12-26 | 2020-05-12 | 杭州鲁尔物联科技有限公司 | Geological disaster prediction method, device and equipment |
| CN111259219A (en) * | 2020-01-10 | 2020-06-09 | 北京金睛云华科技有限公司 | Malicious webpage identification model, identification model establishing method, identification method and identification system |
| CN113128106A (en) * | 2021-04-06 | 2021-07-16 | 汕头大学 | Method for determining surface subsidence caused by shield construction of karst stratum |
| CN113239439A (en) * | 2021-05-21 | 2021-08-10 | 上海大学 | Shield construction ground surface settlement prediction system and method |
| CN113435056A (en) * | 2021-07-08 | 2021-09-24 | 上海交通大学 | Shield utilization rate prediction and operation parameter optimization method and system based on SVR and PSO |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115018164A (en) * | 2022-06-09 | 2022-09-06 | 广州市第三市政工程有限公司 | Method, system, equipment and storage medium for predicting sedimentation of slurry balance pipe jacking construction |
| CN117592163A (en) * | 2023-12-04 | 2024-02-23 | 南宁轨道交通建设有限公司 | Auxiliary decision method for treating longitudinal differential settlement of shield tunnel |
| CN117592163B (en) * | 2023-12-04 | 2024-04-16 | 南宁轨道交通建设有限公司 | Auxiliary decision method for treating longitudinal differential settlement of shield tunnel |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111858803B (en) | Landslide land disaster risk zoning map generation method | |
| KR101642951B1 (en) | GIS-based real time earthquake prediction method | |
| CN114219134A (en) | Method and system for predicting ground settlement caused by shield construction | |
| CN104899437A (en) | Early-warning method for heavy-rainfall type landslide hazard | |
| Fontul et al. | Railway track condition assessment at network level by frequency domain analysis of GPR data | |
| CN105842341B (en) | A kind of test of intelligentized sound wave transmission method detection foundation pile integrity and analysis method | |
| CN105372631A (en) | Polarizing radar inversion method based on genetic-partial least square algorithm, and application of polarizing radar inversion method | |
| Xiong et al. | Use of NDT systems to investigate pavement reconstruction needs and improve maintenance treatment decision-making | |
| Hassan et al. | Predictive models to evaluate the interaction effect of soil-tunnel interaction parameters on surface and subsurface settlement | |
| Chiaia et al. | Deep convolutional neural network for multi-level non-invasive tunnel lining assessment | |
| CN110633557B (en) | Identification method for favorable area of coal bed gas structure | |
| Bonetto et al. | New application of open source data and Rock Engineering System for debris flow susceptibility analysis | |
| Schafer | Kinematics and controlling mechanics of the slow moving Ripley Landslide | |
| Guo et al. | A new method for fracturing wells reservoir evaluation in fractured gas reservoir | |
| Rashed et al. | GIS as a tool for expansive soil detection at Sulaymaniyah City | |
| CN115930852A (en) | Method and device for estimating thickness of active layer, storage medium and electronic device | |
| Xiong et al. | Automatic detection and location of pavement internal distresses from ground penetrating radar images based on deep learning | |
| CN114236624B (en) | Method and system for estimating fracturing modification space volume based on electromagnetic method | |
| CN114663840A (en) | Tunnel environment inspection equipment monitoring method and system | |
| CN110080197B (en) | Lanzhou red sandstone classification method and corresponding engineering measure selection method | |
| CN114066271A (en) | Tunnel water inrush disaster monitoring and management system | |
| CN116009096A (en) | Shale gas dessert prediction method and equipment for multi-parameter fusion inversion | |
| CN113449362A (en) | Road reconnaissance design method and system based on 5G mobile network | |
| Wu et al. | Prediction of limit pressure and pressuremeter modulus using artificial neural network analysis based on CPTU data | |
| CN116263511A (en) | Quantitative evaluation method for running and sliding fracture activity strength |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220322 |
|
| RJ01 | Rejection of invention patent application after publication |