CN113722995A - Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock - Google Patents

Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock Download PDF

Info

Publication number
CN113722995A
CN113722995A CN202111006549.8A CN202111006549A CN113722995A CN 113722995 A CN113722995 A CN 113722995A CN 202111006549 A CN202111006549 A CN 202111006549A CN 113722995 A CN113722995 A CN 113722995A
Authority
CN
China
Prior art keywords
rock
compressive strength
uniaxial compressive
model
density
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
Application number
CN202111006549.8A
Other languages
Chinese (zh)
Inventor
谢学斌
李少乾
董世华
郑攻关
苏卫宏
过江
周贵斌
余茂杰
汪令辉
孟稳权
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Central South University
Tongling Nonferrous Metals Group Co Ltd
Original Assignee
Central South University
Tongling Nonferrous Metals Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Central South University, Tongling Nonferrous Metals Group Co Ltd filed Critical Central South University
Priority to CN202111006549.8A priority Critical patent/CN113722995A/en
Publication of CN113722995A publication Critical patent/CN113722995A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N3/00—Investigating strength properties of solid materials by application of mechanical stress
    • G01N3/08—Investigating strength properties of solid materials by application of mechanical stress by applying steady tensile or compressive forces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Evolutionary Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Genetics & Genomics (AREA)
  • Physiology (AREA)
  • Geometry (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Computer Hardware Design (AREA)
  • Molecular Biology (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)

Abstract

The invention discloses a method, a system, a terminal and a readable storage medium for measuring an elastic deformation energy index of a rock, wherein the method comprises the following steps: measuring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample; inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into a machine learning model to obtain a uniaxial compressive strength prediction model; measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model; and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested to obtain the elastic deformation energy index of the rock to be tested. According to the method, the density and the longitudinal wave speed are used as input characteristics to carry out model training to obtain the uniaxial compressive strength prediction model, so that the uniaxial compressive strength of the rock can be accurately predicted, the unloading point position of the follow-up rock to be tested can be accurately predicted based on the accurate uniaxial compressive strength, and the problem that the unloading point cannot be accurately judged in the existing rock elastic deformation energy index test is solved.

