CN118940013B - Rock mass integrity evaluation method and system based on fluctuation characteristics of drilling parameters - Google Patents
Rock mass integrity evaluation method and system based on fluctuation characteristics of drilling parameters Download PDFInfo
- Publication number
- CN118940013B CN118940013B CN202411418780.1A CN202411418780A CN118940013B CN 118940013 B CN118940013 B CN 118940013B CN 202411418780 A CN202411418780 A CN 202411418780A CN 118940013 B CN118940013 B CN 118940013B
- Authority
- CN
- China
- Prior art keywords
- rock
- rock mass
- drilling
- model
- integrity
- 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.)
- Active
Links
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V11/00—Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Mining & Mineral Resources (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Geophysics (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Geochemistry & Mineralogy (AREA)
- Investigation Of Foundation Soil And Reinforcement Of Foundation Soil By Compacting Or Drainage (AREA)
Abstract
The invention belongs to the field of advanced drilling, and provides a rock mass integrity evaluation method and system based on fluctuation characteristics of parameters while drilling, wherein the rock mass integrity evaluation method comprises the steps of obtaining related parameters for rock mass integrity evaluation, namely original parameters while drilling, rock mass longitudinal wave velocity and rock longitudinal wave velocity, and calculating to obtain a rock mass integrity coefficient based on the rock mass longitudinal wave velocity and the rock longitudinal wave velocity; the method comprises the steps of preprocessing original while-drilling parameters, extracting fluctuation characteristics, namely time domain characteristics and frequency domain characteristics, selecting the characteristics of time domain characteristics and the characteristics of frequency domain characteristics, finally determining optimal input characteristics for training a rock integrity evaluation model, taking the optimal input characteristics as model input, taking corresponding rock integrity coefficients as model output, and training the rock integrity evaluation model. According to the invention, the rock integrity evaluation is carried out by fusing the time domain features and the frequency domain features of the while-drilling parameters, so that the rock integrity evaluation can be more refined and accurate.
Description
Technical Field
The invention belongs to the technical field of advanced drilling, and particularly relates to a rock mass integrity evaluation method and system based on parameter fluctuation characteristics while drilling.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
In the field of tunnel construction, natural rock mass often contains a large number of joint cracks due to geological structure action, weathering action, erosion action and the like, so that engineering disasters such as rock mass strength and stability are easily caused to be reduced, rock burst, collapse, water burst and mud burst are caused to be caused, and how to quickly and accurately obtain the rock mass integrity is an effective guarantee for carrying out geotechnical engineering safety and efficient construction.
In the prior art, physical prospecting, drilling coring and the like are adopted to qualitatively identify the integrity condition of surrounding rock, but the technology is time-consuming and laborious, has complex flow and affects the on-site construction period progress. The research finds that the while-drilling parameters have good correlation with the integrity of the rock mass, a large number of students also conduct correlation analysis between the while-drilling parameters and the integrity of the rock mass, and propose corresponding prediction models and schemes, but most of the schemes directly utilize the while-drilling parameters to conduct the integrity prediction of the rock mass, and neglect the fluctuation response characteristics of the while-drilling parameters in the process of crossing fractured rock mass and complete rock mass.
The inventor combines the existing drilling test research to find that when a complete rock mass and a fractured rock mass are drilled, response characteristics such as abrupt change, fluctuation and the like of drilling speed and torque can be generated, and fluctuation response characteristics can be different along with the different fracture properties, so that the method has research significance for evaluating the rock mass integrity by utilizing the fluctuation characteristics of the parameters while drilling. In summary, the influence on the rock integrity of the fluctuation characteristics of the while-drilling parameters when drilling the completed rock and fractured rock is ignored when rock integrity prediction is performed.
Disclosure of Invention
In order to solve the problems, the invention provides a rock integrity evaluation method and a rock integrity evaluation system based on the fluctuation characteristics of parameters while drilling, the invention comprehensively analyzes the fluctuation characteristics of the parameters while drilling in the time domain and the fluctuation characteristics of the parameters in the frequency domain, establishes a rock integrity evaluation model, the method can be used for evaluating the integrity condition of the front non-excavated rock mass in real time by combining the while-drilling parameters, and timely adjusting the supporting and construction scheme according to the on-site rapid evaluation result, so that the construction safety is improved, the construction cost can be reduced, and the construction efficiency is improved.
