CN117572503A - Rock wave velocity testing method - Google Patents
Rock wave velocity testing method Download PDFInfo
- Publication number
- CN117572503A CN117572503A CN202410053198.3A CN202410053198A CN117572503A CN 117572503 A CN117572503 A CN 117572503A CN 202410053198 A CN202410053198 A CN 202410053198A CN 117572503 A CN117572503 A CN 117572503A
- Authority
- CN
- China
- Prior art keywords
- real
- parameter
- rock
- low
- test sample
- 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.)
- Granted
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 205
- 239000011435 rock Substances 0.000 title claims abstract description 106
- 230000007613 environmental effect Effects 0.000 claims abstract description 40
- 238000005259 measurement Methods 0.000 claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 38
- 230000008569 process Effects 0.000 claims abstract description 26
- 230000009467 reduction Effects 0.000 claims abstract description 8
- 238000013507 mapping Methods 0.000 claims abstract description 7
- 238000010586 diagram Methods 0.000 claims description 30
- 239000013598 vector Substances 0.000 claims description 30
- 239000000203 mixture Substances 0.000 claims description 5
- 230000004044 response Effects 0.000 description 5
- 238000002791 soaking Methods 0.000 description 4
- 239000007788 liquid Substances 0.000 description 3
- 238000010183 spectrum analysis Methods 0.000 description 3
- 238000002441 X-ray diffraction Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000007872 degassing Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 239000011148 porous material Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000010998 test method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
Abstract
The application relates to the technical field of geological exploration, in particular to a rock wave velocity testing method, which comprises the following steps: collecting real-time environment parameters, and mapping the real-time environment parameters to a low-dimensional parameter space by using a dimension reduction algorithm; constructing a test sample graph, wherein the test sample graph comprises historical test results of each position point in the low-dimensional parameter space in the historical rock wave speed test process; querying the test sample graph based on the location points of the real-time environmental parameters in the low-dimensional parameter space to determine measurement complexity and wave speed predicted values of the real-time environmental parameters; collecting wave velocity measurement values of the real-time environment parameters; and carrying out weighted summation on the wave speed measured value and the wave speed predicted value based on the measurement complexity to obtain a wave speed test result. According to the technical scheme, the wave velocity test result of the rock can be accurately obtained.
Description
Technical Field
The present application relates generally to the field of geological exploration technology, and more particularly to a rock wave velocity testing method.
Background
The rock acoustic wave test is widely applied to the design and construction of rock engineering and the exploration and exploitation of underground resources, and can evaluate the strength, stability and deformation characteristics of the rock by measuring the propagation speed of acoustic waves or ultrasonic waves in the rock, thereby providing scientific basis for the design construction of the rock engineering and the judgment of the physical characteristics of a reservoir.
Currently, a method, a device, a computer device and a storage medium for predicting a longitudinal and transverse wave speed are disclosed in a patent application document with a publication number of CN115685335a, wherein the method comprises: constructing a deep neural network model according to the rock physical model and training; obtaining target logging response parameters based on the optimized logging interpretation solution; inputting the target logging response parameters into the trained deep neural network model to predict and obtain a target longitudinal wave speed value and a target transverse wave speed value; the target logging response parameters include density, natural gamma, resistivity, and compensating neutrons, among others.
However, the method predicts the longitudinal wave velocity value and the transverse wave velocity value in the rock according to the target logging response parameters such as density, natural gamma, resistivity, compensated neutrons and the like; however, in the rock wave velocity testing process, a small gap between temperature and humidity and rock characteristics may cause different wave velocity testing results, and the rock wave velocity is directly predicted according to logging response parameters such as density, natural gamma, resistivity, compensated neutrons and the like, so that the accuracy of the rock wave velocity cannot be ensured, and the rock wave velocity testing results are inaccurate.
Disclosure of Invention
In order to solve the technical problems in the prior art, the application provides a rock wave velocity testing method which can accurately acquire a rock wave velocity testing result.
