CN110737977A - tunnel surrounding rock deformation prediction method and prediction device - Google Patents
tunnel surrounding rock deformation prediction method and prediction device Download PDFInfo
- Publication number
- CN110737977A CN110737977A CN201910960267.8A CN201910960267A CN110737977A CN 110737977 A CN110737977 A CN 110737977A CN 201910960267 A CN201910960267 A CN 201910960267A CN 110737977 A CN110737977 A CN 110737977A
- Authority
- CN
- China
- Prior art keywords
- deformation
- tunnel surrounding
- tunnel
- surrounding rocks
- measuring points
- 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
- 239000011435 rock Substances 0.000 title claims abstract description 211
- 238000000034 method Methods 0.000 title claims abstract description 45
- 239000013598 vector Substances 0.000 claims abstract description 132
- 238000012544 monitoring process Methods 0.000 claims abstract description 63
- 238000012549 training Methods 0.000 claims description 71
- 238000010276 construction Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 9
- 230000015654 memory Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000005259 measurement Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention provides tunnel surrounding rock deformation prediction methods and prediction devices, wherein the prediction methods comprise the steps of collecting deformation monitoring data of a plurality of measuring points of each sections of a plurality of tunnel surrounding rocks, wherein a plurality of measuring points are arranged on each sections of each tunnel surrounding rocks, inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model trained in advance, and outputting the predicted deformation amount of the surrounding rocks of each tunnel surrounding rocks.
Description
Technical Field
The invention belongs to the technical field of tunnel engineering construction, and particularly relates to a prediction method and a prediction device for kinds of tunnel surrounding rock deformation.
Background
In the construction of the tunnel surrounding rock, the safety of the tunnel surrounding rock is very important, so that the safety monitoring of the tunnel surrounding rock is very important, and the safety of the tunnel surrounding rock is also characterized by monitoring the deformation of the tunnel surrounding rock.
The tunnel surrounding rock deformation monitoring means that instrument meters and instruments such as a total station instrument, a level gauge, a tower staff, a convergence gauge and the like are used for monitoring and measuring items such as tunnel periphery convergence, vault subsidence, earth surface subsidence and the like. On the basis, deformation data are analyzed and predicted, guidance can be provided for subsequent construction, tunnel construction safety can be guaranteed, construction accidents are reduced, and construction quality is improved.
At present, -popular methods such as a nonlinear regression model, a polynomial model, an ARIMA (autoregressive differential moving average model), a gray model and the like are used, the methods are all used for establishing a tunnel surrounding rock deformation prediction model based on measuring point history data and predicting measuring point deformation, however, when the tunnel surrounding rock deformation prediction model is established, historical monitoring data of a few measuring points are used for training, and single factors of the historical monitoring data are considered, so that the prediction result of the tunnel surrounding rock deformation prediction model established in the way on the deformation of the tunnel surrounding rock is not accurate enough.
Disclosure of Invention
In order to overcome the problem that the prediction result is inaccurate due to the conventional consideration of or at least partially solve the problem, embodiments of the present invention provide methods and apparatuses for predicting tunnel surrounding rock deformation.
According to th aspect of the embodiment of the invention, tunnel surrounding rock deformation prediction methods are provided, which comprise the following steps:
acquiring deformation monitoring data of a plurality of measuring points of sections of a plurality of tunnel surrounding rocks, wherein the measuring points are arranged on sections of tunnel surrounding rocks;
inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model trained in advance, and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks.
On the basis of the technical scheme, the invention can be improved as follows.
, the acquiring deformation monitoring data at multiple measuring points of each sections of multiple tunnel surrounding rocks further comprises:
and training a pre-constructed support vector regression model by using a training sample set, wherein the training sample set comprises deformation monitoring data of a plurality of measuring points of each tunnel surrounding rocks.
And , extracting the feature vectors of the deformation monitoring data of each tunnel surrounding rock at each measuring points in the training sample set and the actually measured surrounding rock deformation at each measuring points to form a training sample feature vector set.
, the training the pre-constructed support vector regression model with the training sample set includes:
inputting every feature vectors in the training sample feature vector set into a pre-constructed support vector regression model, and outputting the predicted deformation of the surrounding rock corresponding to every feature vectors;
calculating the loss between the predicted deformation amount of the surrounding rock corresponding to each feature vectors output by the support vector regression model and the actually measured deformation amount of the surrounding rock;
and adjusting the regression parameter values of the support vector regression model until the loss corresponding to each feature vectors is less than a preset threshold value and the sum of the losses corresponding to each feature vectors in the training sample set is minimum.
, inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model trained in advance, and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks includes:
for any tunnel surrounding rocks, extracting characteristic vectors of deformation monitoring data of a plurality of measuring points of any tunnel surrounding rocks;
inputting the feature vectors corresponding to each measuring points into a support vector regression model which is trained in advance, and outputting the predicted deformation amount of the surrounding rock corresponding to the feature vectors corresponding to each measuring points;
and obtaining the predicted deformation amount of the surrounding rock of any tunnel surrounding rocks according to the predicted deformation amount of the surrounding rock corresponding to each measuring points of the tunnel surrounding rocks.
And , the characteristic vector of the deformation monitoring data at each measuring points comprises the surrounding rock grade, the measuring point type, the measuring point depth, the measuring point temperature and the construction days of the measuring points.
Further , the support vector regression model is:
wherein, w is a weight vector,
for mapping implicit functions, x is the feature vector and b is the offset.
According to a second aspect of the embodiments of the present invention, there is provided kinds of tunnel surrounding rock deformation prediction devices, including:
the acquisition module is used for acquiring deformation monitoring data of a plurality of measuring points of sections of a plurality of tunnel surrounding rocks, wherein the measuring points are arranged on sections of tunnel surrounding rocks;
and the prediction module is used for inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into the support vector regression model trained in advance and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks.
According to a third aspect of the embodiments of the present invention, there are also provided electronic devices, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor calls the program instructions to be able to execute the method for predicting deformation of a tunnel surrounding rock provided by any possible implementations of the possible implementations.
According to a fourth aspect of the embodiments of the present invention, there are also provided non-transitory computer-readable storage media storing computer instructions for causing the computer to execute the method for predicting deformation of tunnel surrounding rock provided by any of the possible implementations.
The embodiment of the invention provides tunnel surrounding rock deformation prediction methods and prediction devices, wherein the method collects deformation monitoring data of a plurality of measuring points of a plurality of sections of a plurality of tunnel surrounding rocks, trains a regression model according to the deformation monitoring data of the plurality of measuring points, and considers a plurality of factors influencing the deformation of the tunnel surrounding rocks so that the prediction result of the deformation of the surrounding rocks output by the finally trained model is more accurate because of the difference of data characteristic vectors of the measuring points of different sections of the tunnel surrounding rocks.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, is briefly introduced in the drawings required in the description of the embodiments or the prior art, obviously, the drawings in the following description are embodiments of the present invention, and other drawings can be obtained according to the drawings without creative efforts for those skilled in the art.
Fig. 1 is a schematic overall flow chart of a tunnel surrounding rock deformation prediction method provided by embodiments of the present invention;
FIG. 2 is a flow chart of a method for training a support vector regression model;
FIG. 3 is a flowchart of a method for predicting deformation of tunnel surrounding rock by using a trained support vector regression model;
FIG. 4 is a schematic view of the types of the measuring points of the surrounding rocks;
FIG. 5 is a connecting block diagram of a tunnel surrounding rock deformation prediction device according to embodiments of the invention;
FIG. 6 is a block diagram of tunnel surrounding rock deformation prediction devices according to another embodiments of the present invention;
FIG. 7 is a block diagram of the internal connections of the training module of FIG. 6;
fig. 8 is a schematic view of an overall structure of an electronic device according to embodiments of the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, is briefly introduced in the drawings required in the description of the embodiments or the prior art, obviously, the drawings in the following description are embodiments of the present invention, and other drawings can be obtained according to the drawings without creative efforts for those skilled in the art.
tunnel surrounding rock deformation prediction methods are provided in embodiments of the invention, and fig. 1 is an overall flow schematic diagram of the tunnel surrounding rock deformation prediction method provided in the embodiment of the invention, and the method includes the steps of collecting deformation monitoring data of a plurality of measuring points of each sections of a plurality of tunnel surrounding rocks, wherein a plurality of measuring points are arranged on each sections of each tunnel surrounding rocks, inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model which is trained in advance, and outputting the predicted deformation amount of each tunnel surrounding rocks.
