CN114529110A - Lithofacies inversion method and system based on deep neural network model - Google Patents
Lithofacies inversion method and system based on deep neural network model Download PDFInfo
- Publication number
- CN114529110A CN114529110A CN202011209672.5A CN202011209672A CN114529110A CN 114529110 A CN114529110 A CN 114529110A CN 202011209672 A CN202011209672 A CN 202011209672A CN 114529110 A CN114529110 A CN 114529110A
- Authority
- CN
- China
- Prior art keywords
- lithofacies
- training
- model
- data
- neural 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
- 238000003062 neural network model Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012549 training Methods 0.000 claims abstract description 105
- 238000012360 testing method Methods 0.000 claims abstract description 67
- 238000009499 grossing Methods 0.000 claims abstract description 21
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 230000008859 change Effects 0.000 claims abstract description 11
- 238000013507 mapping Methods 0.000 claims abstract description 9
- 238000004590 computer program Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 239000011435 rock Substances 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 8
- 208000035126 Facies Diseases 0.000 claims description 7
- 230000000704 physical effect Effects 0.000 claims description 5
- 230000008021 deposition Effects 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 15
- 238000013528 artificial neural network Methods 0.000 description 11
- 239000007789 gas Substances 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 238000013135 deep learning Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000009826 distribution Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 239000012530 fluid Substances 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 238000005553 drilling Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- BVKZGUZCCUSVTD-UHFFFAOYSA-L Carbonate Chemical compound [O-]C([O-])=O BVKZGUZCCUSVTD-UHFFFAOYSA-L 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000003209 petroleum derivative Substances 0.000 description 1
- 238000000614 phase inversion technique Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000003892 spreading Methods 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Operations Research (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
Abstract
The invention provides a lithofacies inversion method and a lithofacies inversion system based on a deep neural network model, wherein the lithofacies inversion method comprises the following steps: collecting logging data; preprocessing logging data to obtain a training set and a testing set; the method comprises the following steps that a training set is a training sample formed by combining seismic attributes and lithofacies indexes, and unlabeled logging data are selected from a testing set; performing first learning by utilizing the aboveground lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge to perform mapping learning to obtain a first training model; classifying and training the first training model by using a training set, importing a test set into the trained model for testing, and predicting a probability body of each lithofacies observed along a borehole by using the tested model to obtain final lithofacies data; and smoothing the final lithofacies data, and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
Description
Technical Field
The invention belongs to the technical field of petroleum and natural gas seismic exploration, and particularly relates to a lithofacies inversion method and system based on a deep neural network model.
Background
In the exploration and development process of complex lithologic oil and gas reservoirs, seismic facies analysis (seismic waveform classification) is an effective and quick method for predicting spatial distribution characteristics of reservoirs, but is only based on a macroscopic reflection of waveform change and oil and gas distribution, namely a qualitative prediction process. And the reservoir prediction is realized by utilizing an inversion technology. The current population of industrial inversion can be divided into two broad categories, model-based deterministic inversion and geostatistical stochastic inversion.
The former is mainly from earthquake, and obtains relative impedance by solving reflection coefficient by wavelet with convolution model, and inversion resolution is low; the reservoir space variability random simulation method is an inversion method combining a random simulation theory and seismic inversion starting from a well, the geostatistical thought is utilized, the reservoir space variability is represented through a variation function, random simulation constrained by seismic impedance is carried out on the basis, the high-resolution distribution rule of the reservoir is estimated, and the result is a group of equal probability random simulation solutions.
Well log information is the primary source of information for lithology and fluid content, and the key to accurate prediction of lithology and fluid is accurate and careful analysis of the well. The problem that how to research the transverse and vertical anisotropy of the reservoir, how to analyze the connectivity of the reservoir and how to research the transverse spreading rule of the reservoir aiming at the complex reservoir with varied lithofacies diversity is needed to be deeply researched. The conventional fluid prediction method is used for identifying reservoir lithology and fluid by using a prestack AVO attribute, but lithology classification is difficult to obtain from a well logging intersection map; the other method is to use a characteristic curve as a target and establish a mapping relation through a BP neural network to predict lithofacies, but the problems of local convergence, mismatching of well-to-seismic and the like often exist.
In view of the above, a technical solution capable of overcoming the defects in the prior art and improving the accuracy of the lithofacies inversion result is needed.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a lithofacies inversion method and a lithofacies inversion system based on a deep neural network model, which are used for establishing a well-seismic relation between a seismic waveform and a lithofacies complex well through selection of well data and seismic sample points, optimization of a sample set and deep learning of a Deep Neural Network (DNN) model, so that reservoir prediction work is more accurately carried out, the prediction precision of an oil-gas reservoir is improved, the drilling risk of a research target is reduced, and reliable data are provided for efficient exploration and development of an oil-gas field.
In a first aspect of an embodiment of the present invention, a method for rock phase inversion based on a deep neural network model is provided, where the method includes:
collecting logging data;
preprocessing the logging data to obtain a training set and a testing set; the test set selects unlabeled logging data;
performing first learning by utilizing the aboveground lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge to perform mapping learning to obtain a first training model;
classifying and training the first training model by using the training set, importing the test set into the trained model for testing, and predicting a probability body of each lithofacies observed along the borehole by using the tested model to obtain final lithofacies data;
and smoothing the final lithofacies data, and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
In a second aspect of the embodiments of the present invention, a lithofacies inversion system based on a deep neural network model is provided, the system including:
the data acquisition module is used for acquiring logging data;
the preprocessing module is used for preprocessing the logging data to obtain a training set and a testing set; the test set selects unlabeled logging data;
the model learning module is used for carrying out first learning by utilizing the aboveground lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge for mapping learning to obtain a first training model;
the classification training module is used for performing classification training on the first training model by using the training set, importing the test set into the trained model for testing, and predicting a probability body of each lithofacies observed along the borehole by using the tested model to obtain final lithofacies data;
and the lithofacies inversion module is used for smoothing the final lithofacies data and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
In a third aspect of the embodiments of the present invention, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the processor implements a deep neural network model-based rock phase inversion method.
In a fourth aspect of embodiments of the present invention, a computer-readable storage medium is presented, which stores a computer program that, when executed by a processor, implements a method for deep neural network model-based petrographic phase inversion.
According to the lithofacies inversion method and system based on the deep neural network model, well data and seismic sample point selection, sample set optimization and deep learning of the deep neural network model are carried out, the well seismic relation of seismic waveform and lithofacies complex wells is established, reservoir prediction work can be carried out more accurately, the problems existing in the traditional inversion method and the influence of the traditional inversion method on reservoir prediction precision are avoided, the oil and gas reservoir prediction precision is improved, drilling risks of research targets are reduced, and reliable data are provided for efficient exploration and development of oil and gas fields.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a deep neural network model-based lithofacies inversion method according to an embodiment of the present invention.
Fig. 2 is a relationship diagram of a deep neural network model deep learning-based lithofacies inversion according to an embodiment of the present invention.
FIG. 3 is a schematic diagram illustrating the effect of the uphole facies and uphole classification training in accordance with an embodiment of the present invention.
FIG. 4 is a schematic diagram of a deep neural network model-based lithofacies inversion, according to an embodiment of the present invention.
FIG. 5 is a schematic diagram of a deep neural network model-based lithofacies inversion system architecture according to an embodiment of the present invention.
Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The principles and spirit of the present invention will be described with reference to several exemplary embodiments. It is understood that these embodiments are given solely for the purpose of enabling those skilled in the art to better understand and to practice the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As will be appreciated by one skilled in the art, embodiments of the present invention may be embodied as a system, apparatus, device, method, or computer program product. Accordingly, the present disclosure may be embodied in the form of: entirely hardware, entirely software (including firmware, resident software, micro-code, etc.), or a combination of hardware and software.
According to the embodiment of the invention, a lithofacies inversion method and a lithofacies inversion system based on a deep neural network model are provided, the reservoir mode establishment is carried out by utilizing the lithofacies information on the well, the parameters such as the physical property and the thickness of the reservoir are further predicted, the horizontal continuity advantages of the earthquake are combined according to the information such as the logging curve form and the lithology combination characteristics, and finally the different reservoir modes and the plane distribution rules corresponding to different lithofacies are predicted. The method not only can well utilize the local reservoir characteristics of the logging, but also establishes a bridge between wells, and truly achieves the quantitative characterization of the well seismic depth synergy. The method can solve the problem of lithofacies identification, simultaneously meets the precision of thin layer prediction, and has very important significance on the seismic exploration technology of complex oil and gas reservoirs in future.
The principles and spirit of the present invention are explained in detail below with reference to several representative embodiments of the invention.
FIG. 1 is a schematic flow chart of a deep neural network model-based lithofacies inversion method according to an embodiment of the present invention. As shown in fig. 1, the method includes:
s101, collecting logging data;
step S102, preprocessing the logging data to obtain a training set and a testing set; the test set selects unlabeled logging data;
s103, performing first learning by utilizing the aboveground lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge to perform mapping learning to obtain a first training model;
step S104, carrying out classification training on a first training model by using the training set, importing the test set into the trained model for testing, and predicting a probability body of each lithofacies observed along a borehole by using the model passing the test to obtain final lithofacies data;
and S105, smoothing the final lithofacies data, and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
In order to clearly explain the above lithofacies inversion method based on the deep neural network model, the following description is made in detail with reference to each step.
Fig. 2 is a schematic diagram of a relationship of a lithofacies inversion based on deep learning of a deep neural network model according to an embodiment of the present invention. As shown in fig. 2, the specific process is as follows:
step S1, collecting logging data: and selecting logging data for training a deep neural network model according to the information in the thickness and physical property characteristics of the thin layer and the logging curve form and lithology combination characteristics according to the regional deposition background.
The well log data may include, among other things, electrical curves, lithology curves, and some seismic attributes that describe the rock. Seismic attributes are mainly used for probability estimation and lithofacies classification at the time of later training.
Step S2, constructing a training set and a testing set:
preprocessing the logging data, defining lithofacies data, marking a seismic label number on each waveform data, and constructing a training set; and selecting unmarked logging data as a test set.
This step is to generate a data set for training the neural network that can adjust the predictive power of the network. Where the training set is the core data from which the neural network is trained. In the rock type classification process, neural networks are trained based on a combination of seismic pre-or post-stack attributes and discrete log data (rock type, facies, lithology index).
The training set extracted along the shaft is composed of a group of training samples formed by combining seismic attributes and lithofacies indexes; because well data and seismic data have different resolutions (logging is a vertical sampling), lithofacies or lithology logs typically contain information on thin layers that cannot be detected by seismic information. Therefore, a corresponding mode of seismic data and lithology needs to be established, the mode is represented by a series of neural network vectors, and reservoir change conditions are researched by using a probability phase method; then, carrying out independent neural network process for many times; different neural network processes obtain different corresponding modes of seismic data and lithology.
Step S3, first learning:
the method comprises the steps of conducting first learning by utilizing the information of the lithofacies on the well to obtain an initial depth neural network model, adding seismic information into the initial depth neural network model to serve as new knowledge to conduct mapping learning to obtain a first training model, wherein the model can be used for predicting and calculating the lithofacies through a mature network model.
Step S4, classification training and testing:
and carrying out classification training on the first training model by using the training set, introducing the test set into the trained model for testing, and predicting the probability body of each lithofacies observed along the borehole by using the model passing the test to obtain final lithofacies data.
The specific process comprises the following steps:
step S41, training a first training model by using the training set for classification training;
step S42, importing the test set into the trained model for testing to obtain a test result, wherein the test result comprises a probability body of each lithofacies;
step S43, comparing the prediction result with an ideal output result; wherein the content of the first and second substances,
if the error exceeds the range of the preset value, the test fails, and a training set is continuously selected for model enhancement training;
and if the error is within the preset value range, ending the training.
Through classification training, phasors or maps and related probabilities can be obtained; training with a user-selected number of independent neural networks to provide a set of neural network models, and further, using unlabeled data (pre-stack attribute volumes), the trained neural networks can estimate the probability volume for each facies observed along the borehole; quality control and parametric testing can be performed during this process.
Step S5, smoothing, lithofacies inversion:
and smoothing the final lithofacies data, and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
The lithofacies inversion method provided by the invention is utilized to solve the problems of lithology identification and lithofacies thickness, and the amplitude attribute of the input seismic data is the best choice according to the weak anisotropy concept, namely the rule that the longitudinal wave reflection coefficient of the anisotropic medium in the Ruger formula changes along with the azimuth angle and the incidence angle.
The AI inversion is inversion by using the law of amplitude-thickness anisotropic change (namely AVAZ), and is mapped by using nonlinear machine learning methods such as multi-attribute data and well logging lithofacies data, deep learning and the like, and nonlinear inversion with a lithofacies curve as a target. In addition, uniformly extracting seismic channels of a rock facies body (obtained by modeling or inversion) as well curves of the pseudo-wells, and expanding a sample plate set; and the situation that some lithofacies zones on the plane are not extracted to the template, so that the completeness of data is influenced is avoided.
It should be noted that although the operations of the method of the present invention have been described in the above embodiments and the accompanying drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the operations shown must be performed, to achieve the desired results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.
For a clearer explanation of the above lithofacies inversion method based on a deep neural network model, a specific embodiment is described below, but it should be noted that the embodiment is only for better explaining the present invention and does not constitute an undue limitation to the present invention.
By taking the carbonate rock reservoir prediction in the west region of Sichuan basin as an example, the method of the invention can fully utilize the characteristic of rich waveform characteristics of seismic data in a research region and the vertical high-resolution characteristic of a well, accurately describe lithofacies by adopting a recurrent neural network algorithm, and obtain a lithofacies prediction graph conforming to geological knowledge through the transverse change of seismic signals of a target layer, so as to be used for quantitative prediction of different types of reservoirs.
Referring to fig. 3, the effect of the uphole facies and uphole classification training is illustrated. Referring to fig. 4, a schematic diagram of a lithofacies inversion based on a deep neural network model is shown. As shown in fig. 3 and 4, effective prediction is carried out on reservoir distribution with different lithologies and different depths, and the well coincidence rate is verified to be higher than 80%; the prediction result is consistent with the macroscopic geological knowledge of the area, the matching degree with the single-well logging comprehensive interpretation mode is high, and geological basis is provided for increasing storage and increasing production in the western region of the Sichuan basin.
Having described the method of an exemplary embodiment of the present invention, a deep neural network model-based lithofacies inversion system of an exemplary embodiment of the present invention is next described with reference to FIG. 5.
The implementation of the lithofacies inversion system based on the deep neural network model can be referred to the implementation of the method, and repeated details are omitted. The term "module" or "unit" used hereinafter may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Based on the same inventive concept, the invention also provides a lithofacies inversion system based on the deep neural network model, as shown in fig. 5, the system comprises:
a data acquisition module 510 for acquiring logging data;
a preprocessing module 520, configured to preprocess the logging data to obtain a training set and a test set; the test set selects unlabeled logging data;
the model learning module 530 is used for performing first learning by using the uphole lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge for mapping learning to obtain a first training model;
a classification training module 540, configured to perform classification training on the first training model using the training set, introduce the test set into the trained model for testing, and predict a probability body of each lithofacies observed along the borehole using the model that passes the testing, so as to obtain final lithofacies data;
and a lithofacies inversion module 550, configured to smooth the final lithofacies data, and perform lithofacies inversion on the lateral variation of the seismic signal of the target layer to obtain a lithofacies inversion result.
In an embodiment, the data collecting module 510 is specifically configured to:
and selecting logging data for training the deep neural network model according to the information in the thickness and physical property characteristics of the thin layer and the logging curve form and lithology combination characteristics according to the regional deposition background.
Wherein the collected logging data comprises: electrical, lithological, and seismic properties of the rock.
In an embodiment, the preprocessing module 520 is specifically configured to:
preprocessing the logging data, defining lithofacies data, marking a seismic label number on each waveform data, and constructing a training set;
and selecting unmarked logging data as a test set.
In an embodiment, the classification training module 540 is specifically configured to:
training a first training model by using the training set to carry out classification training;
importing the test set into the trained model for testing to obtain a test result, wherein the test result comprises a probability body of each lithofacies;
comparing the predicted outcome to an ideal output outcome; wherein the content of the first and second substances,
if the error exceeds the range of the preset value, the test fails, and a training set is continuously selected for model enhancement training;
and if the error is within the preset value range, ending the training.
In an embodiment, the classification training module 540 is further configured to:
and when the test fails, selecting a new training set for model training, adjusting model parameters and optimizing the model until the condition that the error is within a preset value range is met.
In an embodiment, the lithofacies inversion module 550 is further configured to:
and performing point-by-point processing on the final lithofacies data, obtaining continuity data through structural smoothing, performing smoothing processing on each probability quantity by using the attributes of the dip angle and the azimuth angle, and performing smoothing processing by adopting mean value filtering to obtain the lithofacies data after smoothing processing.
It should be noted that although several modules of a deep neural network model based lithofacies inversion system are mentioned in the above detailed description, such partitioning is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the modules described above may be embodied in one module according to embodiments of the invention. Conversely, the features and functions of one module described above may be further divided into embodiments by a plurality of modules.
Based on the aforementioned inventive concept, as shown in fig. 6, the present invention further proposes a computer device 600, which includes a memory 610, a processor 620 and a computer program 630 stored on the memory 610 and executable on the processor 620, wherein the processor 620 executes the computer program 630 to implement the aforementioned method for rock phase inversion based on a deep neural network model.
Based on the foregoing inventive concept, the present invention proposes a computer-readable storage medium storing a computer program which, when executed by a processor, implements the aforementioned deep neural network model-based lithofacies inversion method.
According to the lithofacies inversion method and system based on the deep neural network model, well data and seismic sample point selection, sample set optimization and deep learning of the deep neural network model are carried out, the well seismic relation of seismic waveform and lithofacies complex wells is established, reservoir prediction work can be carried out more accurately, the problems existing in the traditional inversion method and the influence of the traditional inversion method on reservoir prediction precision are avoided, the oil and gas reservoir prediction precision is improved, drilling risks of research targets are reduced, and reliable data are provided for efficient exploration and development of oil and gas fields.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (16)
1. A lithofacies inversion method based on a deep neural network model is characterized by comprising the following steps:
collecting logging data;
preprocessing the logging data to obtain a training set and a testing set; the test set selects unlabeled logging data;
performing first learning by utilizing the aboveground lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge to perform mapping learning to obtain a first training model;
classifying and training the first training model by using the training set, importing the test set into the trained model for testing, and predicting a probability body of each lithofacies observed along the borehole by using the tested model to obtain final lithofacies data;
and smoothing the final lithofacies data, and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
2. The deep neural network model-based lithofacies inversion method of claim 1, wherein collecting well log data comprises:
and selecting logging data for training a deep neural network model according to the information in the thickness and physical property characteristics of the thin layer and the logging curve form and lithology combination characteristics according to the regional deposition background.
3. The method of claim 2, wherein the collected logging data comprises:
electrical, lithological, and seismic properties of the rock.
4. The deep neural network model-based lithofacies inversion method of claim 1, wherein preprocessing the well log data to obtain a training set and a test set comprises:
preprocessing the logging data, defining lithofacies data, marking a seismic label number on each waveform data, and constructing a training set;
and selecting unmarked logging data as a test set.
5. The deep neural network model-based lithofacies inversion method of claim 4, wherein the training set is used to perform classification training on a first training model, the test set is imported into the trained model for testing, and the model passing the testing is used to predict a probability body of each facies observed along a borehole to obtain final lithofacies data, comprising:
training a first training model by using the training set to carry out classification training;
importing the test set into the trained model for testing to obtain a test result, wherein the test result comprises a probability body of each lithofacies;
comparing the predicted outcome with an ideal output outcome; wherein the content of the first and second substances,
if the error exceeds the range of the preset value, the test fails, and a training set is continuously selected for model enhancement training;
and if the error is within the preset value range, ending the training.
6. The method of deep neural network model-based lithofacies inversion of claim 5, further comprising:
and when the test fails, selecting a new training set for model enhancement training, adjusting model parameters and optimizing the model until the condition that the error is within the range of a preset value is met.
