CN115685335A - Longitudinal and transverse wave velocity prediction method and device, computer equipment and storage medium - Google Patents
Longitudinal and transverse wave velocity prediction method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN115685335A CN115685335A CN202211419409.8A CN202211419409A CN115685335A CN 115685335 A CN115685335 A CN 115685335A CN 202211419409 A CN202211419409 A CN 202211419409A CN 115685335 A CN115685335 A CN 115685335A
- Authority
- CN
- China
- Prior art keywords
- neural network
- deep neural
- target
- model
- logging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The embodiment of the invention discloses a method and a device for predicting longitudinal and transverse wave speeds, computer equipment and a storage medium. The method comprises the following steps: constructing a deep neural network model according to the rock physical model and training; solving based on the optimized logging explanation to obtain a target logging response parameter; and inputting the target logging response parameters into the trained deep neural network model to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value. According to the technical scheme provided by the embodiment of the invention, by integrating the advantages of a deep neural network and rock physics, a network structure is designed according to a rock physics model, the rock physics model is used for driving the prediction of longitudinal wave information and transverse wave information, different from an accurate physical model, only a fuzzy solution set needs to be defined, and compared with the conventional method, the method has the advantages of high efficiency and accuracy, and the application threshold is greatly reduced. Therefore, the reservoir prediction and construction fine implementation can be efficiently guided, and the development of the intelligent exploration field is promoted.
Description
Technical Field
The embodiment of the invention relates to the technical field of geological exploration, in particular to a method and a device for predicting longitudinal and transverse wave speeds, computer equipment and a storage medium.
Background
The lack of longitudinal wave information and transverse wave information can restrict the development of drilling decision, construction fine implementation, reservoir prediction and other works, the reservoir fluid prediction precision can be effectively improved through the combination of the longitudinal wave information and the transverse wave speed information, and the multi-solution of seismic amplitude interpretation is reduced. Many studies have been conducted on the prediction of longitudinal and transverse waves, and Gardner proposed an empirical formula between density and velocity of longitudinal waves in 1974 for some specific layers and environments; han, castagna and the like are based on a large number of rock physical experiments, and the empirical relationship between the longitudinal wave speed and the transverse wave speed of the sand shale is counted; xu-White and the like are combined with a Kuster formula, a differential equivalent medium theory and a Gassmann equation to establish a theoretical model of the sandstone containing mud for predicting longitudinal and transverse waves; and the Wangxiang light adopts an adaptive BP neural network to predict the transverse wave speed. In the above, the conventional longitudinal and transverse wave prediction methods can be summarized into three types, namely empirical formulas, neural networks and rock physical modeling.
The traditional empirical formula realizes speed prediction by fitting the relation between speed and density, but the considered factors are simple, and the accuracy is low; the traditional neural network algorithm obtains the mapping relation by fitting a function between a plurality of curves and the speed, but requires a certain number of wells, and has low precision and unknown physical significance; although the physical basis of the rock physical method is complete, the rock components, the pore fluid components, the volume modulus, the shear modulus and other elastic moduli are fully considered, the speciality is too strong, the efficiency is lower, the requirements on well logging interpretation and mineral modulus accuracy are higher, and certain limitations are realized.
Disclosure of Invention
The embodiment of the invention provides a method and a device for predicting longitudinal and transverse wave speeds, computer equipment and a storage medium, which are used for improving the prediction precision and the prediction efficiency and reducing the application threshold.
In a first aspect, an embodiment of the present invention provides a method for predicting a velocity of a longitudinal wave and a transverse wave, where the method includes:
constructing a deep neural network model according to the rock physical model and training;
solving based on the optimized logging explanation to obtain a target logging response parameter;
and inputting the target logging response parameters into the trained deep neural network model to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value.
Optionally, the deep neural network model is four layer network structure, includes input layer, first hidden layer, second hidden layer and output layer in proper order, the input layer is the logging optimization solution, first hidden layer is fluid, mineral mixture, the second hidden layer is the fluid replacement, the output layer is the vertical, shear wave velocity value of finally seeking.