According to some embodiments, the first scheme of the invention provides a rock mass integrity evaluation method based on the fluctuation characteristics of parameters while drilling, which adopts the following technical scheme:
the rock mass integrity evaluation method based on the fluctuation characteristics of the parameter while drilling comprises the following steps:
acquiring related parameters for rock mass integrity evaluation, wherein the related parameters comprise original while-drilling parameters, rock mass longitudinal wave velocity and rock longitudinal wave velocity, and calculating to obtain a rock mass integrity coefficient based on the rock mass longitudinal wave velocity and the rock longitudinal wave velocity;
preprocessing the original while-drilling parameters to obtain preprocessed while-drilling parameters;
extracting fluctuation characteristics closely related to the integrity of the rock mass based on the preprocessed while-drilling parameters, namely, time domain characteristics and frequency domain characteristics;
Performing feature selection on the extracted time domain features and frequency domain features by combining a feature filtering algorithm and a Pearson correlation coefficient, and finally determining optimal input features for training a rock mass integrity evaluation model;
Taking the optimized optimal input characteristics as model input, taking the corresponding rock mass integrity coefficients as model output, and training a rock mass integrity evaluation model to obtain a trained rock mass integrity evaluation model;
And acquiring while-drilling parameters of the non-excavated section, and evaluating the integrity of the non-excavated rock by using a trained rock integrity evaluation model.
Further, the preprocessing is performed on the original while-drilling parameters to obtain preprocessed while-drilling parameters, which specifically include:
fixing the sampling frequency of the original while-drilling parameter by adopting an interpolation method;
performing data rejection and data noise reduction on original while-drilling parameters, rejecting drilling parameters without interpretation value, and performing signal denoising on null values, transient shock values and drilling data with irregularity and roughness;
and comprehensively considering timeliness, operability and forecasting precision of rock mass structure division, selecting interval of paragraph division, and carrying out paragraph division on the removed and noise-reduced while-drilling parameters to obtain the preprocessed while-drilling parameters.
Further, the time domain features reflect the change rule of the drilling signal in the time domain, including maximum value, minimum value, peak-to-peak value, average value, root mean square value, standard deviation and signal total value of the time domain drilling signal strength features;
the kurtosis, skewness, peak factor, pulse factor, waveform index and margin factor of the waveform and change characteristics of the time domain drilling signal are characterized.
Further, converting the time domain drilling signal into a frequency domain drilling signal by using a time-frequency analysis method, acquiring a corresponding spectrogram, and then extracting frequency domain characteristics to obtain the frequency domain characteristics of the drilling signal;
The frequency domain characteristics of the drilling signal include a frequency domain amplitude average, a center of gravity frequency, a mean square frequency, and a frequency variance.
Further, the feature filtering algorithm and the pearson correlation coefficient are combined to perform feature selection on the extracted time domain features and frequency domain features, and finally, optimal input features for training a rock mass integrity evaluation model are determined, wherein the optimal input features are specifically as follows:
Determining an initial input feature index and normalizing a data set, calculating and ranking the influence of each input feature on the output feature by using a feature filtering algorithm, namely, the weight of the input feature index, determining a threshold value and eliminating the input feature index lower than the threshold value;
calculating pearson correlation coefficients between the rest input feature indexes, and indicating that the input features have strong correlation when the pearson correlation coefficients are larger than a coefficient threshold value;
deleting one input characteristic in the characteristics with strong correlation, and finally determining the optimal input characteristic for training the rock mass integrity evaluation model;
wherein the optimal input features include optimal time domain features and optimal frequency domain features.
Further, the rock mass integrity evaluation model comprises two layers, wherein the first layer is a bottom layer model with several different mathematical principles, and the second layer is a top layer model;
Wherein the bottom model and the top model are both machine learning algorithms.
Further, the optimized input characteristics are used as model input, the corresponding rock integrity coefficients are used as model output, and the rock integrity evaluation model is trained to obtain a trained rock integrity evaluation model, which specifically comprises the following steps:
Constructing an initial training set and a testing set by utilizing optimal input characteristics and rock integrity coefficients;
based on the initial training set and the testing set, training each bottom layer model by adopting k-fold cross validation, and combining based on a plurality of prediction results so as to determine a new training set and a new testing set;
training the top layer model by using a new training set, and testing by using a new testing set to obtain a trained top layer model;
obtaining a trained rock integrity evaluation model according to the trained bottom layer models and the trained top layer model;
And selecting various evaluation indexes to evaluate the performance of the trained rock mass integrity evaluation model.
Further, the performance of the trained rock mass integrity evaluation model is evaluated by selecting various evaluation indexes, specifically:
And comparing the rock integrity predicted by the rock integrity evaluation model with the actual rock integrity by taking the determination coefficient, the root mean square error and the average absolute percentage error as evaluation indexes, and further evaluating the performance of the trained rock integrity evaluation model.