The application provides a rock wave velocity testing method, which comprises the following steps: collecting real-time environment parameters, and mapping the real-time environment parameters to a low-dimensional parameter space by using a dimension reduction algorithm, wherein the real-time environment parameters comprise temperature and humidity, rock porosity, rock density, rock saturation and rock composition; constructing a test sample graph, wherein the test sample graph comprises historical test results of each position point in the low-dimensional parameter space in the historical rock wave speed test process; querying the test sample graph based on the location points of the real-time environmental parameters in the low-dimensional parameter space to determine measurement complexity and wave speed predicted values of the real-time environmental parameters; collecting wave velocity measurement values of the real-time environment parameters; and carrying out weighted summation on the wave speed measured value and the wave speed predicted value based on the measurement complexity to obtain a wave speed test result.
In some embodiments, the dimension-reduction algorithm is a self-encoding network, and the mapping the real-time environmental parameters to a low-dimensional parameter space using the dimension-reduction algorithm comprises: building a self-coding network, wherein the self-coding network comprises a parameter encoder and a parameter decoder, and the output of the parameter encoder is a low-dimensional vector; collecting a plurality of groups of environment parameter samples, sequentially inputting the plurality of groups of environment parameter samples into the parameter encoder to obtain a low-dimensional vector, and inputting the low-dimensional vector into the parameter decoder to obtain a decoding result; calculating a reconstruction loss value based on the decoding result and the environmental parameter samples; back-propagating according to the reconstruction loss value to update the parameter encoder and the parameter decoder; iteratively updating the parameter encoder and the parameter decoder until the reconstruction loss value is smaller than a preset loss value, thereby obtaining a trained parameter encoder; inputting the real-time environment parameters into a trained parameter encoder to output low-dimensional vectors corresponding to the real-time environment parameters, wherein the low-dimensional vectors correspond to the position points of the real-time environment parameters in the low-dimensional parameter space; wherein the low-dimensional vector is a two-dimensional vector or a three-dimensional vector.
In some embodiments, the reconstruction loss value satisfies the relationship:
wherein,for the number of said sets of environmental parameter samples, +.>Is->Group environmental parameter sample, ++>Is->Decoding result of group environmental parameter samples, +.>Reconstructing the loss value.
In some embodiments, the constructing a test sample graph comprises: constructing a test sample initial diagram, wherein the test sample initial diagram comprises all position points in the low-dimensional parameter space, and the numerical value of each position point is 0; in the random historical rock wave speed testing process, acquiring position points of historical environmental parameters in the initial diagram of the test sample, and updating the numerical values of the position points according to the historical test results, wherein the historical test results are rock wave speeds under the historical environmental parameters; traversing all the wave speed testing processes of the historical rock, and continuously updating the numerical value of each position point in the initial diagram of the test sample; and in the updated initial diagram of the test sample, marking any position point with the value of 0 as a zero position point, and determining the value of the zero position point by using an interpolation algorithm until all the zero position points in the updated initial diagram of the test sample are traversed, so as to obtain the diagram of the test sample.
In some embodiments, the updating the value of the location point according to the historical test result comprises: and taking the numerical value before updating the position point and the average value of the historical test result as the numerical value after updating the position point.
In some embodiments, the updating the value of the location point according to the historical test result comprises: taking all the historical test results corresponding to the position points as a test result set of the position points; and taking the mode in the test result set as the numerical value after the position point is updated.
In some embodiments, querying the test sample graph based on the location points of the real-time environmental parameter in the low-dimensional parameter space to determine the measurement complexity and wave velocity predictors of the real-time environmental parameter comprises: taking the numerical value of a target position point in the test sample graph as a wave velocity predicted value of the real-time environment parameter, wherein the target position point is a position point of the real-time environment parameter in the low-dimensional parameter space; calculating variances of all historical test results in a neighborhood range of target position points in the test sample graph to serve as measurement complexity of the real-time environment parameters; the neighborhood range of the target position point is a rectangular range with the target position point as the center and the set size.
In some embodiments, the wave speed test results satisfy the relationship:
wherein,for the measurement complexity of said real-time environmental parameter, < >>For wave velocity measurement, +.>As the predicted value of the wave velocity,and the wave speed test result is a transverse wave speed or a longitudinal wave speed.
In some embodiments, the test method further comprises: and updating the numerical value of the target position point in the test sample graph according to the wave speed test result.