Optionally, the target logging response parameter includes: density, natural gamma, resistivity, and compensated neutrons.
Optionally, the objective function of the optimized well logging interpretation is:
the volume model of the optimized well log interpretation is:
wherein F represents the objective function,
V j represents the relative content of the jth mineral, and the total n minerals,
to representFluid content, P i,j The ith logging response parameter of the jth mineral is shown, the logging response parameters are m in total,
i-th log response parameter, f, representing the fluid i Representing the i-th actual log measurement, f, after environmental correction i (x) Theoretical calculation, σ, representing the ith log response equation i Error of the ith logging response equation is represented.
Optionally, the functional relationship of the petrophysical model includes:
M＝(M V +M R )/2
wherein ρ sat Representing the fluid saturated rock density, p ma Which is indicative of the density of the rock matrix,
denotes porosity, p fl Denotes the pore fluid density, M denotes the equivalent matrix modulus, M V Denotes the Voigt mean value, M R Represents the Reuss mean.
Optionally, the constructing and training the deep neural network model according to the petrophysical model includes:
acquiring a rock electricity parameter rule of a region to be detected;
and initializing the network weight of the deep neural network model according to the lithoelectric parameter rule.
Optionally, the constructing and training the deep neural network model according to the petrophysical model by using an elu activation function includes:
and training the deep neural network model by adopting an RMSprop method.
In a second aspect, an embodiment of the present invention further provides a device for predicting a longitudinal wave velocity, where the device includes:
the prediction model construction module is used for constructing a deep neural network model according to the rock physical model and training the deep neural network model;
the logging parameter solving module is used for solving and obtaining target logging response parameters based on the optimized logging interpretation;
and the target speed prediction module is used for inputting the target logging response parameters into the trained deep neural network model so as to predict and obtain a target longitudinal wave speed value and a target transverse wave speed value.
In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a method of compressional velocity prediction as provided by any embodiment of the present invention.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for predicting a velocity of a longitudinal wave and a transverse wave provided in any embodiment of the present invention.
The embodiment of the invention provides a longitudinal wave velocity prediction method and a transverse wave velocity prediction method. According to the method for predicting the longitudinal wave velocity and the transverse wave velocity, the advantages of a deep neural network and rock physics are integrated, a network structure is designed according to a rock physics model, the rock physics model is used for driving prediction of longitudinal wave information and transverse wave information, the method is different from an accurate physics model, only a fuzzy solution set needs to be defined, compared with the conventional method, the method has the advantages of high efficiency and accuracy, and an application threshold is greatly reduced. Therefore, the reservoir prediction and construction fine implementation can be efficiently guided, and the development of the intelligent exploration field is promoted.
Drawings
Fig. 1 is a flowchart of a method for predicting a velocity of a longitudinal wave and a transverse wave according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a longitudinal-transverse wave velocity prediction apparatus according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a computer device according to a third embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the steps as a sequential process, many of the steps can be performed in parallel, concurrently or simultaneously. In addition, the order of the steps may be rearranged. The process may be terminated when its operations are completed, but could have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
Example one
Fig. 1 is a flowchart of a method for predicting a velocity of a longitudinal wave and a transverse wave according to an embodiment of the present invention. The method can be executed by the longitudinal and transverse wave velocity prediction device provided by the embodiment of the invention, the device can be realized by hardware and/or software, and the device can be generally integrated in computer equipment. As shown in fig. 1, the method specifically comprises the following steps:
s11, constructing a deep neural network model according to the rock physical model and training.