Further, the parameters while drilling of the non-excavated section are collected, and the integrity of the non-excavated rock is evaluated by using a trained rock integrity evaluation model, specifically:
collecting while-drilling parameters of an unexcavated section, and preprocessing the while-drilling parameters;
Extracting optimal input characteristics based on the preprocessed while-drilling parameters as input parameters, wherein the optimal input characteristics comprise optimal time domain characteristics and optimal frequency domain characteristics;
And based on the input parameters, evaluating the integrity of the unexcavated rock mass by using a trained rock mass integrity evaluation model.
According to some embodiments, a second aspect of the present invention provides a rock integrity evaluation system based on a parameter while drilling fluctuation feature, which adopts the following technical scheme:
rock mass integrity evaluation system based on parameter fluctuation characteristics while drilling includes:
the parameter acquisition module is configured to acquire related parameters for rock mass integrity evaluation, including original while-drilling parameters, rock mass longitudinal wave speed and rock longitudinal wave speed, and calculate rock mass integrity coefficients based on the rock mass longitudinal wave speed and the rock longitudinal wave speed;
The drilling parameter preprocessing module is configured to preprocess the original drilling parameters to obtain preprocessed drilling parameters;
the feature extraction module is configured to extract fluctuation features closely related to the integrity of the rock mass, namely time domain features and frequency domain features based on the preprocessed while-drilling parameters;
The feature optimization module is configured to perform feature selection on the extracted time domain features and frequency domain features by combining a feature filtering algorithm and a pearson correlation coefficient, and finally determine optimal input features for training a rock mass integrity evaluation model;
the model training module is configured to train the rock integrity evaluation model by taking the optimized optimal input characteristics as model input and the corresponding rock integrity coefficients as model output to obtain a trained rock integrity evaluation model;
And the rock integrity evaluation module is used for acquiring while-drilling parameters of the non-excavated section and evaluating the integrity of the non-excavated rock by using the trained rock integrity evaluation model.
Compared with the prior art, the invention has the beneficial effects that:
According to the method, the time domain features and the frequency domain features of the while-drilling parameters are fused to perform rock integrity evaluation, and the rock integrity evaluation can be more refined and accurate compared with the rock integrity evaluation directly performed by directly using the while-drilling parameters.
The invention uses the integrated learning algorithm to replace a single machine learning algorithm, combines the advantages of a plurality of bottom models, improves the prediction performance and stability, reduces the risk of overfitting, and has stronger flexibility and adaptability.
The method can evaluate the integrity of the rock mass in real time in the drilling process, and can acquire the integrity of the rock mass by performing an indoor test relative to the coring process, and has the advantages of obvious economy and timeliness, wide application prospect and obvious engineering value.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
FIG. 1 is a flow chart of a rock mass integrity evaluation method based on the fluctuation characteristics of parameters while drilling in an embodiment of the invention;
FIG. 2 is a training flow diagram of a rock mass integrity evaluation method based on the fluctuation characteristics of parameters while drilling in an embodiment of the invention;
FIG. 3 is a flow chart of feature selection in an embodiment of the invention.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. 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 present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
Embodiments of the invention and features of the embodiments may be combined with each other without conflict.
Example 1
As shown in fig. 1, the embodiment provides a rock mass integrity evaluation method based on the fluctuation characteristics of parameters while drilling, which comprises the following steps:
acquiring related parameters for rock mass integrity evaluation, wherein the related parameters comprise original while-drilling parameters, rock mass longitudinal wave velocity and rock longitudinal wave velocity, and calculating to obtain a rock mass integrity coefficient based on the rock mass longitudinal wave velocity and the rock longitudinal wave velocity;
preprocessing the original while-drilling parameters to obtain preprocessed while-drilling parameters;
extracting fluctuation characteristics closely related to the integrity of the rock mass based on the preprocessed while-drilling parameters, namely, time domain characteristics and frequency domain characteristics;
Performing feature selection on the extracted time domain features and frequency domain features by combining a feature filtering algorithm and a Pearson correlation coefficient, and finally determining optimal input features for training a rock mass integrity evaluation model;
Taking the optimized optimal input characteristics as model input, taking the corresponding rock mass integrity coefficients as model output, and training a rock mass integrity evaluation model to obtain a trained rock mass integrity evaluation model;
And acquiring while-drilling parameters of the non-excavated section, and evaluating the integrity of the non-excavated rock by using a trained rock integrity evaluation model.