Description

Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock
Technical Field
The invention belongs to the field of mineral engineering and geotechnical engineering, and particularly relates to a method, a system, a terminal and a readable storage medium for measuring an elastic deformation energy index of a rock.
Background
The rock burst tendency of mineral rock substances is described by the common concept of impact tendency of scholars at home and abroad and engineering circles, and the elastic deformation energy index W of the rock proposed by the Polish scholars A.Q.Kidybinshi is adoptedetMeasured. As shown in fig. 1, the rock elastic deformation energy index WetThe determination method comprises the steps of carrying out uniaxial compression test on a rock test piece, loading the test piece until the stress reaches 80% -90% of uniaxial compressive strength, then unloading to zero to obtain a stress-strain curve, and then calculating the areas of different parts of the stress-strain curve to obtain elastic deformation energy EeAnd loss strain energy EpAnd the ratio of the two is the elastic deformation energy index of the rock test piece.
From the viewpoint of the principle, it is,the index is determined by the ratio of the elastic deformation energy stored in the rock sample to the strain energy lost due to permanent deformation and fragmentation, so that from the calculated curve shown in FIG. 1, W can be obtainedetThe specific calculation formula of (2):
Figure BDA0003237264150000011
through the numerical value of the index, scholars correspond the index to the rock burst intensity degree through a large number of case researches and engineering practical researches to obtain a rock burst tendency criterion table shown in table 1, and W can be seen from the table 1etThe greater the index, the more intense the rock burst will appear.
TABLE 1 criterion table of rock burst tendency of elastic deformation energy index
Rock burst free Weak rockburst Moderate rockburst Strong rock burst
<2.0 2.0-3.5 3.5-5.0 ≥5.0
The key step of testing the elastic deformation energy index of the rock is to judge the uniaxial compressive strength of the rock in advance, and then accurately load the test piece to be 0.8-0.9 times of the uniaxial compressive strength. But of mechanical properties of the rockThe dispersion is large, the intensity of each test piece cannot be reliably and accurately predicted, and the loading is controlled to be unloaded when the intensity reaches 80% -90% of the intensity of the test piece, so that the W is limitedetAccuracy of the index. If the selected unloading point is less than 80% of the strength of the test piece, this can lead to premature fall of the curve, in the elastic deformation phase of the rock, so that EeGreater, EpSmaller, resulting in larger elastic deformation energy index. When the unloading point is higher than 0.9 times of the strength of the test piece, the test piece is easy to damage, so that the test piece loading process is finished, the complete unloading curve in the figure 1 cannot be obtained, and the W cannot be obtainedetAnd (4) index. Therefore, a control means capable of instantly forecasting that the load reaches 80% -90% of the limit bearing capacity in the loading process is sought, and the method has very practical application value in testing the elastic deformation energy index of the rock.
At present, the average value substitution method is mainly adopted for the pre-judgment of the uniaxial compressive strength of the rock test piece. The method comprises the steps of firstly testing a batch of rock test pieces to obtain uniaxial compressive strength, then taking the strength average value of the batch of test pieces as the strength of subsequent test pieces, and loading and unloading by taking the strength value as a reference to obtain a stress-strain curve. The average value method has great contingency and error for different rock test pieces, because the strength values of different rock test pieces have great discreteness, and the strength values of different rock test pieces cannot be exactly represented by the average value.
In addition, the patent with publication number CN 109238846 a proposes a method for measuring rock burst elastic deformation energy index by a wave velocity method, which indirectly establishes a relation curve of stress-wave velocity through 'stress-time' and 'wave velocity-time', and finds out a rock wave velocity value corresponding to the stress reaching 80% -90% of the strength of the test piece through analysis. As shown in figure 2, the core of the invention is that the wave velocity value reduction point is connected with 80% of the rock strength, and whether the load reaches 80% of the rock strength can be judged by observing whether the wave velocity value of the sample perpendicular to the stress direction is obviously reduced.
However, the researchers also found through experimental studies that the wave velocity value drop point between the rock elastic wave and the uniaxial compressive strength is not 80% of that of the patent. There is a distinct drop behavior point in the wave velocity-strain curve, where the wave velocity drops suddenly and the corresponding ratio λ of stress to peak pressure also reaches a certain value. In document 1, "research on mechanical and acoustic experiments on dense carbonate rocks", a uniaxial compression test is performed on limestone, and it is found that the transverse wave velocity increases at the initial stage of loading, and then the wave velocity is accelerated and decreased at λ of 75%; the granite test is carried out in literature 2 "rule of sound wave propagation speed in granite under different conditions", and the rock wave speed is found to be accelerated and reduced when the lambda is 76.4%; document 3, "experimental study of acoustic wave velocity variation law in rock loading process" obtains λ of gneiss as 72.8%; document 4, "wave velocity characteristic analysis of rock brittle critical failure", finally concludes, based on relevant research statistics and analysis: the average value of λ is 74.8% for most rocks.
The above-mentioned research results show that the determination of λ 80% is not very accurate, and the measurement method can result in the tested WetThe exponent is large, obvious system error exists, and accurate W cannot be obtainedetAnd (4) index. Meanwhile, the testing process of the invention is dynamic, namely a transmitting transducer and a receiving transducer are required to be arranged on a test piece and receive wave velocity signals while a uniaxial compression test is carried out, in the process, a pressure machine and a rock sound wave parameter tester are operated in parallel, and the height of a rock test piece is generally small, so that the rock wave velocity measurement is easy to be disordered and data fluctuation.
Therefore, the prior art means cannot accurately measure the elastic deformation energy index, wherein in order to realize accurate measurement, it is important to instantly predict that the load reaches 80% -90% of the uniaxial compressive strength in the loading process, and therefore, accurate measurement of the uniaxial compressive strength of each rock is also a critical technology.
Disclosure of Invention
The invention aims to provide a method, a system, a terminal and a readable storage medium for measuring an elastic deformation energy index of a rock, aiming at the problem that the elastic deformation energy index of the rock cannot be accurately measured in the prior art. According to the method, the advantages of machine learning are utilized, model training is carried out by taking the density and the longitudinal wave speed as input characteristics to obtain the uniaxial compressive strength prediction model, so that the uniaxial compressive strength of the rock can be accurately predicted, the position of the unloading point of the follow-up rock to be tested can be accurately predicted based on the accurate uniaxial compressive strength, the problem that the unloading point cannot be accurately judged in the rock elastic deformation energy index test in the prior art is solved, and finally, a more accurate rock elastic deformation energy index can be tested.
In one aspect, the invention provides a method for determining an elastic deformation energy index of a rock, which comprises the following steps:
step 1: obtaining a rock sample, and measuring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
step 2: inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into a machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
and step 3: measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model;
and 4, step 4: and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested to obtain the elastic deformation energy index of the rock to be tested.
Optionally, the machine learning model is a GA-SVM interpretation model, and the uniaxial compressive strength prediction model based on the GA-SVM interpretation model is constructed as follows:
2-1: initializing GA algorithm parameters by the parameters;
2-2: initializing a particle population, wherein values of a punishment factor C and a sensitive parameter g in a SVM (support vector machine) model are taken as optimization targets, each group of punishment factor C and sensitive parameter g represents a particle position, and values are taken in the value ranges of the punishment factor C and the sensitive parameter g to obtain the initialized particle population;
2-3: based on a Support Vector Machine (SVM) model and particle values, performing SVM model training by using rock samples, and calculating the fitness corresponding to each particle based on the trained model;
2-4: removing individuals which do not meet the requirements according to the fitness of the particles, performing crossing, variation and selection on the individuals in the population, returning to the step 2-3, and performing iteration updating in a circulating manner, otherwise, stopping current iteration updating until an iteration termination condition is met, and executing the step 2-5;
2-5: selecting an optimal particle based on the fitness of each current particle, wherein a penalty factor C and a sensitive parameter g corresponding to the optimal particle are the optimal penalty factor C and the sensitive parameter g;
