CN113064216A - Reservoir porosity prediction method, system, equipment and readable storage medium - Google Patents
Reservoir porosity prediction method, system, equipment and readable storage medium Download PDFInfo
- Publication number
- CN113064216A CN113064216A CN202110211673.1A CN202110211673A CN113064216A CN 113064216 A CN113064216 A CN 113064216A CN 202110211673 A CN202110211673 A CN 202110211673A CN 113064216 A CN113064216 A CN 113064216A
- Authority
- CN
- China
- Prior art keywords
- prediction
- porosity
- training
- reservoir
- dragonfly
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V11/00—Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
Landscapes
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a reservoir porosity prediction method, a reservoir porosity prediction system, reservoir porosity prediction equipment and a readable storage medium, wherein the reservoir porosity prediction method comprises the steps of preprocessing acquired original logging data, eliminating redundant data obtained from the original logging data, carrying out parameter use correlation analysis on each parameter, and selecting a parameter with high porosity correlation as a training sample input; the step length of the dragonfly algorithm is optimized by adopting a self-adaptive strategy, and the convergence and randomness of the dragonfly algorithm are improved, so that the problem that the algorithm is easy to fall into a local optimal solution is solved, the solving speed and the solution precision are improved, the reservoir parameter prediction work is more convenient and efficient, a favorable tool is provided for geologists, a support vector regression prediction model is selected, and less parameter adjustment and faster convergence speed can be realized in the process of processing small sample data; the training data is input into the model, so that the complexity of the model is effectively limited, and a better prediction result is obtained.
Description
Technical Field
The invention relates to the technical field of petroleum logging, relates to the field of reservoir parameter prediction, and particularly relates to a reservoir porosity prediction method, a reservoir porosity prediction system, reservoir porosity prediction equipment and a readable storage medium.
Background
The reservoir parameter prediction is to comprehensively utilize earthquake and logging data to evaluate and predict reservoir properties such as lithology, geometric morphology, physical parameters, oil and gas properties of a reservoir, accurately predict the parameters, particularly the porosity and permeability, provide important basis for geological comprehensive evaluation, well location deployment, reserve estimation and the like, and have very important significance for petroleum exploration and development. In the step of well placement decision making, reservoir description is essential and physical parameter estimation (porosity and permeability) is the basic requirement of the workflow. Of these two geophysical parameters, porosity describes the percentage of rock pore space that is occupied by a fluid, such as oil or gas, that may be contained in the pores. The more pores of the rock, the more oil or gas is retained in the pore space. Permeability is a measure of how open the rock allows fluid to pass through. The more permeable the rock, the easier it is for oil or gas to flow through. These two types of reservoir parameters are, to some extent, important factors in the estimation of oil and gas production reserves. At present, more than 40% of world oil production comes from natural fractured carbonate reservoirs, and evaluation of natural fractured carbonate reservoirs is always a primary consideration for researchers in the oil and gas field, but carbonate is concerned due to the highly complex interrelation of porosity, permeability and other reservoir properties. Accurate estimates of reservoir porosity and permeability are critical in both exploration and production planning schemes for oil and gas engineering.
In practice, the estimation of porosity and permeability is very difficult, and many uncertainties affect the estimation accuracy, such as sedimentary formations, lithologic mineral composition, measurement tools, data quality, calculation methods, and damage to the core sample caused by invasion of reservoir fluids and drilling fluids. In the working process of reservoir research, the physical properties of a reservoir mainly comprise porosity, permeability and saturation, and the physical properties are mainly obtained by the following two methods: the first method is laboratory core analysis, which can accurately determine porosity and permeability under strict core experiment principles. The results thus obtained, although reliable, are a time-consuming and expensive process. The second approach is to combine conventional experimental test data with well log data to perform longitudinal and transverse studies on the field by building mathematical models, but these mathematical equations are regression models based on the correlation between geophysical logs and the reservoir parameters of the core measurements, so the estimation results depend to a large extent on the equations or related models. Both of these approaches have significant limitations and drawbacks for addressing complex geological problems. At the same time, the relationship between the well logging of the rock and the geophysical parameters is non-linear and complex, making it difficult to obtain a universal solution for all wells in one survey. Therefore, there is a need for a fast and cost-effective method for determining the porosity of well logs in reservoir description and evaluation, and establishing an efficient and accurate reservoir parameter prediction method is an urgent need in current oil and gas geological research.
