CN110889440A - Rockburst grade prediction method and system based on principal component analysis and BP neural network - Google Patents
Rockburst grade prediction method and system based on principal component analysis and BP neural network Download PDFInfo
- Publication number
- CN110889440A CN110889440A CN201911122028.1A CN201911122028A CN110889440A CN 110889440 A CN110889440 A CN 110889440A CN 201911122028 A CN201911122028 A CN 201911122028A CN 110889440 A CN110889440 A CN 110889440A
- Authority
- CN
- China
- Prior art keywords
- rock burst
- principal component
- index
- variables
- rockburst
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses a rockburst grade prediction method and a rockburst grade prediction system based on principal component analysis and a BP neural network, wherein the rockburst grade prediction method comprises the following steps: grading rock burst grades according to the strength and the weakness of the rock burst; determining all influence factor indexes of rock burst classification in high ground stress areas; acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value; performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst; using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade; and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
Description
Technical Field
The invention belongs to the technical field of rockburst prediction, and particularly relates to a rockburst grade prediction method and system based on principal component analysis and a BP neural network.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
In deep-buried tunnels with high ground stress and the like, rock burst is one of the most common geological disasters. In the process of underground engineering excavation under the condition of high ground stress, hard and brittle surrounding rocks cause stress differentiation of a tunnel wall due to excavation unloading, so that elastic strain energy stored in the rock mass is suddenly released, the phenomena of cracking, loosening, peeling, catapulting and even throwing are generated, the safety of constructors and equipment is seriously threatened, and the construction progress is influenced. Therefore, in the actual construction process of the deep-buried tunnel, the influence of the rock burst on the construction progress and the safety of personnel must be considered, and the grade prediction of the rock burst also becomes the key point to be considered in the field engineering.
At present, a series of problems which are difficult to deal with are faced in the aspect of rock burst grade prediction of a deep-buried tunnel in a high ground stress area, and the specific problems are as follows:
the existing single index empirical criterion method is mostly based on energy, strength and lithology, but rock burst is generated by comprehensive influence of various factors, and the single index criterion has one-sidedness.
Uncertainty and complexity of rock mass conditions are main factors for restricting accuracy of rock burst grade prediction.
Most rock burst prediction indexes in the conventional multi-index comprehensive prediction method have certain correlation, so that data redundancy and calculation burden are caused.
How to establish accurate and effective multi-index comprehensive rock burst grade prediction is also a big difficulty.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides the rockburst grade prediction method based on principal component analysis and a BP neural network, the construction site data is deeply mined based on a statistical method, the evaluation information is rich, the result accuracy is high, and the method is simple, efficient, convenient and quick.
In order to achieve the above object, one or more embodiments of the present invention provide the following technical solutions:
the rockburst grade prediction method based on principal component analysis and BP neural network comprises the following steps:
grading rock burst grades according to the strength and the weakness of the rock burst;
determining all influence factor indexes of rock burst classification in high ground stress areas;
acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
According to the further technical scheme, the main components of the test data are extracted to serve as input variables of the BP neural network, the corresponding rock burst level serves as an output value, and the accuracy of the established model is verified by the established mathematical model.
According to the further technical scheme, the rockburst grade is classified according to the rockburst strength:
the first level of [1, 0, 0, 0] represents no rock burst activity, the second level of [0, 1, 0, 0] represents low rock burst, the third level of [0, 0, 1, 0] represents medium rock burst, the fourth level of [0, 0, 0, 1] represents severe rock burst, and the rock burst is stronger when the grade is higher.
According to the further technical scheme, the determination of all influence factor indexes of the rock burst grade in the high ground stress area comprises the following steps:
compressive strength R of single axiscDenoted x 1; ratio R of rock strength to ground stressc/σ1Denoted x 2; ratio R of compressive strength to tensile strength of rockc/σtDenoted x 3; elastic strain energy index WetDenoted x 4; ratio of tangential stress to compressive rock strength σθ/RcDenoted x 5; the tunnel buried depth H is marked as x 6; integrity of rock mass KvAnd is denoted as x 7.
