CN115542384A - Earthquake inversion method and device based on multitask reversible network - Google Patents
Earthquake inversion method and device based on multitask reversible network Download PDFInfo
- Publication number
- CN115542384A CN115542384A CN202211228562.2A CN202211228562A CN115542384A CN 115542384 A CN115542384 A CN 115542384A CN 202211228562 A CN202211228562 A CN 202211228562A CN 115542384 A CN115542384 A CN 115542384A
- Authority
- CN
- China
- Prior art keywords
- data
- seismic
- reversible
- inversion
- network
- 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
- 230000002441 reversible effect Effects 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000011160 research Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000008878 coupling Effects 0.000 claims abstract description 8
- 238000010168 coupling process Methods 0.000 claims abstract description 8
- 238000005859 coupling reaction Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 32
- 238000005457 optimization Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 230000002457 bidirectional effect Effects 0.000 claims 1
- 238000013507 mapping Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/288—Event detection in seismic signals, e.g. microseismics
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/282—Application of seismic models, synthetic seismograms
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Environmental & Geological Engineering (AREA)
- Acoustics & Sound (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a seismic inversion method and device based on a multitask reversible network, and relates to the technical field of seismic exploration. The method comprises the steps of firstly obtaining well seismic data of a research area, processing the well seismic data to construct a data set, then establishing a multi-task reversible network model, training the constructed data set to obtain a complex relation model between target data and seismic data, and finally performing inversion on actual data to obtain a target data volume. The invention can solve the problem that the conventional inversion method is difficult to fully utilize the interdependence relationship between input and output data and the mutual coupling relationship between different output data, and improve the accuracy of the inversion result.
Description
Technical Field
The invention relates to the technical field of seismic exploration, in particular to a seismic inversion method and device based on a multitask reversible network.
Background
Seismic inversion is one of the important contents of geophysical inversion and plays a crucial role in oil and gas seismic exploration. Scholars at home and abroad develop effective seismic inversion methods through continuous exploration, and obtain good application effects under certain conditions, but have certain limitations. Most of AVO inversion commonly used at present is to establish a model by utilizing an approximate expression of a Zoepppritz equation, and the approximate expression is established under certain assumption, so that a satisfactory result is sometimes difficult to obtain in complex reservoir exploration. In addition, conventional inversion methods often ignore interdependencies between target data and seismic data and intercoupling relationships between different target data.
Disclosure of Invention
The invention provides a seismic inversion method and a seismic inversion device based on a multitask reversible network, and aims to solve or partially solve the problem that the mutual dependency relationship of a forward inversion process and the mutual coupling relationship between different target data are difficult to be fully utilized in a conventional inversion method, and improve the accuracy and efficiency of an inversion result.
In order to achieve the above purpose and achieve the above technical effects, the present invention provides the following technical solutions:
according to one aspect of the invention, a seismic inversion method based on a multitask reversible network is provided, and the method comprises three main steps of data acquisition, model establishment and data inversion:
acquiring data, namely acquiring seismic and well logging data of a research area, and processing the data to construct a data set; the method comprises the following steps: acquiring and processing seismic and well logging data of a research area; converting the processed seismic data into angle gather data to obtain a common angle gather stacking data volume; based on the processed logging data and the data volume overlapped by the gather of different angles, the geological information is synthesized to finely calibrate the horizon; obtaining target data based on the logging data and carrying out scale transformation according to the calibration horizon; constructing a data set based on the obtained target data and the angle gather superposition data;
establishing a model, namely establishing a multi-task reversible network model and training by using the established data set to obtain a complex geophysical relation model between target data and seismic data; the method comprises the following steps: determining the input and output dimensionality of the model according to the multitask target and establishing a reversible module of a reversible network; constructing a multitask reversible network model based on the established reversible module; setting network model parameters; calculating an optimized target value in the network training process; updating network model parameters according to the calculated optimization target value; saving the network model parameters; acquiring a multi-task reversible network inversion model;
data inversion, namely inverting the field seismic data by using the trained multi-task reversible network model to obtain a target data volume of a research area; the method comprises the following steps: and inputting implicit variables output in network training and the extracted angle seismic data into an inversion process of the trained multi-task reversible network model, and performing inversion to obtain a target data volume of a research area, wherein the dimension of forward output needs to be determined according to a target task.
Further preferably, the processing for acquiring the seismic and logging data includes performing trace editing, amplitude compensation, denoising, pre-stack migration, and the like on the seismic data, and performing outlier rejection, correction, lateral normalization, and the like on the logging data.