Disclosure of Invention
The invention aims to provide a reservoir porosity prediction method, a reservoir porosity prediction system, reservoir porosity prediction equipment and a readable storage medium, so as to overcome the defects of the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a reservoir porosity prediction method comprising the steps of:
s1, preprocessing the collected original logging data, eliminating redundant data in the original logging data, then performing parameter use correlation analysis, and preferably selecting a parameter with high porosity correlation as a training sample input;
s2, optimizing the step length of the dragonfly algorithm by adopting a self-adaptive strategy, and selecting dragonfly individuals with high adaptive values by adopting an elite strategy during sample training; and then optimizing the support vector regression parameters by using the optimized dragonfly algorithm, and finally establishing an ADA-SVR prediction model by using the optimized support vector regression parameters to realize the prediction of the porosity.
Furthermore, all the original logging data are converted into dimensionless index values, all the index values are in the same quantity level, then comprehensive evaluation analysis is carried out, the relevance of each parameter participating in the porosity is obtained, and the first N parameters with high relevance to the porosity are preferably used as training sample input.
Further, the optimized regression function f (x) is:
wherein w is a weight vector; b is an offset;
is the mapping of the input feature space to the high-dimensional feature space, and epsilon is the loss function.
Furthermore, support vector regression is realized from regression expanded by the support vector machine, and a radial basis kernel function is adopted as a kernel function of the support vector machine.
Furthermore, the step length of the adaptive strategy of the dragonfly algorithm is as follows:
where fit (t) is the function fitness value, bestf (t) and worst fitness value, respectively, and t represents the current number of iterations.
Further, when the sample is trained, the error corresponding to each group of parameters is calculated and is used as the fitness function value of each dragonfly individual in the population.
Further, before each iteration, copying the current optimal solution into the next generation, when two dragonflies which are the least suitable or the worst appear, the adaptability of the dragonflies is replaced by the dragonflies which are highly suitable, then comparing the individual fitness function value of each dragonflies calculated by each group of parameters with the error precision, if the individual fitness function value of the dragonflies is less than the error precision, finishing the training and outputting the optimal parameter value; if the dragonfly individual fitness function value is larger than the error precision, iterative training is carried out, the individual, neighborhood radius and position are updated, the training is finished until the error precision is met or the maximum iteration number is reached, the optimal parameter value is output, and the optimal solution generated in the evolution process can be prevented from being damaged.
A reservoir porosity prediction system comprises a prediction module and a training module, wherein the training module is used for carrying out model training according to a stored prediction model and logging training data and storing the trained model to the prediction module, the prediction model is established according to a support vector regression method after optimization is carried out on support vector regression parameters by a dragonfly algorithm after self-adaptive strategy optimization is adopted, and the prediction module is used for predicting porosity according to collected data.
A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method of any one of claims 1 to 7 when executing the computer program.
A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Compared with the prior art, the invention has the following beneficial technical effects:
the invention relates to a reservoir porosity prediction method, which comprises the steps of preprocessing collected original logging data, eliminating redundant data obtained from the original logging data, then carrying out parameter use correlation analysis, and preferably selecting a parameter with high porosity correlation as a training sample input; the step length of the dragonfly algorithm is optimized by adopting a self-adaptive strategy, and the convergence and randomness of the dragonfly algorithm are improved, so that the problem that the algorithm is easy to fall into a local optimal solution is solved, the solving speed and the solution precision are improved, the reservoir parameter prediction work is more convenient and efficient, a favorable tool is provided for geologists, a support vector regression prediction model is selected, and less parameter adjustment and faster convergence speed can be realized in the process of processing small sample data; the training data is input into the model, so that the complexity of the model is effectively limited, a better prediction result is obtained, the prediction accuracy is 96.3, and the method can be used as an effective tool for predicting other reservoir parameters.
Furthermore, the RBF kernel function is selected in the invention, so that the model can fully learn the internal rules of the training data sample, thereby improving the accuracy and precision of the model, and having better generalization capability than other methods.
The reservoir porosity prediction system is simple in structure and easy to implement.
Drawings
FIG. 1 is a flow chart of a porosity prediction method according to an embodiment of the present invention.
FIG. 2 is a schematic structural diagram of support vector regression in the embodiment of the present invention.
FIG. 3 is a flow chart of the dragonfly algorithm in an embodiment of the present invention.
Fig. 4 is a comparison graph of a prediction result of the prediction method and a core porosity in the embodiment of the present invention.