The rock physical model is one of important bases and main methods for rock physical research, fully considers rock mineral components and pore structures, and idealized key factors such as geometric shapes and arrangement modes of particles and pores in actual rocks through certain assumed conditions, and then establishes a universal functional relation by adopting a physical principle and a mathematical method based on the theoretical model. The method specifically can adopt an Xu-White rock physical model, which essentially separates the influence of pore fluid on rock elasticity, and correspondingly, optionally, the functional relationship of the rock physical model includes:
M＝(M V +M R )/2
wherein ρ sat Representing the fluid saturated rock density, p ma Which is indicative of the density of the rock matrix,
denotes porosity, p fl Denotes pore fluid density, M denotes equivalent matrix modulus, M V Denotes the Voigt mean value, M R Represents the Reuss mean. That is, the fluid saturation rock density can be obtained from the volume weighted average of the rock matrix (constituent mineral) density and porosity, while the bulk modulus and shear modulus of the rock matrix can be calculated from the Voigt-reus-Hill average. As the functional relation is described, the Xu-White rock physical model comprises the processes of fluid mixing, skeleton mixing, pore addition, fluid replacement and the like, the physical basis is complete, various theoretical models are involved, and the modeling process is also very suitable for being characterized by a neural network.
Unlike the current situation of the deep network structure determined by experiments in the industry, the present embodiment may integrate the above physical processes into the deep neural network structure based on the contents of the petrophysical model. Optionally, the deep neural network model is four layer network structure, includes input layer, first hidden layer, the hidden layer of second and output layer in proper order, the input layer is the logging optimization solution, first hidden layer is fluid, mineral mixture, the hidden layer of second is the fluid replacement, the output layer is for finally seeking indulging, the shear wave velocity value. Through tests, compared with a traditional network structure determination mode, the stability and the precision of a prediction result are obviously improved, and the correlation coefficient between a target longitudinal wave speed value obtained through final prediction and a corresponding actual measurement result is also improved from 68% to 78%.
Optionally, the constructing and training a deep neural network model according to the petrophysical model includes: acquiring a rock electricity parameter rule of a region to be detected; and initializing the network weight of the deep neural network model according to the lithoelectric parameter rule. Specifically, the currently common neural network weight coefficient initialization method is random initialization, and for the network structure designed in this embodiment, the initialization parameters themselves have geological significance, so that the network weight of the deep neural network model can be initialized based on the guidance of the regional geological law, so as to reduce the initial error and improve the learning efficiency. The law of the electrical parameters of the rock can include the density, natural gamma and resistivity of some minerals and fluids (such as clay, quartz, limestone, oil and water), and the initial error is reduced from 0.349 to 0.276 compared with the conventional initialization method through tests.
Further optionally, the constructing and training the deep neural network model according to the petrophysical model by using an elu activation function includes: and training the deep neural network model by adopting an RMSprop method. Specifically, due to the adoption of the back propagation mode for training, the network performance of the traditional neural network is rapidly reduced along with the increase of the number of network layers. During training, the gradient signals must be propagated back from the topmost to the bottommost layer of the network to ensure that the network itself can be updated correctly. In a conventional network, the gradient signal is slightly reduced as it passes through each layer of the network, and the more layers of the network, the faster the gradient signal decays. Therefore, the deep neural network model designed by the embodiment can use the elu activation function to reduce the gradient disappearance problem, and the training process can also adopt the RMSprop method to improve the learning efficiency of the deep neural network model so as to further reduce the residual error. Through tests, compared with the traditional training process, the back iteration speed can be doubled, the residual error is reduced by 40%, and a better effect is achieved.
And S12, solving to obtain target logging response parameters based on the optimized logging interpretation.
The optimized logging interpretation is a linear solving process, and optionally, the objective function of the optimized logging interpretation is:
the volume model of the optimized well log interpretation is:
wherein F represents the objective function,
V j represents the relative content of the jth mineral, and the total n minerals,
denotes the fluid content, P i,j The ith logging response parameter of the jth mineral is shown, the logging response parameters are m in total,
i-th log response parameter, f, representing the fluid i Representing the i-th actual log measurement, f, after environmental correction i (x) Theoretical calculation, σ, representing the ith log response equation i Error of the ith logging response equation is represented. Specifically, for a certain region to be measured, the optimal values of the logging response parameters can be obtained by calculation based on the objective function and the volume model, and then the objective logging response parameters are selected as the designed deep neural network model input layerAnd (5) required logging optimization solution. Optionally, the target logging response parameter includes: density, natural gamma, resistivity, and compensated neutrons.
And S13, inputting the target logging response parameters into the trained deep neural network model to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value.