Preferably, the comprehensive geological information fine calibration layer is calibrated by combining the synthetic seismic records and the well-crossing seismic data, and the geological information and other logging data are checked and adjusted one by one.
As a further preferred aspect of the present invention, the target data may include a longitudinal and transverse wave velocity, a density, and the like.
Further preferably, the scaling is performed by converting the obtained target data into a time domain by time-depth conversion, smoothing the converted data after the time-depth conversion, and down-sampling the smoothed data to have the same size as the angular seismic data.
As a further preferred aspect of the present invention, the constructed data set includes a training set and a test set, wherein the training set is used for training the multitask reversible network model, and the test set is used for testing the performance of the model; the training set is composed of target data and angle gather data, wherein the dimension of forward output needs to be determined according to a target task.
Preferably, in the multitask, a network model is trained in a multitask joint learning mode in a modeling process, and mutual coupling relations among different target data are mined while complex geophysical relations among the characteristic of seismic amplitude and the target data are learned and represented.
As a further preference of the invention, the calculated optimal target value comprises three loss functions, one supervised loss function and two unsupervised loss functions. Wherein the forward process uses a supervised loss function to measure the error between observed and predicted seismic data; the first unsupervised loss function is intended to capture the loss information around the input data in the forward process, introducing additional potential output variables, since some information is lost in the forward process, trained to capture information related to the forward input but not included in the forward output. In addition, a second unsupervised loss function further trains the model by comparing the distribution of the backward prediction and the forward input.
According to another aspect of the present invention, there is provided a seismic inversion apparatus based on a multitasking reversible network, comprising:
the data acquisition module is used for acquiring seismic and well logging data of a research area and processing the data to construct a data set; the method comprises the following steps: acquiring and processing seismic and well logging data of a research area; converting the processed seismic data into angle gather data to obtain a common angle gather stacking data volume; based on the processed logging information and the gather superposition data volume with different angles, the horizon is finely calibrated by integrating geological information; obtaining target data based on the logging data and carrying out scale transformation according to the calibration horizon; constructing a data set based on the obtained target data and the angle gather overlapping data;
the model establishing module is used for establishing a multi-task reversible network model and training by using the established data set to obtain a complex geophysical relation model between target data and seismic data; the method comprises the following steps: determining the input and output dimensionality of the model according to the multitask target and establishing a reversible module of a reversible network; constructing a multitask reversible network model based on the established reversible module; setting network model parameters; calculating an optimization target value in the network training process; updating network model parameters according to the calculated optimization target value; saving the network model parameters; acquiring a multi-task reversible network inversion model;
the data inversion module is used for inverting the field seismic data by utilizing the trained multi-task reversible network model to obtain a target data volume of the research area; the method comprises the following steps: and inputting implicit variables output in network training and the extracted angle seismic data into an inversion process of the trained multi-task reversible network model, and performing inversion to obtain a target data volume of a research area, wherein the dimension of forward output needs to be determined according to a target task.
Compared with the prior art, the invention has the beneficial effects that: the method combines the advantages of multi-task learning and a reversible network, can make up the problem that the conventional inversion method cannot fully utilize the interdependence relation between the seismic data and the target data and the intercoupling relation between different target data, and improves the accuracy and efficiency of the inversion result.
Drawings
FIG. 1 is a flowchart of a seismic inversion method based on a multi-task reversible network according to an embodiment of the present invention;
FIG. 2 is a diagram comparing a single task based learning process with a multi-task based learning process according to an embodiment of the present invention;
FIG. 3 is a diagram of a reversible network provided by an embodiment of the present invention;
fig. 4 is a comparison diagram of different inversion processes provided by the embodiment of the present invention.
Fig. 5 is a structural diagram of a seismic inversion device based on a multitask reversible network according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the specification, and not all of the embodiments.
The multitasking learning and reversible network to which the present application relates is first briefly described as follows:
the multi-task learning aims to utilize the similarity between different tasks to carry out common learning on implicit information in a plurality of related tasks, solve a plurality of different tasks and improve the performance of a single task.
Fig. 2 is a diagram based on a comparison between a single task and a multitask process, where the single task is to treat different target data as isolated tasks, and perform modeling inversion respectively, each network model has only one output target for the same input, there is no connection between different models, and the internal structure of the network models trained with each other has no influence on other network models. Multiple tasks pay attention to the mutual relation among different tasks, rich associated information among different target data can be better mined, and the problem that the mutual coupling relation among different target data cannot be well considered by a single task is solved.
