CN114741944A - Method for predicting logging curve by using machine learning and deep learning algorithm - Google Patents
Method for predicting logging curve by using machine learning and deep learning algorithm Download PDFInfo
- Publication number
- CN114741944A CN114741944A CN202210154668.6A CN202210154668A CN114741944A CN 114741944 A CN114741944 A CN 114741944A CN 202210154668 A CN202210154668 A CN 202210154668A CN 114741944 A CN114741944 A CN 114741944A
- Authority
- CN
- China
- Prior art keywords
- curve
- machine learning
- log
- target
- predicting
- 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
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 82
- 238000010801 machine learning Methods 0.000 title claims abstract description 44
- 238000013135 deep learning Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000007637 random forest analysis Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 239000011435 rock Substances 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000003908 quality control method Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 239000004215 Carbon black (E152) Substances 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005755 formation reaction Methods 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 229930195733 hydrocarbon Natural products 0.000 description 2
- 150000002430 hydrocarbons Chemical group 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 239000003208 petroleum Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007596 consolidation process Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000011707 mineral Substances 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
- 239000003209 petroleum derivative Substances 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
- G06N3/048—Activation functions
-
- 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
- G06N3/08—Learning methods
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Animal Husbandry (AREA)
- Economics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Agronomy & Crop Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for predicting a logging curve by utilizing a machine learning and deep learning algorithm, which comprises the following steps of 1) selecting a modeling well containing a complete logging sequence curve, selecting a target logging curve to be predicted from the logging sequence curve and establishing a characteristic logging curve of the deep learning algorithm; 2) directly establishing a deep learning neural network between the target logging curve and the characteristic logging curve by using a neural network algorithm; 3) training through machine learning based on the neural network, and comparing different machine learning algorithms, thereby selecting a final prediction algorithm model; and 4) applying the final prediction algorithm model to the target well, and predicting a target logging curve of the target well.
Description
Technical Field
The invention relates to the technical field of petroleum and natural gas, in particular to a method for predicting a logging curve by utilizing a machine learning and deep learning algorithm.
Background
In the current field of oil and gas exploration, oil and gas well data are the data that most directly reflect the characteristics of underground hydrocarbon-bearing formations, and well logs are one of the most important basic research data. Because many wells do not have a complete log sequence for historical reasons or cost factors, some logs need to be predicted. All current well log predictions are based on petrophysical models, which may be referred to as "model-driven" data prediction methods. This approach has some drawbacks.
The limitations of the prior art methods are further described below by way of example of shear wave curve prediction. In rock mechanics, it is necessary to calculate relevant mechanical parameters using the velocity of longitudinal waves and shear waves, and therefore, the velocity of shear waves is also very important. However, in an oil field site, only a few wells in a work area may have actually measured shear wave velocity curves, especially shear wave data in an old well area is lacked, and the conventional method mainly uses various rock physical models, such as a Biot-Gassmann fluid substitution model, an Xu-White model and the like, as a basis to introduce consolidation coefficients to establish a pore medium rock physical model method for predicting the shear wave velocity. The main disadvantages based on petrophysical models include the following: 1. the rock physical model is very complex and has many parameters, so that the parameter values are difficult to accurately set, and the logging curve prediction result is inaccurate, for example, the underground pressure and temperature change along with the depth of a well, the mineral structure of rock is also very complex, and the average value of a certain interval is difficult to represent the characteristics of the whole interval; 2. the petrophysical model itself is inaccurate. At present, all rock physical models are based on laboratory test data, and then regression fitting is carried out by utilizing a rock physical quantity plate according to data points obtained by testing, wherein rock samples come from different oil fields, most of the rock samples are foreign oil fields and cannot necessarily represent rock characteristics of the Chinese oil field. 3. Labor costs are high, for example, a high-grade petrophysicist needs at least 3 days to complete a well for shear curve prediction.
There is therefore a need for new techniques and methods to at least fill the gap in the prior art or eliminate the deficiencies of the prior art.
Disclosure of Invention
The invention aims to provide a method for predicting a logging curve by utilizing a machine learning and deep learning algorithm, which is completely different from all transverse wave prediction software and technologies in the current market, is a set of logging curve prediction technology based on an artificial intelligence technology (mainly machine learning and deep learning), and has the basic technology and theoretical basis not based on Model drive (Model drive) but Data drive (Data drive). The basic theory assumes that different log sequences contain different properties of the subsurface rock (such as lithology, porosity, water saturation, brittleness, hydrocarbon content, etc.), and thus the different logs, although measuring data in different categories, are closely related to each other because they are all the same formation being measured. Therefore, the oil data (such as well data, geological data and geophysical data) can be more efficiently mined to find more oil resources for the country by utilizing the data-driven artificial intelligence technology, such as machine learning and deep learning, so as to serve the energy safety and double-carbon transformation of the country.
According to an aspect of the invention, there is provided a method of predicting a well log using machine learning and deep learning algorithms, comprising
1) Selecting a modeling well containing a complete logging sequence curve, selecting a target logging curve to be predicted from the logging sequence curves and establishing a characteristic logging curve of a deep learning algorithm;
2) directly establishing a deep learning neural network between the target logging curve and the characteristic logging curve by using a neural network algorithm;
3) training through machine learning based on the neural network, and comparing different machine learning algorithms, thereby selecting a final prediction algorithm model; and
4) and applying the final prediction algorithm model to the target well to predict a target logging curve of the target well.
According to an embodiment of the invention, wherein the different machine learning algorithms comprise random forest algorithms, KNN, XGBoost, SVM and ANN algorithms.
According to an embodiment of the present invention, the method for predicting a well log by using machine learning and deep learning algorithms further comprises performing quality analysis on the predicted target well log, for example, in terms of final comparison and quality control of the predicted data and the measured data, comparison may be performed by other technical means, such as curve superposition display, relative error analysis, and the like.
According to an embodiment of the invention, wherein the log sequence curves comprise density, sonic, natural potential, GR, resistivity, porosity curves.
According to an embodiment of the present invention, wherein the neural network comprises an input layer, a plurality of hidden layers and an output layer, the hidden layers are composed of at least four nodes.
According to an embodiment of the present invention, wherein the porosity curve is a target log and the sonic, natural potential, resistivity, GR curve is a characteristic log.
Compared with the prior art, the invention can realize the beneficial technical effects that: 1. the labor cost is very low; for example, currently, at least one senior petrophysicist needs 3 days to complete the transverse wave curve prediction of one well, and the prediction by the method can be basically completed within 1 hour by adopting computer calculation (the specific operation time is determined according to the performance of the computer), so that labor and time are saved; 2. the accuracy is very high, and the correlation coefficient of a predicted curve and a real curve exceeds 90% by verifying real logging data of a certain oil field; 3. by utilizing the data-driven artificial intelligence technology, such as machine learning and deep learning technology, the petroleum data (such as well data, geological data and geophysical data) are more efficiently mined, so that more petroleum resources can be found, and the energy safety and double-carbon conversion service of the country can be realized. Therefore, the technology of the invention has great innovation and practicability.
Drawings
FIG. 1 is a schematic flow diagram of a method for predicting a well log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention;
FIG. 2 is a schematic diagram of a constructed deep-learning neural network, according to one embodiment of the present invention;
FIG. 3 is a graph of data readings for an example application of a method for predicting a well log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention;
FIG. 4 is a graph of data reading analysis results of an example application of a method for predicting a well log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention;
FIG. 5 is a graphical illustration of the results of various machine learning algorithms of an example application of a method for predicting a well log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention;
FIG. 6 is a graph of predicted porosity log results for an example application of a method for predicting a log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention;
FIG. 7 is a graph of predicted porosity log result validation for an example application of a method of predicting a log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention; and
FIG. 8 is a graph comparing a predicted porosity log to an actual measured porosity log for an example application of a method for predicting a log using machine learning and deep learning algorithms in accordance with one embodiment of the present invention.
Detailed Description
The invention will be better understood from the following examples and the accompanying drawings. However, those skilled in the art will readily appreciate that the description of the embodiments is only for the purpose of illustration and should not be taken as a limitation of the invention.
It should be understood that the machine learning and deep learning algorithms referred to in the present invention, such as neural network algorithms, random forest algorithms, KNN, XGBoost, SVM algorithms, etc., are known per se, such as sub-modules of the model, various parameters, operation principle mechanisms, etc., and thus the present invention focuses on the combination of well logs, machine learning and deep learning algorithms, etc.
Referring to FIG. 1, according to an embodiment of the present invention, there is provided a method of predicting a well log using machine learning and deep learning algorithms, which may include the steps of:
step 1), selecting a well containing a complete logging sequence curve, selecting a target logging curve to be predicted and establishing a characteristic logging curve of a deep learning algorithm. Because logs have a certain sampling rate and are usually limited by economic considerations, some logs only measure the interval of interest. Thus, multiple such wells may be selected to determine that the input data has sufficient sample points to build a reasonable machine learning and deep learning algorithm model. The log sequence curves may include, for example, density, sonic, natural potential, GR, resistivity, porosity curves, and the like. For example, a porosity curve may be selected as the target log, and other curves such as density, sonic, natural potential, GR, resistivity, etc. may be used as the characteristic logs. It should be understood that the wells selected herein are used to create a predictive model, which may be referred to as modeled wells, to distinguish them from the target wells subsequently used to predict the subject using the modeling method created by the present invention. After the target well log and the characteristic well log are determined, the data may be further detected and analyzed, for example, basic analysis and data quality control, such as checking whether there are abnormal points in the data, or whether the data is unreasonable, etc., to improve the data quality.
And 2) directly establishing a deep learning neural network on the target logging curve and the characteristic logging curve by using a neural network algorithm. Fig. 2 is a schematic diagram of a deep learning neural network constructed according to an embodiment of the present invention. As shown, the fully-connected neural network may include an input layer, a hidden layer and an output layer, the hidden layer being composed of four nodes; of course, multiple hidden layers, and more nodes, may be included. The input layer data can be received by establishing a neural network mapping relation, such as a hidden layer, weight summation is distributed to each dimension Xi of the data, weighting summation is carried out on the data and bias b are summed, and finally a nonlinear function f is introduced to serve as an activation function to be output. Through the operation, the neural network can complete a relatively complex nonlinear mapping task. In this embodiment, the input layer corresponds to the characteristic log data, and the output layer corresponds to the target log.
And 3) after the deep learning algorithm is constructed, optimizing the machine learning algorithm, namely comparing different machine learning algorithms, and selecting a final prediction algorithm model. More commonly used algorithms may include Random Forest (Random Forest) algorithms, KNN, XGBoost, SVM, and ANN algorithms, among others. In training by machine learning, the data of the modeled well may be divided into two parts, one part being training data, which may be 80% data for example, and the other part being verification data, which may be 20% data for example.
The XGboost algorithm is used as an example for explanation, is an optimized distributed gradient enhancement library and aims to achieve high efficiency, flexibility and portability. The method realizes a machine learning algorithm under a Gradient Boosting framework. XGBoost provides parallel tree lifting (also known as GBDT, GBM) that can quickly and accurately solve many data science problems. The same code runs on the main distributed environment (Hadoop, SGE, MPI) and can solve the problem of over billions of samples. XGBoost takes advantage of extra-core computing and enables data scientists to process hundreds of millions of sample data on one host. Finally, these techniques are combined to make an end-to-end system that extends to larger data sets with minimal clustering.
The method for well logging curve prediction based on the XGboost algorithm can comprise the following steps:
(1) and establishing a logging curve prediction neural network structure, and carrying out normalization preprocessing on input data. Normalization techniques are well known in the art and will not be described in detail herein;
(2) the optimized objective function of XGboost is
(3) And continuously iterating by a global optimization method, and searching an optimal value to meet the optimization objective function to obtain an optimal solution. The optimization method mainly comprises a precise Greedy Algorithm (Basic Exact Greedy Algorithm) and an approximation Algorithm (Approximate Algorithm). The invention can adopt a Basic Exact Greedy Algorithm (Basic Exact Greedy Algorithm), namely, all values of all characteristics are sequenced, and then Gini of each point is compared to find out the node with the largest change. When the feature is a continuous feature, discretizing a continuous value, and taking the average value of two points as a segmentation node. And traversing all the characteristics of the whole sample by a sorting algorithm, and sorting to finally obtain a global optimal solution.
(4) And fixing the obtained related parameters, training a model, and finally obtaining a logging curve prediction model.
Meanwhile, the quality control can be performed by performing blind test (blid test) on other sampling test sample points of the training well which do not participate in the training.
(5) And applying the final prediction algorithm model to the target well to predict a target logging curve of the target well. The selected target well lacks a target logging curve, so that the target logging curve, such as a shear wave curve and the like, can be predicted based on the existing characteristic curve in the target well by using the prediction algorithm model of the invention.
According to an embodiment of the present invention, the method may further comprise performing a quality analysis on the predicted target well log of the target well. For example, the comparison between the predicted data and the measured data and the quality control can be performed by other technical means, such as curve superposition display and relative error analysis.
The following is a further description with reference to specific examples.
In the research, a certain experimental work area is selected as a research area, wherein a plurality of wells have complete logging sequence curves, including density, sound wave, natural potential, resistivity, GR curve and the like. The porosity curve is then selected as our target log and the other curves as our characteristic log. The well log is first input and normalized, and the result is shown in fig. 3. All curve data read are shown in fig. 4, and the data-missing interval is analyzed. Based on this, it can be determined, for example, whether the number of samples is sufficient, and if not, it can be supplemented.
In this study, the natural potential, GR curve, sonic curve, and resistivity RT curve were selected as the characteristic log. And selecting the porosity curve as a target logging curve, and constructing a machine learning and depth learning algorithm by using the target logging curve and the characteristic logging curve of the known well (modeling well). Fig. 5 is a schematic diagram showing comparison of results of various machine learning algorithms in an application example of a method for predicting a logging curve by using machine learning and deep learning algorithms according to an embodiment of the present invention, and a final prediction algorithm model is constructed by selecting a random Forest algorithm as a preferred algorithm from a random Forest (Radom Forest) algorithm, a KNN algorithm, an SVR algorithm, and an XGBoost algorithm according to simulation results of various machine learning algorithms. More specifically, as shown in fig. 5, the results of the various methods include two columns, the left column showing the correlation coefficient between training data sample points and real data, and the right column showing the correlation coefficient between data sample points not involved in training and real data. As can be seen from the figure, the correlation coefficient of the random forest algorithm is relatively good and thus preferred.
The prediction algorithm model is applied to the well (target well) missing the target log curve, predicting the porosity curve. The results are shown in FIG. 6. At the same time, the verification was performed using the measured values (true values), and the results are shown in fig. 7 and 8. The left graph in fig. 7 shows the correlation coefficient between the training data sample point and the real data, and the right graph shows the correlation coefficient between the prediction data sample point and the real data, both of which are much larger than 0.9. The predicted porosity curve and the actual porosity curve are displayed in parallel, so as to further determine the accuracy of the prediction result of the invention, as shown in fig. 8, a black curve is a true porosity log curve, and a red curve is a predicted porosity log curve. It can be seen that the similarity between the two is extremely high.
The specific embodiments are given above, but the present invention is not limited to the above-described embodiments. The basic idea of the present invention lies in the above basic scheme, and it is obvious to those skilled in the art that no creative effort is needed to design various modified models, formulas and parameters according to the teaching of the present invention. Variations, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the invention, and still fall within the scope of the invention.
Claims (6)
1. A method for predicting a well log using machine learning and deep learning algorithms, comprising,
1) selecting a modeling well containing a complete logging sequence curve, selecting a target logging curve to be predicted from the logging sequence curves and establishing a characteristic logging curve of a deep learning algorithm;
2) directly establishing a deep learning neural network between the target logging curve and the characteristic logging curve by using a neural network algorithm;
3) training through machine learning based on the neural network, and comparing different machine learning algorithms, thereby selecting a final prediction algorithm model; and
4) and applying the final prediction algorithm model to the target well to predict a target logging curve of the target well.
2. The method of predicting a well log using machine learning and deep learning algorithms of claim 1, wherein the different machine learning algorithms comprise random forest, KNN, XGBoost, SVM and ANN algorithms.
3. The method for predicting well logs using machine learning and deep learning algorithms of claim 1, further comprising performing a quality analysis of the predicted target well log.
4. The method of predicting a log curve using machine learning and depth learning algorithms of claim 1, wherein the log sequence curve comprises density, sonic, natural potential, GR, resistivity, porosity curve.
5. The method of predicting a log using machine learning and deep learning algorithms of claim 1, wherein the neural network comprises an input layer, a plurality of hidden layers, and an output layer, the hidden layers comprising at least four nodes.
6. The method of predicting a log using machine learning and deep learning algorithms of claim 4, wherein the porosity curve is a target log and the sonic, natural potential, resistivity, GR curves are characteristic logs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210154668.6A CN114741944A (en) | 2022-02-21 | 2022-02-21 | Method for predicting logging curve by using machine learning and deep learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210154668.6A CN114741944A (en) | 2022-02-21 | 2022-02-21 | Method for predicting logging curve by using machine learning and deep learning algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114741944A true CN114741944A (en) | 2022-07-12 |
Family
ID=82275893
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210154668.6A Pending CN114741944A (en) | 2022-02-21 | 2022-02-21 | Method for predicting logging curve by using machine learning and deep learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114741944A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115576028A (en) * | 2022-12-01 | 2023-01-06 | 武汉盛华伟业科技股份有限公司 | Geological feature layer prediction method and system based on support vector machine |
| CN116259168A (en) * | 2023-05-16 | 2023-06-13 | 陕西天成石油科技有限公司 | Alarm method and device for oilfield logging |
-
2022
- 2022-02-21 CN CN202210154668.6A patent/CN114741944A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115576028A (en) * | 2022-12-01 | 2023-01-06 | 武汉盛华伟业科技股份有限公司 | Geological feature layer prediction method and system based on support vector machine |
| CN115576028B (en) * | 2022-12-01 | 2023-03-14 | 武汉盛华伟业科技股份有限公司 | Geological feature layer prediction method and system based on support vector machine |
| CN116259168A (en) * | 2023-05-16 | 2023-06-13 | 陕西天成石油科技有限公司 | Alarm method and device for oilfield logging |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Solanki et al. | Artificial intelligence: New age of transformation in petroleum upstream | |
| Syed et al. | Application of ML & AI to model petrophysical and geomechanical properties of shale reservoirs–A systematic literature review | |
| Syed et al. | Smart shale gas production performance analysis using machine learning applications | |
| CN111596978A (en) | Web page display method, module and system for lithofacies classification by artificial intelligence | |
| Rajabi et al. | Predicting shear wave velocity from conventional well logs with deep and hybrid machine learning algorithms | |
| Liu et al. | Permeability predictions for tight sandstone reservoir using explainable machine learning and particle swarm optimization | |
| Tariq et al. | A holistic approach to develop new rigorous empirical correlation for static Young's modulus | |
| CN114741944A (en) | Method for predicting logging curve by using machine learning and deep learning algorithm | |
| CN113534261A (en) | Reservoir gas content detection method and device based on intelligent optimization integrated network | |
| Li et al. | A data-driven field-scale approach to estimate the permeability of fractured rocks | |
| Da Silva et al. | Support vector regression for petroleum reservoir production forecast considering geostatistical realizations | |
| Zhong et al. | Coal identification using neural networks with real-time coalbed methane drilling data | |
| Hou et al. | Data-driven optimization of brittleness index for hydraulic fracturing | |
| Yang et al. | Deep-learning missing well-log prediction via long short-term memory network with attention-period mechanism | |
| Safaei-Farouji et al. | An advanced computational intelligent framework to predict shear sonic velocity with application to mechanical rock classification | |
| Roudini et al. | Proxy-based Bayesian inversion of strain tensor data measured during well tests | |
| Wood | Bakken stratigraphic and type well log learning network exploited to predict and data mine shear wave acoustic velocity | |
| Mutalova et al. | Machine learning on field data for hydraulic fracturing design optimization | |
| Khilrani et al. | Contrasting machine learning regression algorithms used for the estimation of permeability from well log data | |
| Khalili et al. | Reservoir modeling & simulation: Advancements, challenges, and future perspectives | |
| Sarkheil et al. | The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements | |
| Tariq et al. | An Intelligent Safe Well Bottom-Hole Pressure Monitoring of CO2 Injection Well into Deep Saline: A coupled Hydro-Mechanical Approach | |
| Gonçalves et al. | Predicting oil field production using the Random Forest algorithm | |
| Huang et al. | Quantitative analysis of the main controlling factors of oil saturation variation | |
| Wood | Bakken stratigraphic and type well-log learning network for transparent prediction and rigorous data mining |
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 |