2-6: and based on the optimal punishment factor C, the sensitive parameter g and a Support Vector Machine (SVM) model, carrying out SVM model training by using a rock sample to obtain a uniaxial compressive strength prediction model.
Optionally, a model function of a hidden layer in the SVM model is:
Figure BDA0003237264150000031
the corresponding model solution is obtained as:
Figure BDA0003237264150000032
wherein f (x) is a solution of the uniaxial compressive strength prediction model and corresponds to a predicted value of the uniaxial compressive strength; k (x, x)i) In order to be a kernel function, the kernel function,
Figure BDA0003237264150000041
xicorresponding to the ith rock sample, x is the independent variable of the model, sigma is the bandwidth of the Gaussian kernel function, and alphaiIntroducing relaxation variable xi for ith sampleiLagrange multiplier, alpha, of time correspondencei *The ith sample introduces a relaxation variable
Figure BDA0003237264150000042
Corresponding Lagrange multiplier, C is penalty factor, xiiAnd
Figure BDA0003237264150000043
in order to be a function of the relaxation variable,
Figure BDA0003237264150000044
Figure BDA0003237264150000045
dividing a hyperplane normal vector for sample data, wherein a sensitive parameter g exists:
Figure BDA0003237264150000046
b is the bias constant and m is the sample volume.
Optionally, in step 2-6, based on the optimal penalty factor C, the sensitivity parameter g, and the support vector machine SVM model, the process of obtaining the uniaxial compressive strength prediction model by performing SVM model training using the rock sample is as follows:
firstly, obtaining a test set of rock samples, and carrying out SVM model training by using the rock samples in the test set based on the optimal punishment factor C and the sensitive parameter g to obtain a uniaxial compressive strength prediction model;
and then, obtaining a prediction set of the rock samples, and performing model verification and adjustment on a uniaxial compressive strength prediction model by using the rock samples in the prediction set.
Optionally, before the rock sample is a standard rock specimen and the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample are input into the machine learning model as input data in step 2, the method further comprises: carrying out standardization processing on the data;
wherein the density, the longitudinal wave velocity and the uniaxial compressive strength of any one rock sample are expressed as:
Figure BDA0003237264150000047
the normalization process formula is as follows:
Figure BDA0003237264150000048
in the formula, XiA data matrix, X, formed from the density, longitudinal wave velocity and uniaxial compressive strength of the rock sample ii *Corresponding to the standardized data matrix for the rock sample i; rhoi、vi、RiThe density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample i are obtained;
Figure BDA0003237264150000049
the average values of the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample are obtained; sρ、Sv、SRThe standard deviation of the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample.
Optionally, the density, the velocity of longitudinal waves and the uniaxial compressive strength of the rock sample are determined by the following steps:
measuring the volume V and the mass m of the rock sample, and calculating the density by utilizing the volume V and the mass m;
smearing a couplant on the surface of a rock sample, arranging an energy transducer by adopting a direct transmission method, applying pressure to the energy transducer, measuring and reading the traveling time of longitudinal waves in the rock sample, and calculating the velocity of the longitudinal waves according to the following formula;
Figure BDA0003237264150000051
wherein v is the longitudinal wave velocity in m/s; l is the distance between the centers of the transmitting transducer and the receiving transducer, m; t is tpTime of travel of longitudinal wave in rock sample, unit s, t0Is zero delay of the instrumentation system, in units of s
And carrying out a uniaxial compression test on the rock sample to obtain uniaxial compressive strength.
In a second aspect, the present invention provides a measurement system based on the measurement method, including:
the parameter acquisition module is used for acquiring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
the model construction module is used for inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into the machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
the measuring module is used for measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model; and the elastic deformation energy index of the rock to be tested is obtained by carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested.
In a third aspect, the present invention provides a terminal, comprising:
one or more processors;
a memory storing one or more programs;
the processor calls the program to implement:
the method for measuring the elastic deformation energy index of the rock comprises the following steps.
In a fourth aspect, the present invention provides a readable storage medium storing a computer program for invocation by a processor to implement:
the method for measuring the elastic deformation energy index of the rock comprises the following steps.
Advantageous effects
1. The method provided by the invention utilizes the advantages of machine learning, selects the density and the longitudinal wave speed as input characteristics to perform model training to obtain the uniaxial compressive strength prediction model, so that the uniaxial compressive strength of the rock can be accurately predicted, further, the unloading point position of the subsequent rock to be tested can be accurately predicted based on the accurate uniaxial compressive strength, the problem that the unloading point cannot be accurately judged in the rock elastic deformation energy index test in the prior art is solved, and finally, a more accurate rock elastic deformation energy index can be tested.
2. According to the invention, the selection of the parameter combination of the density and the longitudinal wave velocity considers that the test of the density and the longitudinal wave velocity is lossless, no additional disturbance is generated on the test piece, and the obtained prediction result is more accurate.
3. In a further preferred embodiment of the invention, a GA-SVM interpretation model is selected. The Support Vector Machine (SVM) model can approach any nonlinear function in a global sense theoretically, the generalization capability of the model is improved according to the structural risk minimization principle, and a better statistical rule can be obtained under the condition of less statistical sample quantity. The support vector machine model converts the problem into a high-dimensional feature space through nonlinear transformation, thereby remarkably avoiding the problems frequently occurring in the prediction process, such as a disaster data phenomenon, local minimization and the like. The rock test has the property of a small sample, the number of times of the test piece is small, and therefore the method is very suitable for solving the parameter prediction problem related to the rock test piece by using the support vector machine model.
The Genetic Algorithm (GA) enables the population to evolve towards a certain direction continuously based on the evolution principle of the advantages and the disadvantages of population individuals, and meanwhile, the optimal individuals in the population can be searched according to the whole situation, so that the optimal solution meeting the requirements is obtained. The method has extremely high compatibility and intrinsic parallelism, can directly operate a structural object, can adaptively adjust the search direction without requiring the continuity and derivation of functions until the iterative process tends to converge, and thus, the optimal values of the penalty factor C and the sensitive parameter g are searched in the global state more quickly.
Drawings
FIG. 1 is a schematic diagram of a rock elastic energy index calculation model;
FIG. 2 is a graph showing stress-wave velocity curves obtained in CN 201811006283.5;
FIG. 3 is a schematic diagram of the algorithm flow of the GA-SVM interpretation model provided by the present invention;
FIG. 4 is a schematic diagram of a network architecture of a uniaxial compressive strength prediction model provided by the present invention;
FIG. 5 is a GA-SVM interpretation model diagram of uniaxial compressive strength of rock of training set, wherein (a), (b) and (c) are respectively application A10GA-SVM interpretation model constructed by training set and application A15GA once constructed from training setSVM interpretation model, application A20A GA-SVM interpretation model constructed by the training set;
FIG. 6 is a comparison graph of prediction results of prediction sets, wherein (a), (b), and (c) are the results obtained by using A10Trained interpretation model predicts uniaxial compressive strength-Predict _ a10Use of A15Trained interpretation model predicts uniaxial compressive strength-Predict _ a15Use of A20Trained interpretation model predicts uniaxial compressive strength Predict _ a20；
FIG. 7 is a graph comparing the predicted value and the actual value of the uniaxial compressive strength of the BY1-BY5 test pieces;
FIG. 8 is a test piece loading and unloading curve diagram, wherein (a), (b), (c), (d) and (e) are a BY-1 test piece loading and unloading curve diagram, a BY-2 test piece loading and unloading curve diagram, a BY-3 test piece loading and unloading curve diagram, a BY-4 test piece loading and unloading curve diagram and a BY-5 test piece loading and unloading curve diagram respectively;
fig. 9 is a schematic flow chart corresponding to embodiment 1 of the present invention.
Detailed Description
The invention provides a method, a system, a terminal and a readable storage medium for measuring an elastic deformation energy index of a rock, which are used for accurately measuring the elastic deformation energy index of the rock. The present invention will be further described with reference to the following examples.
Example 1:
the embodiment provides a method for measuring an elastic deformation energy index of a rock, which comprises the following steps:
step 1: and obtaining a rock sample, and measuring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample.
The rock to be tested is prepared into a standard rock test piece meeting the requirements of relevant test specifications and regulations, and the test piece is labeled to be used as a rock sample. The density, longitudinal wave velocity and uniaxial compressive strength of the rock sample were then determined as follows.
Testing the density rho of the standard rock test piece: firstly, measuring the volume V of each standard rock test piece, then measuring the mass m (weighing to be accurate to 0.01g) of each standard rock test piece on a precision balance, calculating the natural density of the ore rock according to the following formula, and recording the obtained rock density in a data table:
Figure BDA0003237264150000071
wherein rho is the natural density of rock and the unit is g/cm3(ii) a m is the rock mass in g; a is the sectional area of the standard rock specimen in cm2(ii) a H is the height of the standard rock specimen in cm.
Testing the longitudinal wave velocity v of the standard rock test piece: smearing a coupling agent on the surface of a standard rock test piece, arranging an energy transducer by adopting a direct transmission method, applying pressure to the energy transducer, measuring and reading the time of longitudinal waves walking in the standard rock test piece, calculating the longitudinal wave speed of a rock block according to the following formula, and correspondingly recording the data into a table:
Figure BDA0003237264150000072
wherein v is the longitudinal wave velocity in m/s; l is the distance between the centers of the transmitting transducer and the receiving transducer, m; t is tpIs the time of the longitudinal wave traveling in the standard rock specimen in units of s, t0Is the zero delay of the instrument system, in units of s.
Testing the uniaxial compressive strength R of a standard rock test piece: and (4) carrying out a conventional uniaxial compression test, obtaining the uniaxial compressive strength R of the standard rock test piece on a material testing machine, and correspondingly filling the uniaxial compression test data result of the standard rock test piece into the table. It will be appreciated that other ways of determining the density, compressional velocity and uniaxial compressive strength of a rock sample are possible in the art and are suitable for use in the present invention.
Step 2: inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into a machine learning model to obtain a uniaxial compressive strength prediction model; wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock.
In this embodiment, it is preferable that the data is normalized before being input to the machine learning model. Wherein the density, the longitudinal wave velocity and the uniaxial compressive strength of any one rock sample are expressed as:
Figure BDA0003237264150000081
the normalization process formula is as follows:
Figure BDA0003237264150000082
in the formula, XiA data matrix, X, formed from the density, longitudinal wave velocity and uniaxial compressive strength of the rock sample ii *Corresponding to the standardized data matrix for the rock sample i; rhoi、vi、RiThe density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample i are obtained;
Figure BDA0003237264150000083
the average values of the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample are obtained; sρ、Sv、SRThe standard deviation of the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample.
In other possible embodiments, the process of normalizing the data is not limited. The machine learning model selected in this embodiment is a GA-SVM interpretation model. The GA-SVM interpretation model is essentially to optimize the parameters of the support vector machine SVM by using a GA genetic algorithm. The method for predicting the uniaxial compressive strength of the rock loading by using the GA-SVM model has the advantages of small error, high stability and high accuracy of the model, no need of a large number of rock loading tests, and capability of remarkably reducing the test cost during the test.
Wherein, the construction process of the uniaxial compressive strength prediction model based on the GA-SVM interpretation model is as follows:
2-1: initializing GA algorithm parameters by the parameters;
2-2: initializing a particle population, wherein values of a punishment factor C and a sensitive parameter g in a SVM (support vector machine) model are taken as optimization targets, each group of punishment factor C and sensitive parameter g represents a particle position, and values are taken in the value ranges of the punishment factor C and the sensitive parameter g to obtain the initialized particle population;
2-3: and based on the SVM model and the particle value, performing SVM model training by using the rock sample, and calculating the fitness corresponding to each particle based on the trained model.
In this embodiment, a fitness function is not specifically limited, but the fitness function should be related to the accuracy and precision of a model, and is used to characterize the matching between a support vector machine SVM model determined by a set of penalty factors C and a set of sensitive parameters g and the present invention, that is, characterize the performance of the support vector machine SVM model determined by a set of penalty factors C and a set of sensitive parameters g.
2-4: removing individuals which do not meet the requirements according to the fitness of the particles, performing crossing, variation and selection on the individuals in the population, returning to the step 2-3, and performing iteration updating in a circulating manner, otherwise, stopping current iteration updating until an iteration termination condition is met, and executing the step 2-5;
2-5: and selecting the optimal particle based on the fitness of each current particle, wherein the penalty factor C and the sensitive parameter g corresponding to the optimal particle are the optimal penalty factor C and the sensitive parameter g.
If the higher the fitness value is, the better the model performance is, selecting the example with the maximum fitness value as the optimal particle; if the lower the fitness value is, the better the model performance is, the particle with the smallest fitness value is selected as the best particle.
2-6: and based on the optimal punishment factor C, the sensitive parameter g and a Support Vector Machine (SVM) model, carrying out SVM model training by using a rock sample to obtain a uniaxial compressive strength prediction model.
In the steps 2-6, based on the optimal punishment factor C, the sensitive parameter g and the SVM model, the process of obtaining the uniaxial compressive strength prediction model by performing SVM model training by using the rock sample is as follows:
firstly, obtaining a test set of rock samples, and carrying out SVM model training by using the rock samples in the test set based on the optimal punishment factor C and the sensitive parameter g to obtain a uniaxial compressive strength prediction model;
and then, obtaining a prediction set of the rock samples, and performing model verification and adjustment on a uniaxial compressive strength prediction model by using the rock samples in the prediction set. The adjusting process can select to adjust the model parameters of the support vector machine SVM according to the conventional adjusting mode; or selecting and returning to optimize the penalty factor C and the sensitive parameter g by utilizing the GA algorithm in an iterative way.
In this embodiment, the prediction accuracy of the uniaxial compressive strength of the rock sample in the test set is higher than 90%, the GA-SVM interpretation model constructed by using the training set is considered to be feasible and effective, and if the accuracy is lower than 90%, the requirement is considered to be not met, and adjustment is required. In other possible embodiments, the conditions may be adaptively adjusted according to actual precision requirements, and the like, which is not specifically limited in the present invention.
And step 3: and determining the uniaxial compressive strength of the rock to be detected by using the uniaxial compressive strength prediction model.
The method comprises the steps of obtaining the density and the longitudinal wave velocity of a rock to be measured, and inputting a value uniaxial compressive strength prediction model to measure the uniaxial compressive strength. It will be appreciated that the trained model is certainly best for use on the same type of rock, but may also be used in rock types where the uniaxial compressive strength difference is not significant.
And 4, step 4: and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested to obtain the elastic deformation energy index of the rock to be tested.
After the uniaxial compressive strength of each rock to be tested is obtained in the step 4, the test piece can be accurately loaded until the stress reaches 80% -90% of the uniaxial compressive strength, then the stress is unloaded to zero to obtain a stress-strain curve, and the elastic deformation energy index of the rock test piece is calculated.
Example 2:
the present embodiment provides a measurement system based on the above measurement method, including: the device comprises a parameter acquisition module, a model construction module and a determination module.
The parameter acquisition module is used for acquiring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
the model construction module is used for inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into the machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
and the determination module is used for determining the uniaxial compressive strength of the rock to be determined by using the uniaxial compressive strength prediction model and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be determined to obtain the elastic deformation energy index of the rock to be determined.
It should be understood that embodiment 2 corresponds to embodiment 1, and therefore, the implementation process of each module may refer to the relevant statement of embodiment 1.