As shown in fig. 2, the present embodiment provides a training flow of a rock mass integrity evaluation method based on the fluctuation characteristics of parameters while drilling, which includes the following steps:
Step one, preprocessing parameters while drilling:
firstly, fixing sampling frequency of original while-drilling parameters by adopting an interpolation method, facilitating subsequent time domain and frequency domain analysis and feature extraction of fluctuation signal features of each while-drilling parameter, then carrying out data rejection and data noise reduction to improve data quality of the while-drilling parameters, and finally carrying out paragraph division.
The data rejection is to reject drilling data without interpretation value, such as data in the initial stage of drilling, and the acquired drilling data does not have interpretation value under the influence of factors such as idling of a drilling machine, tapping and passing through a primary spraying concrete section;
And the data denoising is to perform signal denoising on null values, transient shock values and drilling data with irregularity and roughness so as to improve the data quality of the parameters while drilling.
The paragraph division is to comprehensively consider timeliness, operability and forecasting precision of rock mass structure division, select intervals of paragraph division, and equally divide data while drilling so as to study drilling data characteristics and corresponding surrounding rock integrity characteristics in a segmented mode.
During initial data acquisition, besides original while-drilling parameters, rock mass longitudinal wave velocity and rock longitudinal wave velocity are also required to be acquired, and rock mass integrity coefficients are calculated by using the rock mass longitudinal wave velocity and the rock longitudinal wave velocity and serve as output parameters of a rock mass integrity evaluation model.
Extracting fluctuation characteristics closely related to rock mass integrity evaluation:
Comprehensively considering the influence of the time domain features and the frequency domain features of the drilling signals on the rock mass integrity evaluation, extracting fluctuation features closely related to the rock mass integrity based on the preprocessed while-drilling parameters, namely the time domain features and the frequency domain features, and selecting the time domain features and the frequency domain features of the drilling signals closely related to the rock mass integrity.
The time domain features reflect the change rule of the drilling signal in the time domain, and comprise maximum value, minimum value, peak-to-peak value, average value, root mean square value, standard deviation and signal total value which describe the intensity features of the drilling signal;
the technical scheme of the frequency domain feature extraction is that a time-frequency analysis method is used for converting a time domain drilling signal into a frequency domain, and frequency domain feature extraction is carried out after a corresponding spectrogram is obtained to obtain the frequency domain feature of the drilling signal;
The frequency domain characteristics of the drilling signal include a frequency domain amplitude average, a center of gravity frequency, a mean square frequency, and a frequency variance.
The extraction of fluctuation features is to extract initial input features of a rock mass integrity evaluation model, but whether the features influence the rock mass integrity or whether multiple collinearity exists between the features is important, and the final input parameters are determined by feature selection.
Step three, feature selection:
As shown in fig. 3, the initial input features are subjected to validity check and selection by adopting a feature filtering algorithm, the input features with small influence on the uniaxial compressive strength of the rock are removed, and the input features with strong correlation are removed, so that the optimal input features finally used for training the rock integrity evaluation model are determined;
The feature selection is carried out on the extracted time domain features and frequency domain features by combining a feature filtering algorithm and a pearson correlation coefficient, and the optimal input features for training a rock mass integrity evaluation model are finally determined, wherein the optimal input features are specifically as follows:
Determining an initial input feature index and normalizing a data set, calculating and ranking the influence of each input feature on the output feature by using a feature filtering algorithm, namely, the weight of the input feature index, determining a threshold value, and eliminating the input feature index lower than the threshold value, namely, generally deleting the input feature with small influence on the output feature;
calculating pearson correlation coefficients between the rest input feature indexes, and indicating that the input features have strong correlation when the pearson correlation coefficients are larger than a coefficient threshold value;
deleting one input characteristic in the characteristics with strong correlation, and finally determining the optimal input characteristic for training the rock mass integrity evaluation model;
wherein the optimal input features include optimal time domain features and optimal frequency domain features.
For example, including but not limited to calculating and ranking the weights of each input feature index using a feature optimization algorithm, then determining a threshold value and eliminating input feature indexes below the threshold value;
And calculating correlation coefficients between the residual input feature indexes, when the correlation coefficients between the features are larger than a coefficient threshold value, representing that the features have strong correlation, deleting one of the two input features with strong correlation to reduce the complexity of the model, determining an optimal feature index dataset, and finally using the optimal feature index dataset for predicting the input features of model training.