According to the further technical scheme, when principal component analysis is carried out on the obtained normalized index data, a Pearson correlation coefficient matrix among indexes is calculated;
calculating the eigenvalue and eigenvector of the correlation matrix R;
and determining the number of the main components.
According to the further technical scheme, the low-dimensional principal component variables and the corresponding surrounding rock grades are used as input variables and output values of the BP neural network, and the number of nodes of the hidden layer is determined.
In a further technical scheme, the first k principal component variables of the test data are calculated based on the mean and standard deviation of the training data.
The invention also discloses a rockburst grade prediction system based on principal component analysis and BP neural network, comprising:
an influence factor index determination module: grading rock burst grades according to the strength and the weakness of the rock burst; determining all influence factor indexes of rock burst classification in high ground stress areas;
the index data processing module is used for acquiring index variables and corresponding actual rockburst level data in actual engineering and normalizing the mean variance of the index variable values;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
the model establishing module is used for taking the obtained index variables as input indexes and corresponding rock burst grades as output values, training and learning are carried out on data by adopting a BP neural network algorithm, and a mathematical model of each index-rock burst grade is established;
and the prediction module is used for acquiring index variable values near the face of the excavated tunnel, analyzing principal components based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and predicting the rock burst grade by using the acquired mathematical model after obtaining the principal components.
The above one or more technical solutions have the following beneficial effects:
the method is based on a statistical method to deeply mine the site construction data, and has the advantages of rich evaluation information, high result accuracy, simplicity, high efficiency, convenience and rapidness.
The grade of the rockburst drawn by the invention is obtained according to a large number of domestic and foreign engineering practices and based on the actual strength of the rockburst, the grading method of [ x, y, z, w ] is favorable for generalization of the output result, and the position n of the output result closest to 1 represents that the predicted rockburst grade is n grade.
The invention can indirectly predict the rock burst risk level of the high ground stress area by utilizing the variable values of all the indexes, the introduced Sigmoid function can approach any nonlinear continuous function with any precision, and the rock burst prediction problem in the modeling process is highly nonlinear, so that the tunnel construction can be effectively guided to carry out rock burst prediction.
When training, the neural network can extract the regularity knowledge from the input and output data, memorize the regularity knowledge in the weight of the network, and have generalization capability, namely the capability of applying the group of weight to the general situation.
The mean variance normalization is carried out on the original data, so that errors caused by data with large range difference can be effectively reduced, and the accuracy of subsequent rock burst prediction is effectively improved.
Based on training data, the importance of an original database can be effectively reflected by carrying out mean variance normalization on test data, and the original data is a reference for subsequent prediction.
The principal component analysis method can extract the principal components in each influence factor, can reflect a relatively real situation by using a small amount of data, and simultaneously reduces the dimensionality of the influence factors, thereby being beneficial to reducing the calculated amount of the BP neural network and improving the calculation efficiency.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of an embodiment of the present invention.
Detailed Description
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example one
The embodiment discloses a rockburst grade prediction method based on principal component analysis and a BP neural network, which specifically comprises the following steps:
the method comprises the following steps: dividing the rock burst level into four levels according to construction statistics and rock burst strength and weakness data at home and abroad, wherein the rock burst is stronger when the level is higher;
step two: determining all influence factor indexes of rock burst classification in high ground stress areas based on physical and mechanical properties of rock masses and domestic and foreign engineering data examples;
step three: collecting index variables and corresponding actual rockburst level data in actual engineering, and carrying out mean variance normalization on the index variable values;
step four: performing principal component analysis on the excavated index variables by adopting a Principal Component Analysis (PCA) method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
step five: using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
step six: extracting principal components from the test data based on the training data to be used as input variables of the BP neural network, using the corresponding rock burst grade as an output value, and verifying the accuracy of the established model by using the established mathematical model;
step seven: and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
According to the existing engineering example, the rock burst in the high ground stress area is divided into four grades: the first level of [1, 0, 0, 0] represents no rock burst activity, the second level of [0, 1, 0, 0] represents low rock burst, the third level of [0, 0, 1, 0] represents medium rock burst, the fourth level of [0, 0, 0, 1] represents severe rock burst, and the rock burst is stronger when the grade is higher.
In the specific grade division, because the predicted result cannot completely point to a certain fixed grade, the algorithm sets four output results, and the position of the number closest to 1 represents the actual grade, such as [0.182,0.932,0.564,0.235], and the second number 0.932 is closest to 1, so that the predicted grade at this time can be judged to be grade II.
Selecting all the influencing factor indexes according to the construction examples of domestic and foreign projects as follows: compressive strength R of single axisc(x1), ratio R of rock strength to ground stressc/σ1(x2), and the ratio R of compressive strength to tensile strength of rockc/σt(x3) and elastic strain energy index Wet(x4), ratio of tangential stress to rock compressive strength σθ/Rc(x5), tunnel buried depth H (x6), rock integrity Kv(x7) as an index variable.
The acquisition mode can be based on early geological reports, later indoor tests and the like, and the selection reason is that the correlation of the factors on the rock burst level is very large according to domestic and foreign statistical data.
In order to eliminate the adverse effect of the excessive difference of the variable values of the indexes on the prediction result, the mean variance normalization is carried out on the collected original index data x1, x2, x3, x4, x5, x6 and x7, wherein:
In the formulae (1), (2) and (3)
And sjSample data mean and standard respectively representing jth indexAnd (4) poor.
When the obtained normalized index data is subjected to principal component analysis, a Pearson correlation coefficient matrix among indexes is calculated, namely:
R＝(rkl)n×n(k，l＝1，2，…，n) (4)
in the formula (4), rklIs a correlation coefficient between the kth index and the l index, and rkl＝rlkThe specific calculation formula is as follows:
calculating eigenvalues and eigenvectors of a correlation matrix R, where1，λ2，λnSatisfy lambdaiAnd more than or equal to 0(i is 1, 2, …, n), and the unitized feature vector corresponding to the feature value is marked as p1，p2，pn；
And determining the number of the main components. And calculating the accumulated contribution rate of the principal components, and generally taking the first k principal components corresponding to the feature value of more than 1 and the accumulated variance contribution rate of 85-95%.
V in formulae (6) and (7)sIs the variance contribution rate, v, of the s-th principal componentsumkIs the cumulative contribution rate of the first k principal components.
Calculating the corresponding score of the extracted principal component. The principal component coefficient matrix is: u ═ p (p)1，p2，…，pn) If the first k main components are extracted from the original index, the following main components are obtained:
x in the formula (8)*Is a normalized matrix of the original index data,
is the index variable after normalization.
The number k of target principal components obtained by training data is generally less than the number 7 of index variables, low-dimensional principal component variables and corresponding surrounding rock grading are used as input variables and output values of a BP neural network, so far, a unified method is not available for determining the number of nodes of a hidden layer, and the node number is generally judged according to the following empirical formula:
in the formula (9), l is the number of hidden layer nodes, k is the number of input nodes, 4 is the number of output nodes, and a is an adjustment constant, and the range is 1-10.
The output of the neural network is generally the following:
and the activation function adopts a common Sigmoid function to train and learn data and establish a mathematical model.
And calculating the first k principal component variables of the test data based on the mean value and the standard deviation of the training data, reducing errors by taking the training data as a reference, and verifying the accuracy of the established model by using the established neural network mathematical model. And reducing the dimension of the original high-dimensional data by adopting the principal components under the condition of keeping most information of the original data, thereby reducing the quantity of index data.
And acquiring index variable values attached to the face not excavated, extracting principal component variables, and then bringing the principal component variables into a neural network mathematical model to predict the rock burst-adaptive level.
Example two
The present embodiment aims to provide a computing device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the following steps, including:
grading rock burst grades according to the strength and the weakness of the rock burst;
determining all influence factor indexes of rock burst classification in high ground stress areas;
acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
EXAMPLE III
An object of the present embodiment is to provide a computer-readable storage medium.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, performs the steps of:
grading rock burst grades according to the strength and the weakness of the rock burst;
determining all influence factor indexes of rock burst classification in high ground stress areas;
acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
Example four
The invention also discloses a rockburst grade prediction system based on principal component analysis and BP neural network, comprising:
an influence factor index determination module: grading rock burst grades according to the strength and the weakness of the rock burst; determining all influence factor indexes of rock burst classification in high ground stress areas;
the index data processing module is used for acquiring index variables and corresponding actual rockburst level data in actual engineering and normalizing the mean variance of the index variable values;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
the model establishing module is used for taking the obtained index variables as input indexes and corresponding rock burst grades as output values, training and learning are carried out on data by adopting a BP neural network algorithm, and a mathematical model of each index-rock burst grade is established;
and the prediction module is used for acquiring index variable values near the face of the excavated tunnel, analyzing principal components based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and predicting the rock burst grade by using the acquired mathematical model after obtaining the principal components.
The steps involved in the apparatuses of the above second, third and fourth embodiments correspond to the first embodiment of the method, and the detailed description thereof can be found in the relevant description of the first embodiment. The term "computer-readable storage medium" should be taken to include a single medium or multiple media containing one or more sets of instructions; it should also be understood to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by a processor and that cause the processor to perform any of the methods of the present invention.
Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using general purpose computer means, or alternatively, they can be implemented using program code that is executable by computing means, such that they are stored in memory means for execution by the computing means, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps of them are fabricated into a single integrated circuit module. The present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. The rockburst grade prediction method based on principal component analysis and BP neural network is characterized by comprising the following steps:
grading rock burst grades according to the strength and the weakness of the rock burst;
determining all influence factor indexes of rock burst classification in high ground stress areas;
acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
2. The method of claim 1, wherein principal component analysis and BP neural network-based rockburst level prediction is performed on test data by extracting principal components as input variables of the BP neural network, using corresponding rockburst levels as output values, and verifying accuracy of the established model using an established mathematical model.
3. The method according to claim 1, wherein the degree of the rockburst is classified according to the level of the rockburst:
the first level of [1, 0, 0, 0] represents no rock burst activity, the second level of [0, 1, 0, 0] represents low rock burst, the third level of [0, 0, 1, 0] represents medium rock burst, the fourth level of [0, 0, 0, 1] represents severe rock burst, and the rock burst is stronger when the grade is higher.
4. The method for predicting the rockburst level based on principal component analysis and BP neural network as claimed in claim 1, wherein said determining all the influence factor indexes of the rockburst level in the high geostress area comprises:
compressive strength R of single axiscDenoted x 1; ratio R of rock strength to ground stressc/σ1Denoted x 2; ratio R of compressive strength to tensile strength of rockc/σtDenoted x 3; elastic strain energy index WetDenoted x 4; ratio of tangential stress to compressive rock strength σθ/RcDenoted x 5; the tunnel buried depth H is marked as x 6; integrity of rock mass KvAnd is denoted as x 7.
5. The method for predicting the rockburst level based on principal component analysis and BP neural network according to claim 1, wherein when principal component analysis is performed on the obtained normalized index data, a Pearson correlation coefficient matrix between indexes is calculated;
calculating the eigenvalue and eigenvector of the correlation matrix R;
and determining the number of the main components.
6. The method of claim 1, wherein the node number of the hidden layer is determined by using low-dimensional principal component variables and corresponding surrounding rock grades as input variables and output values of the BP neural network.
7. The method of claim 1, wherein the first k principal component variables of the test data are calculated based on a mean and a standard deviation of the training data.
8. Rockburst grade prediction system based on principal component analysis and BP neural network is characterized by comprising:
an influence factor index determination module: grading rock burst grades according to the strength and the weakness of the rock burst; determining all influence factor indexes of rock burst classification in high ground stress areas;
the index data processing module is used for acquiring index variables and corresponding actual rockburst level data in actual engineering and normalizing the mean variance of the index variable values;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
the model establishing module is used for taking the obtained index variables as input indexes and corresponding rock burst grades as output values, training and learning are carried out on data by adopting a BP neural network algorithm, and a mathematical model of each index-rock burst grade is established;
and the prediction module is used for acquiring index variable values near the face of the excavated tunnel, analyzing principal components based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and predicting the rock burst grade by using the acquired mathematical model after obtaining the principal components.
9. A computing device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to perform steps comprising:
grading rock burst grades according to the strength and the weakness of the rock burst;
determining all influence factor indexes of rock burst classification in high ground stress areas;
acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
10. A computer-readable storage medium, having a computer program stored thereon, the program, when executed by a processor, performing the steps of:
grading rock burst grades according to the strength and the weakness of the rock burst;
determining all influence factor indexes of rock burst classification in high ground stress areas;
acquiring an index variable and corresponding actual rockburst grade data in actual engineering, and performing mean variance normalization on the index variable value;
performing principal component analysis on the excavated index variables by adopting a principal component analysis method to obtain a plurality of principal component variables, wherein the principal component variables correspond to rock burst grades determined according to the intensity of rock burst;
using the obtained multiple index variables as input indexes and corresponding rock burst grades as output values, training and learning data by adopting a BP neural network algorithm, and establishing a mathematical model of each index-rock burst grade;
and acquiring index variable values near the face of the unearthed tunnel, performing principal component analysis based on the average value, standard deviation and the like of the training data, extracting corresponding principal component variables, and performing rock burst grade prediction by using the acquired mathematical model after obtaining the principal components.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911122028.1A CN110889440A (en) | 2019-11-15 | 2019-11-15 | Rockburst grade prediction method and system based on principal component analysis and BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911122028.1A CN110889440A (en) | 2019-11-15 | 2019-11-15 | Rockburst grade prediction method and system based on principal component analysis and BP neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110889440A true CN110889440A (en) | 2020-03-17 |
Family
ID=69747686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911122028.1A Pending CN110889440A (en) | 2019-11-15 | 2019-11-15 | Rockburst grade prediction method and system based on principal component analysis and BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110889440A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926267A (en) * | 2021-03-10 | 2021-06-08 | 山东大学 | TBM tunnel rock burst grade prediction method and system based on tunneling parameter inversion |
| CN112987125A (en) * | 2021-02-22 | 2021-06-18 | 中国地质大学（北京） | Shale brittleness index prediction method based on logging data |
| CN113076700A (en) * | 2021-04-27 | 2021-07-06 | 昆明理工大学 | SVM-LDA rock burst machine learning prediction model method based on data analysis principle |
| CN113128031A (en) * | 2021-03-27 | 2021-07-16 | 南京林业大学 | Measuring device and evaluation method for comfort level of seat |
| CN116128382A (en) * | 2023-04-14 | 2023-05-16 | 深圳市宇芯数码技术有限公司 | Chip quality detection system and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732070A (en) * | 2015-02-27 | 2015-06-24 | 广西大学 | Rockburst grade predicting method based on information vector machine |
| CN109635461A (en) * | 2018-12-18 | 2019-04-16 | 中国铁建重工集团有限公司 | A kind of application carrys out the method and system of automatic identification Grades of Surrounding Rock with brill parameter |
-
2019
- 2019-11-15 CN CN201911122028.1A patent/CN110889440A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732070A (en) * | 2015-02-27 | 2015-06-24 | 广西大学 | Rockburst grade predicting method based on information vector machine |
| CN109635461A (en) * | 2018-12-18 | 2019-04-16 | 中国铁建重工集团有限公司 | A kind of application carrys out the method and system of automatic identification Grades of Surrounding Rock with brill parameter |
Non-Patent Citations (2)
| Title |
|---|
| 张钦礼 等: "基于PCA和改进BP组合预测模型的矿岩可爆性研究", 《爆破》 * |
| 赵国彦 等: "岩爆等级预测的PCA-OPF模型", 《矿冶工程》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112987125A (en) * | 2021-02-22 | 2021-06-18 | 中国地质大学（北京） | Shale brittleness index prediction method based on logging data |
| CN112926267A (en) * | 2021-03-10 | 2021-06-08 | 山东大学 | TBM tunnel rock burst grade prediction method and system based on tunneling parameter inversion |
| CN113128031A (en) * | 2021-03-27 | 2021-07-16 | 南京林业大学 | Measuring device and evaluation method for comfort level of seat |
| CN113076700A (en) * | 2021-04-27 | 2021-07-06 | 昆明理工大学 | SVM-LDA rock burst machine learning prediction model method based on data analysis principle |
| CN116128382A (en) * | 2023-04-14 | 2023-05-16 | 深圳市宇芯数码技术有限公司 | Chip quality detection system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110889440A (en) | Rockburst grade prediction method and system based on principal component analysis and BP neural network | |
| CN109635461B (en) | Method and system for automatically identifying surrounding rock grade by using while-drilling parameters | |
| CN111582718B (en) | Cable channel fire risk assessment method and device based on network analytic hierarchy process | |
| CN111222683B (en) | PCA-KNN-based comprehensive grading prediction method for TBM construction surrounding rock | |
| WO2020125682A1 (en) | Method and system for calculating rock strength using logging-while-drilling data | |
| CN112699553B (en) | Intelligent prediction system method for rock burst intensity level | |
| CN112949202A (en) | Bayesian network-based rockburst probability prediction method | |
| CN111915022B (en) | Gaussian process method and device for rapidly identifying stability coefficient of sliding karst dangerous rock | |
| CN111753776B (en) | Structural damage identification method based on echo state and multi-scale convolution combined model | |
| CN103134433B (en) | Method of identifying slip factors causing slope instability by utilizing displacement monitoring | |
| CN115879654A (en) | Method for predicting stability and evaluating reliability of underground roadway | |
| CN112365054A (en) | Comprehensive grading prediction method for deep well roadway surrounding rock | |
| CN117235510A (en) | Joint roughness prediction method and training method of joint roughness prediction model | |
| CN112329255A (en) | Rock burst prediction method based on tendency degree and uncertain measure | |
| CN112926267A (en) | TBM tunnel rock burst grade prediction method and system based on tunneling parameter inversion | |
| CN111026790A (en) | Structure safety assessment and forecasting method based on data mining | |
| CN115358454A (en) | Coal and gas outburst prediction method based on extension-fuzzy hierarchical analysis theory | |
| CN117110584B (en) | Strain type tunnel rock burst intensity level prediction method | |
| CN112257763A (en) | Mine water inrush source distinguishing method based on feature selection and AdaBoost model | |
| CN117390973B (en) | Mine blasting hole utilization rate prediction method based on multilayer perceptron model | |
| CN114676587B (en) | Fatigue life evaluation method based on load spectrum similarity | |
| LIN | An introduction of the Chinese standard for engineering classification of rock masses (GB50218-94) | |
| CN116741385B (en) | Infectious disease cross-border propagation modeling prediction method | |
| Tang et al. | Stability Evaluation of Tunnel Surrounding Rock Based on Data Mining | |
| CN112132416A (en) | PageRank algorithm-based engineering investigation quality refinement evaluation method |
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 |