Specifically, after the target logging response parameters are obtained through solving, the target logging response parameters can be input into the trained deep neural network model, and the output of the model is the target longitudinal wave velocity value and the target transverse wave velocity value which are finally obtained.
On the basis of the technical scheme, the validity of the prediction result can be verified by using actual logging data, and particularly 6 wells can be selected from the shallow layer and the middle-deep layer in the Yangtze river basin (east) for testing. Firstly, an A1 well is taken as a model well, a traditional empirical formula, a traditional neural network, a traditional rock physical model and the model-driven deep learning method provided by the embodiment are respectively adopted to establish a model to predict the sound wave of the B1 well, then the seismic record of the B1 well, the sound wave synthetic record predicted by the Gardner formula method, the sound wave synthetic record predicted by the Emerge neural network method, the sound wave synthetic record predicted by the rock physical model method and the sound wave synthetic record predicted by the deep learning method are respectively recorded, the correlation coefficient between the prediction result of the model-driven deep learning method and the well earthquake of the synthetic record is 77.2%, the accuracy equivalent to that of the rock physical model method is achieved, and the method is greatly better than other methods. Meanwhile, the conditions of sound wave synthesis record calibration and time consumption prediction by different methods are recorded, and aiming at stretching compression, a deep learning method and a rock physical model method are fewer, an empirical formula method is larger, and a traditional neural network method is general; aiming at energy matching, a deep learning method and a rock physical model method are good, an empirical formula method is poor, and a traditional neural network method is general; aiming at time consumption, the rock physical model method needs 1 day of professional personnel to complete, and the deep learning method only needs 1 hour of non-professional personnel to complete. Therefore, the model-driven deep learning method also reduces the application threshold and improves the prediction efficiency. And further applying a model-driven deep learning method to sound wave prediction of drilled wells in the middle-shallow layer and the middle-deep layer, wherein correlation coefficients between a well prediction result obtained by testing and a well logging actual measurement result are respectively 86% and 64%, and the requirements of structure implementation and reservoir prediction are met. The lithology of the middle and deep layers is complex, the well condition is poor, and the quality of an actually measured curve is general, so that the correlation coefficient of the curve is lower than that of the middle and shallow layers.
According to the technical scheme provided by the embodiment of the invention, firstly, a deep neural network model is constructed according to a rock physical model and trained, then, a target logging response parameter is obtained based on optimized logging interpretation solving, and the obtained target logging response parameter is input into the trained deep neural network model so as to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value. By integrating the advantages of the deep neural network and the rock physics, a network structure is designed according to the rock physics model, the rock physics model drives the prediction of longitudinal wave information and transverse wave information, the method is different from an accurate physical model, only a fuzzy solution set needs to be defined, compared with the conventional method, the method has the advantages of high efficiency and accuracy, and the application threshold is greatly reduced. Therefore, the reservoir prediction and construction fine implementation can be efficiently guided, and the development of the intelligent exploration field is promoted.
Example two
Fig. 2 is a schematic structural diagram of a device for predicting a velocity of a longitudinal/transverse wave according to a second embodiment of the present invention, where the device may be implemented by hardware and/or software, and may be generally integrated in a computer device, and is used to execute the method for predicting a velocity of a longitudinal/transverse wave according to any embodiment of the present invention. As shown in fig. 2, the apparatus includes:
the prediction model construction module 21 is used for constructing a deep neural network model according to the rock physical model and training the deep neural network model;
the logging parameter solving module 22 is used for solving and obtaining target logging response parameters based on the optimized logging interpretation;
and the target speed prediction module 23 is configured to input the target logging response parameter into the trained deep neural network model to predict and obtain a target longitudinal wave speed value and a target transverse wave speed value.
According to the technical scheme provided by the embodiment of the invention, firstly, a deep neural network model is constructed according to a rock physical model and trained, then, a target logging response parameter is obtained based on optimized logging interpretation solving, and the obtained target logging response parameter is input into the trained deep neural network model so as to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value. By integrating the advantages of the deep neural network and the rock physics, a network structure is designed according to a rock physics model, the rock physics model is used for driving the prediction of longitudinal wave information and transverse wave information, the method is different from an accurate physical model, only a fuzzy solution set needs to be defined, compared with the conventional method, the method has the advantages of high efficiency and accuracy, and the application threshold is greatly reduced. Therefore, the reservoir prediction and construction fine implementation can be efficiently guided, and the development of the intelligent exploration field is promoted.