Step four, rock mass integrity evaluation model construction:
the rock mass integrity evaluation model comprises two layers, wherein the first layer is a bottom layer model with several different mathematical principles, and the second layer is a top layer model, wherein the bottom layer model and the top layer model are both machine learning algorithms;
taking the optimized optimal input characteristic as model input, taking the corresponding rock mass integrity coefficient as model output, and training a rock mass integrity evaluation model, wherein the method specifically comprises the following steps:
Constructing an initial training set and a testing set by utilizing optimal input characteristics and rock integrity coefficients;
based on the initial training set and the testing set, training each bottom layer model by adopting k-fold cross validation, and combining based on a plurality of prediction results so as to determine a new training set and a new testing set;
training the top layer model by using a new training set, and testing by using a new testing set to obtain a trained top layer model;
obtaining a trained rock integrity evaluation model according to the trained bottom layer models and the trained top layer model;
And selecting various evaluation indexes to evaluate the performance of the trained rock mass integrity evaluation model.
The rock mass integrity evaluation model is built by adopting a machine learning algorithm with a plurality of different mathematical principles, and comprises two layers, wherein the first layer is a bottom layer model with a plurality of different mathematical principles, and the second layer is a simple linear top layer model for avoiding overfitting. The method comprises the steps of obtaining a plurality of prediction results by training a plurality of bottom layer models, combining the prediction results into new features, and training the new features as input of the top layer models, so that a more accurate and stable prediction result is finally obtained.
The method comprises the steps of optimizing super parameters of a bottom model, training the bottom model, obtaining a new training set, a new testing set and a training element learner.
The optimizing the super parameters of the bottom layer model comprises optimizing the super parameters of the bottom layer model by using an optimizing algorithm, and finding out the optimal super parameters of the bottom layer model;
the technical scheme of training the bottom layer models is that each bottom layer model is trained by adopting k-fold cross validation, a new training set and a new testing set are obtained, training of all the bottom layer models is finally completed, and a new feature matrix required by a plurality of groups of top layer models is finally obtained.
The training of the top model is mainly based on the newly constructed training set combination to train the top model, the new testing set combination is predicted, and finally a final result is output.
Step five, model evaluation:
And selecting various evaluation indexes to evaluate the performance of the rock mass integrity evaluation model after training.
Specifically, the method comprises the step of comparing the rock integrity predicted by a rock integrity evaluation model with the actual rock integrity by taking a determination coefficient (R 2), a Root Mean Square Error (RMSE) and an average absolute percentage error (MAPE) as evaluation indexes.
R 2 represents the accuracy of model fitting data, and the closer to 1, the better the fitting effect is;
the RMSE represents the deviation degree between the model predicted value and the measured value, and the smaller the value is, the better the model predicted performance is;
the MAPE represents the average value of absolute errors between the predicted value and the measured value of the model, and the smaller the value is, the better the prediction performance of the model is.
Step six, rock mass integrity evaluation:
the evaluation model after training and evaluation is applied to actual engineering, and the integrity evaluation of the front non-excavated rock mass is carried out, specifically comprising the following steps:
collecting while-drilling parameters of an unexcavated section, and preprocessing the while-drilling parameters;
Extracting optimal input characteristics based on the preprocessed while-drilling parameters as input parameters, wherein the optimal input characteristics comprise optimal time domain characteristics and optimal frequency domain characteristics;
And based on the input parameters, evaluating the integrity of the unexcavated rock mass by using a trained rock mass integrity evaluation model.
In actual engineering, acquiring parameters while drilling, extracting time domain and frequency domain characteristics, and carrying out integrity evaluation of the front unexcavated rock mass by utilizing a rock mass integrity evaluation model after training and evaluation.
Example two
The embodiment provides a rock mass integrity evaluation system based on parameter fluctuation while drilling characteristics, which comprises the following components:
the parameter acquisition module is configured to acquire related parameters for rock mass integrity evaluation, including original while-drilling parameters, rock mass longitudinal wave speed and rock longitudinal wave speed, and calculate rock mass integrity coefficients based on the rock mass longitudinal wave speed and the rock longitudinal wave speed;
The drilling parameter preprocessing module is configured to preprocess the original drilling parameters to obtain preprocessed drilling parameters;
the feature extraction module is configured to extract fluctuation features closely related to the integrity of the rock mass, namely time domain features and frequency domain features based on the preprocessed while-drilling parameters;
The feature optimization module is configured to perform feature selection on the extracted time domain features and frequency domain features by combining a feature filtering algorithm and a pearson correlation coefficient, and finally determine optimal input features for training a rock mass integrity evaluation model;
the model training module is configured to train the rock integrity evaluation model by taking the optimized optimal input characteristics as model input and the corresponding rock integrity coefficients as model output to obtain a trained rock integrity evaluation model;
And the rock integrity evaluation module is used for acquiring while-drilling parameters of the non-excavated section and evaluating the integrity of the non-excavated rock by using the trained rock integrity evaluation model.