The reversible network optimizes the modeling process through the forward direction and the reverse direction, and uses an additional implicit output variable to capture information which is possibly lost in the forward direction, namely after input data is output through forward propagation, initial input data can be obtained through reverse propagation according to the output in the reverse process, and the input data is not lost in the process.
The basic unit of a reversible network is a reversible module with two complementary affine coupling layers, the forward course of which is shown in fig. 3a, where x 1 And x 2 Is divided into two parts by an input x, the two parts being divided by a learning function t i And s i (i =1,2) are converted and coupled in an alternating manner, x 1 And x 2 Y can be obtained by transforming the following formula 1 And y 2 ：
y 1 ＝x 1 ⊙exp(s 2 (x 2 ))+t 2 (x 2 )
y 2 ＝x 2 ⊙exp(s 1 (x 1 ))+t 1 (x 1 )
Wherein |, indicates element-by-element multiplication. t is t 1 、t 2 、s 1 And s 2 May be any complex function, such as a fully connected network or a convolutional network, and the function itself need not be invertible.
The inverse process of the affine coupling layer is shown in fig. 3b, and the reversible process of the above equation can be obtained by the following equation:
x 1 ＝(y 1 -t 2 (x 2 ))⊙exp(-s 2 (x 2 ))
x 2 ＝(y 2 -t 1 (x 1 ))⊙exp(-s 1 (x 1 ))
the embodiment of the invention provides a seismic inversion method based on a multitask reversible network, and with reference to fig. 1, the specific implementation steps of the method are as follows:
101, acquiring data, namely acquiring seismic and well logging data of a research area, and processing the data to construct a data set; the method comprises the following steps: acquiring earthquake and logging data of a research area, performing channel editing, amplitude compensation, denoising and prestack migration on the earthquake data, and performing outlier rejection, correction and transverse standardization processing on the logging data; converting the processed seismic data into angle gather data to obtain a common-angle gather stacked data volume, wherein the embodiment takes longitudinal and transverse wave velocity and density inversion as an example for explanation, so that stacked data volumes corresponding to different incidence angle ranges need to be extracted; based on the processed logging data and the data volume overlapped by the gather of different angles, the geological information is synthesized to finely calibrate the horizon; obtaining target data based on the logging data and carrying out scale transformation by a calibration layer, namely converting the obtained target data to a time domain by time-depth conversion, carrying out smoothing treatment on the target data after the time-depth conversion, and then carrying out down-sampling to make the target data have the same size as angle seismic data; constructing a training set and a test set based on the obtained target data and the angle gather superposition data;
102, establishing a model, namely establishing a multi-task reversible network model and training by using the established data set to obtain a complex geophysical relation model between target data and seismic data; the method comprises the following steps: determining the input and output dimensionality of a model according to the multitask target and establishing a reversible module of a reversible network, wherein the reversible module consists of an affine coupling layer which is mapped in a two-way mode; constructing a multitask reversible network model based on the established reversible module, wherein forward input is target data to be inverted, and forward output is a corresponding superimposed data volume with different incidence angle ranges; setting the depth and the learning function of the multi-task reversible network model, wherein the model depth is the number of layers corresponding to the reversible module, and the learning function can be any complex function, such as a fully-connected network or a convolutional network; calculating an optimized target value in the network training process, wherein the calculated optimized target value comprises a supervised loss function and two unsupervised loss functions; updating the model parameters according to the calculated optimization target value; saving the network model parameters; acquiring a multi-task reversible network inversion model;
103, data inversion is carried out, and the trained multi-task reversible network model is used for carrying out inversion on the field seismic data to obtain a target data volume of the research area; the method comprises the following steps: inputting implicit variables output in network training and extracted real angle seismic data into an inversion process of a trained multi-task reversible network model, and performing inversion to obtain a target data body of a research area, wherein the dimension of forward output needs to be determined according to a target task;
fig. 4 is a comparison of inversion processes of different networks, a common network generally performs a separate modeling on a forward and backward modeling process, that is, a forward relationship between target data and observation data is established or a backward relationship between observation data and target data is established, the former is the basis of the latter, but in the inversion based on the common network, we generally only pay attention to the latter, that is, how to accurately construct a complex geophysical relationship between observation seismic data and target data, and how to obtain various geophysical parameters representing a stratum from the observation seismic data is realized, which is a problem to be solved in practical application. However, in practical situations, there often exists a one-to-many relationship between the observed seismic data and the physical parameters of each earth, i.e. a group of observed seismic data may correspond to a plurality of different target data combinations, resulting in an ill-posed inversion process. The reversible network is adopted for modeling inversion, so that the problem that a forward and backward modeling process of a common network model is independently modeled can be solved to a certain extent, and the accuracy of an inversion result is improved.