On the basis of the technical scheme, the deep neural network model is selectable in a four-layer network structure and sequentially comprises an input layer, a first hidden layer, a second hidden layer and an output layer, wherein the input layer is used for solving the logging optimization, the first hidden layer is formed by mixing fluid and minerals, the second hidden layer is used for replacing the fluid, and the output layer is used for finally solving the longitudinal and transverse wave speed values.
On the basis of the foregoing technical solution, optionally, the target logging response parameter includes: density, natural gamma, resistivity, and compensated neutrons.
On the basis of the above technical solution, optionally, the objective function of the optimized logging interpretation is:
the volume model of the optimized well log interpretation is:
wherein F represents the objective function,
V j represents the relative content of the jth mineral, and the total n minerals,
denotes the fluid content, P i,j The ith logging response parameter of the jth mineral is represented, the logging response parameters are m in number,
i-th log response parameter, f, representing the fluid i Representing the i-th actual log measurement, f, after environmental correction i (x) Theoretical calculation, σ, representing the ith log response equation i Error of the ith logging response equation is expressed.
On the basis of the above technical solution, optionally, the functional relationship of the petrophysical model includes:
M＝(M V +M R )/2
wherein ρ sat Representing the fluid saturated rock density, p ma Which is indicative of the density of the rock matrix,
denotes porosity, p fl Denotes pore fluid density, M denotes equivalent matrix modulus, M V Denotes the Voigt mean value, M R Representing the Reuss mean.
On the basis of the above technical solution, optionally, the prediction model building module 21 includes:
the device comprises a lithoelectric parameter rule obtaining unit, a lithoelectric parameter rule obtaining unit and a control unit, wherein the lithoelectric parameter rule obtaining unit is used for obtaining a lithoelectric parameter rule of a region to be detected;
and the network weight initialization unit is used for initializing the network weight of the deep neural network model according to the lithoelectric parameter rule.
On the basis of the above technical solution, optionally, the deep neural network model uses an elu activation function, and the prediction model building module 21 is specifically configured to:
and training the deep neural network model by adopting an RMSprop method.
The longitudinal and transverse wave velocity prediction device provided by the embodiment of the invention can execute the longitudinal and transverse wave velocity prediction method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
It should be noted that, in the embodiment of the longitudinal and transverse wave velocity prediction apparatus, the included units and modules are merely divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a computer device provided in the third embodiment of the present invention, and shows a block diagram of an exemplary computer device suitable for implementing the embodiment of the present invention. The computer device shown in fig. 3 is only an example, and should not bring any limitation to the function and the scope of use of the embodiments of the present invention. As shown in fig. 3, the computer apparatus includes a processor 31, a memory 32, an input device 33, and an output device 34; the number of the processors 31 in the computer device may be one or more, one processor 31 is taken as an example in fig. 3, the processor 31, the memory 32, the input device 33 and the output device 34 in the computer device may be connected by a bus or in other ways, and the connection by the bus is taken as an example in fig. 3.
The memory 32 is a computer readable storage medium, which can be used to store software programs, computer executable programs, and modules, such as program instructions/modules corresponding to the method for predicting longitudinal and transverse wave velocity in the embodiment of the present invention (for example, the prediction model building module 21, the logging parameter solving module 22, and the target velocity prediction module 23 in the longitudinal and transverse wave velocity prediction apparatus). The processor 31 executes various functional applications and data processing of the computer device by running software programs, instructions and modules stored in the memory 32, that is, implements the above-described method for predicting the velocity of a longitudinal wave and a transverse wave.