The technical scheme of the application has the following beneficial technical effects:
according to the rock wave velocity testing method provided by the embodiment of the application, firstly, the dimension of the real-time environment parameter is reduced to the low-dimension parameter space by using the dimension reduction algorithm, and the position point of the real-time environment parameter in the low-dimension parameter space is obtained; determining a historical test result of each position point in the low-dimensional parameter space based on all historical rock wave velocity test processes, further constructing a test sample diagram, and obtaining a wave velocity predicted value of the position point where the real-time environment parameter is located and a variance of the historical test result in a neighborhood range of the position point where the real-time environment parameter is located by inquiring the test sample diagram; taking the variance of the historical test result in the neighborhood range of the position point as the measurement complexity of the real-time environment parameter, and carrying out weighted summation on the wave velocity measured value and the wave velocity predicted value according to the measurement complexity of the real-time environment parameter, so that an accurate wave velocity test result is obtained, and the accuracy of the wave velocity test result of the rock is improved.
Further, after the wave velocity test result under the real-time environment parameter is obtained, the real-time environment parameter and the wave velocity test result are regarded as a historical rock wave velocity test process, and the numerical value of the position point corresponding to the real-time environment parameter in the test sample graph is updated according to the wave velocity test result, so that the real-time update of the test sample graph is realized, the accuracy of the measurement complexity and the wave velocity predicted value is ensured, and the accuracy of the wave velocity test result is further ensured.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar or corresponding parts and in which:
FIG. 1 is a flow chart of a rock wave velocity testing method according to an embodiment of the present application;
FIG. 2 is a flow chart of constructing a test sample graph according to an embodiment of the present application;
fig. 3 is a schematic diagram of a test sample graph according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
It should be understood that when the terms "first," "second," and the like are used herein, they are merely used to distinguish between different objects and are not used to describe a particular order. The terms "comprises" and "comprising" when used in this application are taken to specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The application provides a rock wave velocity testing method. Referring to fig. 1, a flow chart of a rock wave velocity testing method according to an embodiment of the present application is shown. The order of the steps in the flow diagrams may be changed, and some steps may be omitted, according to different needs.
S11, acquiring real-time environment parameters, and mapping the real-time environment parameters to a low-dimensional parameter space by using a dimension reduction algorithm, wherein the real-time environment parameters comprise temperature and humidity, rock porosity, rock density, rock saturation and rock composition.
In one embodiment, in a rock wave speed testing process, a plurality of rock samples of a research area are collected, and rock wave speed testing is carried out on the rock samples, so that the wave speed of sound waves in the rock samples is obtained, wherein the wave speed is a transverse wave speed or a longitudinal wave speed; wherein, rock samples of the research area can be obtained by drilling, core collection and the like.
And collecting real-time environment parameters in the rock wave speed testing process, wherein the real-time environment parameters are used for reflecting the testing environment and rock properties of the rock wave speed testing process, and the real-time environment parameters comprise temperature and humidity, rock porosity, rock density, rock saturation and rock composition.
The temperature and humidity can be obtained through a temperature sensor and a humidity sensor and are used for reflecting the temperature and humidity of the test environment.
The rock porosity can be obtained by a saturation method, and the concrete process is as follows: the rock sample is firstly subjected to degassing treatment under vacuum or pressure to remove gas in pores, then the rock sample is soaked in liquid to be fully saturated, and the ratio of the mass difference of the rock sample before and after soaking to the mass of the rock sample before soaking is porosity.
The rock saturation can be obtained by a saturation method as well, and the concrete process is as follows: firstly, degassing a rock sample under vacuum or pressure to remove gas in pores, and then soaking the rock sample in liquid to fully saturate the rock sample; after soaking, the rock sample is taken out of the liquid and the wet weight is rapidly measured, then the dry weight of the rock sample is measured by using a drying method, and the saturation of the rock sample is calculated according to the ratio of the wet weight to the dry weight.
The rock components can be obtained by an energy spectrum analysis method or an X-ray diffraction analysis method, wherein the X-ray diffraction analysis method uses an X-ray diffractometer to measure the diffraction pattern of the rock sample, and the mineral components and the content in the rock sample are determined by comparison with a standard library; the energy spectrum analysis method uses a scanning electron microscope to observe a rock sample, and the elemental composition and distribution of the surface of the rock sample are measured through energy spectrum analysis, so that the mineral components and the content in the rock sample are determined.
It will be appreciated that the real-time environmental parameters include a variety of parameters that affect the accuracy of the test during rock wave velocity testing.