In particular, in the construction of the tunnel surrounding rock, the safety of the tunnel surrounding rock is very important, so that the safety monitoring of the tunnel surrounding rock is very important, and generally, the safety of the tunnel surrounding rock is represented by monitoring the deformation of the tunnel surrounding rock.
In the embodiment of the invention, a plurality of measuring points can be selected on each sections of tunnel surrounding rocks, deformation monitoring data of each measuring points are collected, so that each tunnel surrounding rocks correspond to the deformation monitoring data of the plurality of measuring points, the deformation monitoring data of the plurality of measuring points of each tunnel surrounding rocks are input into a pre-trained support vector regression model, and the deformation of the tunnel surrounding rocks is predicted through the trained support vector regression model.
Because the characteristic factors of the deformation monitoring data of the measuring points on different sections of the tunnel surrounding rock are different, the support vector regression model is trained according to the deformation monitoring data of the measuring points on different sections of the tunnel surrounding rock, a plurality of factors influencing the deformation of the tunnel surrounding rock are considered, and the prediction result of the deformation of the surrounding rock output by the finally trained model is more accurate.
On the basis of the above embodiment, in the embodiment of the invention, before acquiring the deformation monitoring data of the multiple measuring points of each sections of the multiple tunnel surrounding rocks, the method further comprises the step of training a pre-constructed support vector regression model by using a training sample set, wherein the training sample set comprises the deformation monitoring data of the multiple measuring points of each tunnel surrounding rocks.
Specifically, in the aspect of deformation prediction of tunnel surrounding rocks, the embodiment of the invention adopts a support vector regression model for prediction, that is, a support vector regression model for predicting deformation prediction of tunnel surrounding rocks is constructed in advance.
Specifically, a plurality of measuring points are arranged on each sections of tunnel surrounding rocks, deformation monitoring data of measuring points are collected to form a training sample set, and the training sample set comprises the deformation monitoring data of tunnel surrounding rocks at the measuring points.
The embodiment of the invention trains a pre-constructed support vector regression model by adopting a training sample set, wherein the training sample set comprises deformation monitoring data of a plurality of measuring points of tunnel surrounding rocks, so that the deformation of tunnel surrounding rocks is predicted by using the trained support vector regression model.
In embodiments of the present invention, when training the constructed support vector regression model by using the training samples in the training data set, first, the feature vectors of each training samples (i.e. deformation monitoring data at each measurement points) in the training sample set and the measured surrounding rock deformation at the measurement points are extracted to form a training sample feature vector set.
Referring to fig. 2, in embodiments of the present invention, training a pre-constructed support vector regression model with a training sample set includes inputting each feature vectors in the training sample feature vector set into the pre-constructed support vector regression model, outputting a predicted deformation amount of a surrounding rock corresponding to each feature vectors, calculating a loss between the predicted deformation amount of the surrounding rock corresponding to each feature vectors output by the support vector regression model and the measured deformation amount of the surrounding rock, and adjusting regression parameter values of the support vector regression model until the loss corresponding to each feature vectors is less than a preset threshold, and the sum of the losses corresponding to each feature vectors in the training sample set is minimum.
The feature vectors of each training samples in the training sample set and the actually measured surrounding rock deformation amount corresponding to each training samples are extracted through the embodiment, the feature vectors of each training samples in the training sample set (namely, deformation monitoring data of each measuring points) are input into a pre-constructed support vector regression model, and surrounding rock predicted deformation amounts of each feature vectors are output through the support vector regression model, in the training samples, for each training samples, the surrounding rock deformation amounts are measured, and for each feature vectors output by the support vector regression model, the corresponding surrounding rock predicted deformation amounts and the actually measured surrounding rock deformation amounts are compared, specifically, the loss between the corresponding surrounding rock predicted deformation amounts and the actually measured surrounding rock deformation amounts of each feature vectors output by the support vector regression model is calculated, when the support vector regression model is trained, the parameter values of the support vector regression model are adjusted, so that the regression feature vectors of each training samples (namely, corresponding to each training sample) in the training sample set are continuously adjusted, and the parameter values of the support vector regression model are adjusted, so that the parameter values of the regression model are the support vector regression process, and the parameter values of the regression model are adjusted, and the parameter values of the regression process of the support vector regression model, which is adjusted, so that the support vector regression process is adjusted.