The memory 32 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the computer device, and the like. Further, the memory 32 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 32 may further include memory located remotely from the processor 31, which may be connected to a computer device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 33 may be used to obtain the laws of the petroelectrical parameters of the area to be measured, and to generate key signal inputs related to user settings and function controls of the computer device, etc. The output device 34 may include a display screen that may be used to present the predicted results to the user, and the like.
Example four
A fourth embodiment of the present invention further provides a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a method for longitudinal and transverse wave velocity prediction, the method comprising:
constructing a deep neural network model according to the rock physical model and training;
solving based on the optimized logging explanation to obtain a target logging response parameter;
and inputting the target logging response parameters into the trained deep neural network model to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value.
The storage medium may be any of various types of memory devices or storage devices. The term "storage medium" is intended to include: mounting media such as CD-ROM, floppy disk, or tape devices; computer system memory or random access memory such as DRAM, DDR RAM, SRAM, EDO RAM, rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the computer system in which the program is executed, or may be located in a different second computer system connected to the computer system through a network (such as the internet). The second computer system may provide the program instructions to the computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations, such as in different computer systems that are connected via a network. The storage medium may store program instructions (e.g., embodied as a computer program) that are executable by one or more processors.
Of course, the storage medium containing the computer-executable instructions provided by the embodiments of the present invention is not limited to the method operations described above, and may also perform related operations in the method for predicting a velocity of a longitudinal and transverse wave provided by any embodiment of the present invention.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It is to be noted that the foregoing description is only exemplary of the invention and that the principles of the technology may be employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in some detail by the above embodiments, the invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the invention, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. A method for predicting a velocity of a longitudinal wave and a transverse wave, comprising:
constructing a deep neural network model according to the rock physical model and training;
solving based on the optimized logging interpretation to obtain a target logging response parameter;
and inputting the target logging response parameters into the trained deep neural network model to predict and obtain a target longitudinal wave velocity value and a target transverse wave velocity value.
2. The method for predicting the velocity of longitudinal and transverse waves according to claim 1, wherein the deep neural network model is a four-layer network structure and sequentially comprises an input layer, a first hidden layer, a second hidden layer and an output layer, the input layer is used for solving logging optimization, the first hidden layer is a fluid and mineral mixture, the second hidden layer is used for fluid replacement, and the output layer is a finally obtained velocity value of longitudinal and transverse waves.
3. The method of compressional velocity prediction according to claim 1, wherein the target logging response parameters include: density, natural gamma, resistivity, and compensated neutrons.
4. The method of predicting compressional and shear velocity of claim 1, wherein the objective function of the optimized log interpretation is:
the volume model of the optimized logging interpretation is as follows:
where F denotes the objective function, i =1,2, \8230;, m, j =1,2, \8230;, n,
V j represents the relative content of the jth mineral, and the total n minerals,
denotes the fluid content, P i,j The ith logging response parameter of the jth mineral is shown, the logging response parameters are m in total,
the ith logging response parameter, f, representing the fluid i Representing the i-th actual log measurement, f, after environmental correction i (x) Indicates the ith speciesTheoretical calculation of the log response equation, σ i Error of the ith logging response equation is represented.
5. The method of predicting compressional and shear velocity of claim 1, wherein the functional relationship of the petrophysical model comprises:
M＝(M V +M R )/2
where ρ is sat Representing the fluid saturated rock density, p ma Which is indicative of the density of the rock matrix,
denotes porosity, p fl Denotes pore fluid density, M denotes equivalent matrix modulus, M V Denotes the Voigt mean value, M R Represents the Reuss mean.
6. The method for predicting the speed of the longitudinal waves and the transverse waves according to claim 1, wherein the step of constructing a deep neural network model according to a rock physical model and training the deep neural network model comprises the following steps:
acquiring a rock electricity parameter rule of a region to be detected;
and initializing the network weight of the deep neural network model according to the lithoelectric parameter rule.
7. The method for predicting the velocity of the shear waves according to claim 1, wherein the deep neural network model uses an elu activation function, and the building and training of the deep neural network model according to the petrophysical model comprises the following steps:
and training the deep neural network model by adopting an RMSprop method.