For the specific implementation process of each unit module, refer to the corresponding process of the foregoing method. It should be understood that, the specific implementation process of the above unit module refers to the method content, and the present invention is not described herein in detail, and the division of the above functional module unit is only a division of a logic function, and there may be another division manner in the actual implementation, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or may not be executed. Meanwhile, the integrated unit can be realized in a hardware form, and can also be realized in a software functional unit form.
Example 3:
the present embodiment provides a terminal, which includes: one or more processors and memory storing one or more programs, the processors invoking the programs to implement:
step 1: obtaining a rock sample, and measuring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
step 2: inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into a machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
and step 3: measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model;
and 4, step 4: and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested to obtain the elastic deformation energy index of the rock to be tested.
The specific implementation process of each step refers to the explanation of the foregoing method.
It should be understood that in the embodiments of the present invention, the Processor may be a Central Processing Unit (CPU), and the Processor may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. For example, the memory may also store device type information.
Example 4:
the present embodiments provide a readable storage medium storing a computer program for invocation by a processor to implement:
step 1: obtaining a rock sample, and measuring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
step 2: inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into a machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
and step 3: measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model;
and 4, step 4: and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested to obtain the elastic deformation energy index of the rock to be tested.
The specific implementation process of each step refers to the explanation of the foregoing method.
The readable storage medium is a computer readable storage medium, which may be an internal storage unit of the controller according to any of the foregoing embodiments, for example, a hard disk or a memory of the controller. The readable storage medium may also be an external storage device of the controller, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the controller. Further, the readable storage medium may also include both an internal storage unit of the controller and an external storage device. The readable storage medium is used for storing the computer program and other programs and data required by the controller. The readable storage medium may also be used to temporarily store data that has been output or is to be output.
Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned readable storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Application example:
in order to achieve the purpose, content and advantages of the invention, the invention is described in detail below with reference to a test example of deep mining rock burst tendency research in a lead-zinc mine.
(1) Preparing 25 rock test pieces which meet relevant test standards and specifications from dolomite at the deep part of the mine, and carrying out labeling on the test pieces by 1-25;
(2) and testing the density rho of the dolomite test piece, using vaseline as a coupling agent, obtaining the longitudinal wave velocity v of the test piece by adopting a direct transmission method, and correspondingly recording the data into a table.
(3) And performing a conventional uniaxial compression test to obtain the uniaxial compressive strength R of the test piece, and supplementing the uniaxial compressive strength R into a data table.
(4) The Z-score method was used for parameter normalization. Constructing an original data sample X of density, longitudinal wave velocity and uniaxial compressive strength, standardizing index data in the X, and then obtaining a final sample data matrix X according to the wave velocity and the weight coefficient of the density indexf：
Table 2 final sample data matrix Xf
Figure BDA0003237264150000121
Figure BDA0003237264150000131
(5) And dividing the training set and the prediction set. As shown in tables 3, 4 and 5, in order to find the number of individuals of the optimal training set, real number sets A of 10 groups, 15 groups and 20 groups are respectively divided10、A15、A20As a training set.
TABLE 3 division of 10 groups of data as training set-A10
Figure BDA0003237264150000132
Table 4 division of 15 groups of data as training set-a15
Figure BDA0003237264150000133
Figure BDA0003237264150000141
Table 5 division of 20 groups of data as training set-a20
Figure BDA0003237264150000142
(6) As shown in FIG. 5, the genetic algorithm optimized parameter value C obtained by satisfying the convergence condition1And g1Substituting into hidden layer of training set, and finally respectively establishing A10、A15、A20And (3) training the GA-SVM model of the set.
By A10、A15、A20The GA-SVM interpretation model established by the training set can obtain the corresponding relation between the uniaxial compressive strength R and the static parameter density rho and the wave velocity v, and an interpretation model database of the mining area deep dolomite is formed.
(7) As shown in FIG. 6, application A10、A15、A20The GA-SVM interpretation model trained by the training set can obtain the uniaxial compressive strength R of the prediction set1Prediction value set Predict _ A10、Predict_A15、Predict_A20The individual data in the prediction set were compared to the experimental true values and the results are summarized in table 6.
TABLE 6 prediction of uniaxial compressive strength using the constructed GA-SVM interpretation model
Figure BDA0003237264150000151
The precision of the prediction result is comprehensively evaluated by Mean-Square Error (MSE), and the prediction _ A can be found10、Predict_A15、Predict_A20The mean square errors of the sets are 0.13599, 0.001677, 0.041664, respectively, and it can be seen that Predict _ A15The mean square error of (a) is lowest. The method for predicting the uniaxial compressive strength of the rock has good effect and high accuracy, and the deviation of predicted values is lower than 10%. By comparing Predict _ A10、Predict_A15And Predict _ A20The error rates of the three are smaller, the mean square error of the prediction results is respectively 2.14%, 0.53% and 2.80%, and the prediction _ A is15Mean error of data set is minimal, indicating application A15The GA-SVM model obtained by the training set has the best prediction effect and the highest precision.
The comparison shows that the GA-SVM interpretation model constructed by 10 groups of data has the disadvantage of small data quantity, and the construction of the interpretation model by 20 groups of data consumes more rock test pieces to cause the rise of the test cost, so the comprehensive performance of constructing the interpretation model by 15 groups of data is best, the test cost of the invention can be reduced to the greatest extent, the accuracy is high, and the test parameter test can be accurately and economically completed.
(8) And carrying out subsequent elastic deformation energy index test on the rock to be tested. And preparing 5 dolomite standard test pieces BY using the residual dolomite, numbering the test pieces as BY-1, BY-2, BY-3, BY-4 and BY-5 respectively, and testing the density and the longitudinal wave velocity of the test pieces.
Using A15Predicting the uniaxial compressive strength of the test piece in the loading process by a GA-SVM interpretation model constructed by the training set, calculating the unloading point of each test piece, obtaining a calculation curve of the elastic deformation energy index of the rock after unloading, and finally calculating the W of each test piece by a formulaetIndex and according to the rock burst tendency criterion Table 1, the obtained WetThe index is associated with the engineering practice so as to verify whether the result is consistent with the practice.
As shown in Table 7, after the test for determining the elastic deformation energy index of the rock is completed, the uniaxial compression test is performed on the BY-1 to BY-5 test pieces again to obtain the true values of the uniaxial compressive strength of the test pieces, and the true values are compared with the predicted values, so that whether the performance of the established GA-SVM interpretation model in the actual test is good or not is judged.
TABLE 7 test results of elastic deformation energy index of rock test piece
Figure BDA0003237264150000161
FIG. 7 shows BY1-BY5The predicted value and the true value of the uniaxial compressive strength of the test piece are compared, and FIG. 8 is BY1-BY5And (5) obtaining a rock elastic energy curve after loading and unloading the test piece.
The error rates of the predicted value and the true value of the uniaxial compressive strength predicted by the method in the actual test are respectively 0.9%, 3.1%, 3.2%, 2.4% and 3%, and are all below 5%, so that the actual requirements of engineering are met. The ratio k of the unloading point strength value to the predicted uniaxial compressive strength value is 80-90%, and the finally calculated W isetThe index corresponds to the actual field condition, fully reflects the rock burst tendency degree of the deep ore rock of the lead-zinc mine, and shows that the method has high accuracy and high reliability.
It should be emphasized that the examples described herein are illustrative and not restrictive, and thus the invention is not to be limited to the examples described herein, but rather to other embodiments that may be devised by those skilled in the art based on the teachings herein, and that various modifications, alterations, and substitutions are possible without departing from the spirit and scope of the present invention.

Claims (8)

1. A method for measuring the elastic deformation energy index of a rock is characterized by comprising the following steps: the method comprises the following steps:
step 1: obtaining a rock sample, and measuring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
step 2: inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into a machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
and step 3: measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model;
and 4, step 4: and carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested to obtain the elastic deformation energy index of the rock to be tested.
2. The method for measuring according to claim 1, wherein: the machine learning model is a GA-SVM interpretation model, and the construction process of a uniaxial compressive strength prediction model based on the GA-SVM interpretation model is as follows:
2-1: initializing GA algorithm parameters by the parameters;
2-2: initializing a particle population, wherein values of a punishment factor C and a sensitive parameter g in a SVM (support vector machine) model are taken as optimization targets, each group of punishment factor C and sensitive parameter g represents a particle position, and values are taken in the value ranges of the punishment factor C and the sensitive parameter g to obtain the initialized particle population;
2-3: based on a Support Vector Machine (SVM) model and particle values, performing SVM model training by using rock samples, and calculating the fitness corresponding to each particle based on the trained model;
2-4: removing individuals which do not meet the requirements according to the fitness of the particles, performing crossing, variation and selection on the individuals in the population, returning to the step 2-3, and performing iteration updating in a circulating manner, otherwise, stopping current iteration updating until an iteration termination condition is met, and executing the step 2-5;
2-5: selecting an optimal particle based on the fitness of each current particle, wherein a penalty factor C and a sensitive parameter g corresponding to the optimal particle are the optimal penalty factor C and the sensitive parameter g;
2-6: and based on the optimal punishment factor C, the sensitive parameter g and a Support Vector Machine (SVM) model, carrying out SVM model training by using a rock sample to obtain a uniaxial compressive strength prediction model.
3. The method for measuring according to claim 2, wherein: the model function of the hidden layer in the SVM model is as follows:
Figure FDA0003237264140000011
the corresponding model solution is obtained as:
Figure FDA0003237264140000012
wherein f (x) is a solution of the uniaxial compressive strength prediction model and corresponds to a predicted value of the uniaxial compressive strength; k (x)i,xj) Is a RBF Gaussian kernel function, xiFor the ith rock sample, αiIntroducing relaxation variable xi for ith sampleiLagrange multiplier, alpha, of time correspondencei *The ith sample introduces a relaxation variable
Figure FDA0003237264140000021
Corresponding Lagrange multiplier, C is penalty factor, xiiAnd
Figure FDA0003237264140000022
and b is a relaxation variable, b is a bias constant, m is the sample capacity, and omega is a normal vector of the hyperplane divided by the sample data.
4. The method for measuring according to claim 1, wherein: the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample are measured by the following steps:
measuring the volume V and the mass m of the rock sample, and calculating the density by utilizing the volume V and the mass m;
smearing a couplant on the surface of a rock sample, arranging an energy transducer by adopting a direct transmission method, applying pressure to the energy transducer, measuring and reading the traveling time of longitudinal waves in the rock sample, and calculating the velocity of the longitudinal waves according to the following formula;
Figure FDA0003237264140000023
wherein v is the longitudinal wave velocity in m/s; l is the distance between the centers of the transmitting transducer and the receiving transducer, m; t is tpTime of travel of longitudinal wave in rock sample, unit s, t0Is zero delay of the instrument system, unit s;
and carrying out a uniaxial compression test on the rock sample to obtain uniaxial compressive strength.
5. The method for measuring according to claim 1, wherein: before the rock sample is a standard rock specimen and the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample are input into a machine learning model as input data in step 2, the method further comprises the following steps: carrying out standardization processing on the data;
wherein the density, the longitudinal wave velocity and the uniaxial compressive strength of any one rock sample are expressed as:
Figure FDA0003237264140000024
the normalization process formula is as follows:
Figure FDA0003237264140000025
in the formula, XiA data matrix, X, formed from the density, longitudinal wave velocity and uniaxial compressive strength of the rock sample ii *Corresponding to the standardized data matrix for the rock sample i; rhoi、vi、RiThe density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample i are obtained;
Figure FDA0003237264140000031
the average values of the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample are obtained; sρ、Sv、SRIs a rock sampleStandard deviation of density, longitudinal wave velocity and uniaxial compressive strength of the composite material.
6. An assay system based on the assay method according to claim 1, characterized in that: the method comprises the following steps:
the parameter acquisition module is used for acquiring the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample;
the model construction module is used for inputting the density, the longitudinal wave velocity and the uniaxial compressive strength of the rock sample as input data into the machine learning model to obtain a uniaxial compressive strength prediction model;
wherein the input characteristics of the uniaxial compressive strength prediction model are the density and the longitudinal wave velocity of the rock; the output data is the uniaxial compressive strength of the rock;
the measuring module is used for measuring the uniaxial compressive strength of the rock to be measured by using the uniaxial compressive strength prediction model; and the elastic deformation energy index of the rock to be tested is obtained by carrying out a uniaxial loading and unloading test based on the uniaxial compressive strength of the rock to be tested.
7. A terminal, characterized by: the method comprises the following steps:
one or more processors;
a memory storing one or more programs;
the processor calls the program to implement:
the method for determining an elastic deformation energy index of a rock according to claim 1.
8. A readable storage medium, characterized by: a computer program is stored, which is invoked by a processor to implement:
the method for determining an elastic deformation energy index of a rock according to claim 1.
CN202111006549.8A 2021-08-30 2021-08-30 Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock Pending CN113722995A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111006549.8A CN113722995A (en) 2021-08-30 2021-08-30 Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111006549.8A CN113722995A (en) 2021-08-30 2021-08-30 Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock

Publications (1)

Publication Number Publication Date
CN113722995A true CN113722995A (en) 2021-11-30

Family

ID=78679281

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111006549.8A Pending CN113722995A (en) 2021-08-30 2021-08-30 Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock

Country Status (1)

Country Link
CN (1) CN113722995A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115112521A (en) * 2022-08-29 2022-09-27 泉州海关综合技术服务中心 Tea moisture determination error compensation method, system, electronic device and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102005135A (en) * 2010-12-09 2011-04-06 上海海事大学 Genetic algorithm-based support vector regression shipping traffic flow prediction method
CN107991185A (en) * 2017-11-27 2018-05-04 中南大学 A kind of Rock burst proneness Grade Judgment based on rock material peak value elastic energy index
CN109188529A (en) * 2018-08-31 2019-01-11 湖南率为控制科技有限公司 Multipolar array acoustic tool and its countdown circuit
CN109740119A (en) * 2019-01-23 2019-05-10 山东大学 A kind of TBM driving Wall Rock of Tunnel uniaxial compressive strength Method of fast estimating
CN109816204A (en) * 2018-12-25 2019-05-28 南京理工大学 A kind of safety of subway operation Tendency Prediction method based on GA-SVM
CN109975412A (en) * 2019-03-29 2019-07-05 中国电建集团华东勘测设计研究院有限公司 Rock core uniaxial compressive strength measuring method and device based on the study of ultrasonic spectrum depth migration

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102005135A (en) * 2010-12-09 2011-04-06 上海海事大学 Genetic algorithm-based support vector regression shipping traffic flow prediction method
CN107991185A (en) * 2017-11-27 2018-05-04 中南大学 A kind of Rock burst proneness Grade Judgment based on rock material peak value elastic energy index
CN109188529A (en) * 2018-08-31 2019-01-11 湖南率为控制科技有限公司 Multipolar array acoustic tool and its countdown circuit
CN109816204A (en) * 2018-12-25 2019-05-28 南京理工大学 A kind of safety of subway operation Tendency Prediction method based on GA-SVM
CN109740119A (en) * 2019-01-23 2019-05-10 山东大学 A kind of TBM driving Wall Rock of Tunnel uniaxial compressive strength Method of fast estimating
CN109975412A (en) * 2019-03-29 2019-07-05 中国电建集团华东勘测设计研究院有限公司 Rock core uniaxial compressive strength measuring method and device based on the study of ultrasonic spectrum depth migration

Non-Patent Citations (10)

* Cited by examiner, † Cited by third party
Title
XIE XUE-BIN: "Rockburst prediction method based on grey whitenization weight function cluster theory", 《JOURNAL OF HUNAN UNIVERSITY (NATURAL SCIENCE)》, vol. 34, no. 8, 1 January 2007 (2007-01-01) *
冯涛, 潘长良, 王宏图, 曹平, 王文星: "测定岩爆岩石弹性变形能量指数的新方法", 中国有色金属学报, no. 02, 30 June 1998 (1998-06-30) *
吴景龙;杨淑霞;刘承水;: "基于遗传算法优化参数的支持向量机短期负荷预测方法", 中南大学学报(自然科学版), no. 01, 15 February 2009 (2009-02-15) *
唐晓旭;刘平礼;赵立强;刘丹婷;: "基于GA-SVM的油田注水动态预测模型", 油气田地面工程, no. 06, 20 June 2011 (2011-06-20) *
姜力;贺晓雷;行鸿彦;: "改进GA-SVM的湿度传感器温度补偿研究", 电子测量与仪器学报, no. 09, 15 September 2017 (2017-09-15) *
宫凤强;闫景一;李夕兵;: "基于线性储能规律和剩余弹性能指数的岩爆倾向性判据", 岩石力学与工程学报, no. 09, 23 August 2018 (2018-08-23) *
李文: "基于MLR与LS-SVM的岩石强度预测模型比较", 《矿业研究与开发 》, vol. 36, no. 11, 30 November 2016 (2016-11-30) *
赵佰亭;贾晓芬;: "基于遗传算法寻优的支持向量机图像插值方法", 哈尔滨商业大学学报(自然科学版), no. 02, 15 April 2013 (2013-04-15) *
邓金灿;罗一忠;谢学斌;: "大厂105号矿体深井开采岩爆倾向性研究", 有色金属(矿山部分), no. 05, 15 September 2009 (2009-09-15) *
陈伟根;滕黎;刘军;彭尚怡;孙才新;: "基于遗传优化支持向量机的变压器绕组热点温度预测模型", 电工技术学报, no. 01, 26 January 2014 (2014-01-26) *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115112521A (en) * 2022-08-29 2022-09-27 泉州海关综合技术服务中心 Tea moisture determination error compensation method, system, electronic device and storage medium
CN115112521B (en) * 2022-08-29 2022-11-01 泉州海关综合技术服务中心 Tea moisture determination error compensation method, system, electronic device and storage medium

Similar Documents

Publication Publication Date Title
CN111751070B (en) Tuned liquid damper performance parameter detection method, system, device and medium
Chencho et al. Development and application of random forest technique for element level structural damage quantification
CN103344482A (en) Identification method for dynamic constitutive parameters of concrete materials based on reverse calculation
CN113722995A (en) Method, system, terminal and readable storage medium for measuring elastic deformation energy index of rock
CN111079810A (en) Tunnel surrounding rock grade prediction method based on support vector machine
CN114330042A (en) Program load spectrum compiling method and system based on SN curve and storage medium
Xu et al. A deep learning method for damage prognostics of fiber-reinforced composite laminates using acoustic emission
Ghanbari et al. An improvement in data interpretation to estimate residual stresses and mechanical properties using instrumented indentation: A comparison between machine learning and Kriging model
CN116230136A (en) Machine learning-based titanium alloy mechanical property prediction method and device
Ferreira et al. A contribution to the identification of the critical plane using the maximum variance method
CN109669424A (en) A kind of complex condition material properties of rock experiment control system and method
Balandin et al. Dynamic deformation of soft soil media: experimental studies and mathematical modeling
Liu et al. Uncertainty quantification for granular materials with a stochastic discrete element method
Lawal et al. Reliability assessment of empirical equations, ANN and MARS models for predicting the mode I fracture toughness from non-destructive rock properties
He et al. Similitudes for vibration characteristic of composite laminated plates
Dally 11. Statistical Statistical Analysis of Experimental Data
CN105424554A (en) Method for determining fatigue damage degree of metal material based on specific gravity measurement
Wang et al. Evaluating joint confidence region of hypervolume under ROC manifold and generalized Youden index
CN115423600A (en) Data screening method, device, medium and electronic equipment
US20130332225A1 (en) Risk-profile generation device
CN109710888B (en) Natural gas pipeline damage prediction method and device based on punishment regression
CN111222736A (en) Ammunition storage reliability evaluation method based on mixed correlation vector machine model
Holmquist et al. Using Hertzian indentation to understand the strength and ballistic resistance of silicon carbide
Moon et al. Non-equibiaxial residual stress evaluation methodology using simulated indentation behavior and machine learning
Connolly et al. The implications of constitutive model selection in hyperelastic parameter identification

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