Referring to fig. 3, in embodiments of the invention, inputting deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model after pre-training, and outputting predicted surrounding rock deformation of each tunnel surrounding rocks includes extracting feature vectors of deformation monitoring data of multiple measuring points of any tunnel surrounding rocks for any tunnel surrounding rocks, inputting feature vectors corresponding to each measuring points into the support vector regression model after pre-training, outputting predicted surrounding rock deformation corresponding to each measuring points, and obtaining predicted surrounding rock deformation of any tunnel surrounding rocks according to predicted surrounding rock deformation corresponding to each measuring points of any tunnel surrounding rocks.
Specifically, when the trained support vector regression model is used for predicting the deformation of tunnel surrounding rocks, for tunnel surrounding rocks, the feature vector of deformation monitoring data of measuring points of the tunnel surrounding rocks is extracted, the support vector regression model outputs the predicted deformation of the surrounding rocks corresponding to the feature vector of measuring points, namely predicted deformation of the surrounding rocks are correspondingly output for every measuring points, the predicted deformation of the surrounding rocks of the tunnel surrounding rocks is obtained according to the predicted deformation of the surrounding rocks corresponding to every measuring points of tunnel surrounding rocks, for example, the average value of the predicted deformation of the surrounding rocks corresponding to a plurality of measuring points of tunnel surrounding rocks or the deformation of the whole tunnel surrounding rocks can be obtained in other calculation modes.
In embodiments of the invention, the feature vector of the deformation monitoring data at each measuring points comprises feature parameters such as the surrounding rock grade, the measuring point type, the measuring point depth, the measuring point temperature and the construction days at the measuring points.
Specifically, in the embodiment of the present invention, every feature vectors include 5 feature parameters at the measuring point, such as the surrounding rock grade, the measuring point type, the measuring point depth, the measuring point temperature, and the construction days, wherein in the standard specification of tunnel surrounding rocks, the tunnel surrounding rock grades are divided into 6 categories, specifically as follows:
class I: the rock is fresh and complete, the structural influence is slight, the joint crack does not develop or develops slightly, the rock is closed and does not extend long, no or few weak structural planes exist, the fault zone width is less than 0.1 meter, the rock is nearly orthogonal to the hole, and the rock body is in an integral or blocky masonry structure.
And II, rock is fresh or slightly weathered, is influenced by the structure sea, the joint cracks slightly develop or develop, a small amount of weak structural surfaces and interlayer bonding difference exist, the fracture bandwidth is less than 0.5 m, the fracture bandwidth is oblique or orthogonal to the hole direction, and the rock mass is in a blocky masonry or layered masonry structure.
Class III: the rock is slightly weathered or weakly weathered, the crack development is influenced by the geological structure, the part of the crack is opened and filled with mud, the soft structural surface is distributed more, the fault fracture zone is less than 1 meter, the fault fracture zone is oblique or parallel to the tunnel line, and the rock is in a stone-broken mosaic structure.
And IV: similar to class III, the fracture and weak structural surface is more, the fault fracture zone is less than 2 meters, the fault fracture zone is parallel to the hole, the rock body is of a rubble-shaped mosaic structure, and the local part of the rock body is of a rubble-shaped crushing structure.
And V, type: dispersion: sand layer landslide accumulation and crushed, pebbled and gravelly soil.
Class VI: it is seriously affected by the structure and is in the form of broken stones, gravel, powder and soil.
The types of the measuring points can be divided into three categories, namely surface subsidence measuring points, vault subsidence measuring points and peripheral convergence measuring points, as shown in the attached figure 4; the depth of the measuring point is the vertical distance between the measuring point and the tunnel ground; the temperature of the measuring point refers to the temperature of the measuring point at the moment of measuring the surrounding rock deformation monitoring data; the construction days are the days from the construction of the section where the current measuring point is located to the current measuring point.
On the basis of the above embodiments, in the embodiment of the present invention, the support vector regression model is a linear regression function established in the gaussian feature space, and the linear regression function is:
wherein, w is a weight vector,
for mapping implicit functions, x is the feature vector and b is the offset.
In the process of training the support vector regression model, a constant epsilon is defined as an upper error limit, and a loss function is set as follows:
wherein x isiIs the feature vector of the ith training sample, f (x)i) Predicting deformation, y, for the surrounding rock of the ith training sampleiAnd (3) the measured deformation of the surrounding rock of the ith training sample, namely in the training process, continuously adjusting w and b in the support vector regression model, and ensuring that the loss between the predicted deformation and the measured deformation of the surrounding rock of every training samples is less than a constant epsilon.