Corresponding to the seismic inversion method based on the multitask reversible network, the embodiment of the invention provides a seismic inversion device based on the multitask reversible network. Referring to fig. 5, in some embodiments, the inversion apparatus mainly includes three modules, a data acquisition module 201, a model building module 202 and a data inversion module 203.
The data acquisition module 201 is used for acquiring seismic and well logging data of a research area, processing the data and constructing a data set; the method comprises the following steps: acquiring and processing seismic and well logging data of a research area; converting the processed seismic data into angle gather data to obtain a common angle gather stacking data volume; based on the processed logging information and the gather superposition data volume with different angles, the horizon is finely calibrated by integrating geological information; obtaining target data based on the logging data and carrying out scale transformation according to the calibration horizon; constructing a data set based on the obtained target data and the angle gather superposition data;
the model establishing module 202 is used for establishing a multi-task reversible network model and training by using the established data set to obtain a complex geophysical relation model between target data and seismic data; the method comprises the following steps: determining the input and output dimensions of the model according to the multitask target and establishing a reversible module of a reversible network; constructing a multitask reversible network model based on the established reversible module; setting network model parameters; calculating an optimized target value in the network training process; updating network model parameters according to the calculated optimization target value; saving the network model parameters; acquiring a multi-task reversible network inversion model;
the data inversion module 203 is used for inverting the field seismic data by using the trained multi-task reversible network model to obtain a target data volume of the research area; the method comprises the following steps: and inputting implicit variables output in network training and the extracted angle seismic data into an inversion process of the trained multi-task reversible network model, and performing inversion to obtain a target data volume of a research area, wherein the dimension of forward output needs to be determined according to a target task.
The seismic inversion device based on the multitask reversible network provided by the embodiment of the invention can execute the technical scheme of the method embodiment shown in the figure 1, and the specific implementation process is detailed in the method embodiment and is not described again.
It will be appreciated by those skilled in the art that the above embodiments are only intended to illustrate the benefits of the invention and are not exhaustive. Any modification, equivalent replacement, improvement and the like made without departing from the scope and spirit of the illustrated embodiments should not be excluded from the scope of the present invention.
Claims (4)
1. A seismic inversion method based on a multitask reversible network is characterized by comprising three main steps of data acquisition, model establishment and data inversion:
acquiring data, namely acquiring seismic and well logging data of a research area, and processing the data to construct a data set; the method comprises the following steps: acquiring and processing seismic and well logging data of a research area; converting the processed seismic data into angle gather data to obtain a common angle gather stacking data volume; based on the processed logging data and the data volume overlapped by the gather of different angles, the geological information is synthesized to finely calibrate the horizon; obtaining target data based on the logging data and carrying out scale transformation according to the calibration horizon; constructing a data set based on the obtained target data and the angle gather superposition data;
establishing a model, namely establishing a multi-task reversible network model and training by using the established data set to obtain a complex geophysical relation model between target data and seismic data; the method comprises the following steps: determining the input and output dimensionality of the model according to the multitask target and establishing a reversible module of a reversible network; constructing a multitask reversible network model based on the established reversible module; setting network model parameters; calculating an optimized target value in the network training process; updating network model parameters according to the calculated optimization target value; saving the network model parameters; acquiring a multi-task reversible network inversion model;
data inversion, namely inverting the field seismic data by using the trained multi-task reversible network model to obtain a target data volume of a research area; the method comprises the following steps: and inputting implicit variables output in network training and the extracted angle seismic data into an inversion process of the trained multi-task reversible network model, and performing inversion to obtain a target data volume of a research area, wherein the dimension of forward output needs to be determined according to a target task.
2. The seismic inversion method based on the multitask reversible network according to claim 1, which is characterized in that: according to the method, the inversion model structure is built by introducing the reversible network of the bidirectional mapping, so that the problem that the mutual dependency relationship between target data and seismic data cannot be well utilized by a conventional inversion method can be solved.
3. The seismic inversion method based on the multitask reversible network as claimed in claim 1, characterized in that: the multi-task learning is to train a network model in a multi-task joint learning mode in a modeling process, and mine the mutual coupling relation between different target data while learning and representing the complex geophysical relation between the seismic amplitude characteristics and the target data.
4. A seismic inversion apparatus based on a multitasking reversible network, the apparatus comprising:
the data acquisition module is used for acquiring seismic and well logging data of a research area and processing the data to construct a data set; the method comprises the following steps: acquiring and processing seismic and well logging data of a research area; converting the processed seismic data into angle gather data to obtain a common angle gather stacking data volume; based on the processed logging data and the data volume overlapped by the gather of different angles, the geological information is synthesized to finely calibrate the horizon; obtaining target data based on the logging data and carrying out scale transformation according to the calibration horizon; constructing a data set based on the obtained target data and the angle gather superposition data;
the model establishing module is used for establishing a multi-task reversible network model and training by using the established data set to obtain a complex geophysical relation model between target data and seismic data; the method comprises the following steps: determining the input and output dimensionality of the model according to the multitask target and establishing a reversible module of a reversible network; constructing a multitask reversible network model based on the established reversible module; setting network model parameters; calculating an optimized target value in the network training process; updating network model parameters according to the calculated optimization target value; saving the network model parameters; acquiring a multi-task reversible network inversion model;
the data inversion module is used for inverting the field seismic data by utilizing the trained multi-task reversible network model to obtain a target data volume of the research area; the method comprises the following steps: and inputting implicit variables output in network training and the extracted angle seismic data into the inversion process of the trained multi-task reversible network model, and inverting to obtain a target data body of a research area, wherein the dimension of forward output needs to be determined according to a target task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211228562.2A CN115542384A (en) | 2022-10-08 | 2022-10-08 | Earthquake inversion method and device based on multitask reversible network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211228562.2A CN115542384A (en) | 2022-10-08 | 2022-10-08 | Earthquake inversion method and device based on multitask reversible network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115542384A true CN115542384A (en) | 2022-12-30 |
Family
ID=84734365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211228562.2A Pending CN115542384A (en) | 2022-10-08 | 2022-10-08 | Earthquake inversion method and device based on multitask reversible network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115542384A (en) |
-
2022
- 2022-10-08 CN CN202211228562.2A patent/CN115542384A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102129084B (en) | Method and device for acquiring seismic thin reservoir speed through well control | |
| US7840394B2 (en) | Method for generating a 3D earth model | |
| CN103245970B (en) | Pre-stack seismic wide angle retrieval method | |
| CN112017289A (en) | Well-seismic combined initial lithology model construction method based on deep learning | |
| CN101201409B (en) | Method for revising earthquake data phase | |
| CN111025387B (en) | Pre-stack earthquake multi-parameter inversion method for shale reservoir | |
| CN113031068B (en) | Reflection coefficient accurate base tracking prestack seismic inversion method | |
| CN109001813A (en) | Method, device and system for suppressing multiple waves | |
| CN104914466A (en) | Method for improving seismic data resolution | |
| CN111722283B (en) | Stratum velocity model building method | |
| CN104237937A (en) | Pre-stack seismic inversion method and system thereof | |
| Waheed et al. | Anisotropic eikonal solution using physics-informed neural networks | |
| CN108508481B (en) | A kind of method, apparatus and system of longitudinal wave converted wave seismic data time match | |
| CN105275459A (en) | A method for determining the mobile water volume content of shale formations | |
| Wu et al. | An unsupervised inversion method for seismic brittleness parameters driven by the physical equation | |
| CN114460666A (en) | Seismic data resolution improving method based on deep learning | |
| CN115542384A (en) | Earthquake inversion method and device based on multitask reversible network | |
| CN109459790B (en) | Method and system for establishing seismic velocity field of coal measure strata | |
| CN112147700A (en) | Low-frequency model construction method and system for speed abnormal area | |
| CN117388944A (en) | Multi-physical parameter inversion method of geologic model constraint | |
| CN105467441A (en) | Device using average incidence angle trace gathers to carry out PP save and PS wave combined AVO inversion | |
| CN110333534B (en) | Bayes time-shifting AVO inversion method and system based on Biot theory | |
| Sang et al. | Prestack simultaneous inversion of P-wave impedance and gas saturation using multi-task residual networks | |
| CN115373022A (en) | Elastic wave field Helmholtz decomposition method based on amplitude phase correction | |
| Li et al. | Time-Lapse Seismic Inversion for Predicting Reservoir Parameters Based on a Two-Stage Dual Network |
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 |