The above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the first embodiment. It should be noted that the modules described above may be implemented as part of a system in a computer system, such as a set of computer-executable instructions.
The foregoing embodiments are directed to various embodiments, and details of one embodiment may be found in the related description of another embodiment.
The proposed system may be implemented in other ways. For example, the system embodiments described above are merely illustrative, such as the division of the modules described above, are merely a logical function division, and may be implemented in other manners, such as multiple modules may be combined or integrated into another system, or some features may be omitted, or not performed.
While the foregoing description of the embodiments of the present invention has been presented in conjunction with the drawings, it should be understood that it is not intended to limit the scope of the invention, but rather, it is intended to cover all modifications or variations within the scope of the invention as defined by the claims of the present invention.
Claims (8)
1. The rock mass integrity evaluation method based on the fluctuation characteristics of the parameter while drilling is characterized by comprising the following steps:
acquiring related parameters for rock mass integrity evaluation, wherein the related parameters comprise original while-drilling parameters, rock mass longitudinal wave velocity and rock longitudinal wave velocity, and calculating to obtain a rock mass integrity coefficient based on the rock mass longitudinal wave velocity and the rock longitudinal wave velocity;
preprocessing the original while-drilling parameters to obtain preprocessed while-drilling parameters;
extracting fluctuation characteristics closely related to the integrity of the rock mass based on the preprocessed while-drilling parameters, namely, time domain characteristics and frequency domain characteristics;
The method comprises the steps of combining a feature filtering algorithm and a Pearson correlation coefficient to perform feature selection on extracted time domain features and frequency domain features, and finally determining optimal input features for training a rock mass integrity evaluation model, wherein the rock mass integrity evaluation model comprises two layers, a first layer is a bottom layer model with several different mathematical principles, and a second layer is a top layer model with a simple model, wherein the bottom layer model and the top layer model are both machine learning algorithms;
taking the optimized optimal input characteristics as model input, taking the corresponding rock mass integrity coefficients as model output, and training the rock mass integrity evaluation model to obtain a trained rock mass integrity evaluation model, wherein the method specifically comprises the following steps:
Constructing an initial training set and a testing set by utilizing optimal input characteristics and rock integrity coefficients;
based on the initial training set and the testing set, training each bottom layer model by adopting k-fold cross validation, and combining based on a plurality of prediction results so as to determine a new training set and a new testing set;
training the top layer model by using a new training set, and testing by using a new testing set to obtain a trained top layer model;
obtaining a trained rock integrity evaluation model according to the trained bottom layer models and the trained top layer model;
selecting various evaluation indexes to evaluate the performance of the trained rock mass integrity evaluation model;
And acquiring while-drilling parameters of the non-excavated section, and evaluating the integrity of the non-excavated rock by using a trained rock integrity evaluation model.
2. The rock mass integrity evaluation method based on the fluctuation characteristics of the while-drilling parameters according to claim 1, wherein the preprocessing of the original while-drilling parameters is performed to obtain the preprocessed while-drilling parameters, specifically:
fixing the sampling frequency of the original while-drilling parameter by adopting an interpolation method;
performing data rejection and data noise reduction on original while-drilling parameters, rejecting drilling parameters without interpretation value, and performing signal denoising on null values, transient shock values and drilling data with irregularity and roughness;
and comprehensively considering timeliness, operability and forecasting precision of rock mass structure division, selecting interval of paragraph division, and carrying out paragraph division on the removed and noise-reduced while-drilling parameters to obtain the preprocessed while-drilling parameters.
3. The rock mass integrity evaluation method based on the fluctuation feature of the parameter while drilling as claimed in claim 1, wherein the time domain feature reflects the change rule of the drilling signal in the time domain, including maximum value, minimum value, peak-to-peak value, average value, root mean square value, standard deviation and signal total value of the characteristic describing the time domain drilling signal intensity;
the kurtosis, skewness, peak factor, pulse factor, waveform index and margin factor of the waveform and change characteristics of the time domain drilling signal are characterized.