8. A longitudinal-transverse wave velocity prediction apparatus, comprising:
the prediction model construction module is used for constructing a deep neural network model according to the rock physical model and training the deep neural network model;
the logging parameter solving module is used for solving and obtaining target logging response parameters based on the optimized logging interpretation;
and the target speed prediction module is used for inputting the target logging response parameters into the trained deep neural network model so as to predict and obtain a target longitudinal wave speed value and a target transverse wave speed value.
9. A computer device, comprising:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of compressional and shear velocity prediction as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method for compressional-shear velocity prediction according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211419409.8A CN115685335A (en) | 2022-11-14 | 2022-11-14 | Longitudinal and transverse wave velocity prediction method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211419409.8A CN115685335A (en) | 2022-11-14 | 2022-11-14 | Longitudinal and transverse wave velocity prediction method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115685335A true CN115685335A (en) | 2023-02-03 |
Family
ID=85051317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211419409.8A Pending CN115685335A (en) | 2022-11-14 | 2022-11-14 | Longitudinal and transverse wave velocity prediction method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115685335A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117388919A (en) * | 2023-08-29 | 2024-01-12 | 河海大学 | DNN-based method for predicting transverse wave speed of tight oil reservoir |
-
2022
- 2022-11-14 CN CN202211419409.8A patent/CN115685335A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117388919A (en) * | 2023-08-29 | 2024-01-12 | 河海大学 | DNN-based method for predicting transverse wave speed of tight oil reservoir |
| CN117388919B (en) * | 2023-08-29 | 2024-05-28 | 河海大学 | DNN-based method for predicting transverse wave speed of tight oil reservoir |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210326721A1 (en) | Earth modeling methods using machine learning | |
| US7676349B2 (en) | Integrated anisotropic rock physics model | |
| EP2598915B1 (en) | Inverting geophysical data for geological parameters or lithology | |
| CN103954999B (en) | A kind of S-Wave Velocity Predicted Method being applicable to low-porosity sand-mudstone formation | |
| Wang et al. | S-wave velocity inversion and prediction using a deep hybrid neural network | |
| Shi et al. | Brittleness index prediction in shale gas reservoirs based on efficient network models | |
| US11699099B2 (en) | Confidence volumes for earth modeling using machine learning | |
| CN104698492A (en) | Method for calculating abnormal stratum pressure | |
| Chen et al. | Porosity prediction from well logs using back propagation neural network optimized by genetic algorithm in one heterogeneous oil reservoirs of Ordos Basin, China | |
| CN106014399A (en) | Method for establishing high-precision three-dimensional ground stress model for heterogeneous formation | |
| CN112965103B (en) | Multi-pore reservoir prestack seismic probability multi-channel inversion method | |
| Zahmatkesh et al. | Estimation of DSI log parameters from conventional well log data using a hybrid particle swarm optimization–adaptive neuro-fuzzy inference system | |
| CN115685335A (en) | Longitudinal and transverse wave velocity prediction method and device, computer equipment and storage medium | |
| Naji et al. | Comparison of Estimation Sonic Shear Wave Time Using Empirical Correlations and Artificial Neural Network | |
| Leisi et al. | Investigation and comparison of conventional methods for estimating shear wave velocity from well logging data in one of the sandstone reservoirs in southern Iran | |
| Baouche et al. | Intelligent methods for predicting nuclear magnetic resonance of porosity and permeability by conventional well-logs: a case study of Saharan field | |
| CN104698494A (en) | Method for calculating abnormal stratum pressure | |
| CN116148924A (en) | Shale layer density prediction method based on statistical petrophysical and related equipment | |
| Naji et al. | Prediction of sonic shear wave using artificial neural network | |
| CN112862169B (en) | Method and device for predicting free oil content of continental phase shale | |
| CN104698493A (en) | Method for calculating abnormal stratum pressure | |
| CN110716235B (en) | Sand shale well logging pore structure inversion method | |
| Zhao et al. | Rock-physics-guided machine learning for shear sonic log prediction | |
| Najjar et al. | Time-lapse seismic programme at Gullfaks: value and the road ahead | |
| Naji et al. | Iraqi Geological Journal |
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 |