In solving for w and b, to solve for w and b, a slack variable ξ is introducediAnd establishing the following constraint function:
wherein, the constant C is a punishment coefficient, and N is the quantity of training samples in the training sample set.
To solve the above equations (1), (2) and (3), the largrage function is introduced and transformed into a dual form as follows:
wherein
The optimal solution found in function equation (4) above is such that the sum of the losses for each feature vectors in the training sample set is minimized:
thus, it follows:
m is the number of characteristic parameters in each characteristic vectors, and in the embodiment of the present invention, the characteristic parameters are the surrounding rock grade, the measuring point type, the measuring point depth, the measuring point temperature and the construction day number, so in the embodiment of the present invention, M is 5.
Therefore, the description and the definition in each embodiment of the tunnel surrounding rock deformation prediction method can be used for understanding each execution module in the embodiment of the invention, and fig. 5 is a schematic structural diagram of the whole tunnel surrounding rock deformation prediction device provided by the embodiment of the invention, and the device comprises an acquisition module 31 and a prediction module 32.
The acquisition module 31 is configured to acquire deformation monitoring data at a plurality of measuring points of sections of a plurality of tunnel surrounding rocks, where the plurality of measuring points are arranged at each sections of tunnel surrounding rocks.
And the prediction module 32 is used for inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model trained in advance and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks.
Referring to fig. 6, there are provided another embodiments of the tunnel surrounding rock deformation prediction apparatus of the present invention, which includes an acquisition module 31, a prediction module 32, an extraction module 33, and a training module 34.
The acquisition module 31 is configured to acquire deformation monitoring data at a plurality of measuring points of sections of a plurality of tunnel surrounding rocks, where the plurality of measuring points are arranged at each sections of tunnel surrounding rocks.
And the prediction module 32 is used for inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a support vector regression model trained in advance and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks.
The extracting module 33 is configured to extract feature vectors of deformation monitoring data at each measuring points of each tunnel surrounding rocks in the training sample set and deformation amounts of the actually measured surrounding rocks at each measuring points, so as to form a training sample feature vector set.
The training module 34 is configured to train the pre-constructed support vector regression model by using a training sample set, where the training sample set includes deformation monitoring data of multiple measuring points of each tunnel surrounding rocks.
Referring to fig. 7, the training module 34 includes an input unit 341, a calculation unit 342, and a parameter adjustment unit 343, where the input unit 341 is configured to input every feature vectors in the training sample feature vector set into a pre-constructed support vector regression model, and output a predicted deformation amount of the surrounding rock corresponding to every feature vectors from the support vector regression model.
And a calculating unit 342, configured to calculate a loss between the predicted deformation amount of the surrounding rock corresponding to each feature vectors output by the support vector regression model and the actually measured deformation amount of the surrounding rock.
The parameter adjusting unit 343 is configured to adjust regression parameter values of the support vector regression model until the loss corresponding to each feature vectors is smaller than a preset threshold, and the sum of the losses corresponding to each feature vectors in the training sample set is minimum.
The embodiment provides electronic devices, and fig. 8 is a schematic diagram of an overall structure of an electronic device provided by an embodiment of the present invention, where the device includes at least processors 01, at least memories 02 and a bus 03, where the processors 01 and the memories 02 complete communication with each other through the bus 03, the memories 02 store program instructions executable by the processors 01, and the processor calls the program instructions to be able to execute the method provided by the above embodiments of the method, for example, the method includes collecting deformation monitoring data at multiple measurement points of each sections of multiple tunnel surrounding rocks, where multiple measurement points are arranged at each sections of each tunnel surrounding rocks, inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a pre-trained support vector regression model, and outputting predicted deformation amounts of each tunnel surrounding rocks.
The embodiment provides non-transitory computer readable storage media, wherein the non-transitory computer readable storage media store computer instructions, and the computer instructions enable a computer to execute the methods provided by the above method embodiments, for example, the method includes collecting deformation monitoring data at a plurality of measuring points of each sections of a plurality of tunnel surrounding rocks, wherein a plurality of measuring points are arranged at each sections of each tunnel surrounding rocks, inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a pre-trained support vector regression model, and outputting a predicted deformation amount of each tunnel surrounding rocks.
The embodiment of the invention provides tunnel surrounding rock deformation prediction methods and prediction devices, wherein deformation monitoring data of different measuring points of different sections of tunnel surrounding rocks are collected, a support vector regression model is trained according to the deformation monitoring data of the measuring points, the support vector regression model is trained by using training sample data of the measuring points, and due to the fact that data feature vectors of the measuring points of the different sections of the tunnel surrounding rocks have differences, the regression model is trained by using sample data of the different measuring points of the different sections, so that the prediction result of the surrounding rock deformation output by the finally trained model is more accurate, and for training samples, the extracted feature vectors are multi-parameter feature vectors, and the prediction result reliability and accuracy of the tunnel surrounding rock deformation by using the support vector regression model after the feature vector training are considered.
It will be understood by those skilled in the art that all or part of the steps of implementing the above method embodiments may be implemented by hardware associated with program instructions, and that the program may be stored in computer readable storage media, and when executed, the program performs the steps of the above method embodiments, and the storage media may include various media capable of storing program code, such as ROM, RAM, magnetic or optical disk.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, that is, may be located in places, or may be distributed on a plurality of network units.
Based on the understanding that the above technical solutions essentially or contributing to the prior art can be embodied in the form of a software product that can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing computer devices (which may be personal computers, servers, or network devices, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (10)
1, kinds of tunnel country rock deformation prediction methods, characterized by, including:
acquiring deformation monitoring data of a plurality of measuring points of sections of a plurality of tunnel surrounding rocks, wherein the measuring points are arranged on sections of tunnel surrounding rocks;
inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into a pre-trained support vector regression model, and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks.
2. The method for predicting deformation of tunnel surrounding rocks according to claim 1, wherein the step of acquiring deformation monitoring data of a plurality of measuring points of each sections of a plurality of tunnel surrounding rocks further comprises the following steps:
and training a pre-constructed support vector regression model by using a training sample set, wherein the training sample set comprises deformation monitoring data of a plurality of measuring points of each tunnel surrounding rocks.
3. The method of predicting tunnel wall rock deformation according to claim 2,
and extracting the characteristic vectors of the deformation monitoring data of each measuring points of each tunnel surrounding rocks in the training sample set and the actually measured surrounding rock deformation of each measuring points to form a training sample characteristic vector set.
4. The method for predicting tunnel surrounding rock deformation according to claim 3, wherein the training of the pre-constructed support vector regression model by using the training sample set comprises:
inputting every feature vectors in the training sample feature vector set into a pre-constructed support vector regression model, and outputting the predicted deformation of the surrounding rock corresponding to every feature vectors;
calculating the loss between the predicted deformation amount of the surrounding rock corresponding to each feature vectors output by the support vector regression model and the actually measured deformation amount of the surrounding rock;
and adjusting the regression parameter values of the support vector regression model until the loss corresponding to each feature vectors is less than a preset threshold value and the sum of the losses corresponding to each feature vectors in the training sample set is minimum.
5. The method for predicting deformation of tunnel surrounding rocks according to claim 4, wherein the step of inputting deformation monitoring data corresponding to tunnel surrounding rocks into a pre-trained support vector regression model, and the step of outputting predicted deformation of the surrounding rocks of tunnel surrounding rocks comprises the steps of:
for any tunnel surrounding rocks, extracting characteristic vectors of deformation monitoring data of a plurality of measuring points of any tunnel surrounding rocks;
inputting the feature vectors corresponding to each measuring points into a support vector regression model trained in advance, and outputting the predicted deformation amount of the surrounding rock corresponding to the feature vectors corresponding to each measuring points;
and obtaining the predicted deformation amount of the surrounding rock of any tunnel surrounding rocks according to the predicted deformation amount of the surrounding rock corresponding to each measuring points of the tunnel surrounding rocks.
6. The tunnel surrounding rock deformation prediction method according to claim 3, 4 or 5, characterized in that the feature vector of the deformation monitoring data at each measuring points comprises surrounding rock grade, measuring point type, measuring point depth, measuring point temperature and construction days at the measuring points.
7. The method for predicting deformation of tunnel surrounding rock according to claim 1, 2, 4 or 5, wherein the support vector regression model is:
wherein, w is a weight vector,for mapping implicit functions, x is the feature vector and b is the offset.
8, kind of tunnel country rock deformation prediction device, its characterized in that includes:
the acquisition module is used for acquiring deformation monitoring data of a plurality of measuring points of sections of a plurality of tunnel surrounding rocks, wherein the measuring points are arranged on sections of tunnel surrounding rocks;
and the prediction module is used for inputting the deformation monitoring data corresponding to each tunnel surrounding rocks into the support vector regression model trained in advance and outputting the predicted surrounding rock deformation of each tunnel surrounding rocks.
Electronic device of claim 9, , comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the method of predicting deformation of tunnel wall rock of any of claims 1 to 7 and .
10, non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when being executed by a processor, is adapted to carry out the steps of the method for predicting deformation of tunnel wall rock according to any one of claims 1-7 and .
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910960267.8A CN110737977A (en) | 2019-10-10 | 2019-10-10 | tunnel surrounding rock deformation prediction method and prediction device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910960267.8A CN110737977A (en) | 2019-10-10 | 2019-10-10 | tunnel surrounding rock deformation prediction method and prediction device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110737977A true CN110737977A (en) | 2020-01-31 |
Family
ID=69268699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910960267.8A Pending CN110737977A (en) | 2019-10-10 | 2019-10-10 | tunnel surrounding rock deformation prediction method and prediction device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110737977A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111609805A (en) * | 2020-04-23 | 2020-09-01 | 哈尔滨工业大学 | Tunnel structure state diagnosis method based on full-distribution strain measurement point section curvature |
| CN112085843A (en) * | 2020-08-24 | 2020-12-15 | 北京航天发射技术研究所 | Method and device for extracting and measuring tunnel type target features in real time |
| CN113836617A (en) * | 2021-09-01 | 2021-12-24 | 山东大学 | Method and system for simulating and predicting front fracture of heading face based on time series model |
| CN117194868A (en) * | 2023-11-06 | 2023-12-08 | 中国矿业大学(北京) | Surrounding rock deformation monitoring method and device, electronic equipment and storage medium |
| CN117540481A (en) * | 2024-01-09 | 2024-02-09 | 石家庄铁道大学 | Method and device for predicting lining damage of frozen soil area, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101344389A (en) * | 2008-08-20 | 2009-01-14 | 中国建筑第八工程局有限公司 | Method for estimating tunnel surrounding rock displacement by neural network |
| CN205154263U (en) * | 2015-12-06 | 2016-04-13 | 中铁二十局集团第三工程有限公司 | Structure is monitored with surrouding rock deformation to shallow tunnel construction |
| CN105952495A (en) * | 2016-07-14 | 2016-09-21 | 招商局重庆交通科研设计院有限公司 | Wireless positioning portable tunnel construction safety warning system and wireless positioning portable tunnel construction safety warning method |
| CN110243335A (en) * | 2019-07-16 | 2019-09-17 | 贵州省交通规划勘察设计研究院股份有限公司 | A kind of constructing tunnel wall rock loosening ring deformation auto-monitoring prior-warning device and method |
-
2019
- 2019-10-10 CN CN201910960267.8A patent/CN110737977A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101344389A (en) * | 2008-08-20 | 2009-01-14 | 中国建筑第八工程局有限公司 | Method for estimating tunnel surrounding rock displacement by neural network |
| CN205154263U (en) * | 2015-12-06 | 2016-04-13 | 中铁二十局集团第三工程有限公司 | Structure is monitored with surrouding rock deformation to shallow tunnel construction |
| CN105952495A (en) * | 2016-07-14 | 2016-09-21 | 招商局重庆交通科研设计院有限公司 | Wireless positioning portable tunnel construction safety warning system and wireless positioning portable tunnel construction safety warning method |
| CN110243335A (en) * | 2019-07-16 | 2019-09-17 | 贵州省交通规划勘察设计研究院股份有限公司 | A kind of constructing tunnel wall rock loosening ring deformation auto-monitoring prior-warning device and method |
Non-Patent Citations (4)
| Title |
|---|
| 刘开云等: "高速公路连拱隧道施工变形预测的GA-SVR智能模型研究", 《公路交通科技》 * |
| 徐婷: "基于最小二乘支持向量回归的系统可靠性预测", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 方昱等: "隧道施工围岩变形预测的智能模型", 《铁道标准设计》 * |
| 李晓龙等: "最小二乘支持向量机在隧道围岩变形预测中的应用", 《公路交通科技》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111609805A (en) * | 2020-04-23 | 2020-09-01 | 哈尔滨工业大学 | Tunnel structure state diagnosis method based on full-distribution strain measurement point section curvature |
| CN111609805B (en) * | 2020-04-23 | 2021-06-01 | 哈尔滨工业大学 | Tunnel structure state diagnosis method based on full-distribution strain measurement point section curvature |
| CN112085843A (en) * | 2020-08-24 | 2020-12-15 | 北京航天发射技术研究所 | Method and device for extracting and measuring tunnel type target features in real time |
| CN112085843B (en) * | 2020-08-24 | 2023-11-28 | 北京航天发射技术研究所 | Tunnel class target feature real-time extraction and measurement method and device |
| CN113836617A (en) * | 2021-09-01 | 2021-12-24 | 山东大学 | Method and system for simulating and predicting front fracture of heading face based on time series model |
| CN113836617B (en) * | 2021-09-01 | 2023-10-20 | 山东大学 | Method and system for analogically predicting front fissure of tunneling surface based on time sequence model |
| CN117194868A (en) * | 2023-11-06 | 2023-12-08 | 中国矿业大学(北京) | Surrounding rock deformation monitoring method and device, electronic equipment and storage medium |
| CN117194868B (en) * | 2023-11-06 | 2024-02-27 | 中国矿业大学(北京) | Surrounding rock deformation monitoring method and device, electronic equipment and storage medium |
| CN117540481A (en) * | 2024-01-09 | 2024-02-09 | 石家庄铁道大学 | Method and device for predicting lining damage of frozen soil area, electronic equipment and storage medium |
| CN117540481B (en) * | 2024-01-09 | 2024-03-12 | 石家庄铁道大学 | Method and device for predicting lining damage of frozen soil area, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110737977A (en) | tunnel surrounding rock deformation prediction method and prediction device | |
| Shi et al. | Modular neural networks for predicting settlements during tunneling | |
| EP2649271B1 (en) | A method to improve reservoir simulation and recovery from fractured reservoirs | |
| Lo et al. | Bayesian updating of subsurface spatial variability for improved prediction of braced excavation response | |
| NO339192B1 (en) | Modeling a reservoir using an area model and a geomechanical model | |
| SA517382152B1 (en) | Integrated Well Completions | |
| RU2004116907A (en) | METHOD FOR CONSTRUCTING A REVERSABLE THREE-DIMENSIONAL HYDRODYNAMIC MODEL OF THE EARTH CALIBRATED IN REAL TIME DURING DRILLING | |
| CN107219553B (en) | Underground river based on GR weighted band_wise fills prediction technique | |
| Karaman et al. | A comparative assessment of rock mass deformation modulus | |
| CN103205972A (en) | Method for analyzing relationship between deformation of foundation pit and ground subsidence outside of foundation pit | |
| CN116861704B (en) | Dynamic rapid prediction method for large deformation grade of high-ground-stress soft rock tunnel | |
| Ray et al. | Characterizing and modeling natural fracture networks in a tight carbonate reservoir in the Middle East: A methodology. | |
| CN104153768A (en) | Granite reservoir stratum reservoir performance evaluation method | |
| CN114036831A (en) | Real-time detection method for geotechnical parameters of side slope of engineering field to be detected | |
| Altun et al. | Geostatistical interpolation for modelling SPT data in northern Izmir | |
| CN107290800A (en) | Log Forecasting Methodology before practical brill | |
| CN103953024B (en) | Foundation ditch automatic monitoring disorder data recognition method | |
| CN115563576A (en) | Multi-information fusion horizontal well minimum level main stress prediction method | |
| CN115455791A (en) | Method for improving landslide displacement prediction accuracy rate based on numerical simulation technology | |
| CN108062447A (en) | A kind of method predicted diving precipitation and influenced on building settlement | |
| CN111624662B (en) | Speed field correction method, device and equipment | |
| Risk et al. | Probabilistic characterization of a soft Scandinavian clay supporting a light quay structure | |
| CN107942400B (en) | A kind of method and device for predicting drilling risk | |
| CN113777668A (en) | Geostress calculation method and device for tight gas reservoir of sand-shale interbed | |
| Ebrahimian et al. | Probability-Based Assessment of Jacket-Type Offshore |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200131 |