4. The rock mass integrity evaluation method based on the fluctuation characteristics of the parameters while drilling as claimed in claim 3, wherein the time-frequency analysis method is used for converting the time-domain drilling signal into a frequency-domain drilling signal, and frequency-domain characteristic extraction is carried out after the corresponding spectrogram is obtained to obtain the frequency-domain characteristic of the drilling signal;
The frequency domain characteristics of the drilling signal include a frequency domain amplitude average, a center of gravity frequency, a mean square frequency, and a frequency variance.
5. The rock mass integrity evaluation method based on the parameter fluctuation while drilling feature according to claim 1, wherein the feature selection is performed on the extracted time domain feature and the extracted frequency domain feature by combining a feature filtering algorithm and pearson correlation coefficient, and the optimal input feature for training a rock mass integrity evaluation model is finally determined, which specifically comprises the following steps:
Determining an initial input feature index and normalizing a data set, calculating and ranking the influence of each input feature on the output feature by using a feature filtering algorithm, namely, the weight of the input feature index, determining a threshold value and eliminating the input feature index lower than the threshold value;
calculating pearson correlation coefficients between the rest input feature indexes, and indicating that the input features have strong correlation when the pearson correlation coefficients are larger than a coefficient threshold value;
deleting one input characteristic in the characteristics with strong correlation, and finally determining the optimal input characteristic for training the rock mass integrity evaluation model;
wherein the optimal input features include optimal time domain features and optimal frequency domain features.
6. The rock mass integrity evaluation method based on the fluctuation characteristics of the while-drilling parameters according to claim 1, wherein the selecting a plurality of evaluation indexes evaluates the performance of the trained rock mass integrity evaluation model, specifically:
And comparing the rock integrity predicted by the rock integrity evaluation model with the actual rock integrity by taking the determination coefficient, the root mean square error and the average absolute percentage error as evaluation indexes, and further evaluating the performance of the trained rock integrity evaluation model.
7. The rock mass integrity evaluation method based on the fluctuation characteristics of the while-drilling parameters according to claim 1, wherein the acquiring the while-drilling parameters of the non-excavated section evaluates the integrity of the non-excavated rock mass by using a trained rock mass integrity evaluation model, specifically comprising:
collecting while-drilling parameters of the non-excavated section and preprocessing the while-drilling parameters;
Extracting optimal input characteristics based on the preprocessed while-drilling parameters as input parameters, wherein the optimal input characteristics comprise optimal time domain characteristics and optimal frequency domain characteristics;
And based on the input parameters, evaluating the integrity of the unexcavated rock mass by using a trained rock mass integrity evaluation model.
8. Rock mass integrality evaluation system based on parameter fluctuation characteristics while drilling, characterized by comprising:
the parameter acquisition module is configured to acquire related parameters for rock mass integrity evaluation, including original while-drilling parameters, rock mass longitudinal wave speed and rock longitudinal wave speed, and calculate rock mass integrity coefficients based on the rock mass longitudinal wave speed and the rock longitudinal wave speed;
the drilling parameter preprocessing module is configured to preprocess the original drilling parameters to obtain preprocessed drilling parameters;
the feature extraction module is configured to extract fluctuation features closely related to the integrity of the rock mass, namely time domain features and frequency domain features based on the preprocessed while-drilling parameters;
The feature optimization module is configured to perform feature selection on the extracted time domain features and frequency domain features by combining a feature filtering algorithm and a Pearson correlation coefficient, and finally determine optimal input features for training a rock mass integrity evaluation model, wherein the rock mass integrity evaluation model comprises two layers, a first layer is a bottom layer model with several different mathematical principles, and a second layer is a top layer model with a simple model, wherein the bottom layer model and the top layer model are both machine learning algorithms;
The model training module is configured to train the rock integrity evaluation model by taking the optimized optimal input characteristics as model input and the corresponding rock integrity coefficients as model output, so as to obtain a trained rock integrity evaluation model, and specifically comprises the following steps:
Constructing an initial training set and a testing set by utilizing optimal input characteristics and rock integrity coefficients;
based on the initial training set and the testing set, training each bottom layer model by adopting k-fold cross validation, and combining based on a plurality of prediction results so as to determine a new training set and a new testing set;