In one embodiment, the real-time environment parameter is a high-dimensional information containing multiple parameters, and in order to facilitate calculation of the complexity of subsequent measurement, the real-time environment parameter needs to be mapped to a low-dimensional parameter space by using a dimension-reduction algorithm, and preferably, the low-dimensional parameter space is a two-dimensional space or a three-dimensional space.
The dimension reduction algorithm is a PCA algorithm or a self-coding network. The PCA algorithm (Principal Component Analysis ) is one of the most widely used data dimension reduction algorithms, and will not be described in detail herein.
Specifically, the dimension-reduction algorithm is a self-coding network, and the mapping the real-time environment parameter to a low-dimension parameter space by using the dimension-reduction algorithm includes: building a self-coding network, wherein the self-coding network comprises a parameter encoder and a parameter decoder, and the output of the parameter encoder is a low-dimensional vector; collecting a plurality of groups of environment parameter samples, sequentially inputting the plurality of groups of environment parameter samples into the parameter encoder to obtain a low-dimensional vector, and inputting the low-dimensional vector into the parameter decoder to obtain a decoding result; calculating a reconstruction loss value based on the decoding result and the environmental parameter samples; back-propagating according to the reconstruction loss value to update the parameter encoder and the parameter decoder; iteratively updating the parameter encoder and the parameter decoder until the reconstruction loss value is smaller than a preset loss value, thereby obtaining a trained parameter encoder; inputting the real-time environment parameters into a trained parameter encoder to output low-dimensional vectors corresponding to the real-time environment parameters, wherein the low-dimensional vectors correspond to the position points of the real-time environment parameters in the low-dimensional parameter space. Wherein the low-dimensional vector is a two-dimensional vector or a three-dimensional vector
Wherein the reconstruction loss value satisfies the relationship:
wherein,for the number of said sets of environmental parameter samples, +.>Is->Group environmental parameter sample, ++>Is->Decoding result of group environmental parameter samples, +.>Reconstructing the loss value. The value of the preset loss value is 0.001.
The parameter encoder and the parameter decoder in the self-coding network are all fully-connected neural networks.
As can be appreciated, the low-dimensional vector output by the parameter encoder corresponds to a location point in the low-dimensional parameter space; and responding to the low-dimensional vector as a two-dimensional vector, wherein the low-dimensional parameter space is a two-dimensional space, and responding to the low-dimensional vector as a three-dimensional vector, wherein the low-dimensional parameter space is a three-dimensional space. For example, the real-time environment parameter corresponds to a low-dimensional vector of [0.4,0.5], and the position point of the real-time environment parameter in the low-dimensional parameter space is [0.4,0.5].
Thus, the real-time environment parameters in the rock wave speed testing process are collected, the real-time environment parameters are mapped to the low-dimensional parameter space, and the position points of the real-time environment parameters in the low-dimensional parameter space are obtained.
S12, constructing a test sample graph, wherein the test sample graph comprises historical test results of each position point in the low-dimensional parameter space in the historical rock wave speed test process.
In one embodiment, a plurality of historical environmental parameters and historical test results of each historical environmental parameter are collected during the historical rock wave speed test; the historical test result is the rock wave speed under the historical environment parameters. According to the historical rock wave velocity testing process, historical testing results of each position point in the low-dimensional parameter space can be obtained, and then a testing sample graph is constructed.
As described below, referring to fig. 2, a flowchart of constructing a test sample graph according to an embodiment of the present application is shown. The constructing a test sample graph includes: s21, constructing a test sample initial diagram, wherein the test sample initial diagram comprises all position points in the low-dimensional parameter space, and the numerical value of each position point is 0; s22, in the random historical rock wave speed testing process, acquiring a position point of a historical environment parameter in the initial diagram of the test sample, and updating the numerical value of the position point according to a historical test result, wherein the historical test result is the rock wave speed under the historical environment parameter; s23, traversing all historical rock wave speed testing processes, and continuously updating the numerical value of each position point in the initial diagram of the test sample; and S24, in the updated initial diagram of the test sample, marking any position point with the value of 0 as a zero position point, and determining the value of the zero position point by utilizing an interpolation algorithm until all zero position points in the updated initial diagram of the test sample are traversed, so as to obtain the diagram of the test sample.
Wherein updating the numerical value of the location point according to the historical test result comprises: and taking the numerical value before updating the position point and the average value of the historical test result as the numerical value after updating the position point. The interpolation algorithm adopts a linear interpolation algorithm or a nonlinear interpolation algorithm, and the application is not limited.
In other alternative embodiments, the updating the numerical value of the location point according to the historical test result includes: taking all the historical test results corresponding to the position points as a test result set of the position points; and taking the mode in the test result set as the numerical value after the position point is updated.
Exemplary, please refer to fig. 3, which is a schematic diagram of a test sample graph according to an embodiment of the present application. The low-dimensional parameter space is a two-dimensional space, one position point in the test sample graph corresponds to one environment parameter, and the numerical value of the one position point corresponds to a historical test result under the environment parameter corresponding to the position point; the unit of the numerical value in the test sample graph is kilometers per second, and the numerical value in the first row and the first column in the test sample graph is 3.5, namely, the historical test result is 3.5 kilometers per second under the corresponding environmental parameters of the first row and the first column.
In this way, a test sample diagram is constructed in the low-dimensional parameter space according to all the historical rock wave speed test processes, the test sample diagram comprises all the position points in the low-dimensional parameter space, one position point corresponds to one environment parameter of one rock wave speed test, and the numerical value of each position point can accurately reflect the historical test result under the corresponding environment parameter.
S13, inquiring the test sample graph based on the position points of the real-time environment parameters in the low-dimensional parameter space to determine the measurement complexity and the wave speed predicted value of the real-time environment parameters.
In one embodiment, the real-time environment parameter is mapped to a low-dimensional parameter space in step S11, the location point of the real-time environment parameter in the low-dimensional parameter space is determined, and the measurement complexity and the wave speed predicted value of the real-time environment parameter can be determined by querying a test sample graph based on the location point.
The method for determining the measurement complexity and the wave velocity predicted value of the real-time environment parameter based on the position point of the real-time environment parameter in the low-dimensional parameter space comprises the following steps: taking the numerical value of a target position point in the test sample graph as a wave velocity predicted value of the real-time environment parameter, wherein the target position point is a position point of the real-time environment parameter in the low-dimensional parameter space; and calculating variances of all historical test results in the neighborhood range of the target position point in the test sample graph to serve as the measurement complexity of the real-time environment parameters.
The neighborhood range of the target position point is a rectangular range with the target position point as the center and the set size. For example, when the low-dimensional parameter space is a two-dimensional space, the neighborhood range of the target position point is a rectangular range of 3×3 centered on the target position point; when the low-dimensional parameter space is a three-dimensional space, the neighborhood range of the target position point is targeted the position point is 3× center rectangular range of 3 x 3.
The smaller the variance of all the historical test results in the neighborhood range of the target location point is, the smaller the change of the wave speed test results in the neighborhood range of the target location point is, for example, when the variance is 0, the wave speed test results in the neighborhood range of the target location point are the same; that is, the smaller the variance, the smaller the variation of all the historical test results in the neighborhood of the target location point, the easier it is to obtain accurate wave velocity measurements under real-time environmental parameters, which are less complex to measure.
In this way, the test sample graph is queried according to the position points of the real-time environment parameters in the low-dimensional parameter space, the test sample graph comprises the historical test results of the position points corresponding to the real-time environment parameters, and the measurement complexity and the wave speed predicted value under the real-time environment parameters are determined according to the numerical values of the position points corresponding to the test sample graph and the numerical variance in the neighborhood range.
S14, collecting the wave velocity measured value of the real-time environment parameter.
In one embodiment, the rock wave velocity testing operation is performed under real-time environmental parameters, resulting in wave velocity measurements of the rock sample under real-time environmental parameters.
The rock wave velocity testing operation can be implemented by means of a rock acoustic wave parameter measuring instrument, and the description of the application is omitted.
And S15, carrying out weighted summation on the wave speed measured value and the wave speed predicted value based on the measurement complexity to obtain a wave speed test result.
In one embodiment, the greater the measurement complexity, the more likely the measured wave velocity value representing the real-time environmental parameter is to be in error, and in order to reduce the error of the measured wave velocity value, the weight of the measured wave velocity value should be reduced when determining the final measured wave velocity result, so as to further improve the accuracy of the measured wave velocity result.
The wave speed test result satisfies the relation:
wherein,for the measurement complexity of said real-time environmental parameter, < >>For wave velocity measurement, +.>As the predicted value of the wave velocity,and the wave speed test result is a transverse wave speed or a longitudinal wave speed.
Understandably, the measurement complexity of real-time environmental parametersThe larger the value, the lower the accuracy of the wave speed measurement, at this time the weight of the wave speed measurement is +.>Smaller wave speed predictionWeight of value->The smaller the accuracy of the wave speed test result is ensured.
In an alternative embodiment, after obtaining the wave speed test result, the test method further includes: and updating the numerical value of a target position point in the test sample graph according to the wave speed test result, wherein the target position point is the position point of the real-time environment parameter in the low-dimensional parameter space.
The process of "updating the value of the target position point in the test sample graph according to the wave speed test result" is the same as the process of "updating the value of the position point according to the history test result" in step S22, and will not be described in detail here.
After the wave speed test result under the real-time environment parameter is obtained, the real-time environment parameter and the wave speed test result can be regarded as a historical rock wave speed test process, the numerical value of the corresponding position point of the real-time environment parameter in the test sample graph is updated according to the wave speed test result, the real-time update of the test sample graph is realized, the accuracy of the measurement complexity and the wave speed predicted value is ensured, and the accuracy of the wave speed test result is further ensured.
Thus, according to the measurement complexity, the actually acquired wave speed measured value under the real-time environment parameter and the wave speed predicted value corresponding to the real-time environment parameter in the wave speed testing process of the historical rock are fused, and an accurate wave speed testing result is obtained.
According to the rock wave velocity testing method provided by the embodiment of the application, firstly, a dimension reduction algorithm is utilized to reduce the dimension of real-time environment parameters to a low-dimension parameter space, and position points of the real-time environment parameters in the low-dimension parameter space are obtained; determining a historical test result of each position point in the low-dimensional parameter space based on all historical rock wave velocity test processes, further constructing a test sample diagram, and obtaining a wave velocity predicted value of the position point where the real-time environment parameter is located and a variance of the historical test result in a neighborhood range of the position point where the real-time environment parameter is located by inquiring the test sample diagram; taking the variance of the historical test result in the neighborhood range of the position point as the measurement complexity of the real-time environment parameter, and carrying out weighted summation on the wave velocity measured value and the wave velocity predicted value according to the measurement complexity of the real-time environment parameter, so that an accurate wave velocity test result is obtained, and the accuracy of the wave velocity test result of the rock is improved.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the claims. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application.
Claims (9)
1. A rock wave velocity testing method, the testing method comprising:
collecting real-time environment parameters, and mapping the real-time environment parameters to a low-dimensional parameter space by using a dimension reduction algorithm, wherein the real-time environment parameters comprise temperature and humidity, rock porosity, rock density, rock saturation and rock composition;
constructing a test sample graph, wherein the test sample graph comprises historical test results of each position point in the low-dimensional parameter space in the historical rock wave speed test process;
querying the test sample graph based on the location points of the real-time environmental parameters in the low-dimensional parameter space to determine measurement complexity and wave speed predicted values of the real-time environmental parameters;
collecting wave velocity measurement values of the real-time environment parameters;
and carrying out weighted summation on the wave speed measured value and the wave speed predicted value based on the measurement complexity to obtain a wave speed test result.
2. The rock wave velocity testing method of claim 1, wherein the dimension-reduction algorithm is a self-encoding network, and wherein the mapping the real-time environmental parameters to a low-dimensional parameter space using the dimension-reduction algorithm comprises:
building a self-coding network, wherein the self-coding network comprises a parameter encoder and a parameter decoder, and the output of the parameter encoder is a low-dimensional vector;
collecting a plurality of groups of environment parameter samples, sequentially inputting the plurality of groups of environment parameter samples into the parameter encoder to obtain a low-dimensional vector, and inputting the low-dimensional vector into the parameter decoder to obtain a decoding result;
calculating a reconstruction loss value based on the decoding result and the environmental parameter samples;
back-propagating according to the reconstruction loss value to update the parameter encoder and the parameter decoder;
iteratively updating the parameter encoder and the parameter decoder until the reconstruction loss value is smaller than a preset loss value, thereby obtaining a trained parameter encoder;
inputting the real-time environment parameters into a trained parameter encoder to output low-dimensional vectors corresponding to the real-time environment parameters, wherein the low-dimensional vectors correspond to the position points of the real-time environment parameters in the low-dimensional parameter space;
wherein the low-dimensional vector is a two-dimensional vector or a three-dimensional vector.
3. The rock wave velocity testing method of claim 2, wherein the reconstruction loss value satisfies the relationship:
wherein,for the number of said sets of environmental parameter samples, +.>Is->Group environmental parameter sample, ++>Is->Decoding result of group environmental parameter samples, +.>Reconstructing the loss value.
4. The rock wave velocity testing method of claim 1, wherein constructing the test sample map comprises:
constructing a test sample initial diagram, wherein the test sample initial diagram comprises all position points in the low-dimensional parameter space, and the numerical value of each position point is 0;
in the random historical rock wave speed testing process, acquiring position points of historical environmental parameters in the initial diagram of the test sample, and updating the numerical values of the position points according to the historical test results, wherein the historical test results are rock wave speeds under the historical environmental parameters;
traversing all the wave speed testing processes of the historical rock, and continuously updating the numerical value of each position point in the initial diagram of the test sample;
and in the updated initial diagram of the test sample, marking any position point with the value of 0 as a zero position point, and determining the value of the zero position point by using an interpolation algorithm until all the zero position points in the updated initial diagram of the test sample are traversed, so as to obtain the diagram of the test sample.
5. The method of claim 4, wherein updating the value of the location point based on the historical test results comprises:
and taking the numerical value before updating the position point and the average value of the historical test result as the numerical value after updating the position point.
6. The method of claim 4, wherein updating the value of the location point based on the historical test results comprises:
taking all the historical test results corresponding to the position points as a test result set of the position points;
and taking the mode in the test result set as the numerical value after the position point is updated.
7. The rock wave velocity testing method of claim 5 or 6, wherein querying the test sample graph based on the location point of the real-time environmental parameter in the low-dimensional parameter space to determine the measurement complexity and the wave velocity prediction value of the real-time environmental parameter comprises:
taking the numerical value of a target position point in the test sample graph as a wave velocity predicted value of the real-time environment parameter, wherein the target position point is a position point of the real-time environment parameter in the low-dimensional parameter space;
calculating variances of all historical test results in a neighborhood range of target position points in the test sample graph to serve as measurement complexity of the real-time environment parameters;
the neighborhood range of the target position point is a rectangular range with the target position point as the center and the set size.
8. The rock wave velocity testing method of claim 7, wherein the wave velocity test result satisfies the relationship:
wherein,for the measurement complexity of said real-time environmental parameter, < >>For wave velocity measurement, +.>For the wave velocity prediction value, +.>And the wave speed test result is a transverse wave speed or a longitudinal wave speed.
9. The rock wave velocity testing method according to claim 7, wherein after obtaining the wave velocity test result, the testing method further comprises:
and updating the numerical value of the target position point in the test sample graph according to the wave speed test result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410053198.3A CN117572503B (en) | 2024-01-15 | 2024-01-15 | Rock wave velocity testing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410053198.3A CN117572503B (en) | 2024-01-15 | 2024-01-15 | Rock wave velocity testing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117572503A true CN117572503A (en) | 2024-02-20 |
| CN117572503B CN117572503B (en) | 2024-03-26 |
Family
ID=89884780
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410053198.3A Active CN117572503B (en) | 2024-01-15 | 2024-01-15 | Rock wave velocity testing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117572503B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120272743A1 (en) * | 2011-04-27 | 2012-11-01 | Xiaoqing Sun | Method and Apparatus for Laser-Based Non-Contact Three-Dimensional Borehole Stress Measurement and Pristine Stress Estimation |
| CN104975851A (en) * | 2014-04-10 | 2015-10-14 | 中国石油集团东方地球物理勘探有限责任公司 | Oil reservoir model optimization method for AVO trace gather analysis |
| CN105089615A (en) * | 2014-05-16 | 2015-11-25 | 中国石油化工股份有限公司 | Log data historical retrogression treatment method based on oil reservoir model |
| CN108802814A (en) * | 2018-06-20 | 2018-11-13 | 成都理工大学 | A kind of acquisition methods of tunnel surrounding microseism velocity of wave |
| WO2021232924A1 (en) * | 2020-05-21 | 2021-11-25 | 中国矿业大学 | Deep-ultra-deep rock mechanics parameter prediction method in consideration of temperature effects |
| CN115575505A (en) * | 2022-10-10 | 2023-01-06 | 四川大学 | Method for calculating longitudinal wave velocity and attenuation of rock under stress action condition |
-
2024
- 2024-01-15 CN CN202410053198.3A patent/CN117572503B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120272743A1 (en) * | 2011-04-27 | 2012-11-01 | Xiaoqing Sun | Method and Apparatus for Laser-Based Non-Contact Three-Dimensional Borehole Stress Measurement and Pristine Stress Estimation |
| CN104975851A (en) * | 2014-04-10 | 2015-10-14 | 中国石油集团东方地球物理勘探有限责任公司 | Oil reservoir model optimization method for AVO trace gather analysis |
| CN105089615A (en) * | 2014-05-16 | 2015-11-25 | 中国石油化工股份有限公司 | Log data historical retrogression treatment method based on oil reservoir model |
| CN108802814A (en) * | 2018-06-20 | 2018-11-13 | 成都理工大学 | A kind of acquisition methods of tunnel surrounding microseism velocity of wave |
| WO2021232924A1 (en) * | 2020-05-21 | 2021-11-25 | 中国矿业大学 | Deep-ultra-deep rock mechanics parameter prediction method in consideration of temperature effects |
| CN115575505A (en) * | 2022-10-10 | 2023-01-06 | 四川大学 | Method for calculating longitudinal wave velocity and attenuation of rock under stress action condition |
Non-Patent Citations (2)
| Title |
|---|
| 吕晶 等: "LM-BP神经网络在泥页岩地层横波波速拟合中的应用", 《中国石油大学学报(自然科学版)》, vol. 41, no. 03, 30 June 2017 (2017-06-30), pages 75 - 83 * |
| 张逸凌 等: "高温高压下岩石波速研究进展", 《地球物理学进展》, vol. 38, no. 5, 31 December 2023 (2023-12-31), pages 1999 - 2022 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117572503B (en) | 2024-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Acar et al. | Models to estimate the elastic modulus of weak rocks based on least square support vector machine | |
| CN105444923A (en) | Mechanical temperature instrument error prediction method based on genetic-algorithm optimized least square support vector machine | |
| CN111208565B (en) | KT model-based hole seam parameter inversion method and device and storage medium | |
| CN115935139A (en) | Space field interpolation method for ocean observation data | |
| CN112836789A (en) | Ground connection wall deformation dynamic prediction method based on composite neural network algorithm | |
| CN117572503B (en) | Rock wave velocity testing method | |
| CN116975987B (en) | Deep water shallow geotechnical engineering parameter prediction method and device based on acoustic characteristics | |
| CN111208566B (en) | Hole seam parameter inversion method and device based on SCA model and storage medium | |
| Zhang et al. | Reconstruction of porous media using an information variational auto-encoder | |
| CN113468804A (en) | Underground pipeline identification method based on matrix bundle and deep neural network | |
| Xia et al. | Novel intelligent approach for peak shear strength assessment of rock joints on the basis of the relevance vector machine | |
| CN116960962A (en) | Mid-long term area load prediction method for cross-area data fusion | |
| CN112016235B (en) | Impact load identification method and system for flexible antenna structure | |
| CN114460639B (en) | Shale oil reservoir permeability prediction method and device | |
| CN115047075A (en) | Tree detection method and device and tree detection equipment | |
| Uzundurukan | Prediction of soil–water characteristic curve for plastic soils using PSO algorithm | |
| Jarvis et al. | Empirical realised niche models for British coastal plant species | |
| CN117405175B (en) | Intelligent marine environment monitoring system | |
| CN117935972B (en) | Seawater suspended sand concentration calculation method and device, electronic equipment and storage medium | |
| Pietruszczak et al. | Analysis of Selected Dynamic Properties of Fractional Order Accelerometers for Application in Telematic Equipment | |
| CN117709488B (en) | Dam seepage prediction method based on RUN-XGBoost | |
| CN112577450B (en) | Engineering rock mass structural surface roughness coefficient determination method based on multiple regression analysis | |
| CN116699696A (en) | Reservoir distribution prediction method, device and storage medium | |
| Smith | Visualizing predictability with chaotic ensembles | |
| Kanevski et al. | Mapping of radioactively contaminated territories with geostatistics and artificial neural networks |
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 |