7. The deep neural network model-based lithofacies inversion method of claim 5, wherein smoothing the final lithofacies data comprises:
and performing point-by-point processing on the final lithofacies data, obtaining continuous data through structure smoothing, performing smoothing processing on each probability quantity by using the attributes of the dip angle and the azimuth angle, and performing smoothing processing by adopting mean value filtering to obtain the lithofacies data after smoothing processing.
8. A lithofacies inversion system based on a deep neural network model, the system comprising:
the data acquisition module is used for acquiring logging data;
the preprocessing module is used for preprocessing the logging data to obtain a training set and a testing set; the test set selects unlabeled logging data;
the model learning module is used for carrying out first learning by utilizing the aboveground lithofacies information to obtain an initial deep neural network model, and adding seismic information into the initial deep neural network model as new knowledge for mapping learning to obtain a first training model;
the classification training module is used for performing classification training on the first training model by using the training set, importing the test set into the trained model for testing, and predicting a probability body of each lithofacies observed along the borehole by using the tested model to obtain final lithofacies data;
and the lithofacies inversion module is used for smoothing the final lithofacies data and performing lithofacies inversion on the transverse change of the seismic signal of the target layer to obtain a lithofacies inversion result.
9. The deep neural network model-based lithofacies inversion system of claim 8, wherein the data acquisition module is specifically configured to:
and selecting logging data for training the deep neural network model according to the information in the thickness and physical property characteristics of the thin layer and the logging curve form and lithology combination characteristics according to the regional deposition background.
10. The deep neural network model-based lithofacies inversion system of claim 9, wherein the logging data collected by the data collection module comprises:
electrical, lithological, and seismic properties of the rock.
11. The deep neural network model-based lithofacies inversion system of claim 8, wherein the preprocessing module is specifically configured to:
preprocessing the logging data, defining lithofacies data, marking a seismic label number on each waveform data, and constructing a training set;
and selecting unmarked logging data as a test set.
12. The deep neural network model-based lithofacies inversion system of claim 11, wherein the classification training module is specifically configured to:
training a first training model by using the training set to carry out classification training;
importing the test set into the trained model for testing to obtain a test result, wherein the test result comprises a probability body of each lithofacies;
comparing the predicted outcome to an ideal output outcome; wherein, the first and the second end of the pipe are connected with each other,
if the error exceeds the range of the preset value, the test fails, and a training set is continuously selected for model enhancement training;
and if the error is within the preset value range, ending the training.
13. The deep neural network model-based lithofacies inversion system of claim 12, wherein the classification training module is further configured to:
and when the test fails, selecting a new training set for model training, adjusting model parameters and optimizing the model until the condition that the error is within the range of a preset value is met.
14. The deep neural network model-based lithofacies inversion system of claim 12, wherein the lithofacies inversion module is further configured to:
and performing point-by-point processing on the final lithofacies data, obtaining continuous data through structure smoothing, performing smoothing processing on each probability quantity by using the attributes of the dip angle and the azimuth angle, and performing smoothing processing by adopting mean value filtering to obtain the lithofacies data after smoothing processing.
15. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 7 when executing the computer program.
16. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, implements the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011209672.5A CN114529110A (en) | 2020-11-03 | 2020-11-03 | Lithofacies inversion method and system based on deep neural network model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011209672.5A CN114529110A (en) | 2020-11-03 | 2020-11-03 | Lithofacies inversion method and system based on deep neural network model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114529110A true CN114529110A (en) | 2022-05-24 |
Family
ID=81619614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011209672.5A Pending CN114529110A (en) | 2020-11-03 | 2020-11-03 | Lithofacies inversion method and system based on deep neural network model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114529110A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115421181A (en) * | 2022-07-27 | 2022-12-02 | 北京超维创想信息技术有限公司 | Three-dimensional geological model phase control attribute modeling method based on deep learning |
| CN115587542A (en) * | 2022-11-02 | 2023-01-10 | 四川奥凸水处理系统工程有限公司 | Underground water inversion simulation method, system, equipment and medium based on reinforcement learning |
| CN115877464A (en) * | 2022-12-30 | 2023-03-31 | 中海石油(中国)有限公司深圳分公司 | Lithology identification method and device, computer equipment and storage medium |
| CN115937568A (en) * | 2022-09-29 | 2023-04-07 | 中国地质大学（北京） | Basalt structure background classification method, system and device and storage medium |
-
2020
- 2020-11-03 CN CN202011209672.5A patent/CN114529110A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115421181A (en) * | 2022-07-27 | 2022-12-02 | 北京超维创想信息技术有限公司 | Three-dimensional geological model phase control attribute modeling method based on deep learning |
| CN115421181B (en) * | 2022-07-27 | 2023-10-20 | 北京超维创想信息技术有限公司 | Three-dimensional geological model phase control attribute modeling method based on deep learning |
| CN115937568A (en) * | 2022-09-29 | 2023-04-07 | 中国地质大学（北京） | Basalt structure background classification method, system and device and storage medium |
| CN115937568B (en) * | 2022-09-29 | 2024-05-07 | 中国地质大学（北京） | Basalt structure background classification method, basalt structure background classification system, basalt structure background classification device and storage medium |
| CN115587542A (en) * | 2022-11-02 | 2023-01-10 | 四川奥凸水处理系统工程有限公司 | Underground water inversion simulation method, system, equipment and medium based on reinforcement learning |
| CN115877464A (en) * | 2022-12-30 | 2023-03-31 | 中海石油(中国)有限公司深圳分公司 | Lithology identification method and device, computer equipment and storage medium |
| CN115877464B (en) * | 2022-12-30 | 2024-02-13 | 中海石油(中国)有限公司深圳分公司 | Lithology recognition method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11016214B2 (en) | Dolomite reservoir prediction method and system based on well and seismic combination, and storage medium | |
| CN114529110A (en) | Lithofacies inversion method and system based on deep neural network model | |
| JP3390445B2 (en) | Seismic trace analysis method using statistical calibration techniques to estimate geological properties | |
| CN111596978A (en) | Web page display method, module and system for lithofacies classification by artificial intelligence | |
| CN104514552B (en) | A kind of method that coalbed methane reservoir identification is predicted with abundance | |
| US8255165B2 (en) | Method for predicting differences in subsurface conditions | |
| CN111665560A (en) | Oil-gas reservoir identification method and device, computer equipment and readable storage medium | |
| CN116168224A (en) | Machine learning lithology automatic identification method based on imaging gravel content | |
| US20150369937A1 (en) | Method of modelling a subsurface volume | |
| CN114066084B (en) | Method and system for predicting phase permeation curve based on machine learning | |
| Yin et al. | Enhancement of dynamic reservoir interpretation by correlating multiple 4D seismic monitors to well behavior | |
| CN112578475B (en) | Data mining-based dense reservoir dual dessert identification method | |
| CN112505754B (en) | Method for collaborative partitioning sedimentary microfacies by well-seismic based on high-precision sequence grid model | |
| CN114114414A (en) | Artificial intelligence prediction method for 'dessert' information of shale reservoir | |
| CN115877464B (en) | Lithology recognition method and device, computer equipment and storage medium | |
| CN110118994B (en) | Continental facies hydrocarbon source rock quantitative prediction method based on seismic inversion and machine learning | |
| Sarkheil et al. | The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements | |
| CN112394392B (en) | Method and device for evaluating distribution condition of hydrocarbon source rock | |
| CN115880455A (en) | Three-dimensional intelligent interpolation method based on deep learning | |
| CN112987091A (en) | Reservoir detection method and device, electronic equipment and storage medium | |
| Nath et al. | Prediction and analysis of geomechanical properties using deep learning: A Permian Basin case study | |
| CN113050191A (en) | Shale oil TOC prediction method and device based on double parameters | |
| Topór et al. | Application of machine learning tools for seismic reservoir characterization study of porosity and saturation type | |
| CN108931813B (en) | Method for identifying reef cover, reef core and reef base | |
| Deng et al. | An Automated Data-Driven Workflow for Identifying Fractured Horizontal Well Sweet Spots in Shale Reservoirs |
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 |