training the top layer model by using a new training set, and testing by using a new testing set to obtain a trained top layer model;
obtaining a trained rock integrity evaluation model according to the trained bottom layer models and the trained top layer model;
selecting various evaluation indexes to evaluate the performance of the trained rock mass integrity evaluation model;
And the rock integrity evaluation module is used for acquiring while-drilling parameters of the non-excavated section and evaluating the integrity of the non-excavated rock by using the trained rock integrity evaluation model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411418780.1A CN118940013B (en) | 2024-10-12 | 2024-10-12 | Rock mass integrity evaluation method and system based on fluctuation characteristics of drilling parameters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411418780.1A CN118940013B (en) | 2024-10-12 | 2024-10-12 | Rock mass integrity evaluation method and system based on fluctuation characteristics of drilling parameters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118940013A CN118940013A (en) | 2024-11-12 |
| CN118940013B true CN118940013B (en) | 2025-01-24 |
Family
ID=93352027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411418780.1A Active CN118940013B (en) | 2024-10-12 | 2024-10-12 | Rock mass integrity evaluation method and system based on fluctuation characteristics of drilling parameters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118940013B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779880A (en) * | 2021-09-06 | 2021-12-10 | 广西路桥工程集团有限公司 | Tunnel surrounding rock two-dimensional quality evaluation method based on advanced drilling data |
| CN114961716A (en) * | 2022-08-02 | 2022-08-30 | 中国矿业大学(北京) | Method for evaluating rock mass properties and degradation characteristics while drilling |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8515676B2 (en) * | 2007-11-06 | 2013-08-20 | Csir | Method and apparatus for assessing the integrity of a rock mass |
| CN109635461B (en) * | 2018-12-18 | 2022-04-29 | 中国铁建重工集团股份有限公司 | Method and system for automatically identifying surrounding rock grade by using while-drilling parameters |
| CN117350145A (en) * | 2023-09-01 | 2024-01-05 | 临沂矿业集团菏泽煤电有限公司郭屯煤矿 | Rock mechanical parameter while-drilling prediction method based on vibration signals |
-
2024
- 2024-10-12 CN CN202411418780.1A patent/CN118940013B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779880A (en) * | 2021-09-06 | 2021-12-10 | 广西路桥工程集团有限公司 | Tunnel surrounding rock two-dimensional quality evaluation method based on advanced drilling data |
| CN114961716A (en) * | 2022-08-02 | 2022-08-30 | 中国矿业大学(北京) | Method for evaluating rock mass properties and degradation characteristics while drilling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118940013A (en) | 2024-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104533400B (en) | Method for reconstructing logging curve | |
| CN113792936A (en) | A method, system, equipment and storage medium for intelligent identification of lithology while drilling | |
| CN102520070A (en) | Structural damage detection method based on nonlinear output frequency response function | |
| CN111753776B (en) | Structural damage identification method based on echo state and multi-scale convolution combined model | |
| CN103852525B (en) | Acoustic emission signal recognition methods based on AR-HMM | |
| CN118940013B (en) | Rock mass integrity evaluation method and system based on fluctuation characteristics of drilling parameters | |
| CN119145867A (en) | Method and system for immediately sensing rock mass state in shield tunneling | |
| CN113671564A (en) | Automatic Picking Method of Microseismic Effective Events Based on NARX Dynamic Neural Network | |
| CN110118994B (en) | A quantitative prediction method for continental source rocks based on seismic inversion and machine learning | |
| CN110231410B (en) | Intelligent interpretation method for nondestructive testing data of anchor rod | |
| CN116663126A (en) | A Bridge Temperature Effect Prediction Method Based on Channel Attention BiLSTM Model | |
| Li et al. | Multiparameter-based separation method for acoustic emission of in-service prestressed hollow slab tire–road noise signals | |
| CN119783231A (en) | A bridge structure damage identification method and system based on cross-channel attention | |
| CN118656306B (en) | Engineering site drilling vs30 query system based on shear wave prediction | |
| CN117908126B (en) | A microseismic P-wave first arrival picking method based on UNet3+ | |
| CN114508345B (en) | Well logging information prediction method and device | |
| CN119434350A (en) | Data processing method applied to strength detection of concrete pouring foundation pile | |
| CN118760945A (en) | A method for identifying vibration signals in rail transit environment based on deep learning algorithm | |
| Zhang et al. | Application of artificial neural network for diagnosing pile integrity based on low strain dynamic testing | |
| Nour et al. | Structural health monitoring of civil infrastructures using smart sensor networks | |
| CN119335045A (en) | Concrete bridge pile foundation risk detection method and system | |
| CN117309999A (en) | Method and system for detecting compaction performance of concrete in arch bridge pipe based on complementary fusion of multisource acoustic parameter model | |
| CN119128598A (en) | Intelligent lithology and fluid simultaneous identification method, device and electronic equipment while drilling | |
| CN117590471A (en) | Intelligent identification method for distinguishing bright crystals of shale | |
| CN118940598A (en) | A method for predicting the drillability of a well to be drilled |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |