CN117171852A - Geological structure modeling method and equipment for multi-condition data fusion neural network - Google Patents
Geological structure modeling method and equipment for multi-condition data fusion neural network Download PDFInfo
- Publication number
- CN117171852A CN117171852A CN202311187736.XA CN202311187736A CN117171852A CN 117171852 A CN117171852 A CN 117171852A CN 202311187736 A CN202311187736 A CN 202311187736A CN 117171852 A CN117171852 A CN 117171852A
- Authority
- CN
- China
- Prior art keywords
- data
- condition
- fusion
- neural network
- conditional
- 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
- 230000004927 fusion Effects 0.000 title claims abstract description 108
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 35
- 238000003062 neural network model Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000004088 simulation Methods 0.000 abstract description 36
- 238000010586 diagram Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000013135 deep learning Methods 0.000 description 6
- 238000012512 characterization method Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000002948 stochastic simulation Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The application provides a geological structure modeling method of a multi-condition data fusion neural network, which comprises the following steps: building a multi-condition data fusion neural network model, training, loading condition hard data and condition soft data, inputting the condition hard data and the condition soft data into the trained multi-condition data fusion neural network model, restoring a geological structure model, comparing the restored geological structure model with a reference model, adjusting the training of the multi-condition data fusion neural network model through back propagation, storing the final simulation result of the geological structure model, and completing a simulation flow. And the multi-condition data fusion neural network model is utilized to extract and fuse implicit characteristics in the soft and hard condition data, and a geological structure modeling method based on the multi-condition data fusion neural network model is realized according to the implicit characteristics, so that the reconstruction capability and efficiency of a geological heterogeneous mode are greatly improved.
Description
Technical Field
The application relates to the field of deep learning and geologic modeling, in particular to a geologic structure modeling method and equipment of a multi-condition data fusion neural network.
Background
Description of geologic structures in complex underground spaces is one of the important problems in the field of earth science, and with the continuous progress of geologic resource exploration technology, the description requirements on underground spaces are more and more accurate, and the described geologic structures are more and more complex. By utilizing a combination of multiple condition data, effective data support can be provided for accurate description of subsurface geologic structures.
As an important characterization method of subsurface heterogeneous geological structures and phenomena, geostatistical stochastic simulation methods have been widely applied to the relevant fields of reservoir characterization, geophysical inversion and the like. The geostatistical stochastic simulation method can extract and reproduce heterogeneous subsurface geologic space structures and phenomena by combining various condition data through a mode of maximizing expectation or pattern learning. However, when complex geological phenomena and structures with connectivity characteristics are simulated and constructed, simulation results obtained by the method are often insufficient in global connectivity and require complex parameter settings.
With the vigorous development of artificial intelligence techniques typified by deep learning, various kinds of deep learning techniques are applied to the field of underground structure model construction. Among them, methods such as generation countermeasure networks and variational self-encoders typified by a generation type neural network are most widely used in the field of geologic model construction. And (3) realizing the reconstruction of the geological structure model under the constraint of the condition data by inputting the random variable and the condition data based on the geological structure characterization method for generating the countermeasure network.
But geologic modeling methods based on generating a countermeasure network often require a large number of training samples, which is very difficult in the field of subsurface structure description.
The geologic modeling method based on the variational self-encoder can learn implicit characteristics by using fewer samples and reconstruct to obtain a geologic structure. But such methods have the problem of difficulty in incorporating multiple conditional data constraints.
Disclosure of Invention
The application aims to solve the technical problem that the local connectivity is insufficient and the deep learning modeling method is difficult to reconstruct an underground geological structure under the constraint condition of obeying various condition data in the simulation implementation result of the traditional geostatistics, and provides a geological structure modeling method and equipment of a multi-condition data fusion neural network.
The above object of the present application is achieved by the following technical solutions:
s1: acquiring a training set of conditional hard data and a training set of conditional soft data; the conditional hard data is condition data points randomly extracted from a reference model; the conditional soft data is a probability map established according to the morphology of the reference model;
s2: building a multi-condition data fusion neural network model and training, and specifically comprises the following steps:
inputting the training set of the conditional hard data into a hard data encoder, and extracting implicit characteristics of the conditional hard data;
inputting the training set of the conditional soft data into a soft data encoder, and extracting implicit characteristics of the conditional soft data;
inputting the implicit features of the conditional hard data into a multi-conditional fusion encoder to extract multi-data fusion features;
inputting the multiple data fusion characteristics into a generator, and restoring a geological structure model;
comparing the restored geological structure model with the reference model and determining differences according to a loss function; optimizing parameters of the multi-condition data fusion neural network model through back propagation;
s3: loading hard data and soft data of the condition to be tested;
s4: and inputting the hard data and the soft data of the to-be-detected conditions into the trained multi-condition data fusion neural network model to generate a result of the multi-condition data fusion neural network model, and restoring the geological structure model to be detected.
Optionally, the multi-condition data fusion neural network model includes: hard data encoder, soft data encoder, multi-condition fusion encoder, generator and discriminator;
the hard data encoder is connected with the multi-condition fusion encoder and the generator;
the soft data encoder is connected with the multi-condition fusion encoder and the generator;
the multi-condition fusion encoder is connected with the generator;
the hard data encoder acquires the conditional hard data, encodes the conditional hard data and inputs the encoded conditional hard data to the multi-condition fusion encoder;
the soft data encoder acquires the conditional hard data, encodes the conditional soft data and inputs the encoded conditional soft data to the multi-condition fusion encoder; the multi-condition fusion encoder fuses the coded conditional hard data and the coded conditional soft data with hidden characteristics, generates multi-data fusion characteristics and inputs the multi-data fusion characteristics to the generator; the generator restores a geologic structure model from the input multi-data fusion features.
Optionally, the loss function of the generator is as follows:
wherein R represents a multi-data fusion characteristic extracted by the multi-condition fusion encoder, G (R) represents a generation result of a generator, y represents a reference geological structure, lambda represents an adjustable weight parameter, lambda represents a Manhattan distance,representing mathematical expectations on the distribution of the characteristic R data; d represents a discriminator; e (E) R,y Representing the mathematical expectation of the features R and y.
Optionally, the loss function of the arbiter is as follows:
wherein R represents the multiple data fusion characteristics extracted by the multiple condition fusion encoder, G (R) represents the generation result of the generator, y represents the reference geological structure, lambda represents the adjustable weight parameters,representing mathematical expectations on the characteristic R data distribution, < >>Representing the expectation of y on the data distribution, D (y) representing the output of the arbiter at input y; d (G (R)) represents the output of the arbiter when G (R) is input.
A storage device stores instructions and data for implementing a geological structure modeling method of a multi-condition data fusion neural network.
A geologic structure modeling apparatus of a multi-condition data fusion neural network, comprising: a processor and a storage device; the processor loads and executes instructions and data in the storage device for implementing a geologic structure modeling method of the multi-condition data fusion neural network.
The technical scheme provided by the application has the beneficial effects that:
the implicit characteristics in the conditional hard data are learned by using the hard data encoder, the implicit characteristics in the conditional soft data are learned by using the soft data encoder, the implicit characteristics in the soft and hard data are extracted and fused by using the multi-condition fusion encoder, and the fused characteristics of the multi-condition data can be input into the generator to restore the geological structure model, so that the reconstruction process of the whole geological space is completed.
And utilizing implicit distribution of the learning condition data of a plurality of encoders to fuse and construct a geological structure model, and realizing rapid generation of a heterogeneous space model and a structure model which meet the constraint of a plurality of condition data by using the same trained deep neural network model. The method solves the technical problems that the local connectivity is insufficient and the deep learning modeling method is difficult to reconstruct the underground geological structure under the constraint condition of obeying various condition data in the simulation implementation result of the traditional geostatistics.
Drawings
The application will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a step diagram of a method for modeling a geologic structure of a multi-condition data fusion neural network in an embodiment of the application;
FIG. 2 is an algorithm flow chart of a method for modeling a geologic structure of a multi-condition data fusion neural network in an embodiment of the application;
FIG. 3 is an experimental diagram of a method for modeling a geologic structure of a multi-condition data fusion neural network in an embodiment of the application;
FIG. 4 is a two-dimensional simulation experiment diagram of a geological structure modeling method of a multi-condition data fusion neural network in an embodiment of the application;
FIG. 5 is a three-dimensional simulation experiment diagram of a geological structure modeling method of a multi-condition data fusion neural network in an embodiment of the application;
FIG. 6 is a schematic diagram of the operation of a hardware device in an embodiment of the application.
Detailed Description
For a clearer understanding of technical features, objects and effects of the present application, a detailed description of embodiments of the present application will be made with reference to the accompanying drawings.
The embodiment of the application provides a geological structure modeling method and equipment of a multi-condition data fusion neural network.
Referring to fig. 1, fig. 1 is a step diagram of a geological structure modeling method of a multi-condition data fusion neural network in an embodiment of the application, which specifically includes the following steps:
s1: acquiring a training set of conditional hard data and a training set of conditional soft data; the conditional hard data is condition data points randomly extracted from a reference model; the conditional soft data is a probability map established according to the morphology of the reference model;
s2: building a multi-condition data fusion neural network model and training, and specifically comprises the following steps:
inputting the training set of the conditional hard data into a hard data encoder, and extracting implicit characteristics of the conditional hard data;
inputting the training set of the conditional soft data into a soft data encoder, and extracting implicit characteristics of the conditional soft data;
inputting the implicit features of the conditional hard data into a multi-conditional fusion encoder to extract multi-data fusion features;
inputting the multiple data fusion characteristics into a generator, and restoring a geological structure model;
comparing the restored geological structure model with the reference model and determining differences according to a loss function; optimizing parameters of the multi-condition data fusion neural network model through back propagation;
s3: loading hard data and soft data of the condition to be tested;
s4: and inputting the hard data and the soft data of the to-be-detected conditions into the trained multi-condition data fusion neural network model to generate a result of the multi-condition data fusion neural network model, and restoring the geological structure model to be detected.
Specifically, implicit characteristics of conditional hard data and conditional soft data are respectively extracted by using a multi-condition fusion network model, a complex geological structure is characterized, and meanwhile, the problem that an unnatural space structure appears in an implementation result generated by conventional geostatistical random simulation is avoided.
Specifically, the multi-condition data fusion neural network is used for fusing the implicit characteristics of the condition hard data and the condition soft data, so that the efficient and accurate reconstruction of the geological structure under the constraint of various condition data can be realized, and the problem that the geological modeling method based on deep learning is difficult to reconstruct the underground geological structure under the constraint condition of the data obeying various conditions is solved.
Specifically, the multi-condition data fusion neural network model trained by the application can be popularized and applied in various three-dimensional geological information systems, geographic information systems, geological modeling and simulation systems and other software.
Referring to fig. 2, fig. 2 is an algorithm flow chart of a geological structure modeling method of a multi-condition data fusion neural network according to an embodiment of the present application;
referring to fig. 3, fig. 3 is an experimental diagram of a geological structure modeling method of a multi-condition data fusion neural network according to an embodiment of the present application; fig. 3 (a) shows the conditional hard data, the conditional soft data, the two-dimensional simulation results and the corresponding two-dimensional reference model in the two-dimensional simulation experimental case. Fig. 3 (b) shows the conditional hard data, the conditional soft data, the three-dimensional simulation results and the corresponding three-dimensional reference model in the three-dimensional simulation experimental case.
Referring to fig. 4, fig. 4 is a two-dimensional simulation experiment diagram of a geological structure modeling method of a multi-condition data fusion neural network according to an embodiment of the present application; FIG. 4 (a) is a graph of the overall variogram correspondingly drawn with reference to the two-dimensional geologic structure simulation results, and FIG. 4 (b) is a graph of the spatial connectivity of the two-dimensional geologic structure simulation results in the X-direction
Referring to fig. 5, fig. 5 is a three-dimensional simulation experiment diagram of a geological structure modeling method of a multi-condition data fusion neural network in an embodiment of the application; fig. 5 (a) is an overall variogram correspondingly drawn with reference to the three-dimensional geologic structure simulation result, and fig. 5 (b) is a spatial connectivity curve of the three-dimensional geologic structure simulation result and the corresponding reference model in the X, Y direction.
Fig. 3 shows two-dimensional and three-dimensional experimental cases of the design of the present application. Fig. 3 (a) shows the conditional hard data, the conditional soft data, the simulation results and the corresponding reference model used in the two-dimensional simulation experiment. Wherein the conditional hard data is 100 condition data points for simulation randomly extracted from the reference model; the conditional soft data is a probability map established according to the morphology of the reference model, and the resolution of the conditional soft data, the simulation result and the corresponding reference model is 128×128. The two-dimensional simulation implementation result shows that the simulated river channel has clear texture and similar distribution to the reference model, so that the method provided by the application can better simulate the complex river phase structure. Fig. 3 (b) shows the conditional hard data, the conditional soft data, the simulation results and the corresponding reference model used in the three-dimensional simulation experiment. The three-dimensional conditional hard data is 20 virtual wells randomly extracted from the reference model, and the conditional soft data is S-wave data corresponding to the geological structure. The three-dimensional simulation result shows that the method provided by the application can accurately restore the distribution mode of the three-dimensional complex geological structure.
In order to reveal the differences in statistical characteristics of the simulation results and the reference model in depth, fig. 3 and 4 show the variation function curves and connectivity curves of the reference model and the simulation results in the two-dimensional simulation experiment and the three-dimensional simulation experiment, respectively. Fig. 3 (a) is a variation function curve drawn by 100 different simulation results generated by the method of the present application together with a reference model. Fig. 3 (b) shows the difference in the X-direction connected characteristics of 100 different simulation results from the reference model. From both the variogram and the connectivity function, it can be seen that these 100 simulation results can approach and conform to the variogram and connectivity characteristics in the reference model. It can also be seen from fig. 4 that, similar to the statistical features of the two-dimensional experiment, the 20 three-dimensional simulation results are very close to the attribute distribution pattern in the three-dimensional reference model, both from the variogram (fig. 4 (a)) and the connectivity in the X, Y direction (fig. 4 (b)).
The two experimental cases show that the simulation results of the method provided by the application are close to the corresponding characteristics of the reference model in the aspects of attribute proportion reproduction, space variability depiction, space structure connectivity reconstruction and the like, and the geological structure modeling method based on the multi-condition data fusion neural network also proves that the geological structure modeling method provided by the application has excellent capability of simulating a geological structure model.
The multi-condition data fusion neural network model comprises: hard data encoder, soft data encoder, multi-condition fusion encoder, generator and discriminator;
the hard data encoder is connected with the multi-condition fusion encoder and the generator;
the soft data encoder is connected with the multi-condition fusion encoder and the generator;
the multi-condition fusion encoder is connected with the generator;
the hard data encoder acquires the conditional hard data, encodes the conditional hard data and inputs the encoded conditional hard data to the multi-condition fusion encoder;
the soft data encoder acquires the conditional hard data, encodes the conditional soft data and inputs the encoded conditional soft data to the multi-condition fusion encoder; the multi-condition fusion encoder fuses the coded conditional hard data and the coded conditional soft data with hidden characteristics, generates multi-data fusion characteristics and inputs the multi-data fusion characteristics to the generator; the generator restores a geologic structure model from the input multi-data fusion features.
The loss function of the generator is as follows:
wherein R represents a multi-data fusion characteristic extracted by the multi-condition fusion encoder, G () represents a generation result of a generator, y represents a reference geological structure, lambda represents an adjustable weight parameter, lambda represents a Manhattan distance,representing mathematical expectations on the distribution of the characteristic R data; d represents a discriminator; e (E) R,y Representing the mathematical expectation of the features R and y;
the loss function of the arbiter is as follows:
wherein R represents the multiple data fusion characteristics extracted by the multiple condition fusion encoder, G (R) represents the generation result of the generator, y represents the reference geological structure, lambda represents the adjustable weight parameters,representing mathematical expectations on the characteristic R data distribution, < >>Representing the expectation of y on the data distribution, D (y) representing the output of the arbiter at input y; d (G (R)) represents the output of the arbiter when G (R) is input.
Referring to fig. 6, fig. 6 is a schematic working diagram of a hardware device according to an embodiment of the present application, where the hardware device specifically includes: a geological structure modeling device 401, a processor 402 and a storage device 403 of a multi-condition data fusion neural network.
A geological structure modeling apparatus 401 of a multi-condition data fusion neural network: the geological structure modeling device 401 of the multi-condition data fusion neural network realizes a geological structure modeling method of the multi-condition data fusion neural network.
Processor 402: the processor 402 loads and executes instructions and data in the memory device 403 for implementing a method for modeling geologic structures of a multi-condition data fusion neural network.
Storage device 403: storage device 403 stores instructions and data; the storage device 403 is used to implement a geological structure modeling method of the multi-condition data fusion neural network.
The foregoing is only illustrative of the present application and is not to be construed as limiting thereof, but rather as various modifications, equivalent arrangements, improvements, etc., within the spirit and principles of the present application.
Claims (6)
1. The geological structure modeling method of the multi-condition data fusion neural network is characterized by comprising the following steps of:
s1: acquiring a training set of conditional hard data and a training set of conditional soft data; the conditional hard data is condition data points randomly extracted from a reference model; the conditional soft data is a probability map established according to the morphology of the reference model;
s2: building a multi-condition data fusion neural network model and training, and specifically comprises the following steps:
inputting the training set of the conditional hard data into a hard data encoder, and extracting implicit characteristics of the conditional hard data;
inputting the training set of the conditional soft data into a soft data encoder, and extracting implicit characteristics of the conditional soft data;
inputting the implicit features of the conditional hard data into a multi-conditional fusion encoder to extract multi-data fusion features;
inputting the multiple data fusion characteristics into a generator, and restoring a geological structure model;
comparing the restored geological structure model with the reference model and determining differences according to a loss function; optimizing parameters of the multi-condition data fusion neural network model through back propagation;
s3: loading hard data and soft data of the condition to be tested;
s4: and inputting the hard data and the soft data of the to-be-detected conditions into the trained multi-condition data fusion neural network model to generate a result of the multi-condition data fusion neural network model, and restoring the geological structure model to be detected.
2. A method of modeling a geologic structure of a multi-condition data fusion neural network as defined in claim 1, wherein the multi-condition data fusion neural network model comprises: hard data encoder, soft data encoder, multi-condition fusion encoder, generator and discriminator;
the hard data encoder is connected with the multi-condition fusion encoder and the generator;
the soft data encoder is connected with the multi-condition fusion encoder and the generator;
the multi-condition fusion encoder is connected with the generator;
the hard data encoder acquires the conditional hard data, encodes the conditional hard data and inputs the encoded conditional hard data to the multi-condition fusion encoder;
the soft data encoder acquires the conditional hard data, encodes the conditional soft data and inputs the encoded conditional soft data to the multi-condition fusion encoder; the multi-condition fusion encoder fuses the coded conditional hard data and the coded conditional soft data with hidden characteristics, generates multi-data fusion characteristics and inputs the multi-data fusion characteristics to the generator; the generator restores a geologic structure model from the input multi-data fusion features.
3. A method of modeling a geologic structure of a multi-condition data fusion neural network as defined in claim 1, wherein the generator's loss function is as follows:
wherein R represents a multi-data fusion characteristic extracted by the multi-condition fusion encoder, G (R) represents a generation result of a generator, y represents a reference geological structure, lambda represents an adjustable weight parameter, lambda represents a Manhattan distance,representing mathematical expectations on the distribution of the characteristic R data; d represents a discriminator; e (E) R,y Representing the mathematical expectation of the features R and y.
4. A method of modeling a geologic structure of a multi-condition data fusion neural network as defined in claim 1, wherein the loss function of the arbiter is as follows:
wherein R represents the multiple data fusion characteristics extracted by the multiple condition fusion encoder, F (R) represents the generation result of the generator, y represents the reference geological structure, lambda represents the adjustable weight parameters,representing mathematical expectations on the characteristic R data distribution, < >>Representing the expectation of y on the data distribution, D (y) representing the output of the arbiter at input y; d (G (R)) represents the output of the arbiter when G (R) is input.
5. A memory device, characterized by: the storage device stores instructions and data for implementing a geologic structure modeling method of a multi-condition data fusion neural network of any of claims 1-4.
6. A geological structure modeling device of a multi-condition data fusion neural network is characterized in that: comprising the following steps: a processor and a storage device; the processor loads and executes instructions and data in the storage device for implementing a geologic structure modeling method of the multi-condition data fusion neural network of any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311187736.XA CN117171852A (en) | 2023-09-13 | 2023-09-13 | Geological structure modeling method and equipment for multi-condition data fusion neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311187736.XA CN117171852A (en) | 2023-09-13 | 2023-09-13 | Geological structure modeling method and equipment for multi-condition data fusion neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117171852A true CN117171852A (en) | 2023-12-05 |
Family
ID=88937164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311187736.XA Pending CN117171852A (en) | 2023-09-13 | 2023-09-13 | Geological structure modeling method and equipment for multi-condition data fusion neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117171852A (en) |
-
2023
- 2023-09-13 CN CN202311187736.XA patent/CN117171852A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101133422B (en) | Multiple-point statistics (MPS) simulation with enhanced computational efficiency | |
| US10467357B2 (en) | Geobody continuity in geological models based on multiple point statistics | |
| Strebelle et al. | Reservoir modeling using multiple-point statistics | |
| Arpat et al. | Conditional simulation with patterns | |
| Caers et al. | Stochastic reservoir simulation using neural networks trained on outcrop data | |
| Pyrcz et al. | A library of training images for fluvial and deepwater reservoirs and associated code | |
| US8775142B2 (en) | Stochastic downscaling algorithm and applications to geological model downscaling | |
| CN108416841B (en) | Automatic reconstruction method of multi-point statistical three-dimensional geological model based on local search strategy | |
| CA2875496C (en) | System and method for optimizing the number of conditioning data in multiple point statistics simulation | |
| WO2010057505A1 (en) | A deterministic version of the multiple point geostatistics simulation / reconstruction method wxth. the simulated / reconstructed values are directly taken from the training images without prior estimation of the conditional | |
| CN104504754B (en) | A kind of method and device of oil and gas reservoir multi-point statistic modeling | |
| CN111612906B (en) | Method and system for generating three-dimensional geological model and computer storage medium | |
| Tyler et al. | Modeling heterogeneities in fluvial domains: a review of the influence on production profiles | |
| Cui et al. | Characterization of subsurface hydrogeological structures with convolutional conditional neural processes on limited training data | |
| Schorpp et al. | Automated hierarchical 3D modeling of quaternary aquifers: the ArchPy approach | |
| Hou et al. | Reconstructing Three-dimensional geological structures by the Multiple-point statistics method coupled with a deep neural network: A case study of a metro station in Guangzhou, China | |
| CN117171852A (en) | Geological structure modeling method and equipment for multi-condition data fusion neural network | |
| CN114022632A (en) | Geological model rapid generation method for generating countermeasure network based on deep convolution | |
| Lamuraglia et al. | Application of a global–local full‐waveform inversion of Rayleigh wave to estimate the near‐surface shear wave velocity model | |
| Eskandari et al. | Growthsim–a multiple point framework for pattern simulation | |
| CN111751886A (en) | Shale gas reservoir fracture modeling method based on microseism monitoring data | |
| Gout et al. | Surface fitting of rapidly varying data using rank coding: application to geophysical surfaces | |
| EP3320450B1 (en) | Improved geobody continuity in geological models based on multiple point statistics | |
| CN115393541A (en) | Multi-scale geological model construction method based on convolution condition neural process | |
| Jo et al. | Conditioning stratigraphic, rule-Based models with generative adversarial networks: A deepwater lobe, deep learning example |
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 |