FIG. 5 is a graph comparing the prediction results of the ADA-SVR, DA-SVR, ELM, BP models with core porosity in an example of the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
as shown in fig. 1, a reservoir porosity prediction method includes the following steps:
s1, collecting original logging data, preprocessing the original logging data, eliminating redundant data obtained from the original logging data, analyzing the correlation of each parameter of the original logging data, and preferably selecting a parameter with high correlation with porosity as a training sample input; specifically, original logging data are converted into dimensionless index values, all the index values are in the same quantity level, then comprehensive evaluation analysis is carried out, the relevance between each parameter and porosity is obtained, and the first N parameters with high relevance to the porosity are preferably selected as training sample input;
s2, optimizing the step length of the Dragonfly Algorithm (DA) by adopting a self-adaptive strategy, and improving the convergence and randomness of the dragonfly algorithm, thereby avoiding the problem that the algorithm is easy to fall into a local optimal solution; selecting dragonfly individuals with high adaptive values by adopting an elite strategy during training samples; and then optimizing the parameters of the Support Vector Regression (SVR) by using an optimized Dragonfly Algorithm (DA), and finally establishing an ADA-SVR prediction model by using the optimized parameters of the Support Vector Regression (SVR) to realize the prediction of the porosity. Specifically, the optimized dragonfly algorithm is used for optimizing the parameters of a penalty function C and a kernel function G which support vector regression.
Support Vector Regression (SVR) is a regression method extended from Support Vector Machines (SVMs), and the Support Vector Regression (SVR) is used to build a prediction model, which has many advantages in processing small sample data, such as less parameter adjustment and faster convergence rate, and can be used to solve the problem of nonlinear optimization, as shown in fig. 2, a training sample set is set as { (x)i,yi) I ═ 1,2, …, n }, where n is the number of training samples, Δ Xt+1＝(sSi+aAi+cCi+fFi+eEi)+wΔXtIs the input vector of the i-th training sample, yie.R is the corresponding output value, and the objective of SVR is to determine the following regression function:
the formula for minimizing structural risk is:
introducing a Lagrange function, converting the Lagrange function into a dual form, and finally obtaining an optimized regression function f (x) as follows:
wherein w is a weight vector; b is an offset;
is the mapping of the input feature space to the high-dimensional feature space; c is a penalty factor; epsiloniAnd
are two relaxation variables; ε is a loss function; alpha is alphaiIs a lagrange multiplier;
is the optimal Lagrangian multiplier, b*Is the optimal offset vector;
the present invention employs a Radial Basis Function (RBF) kernel as a kernel of a support vector machine.
The dragonfly algorithm is adaptive to a search space, and the step length of an adaptive strategy is as follows:
wherein fit (t) is the function fitness value, bestf (t) is the best fitness value, worstf (t) is the worst fitness value, and t represents the current iteration number;
the position of the next generation dragonfly is the self-Adaptive Dragonfly Algorithm (ADA):
specifically, a dragonfly algorithm is adopted to optimize parameters supporting vector regression:
the next generation step size of dragonfly is calculated as follows:
ΔXt+1＝(sSi+aAi+cCi+fFi+eEi)+ωΔXt
when there is an adjacent dragonfly, the position of the next generation dragonfly is:
Xt+1＝Xt+ΔXt+1
when there is no neighboring dragonfly, the random walk behavior is set, and the position of the next generation dragonfly is:
Xt+1＝Xt+Le'vy(d)×Xt
wherein S isi、Ai、Ci、Fi、EiThe method comprises the following steps of respectively expressing five behaviors of a dragonfly algorithm:
(1) separation:
(2) alignment:
(3) cohesion:
(4) food attraction force: fi＝X+-X
(5) Repulsive force of natural enemies: ei＝X-+X
Wherein t represents the current number of iterations; siRepresenting the amount of isolation of the ith individual; x is the location of the current individual; xjIs the position of the adjacent individual j; n is the number of adjacent individuals; a. theiRepresenting the alignment of the ith individual; vjThe speed of the jth neighboring individual; ciRepresenting the amount of cohesion of the ith individual; fiIndicating the attraction of the ith individual to food; x+Indicating the location of the food; eiIndicating the distance that the ith individual needs to escape from the natural enemy; x-Indicating the location of the natural enemy. XtRepresenting the current position of the t generation population; Δ Xt+1Representing the next generation population position updating step length; Δ Xi+1Representing the individual position of the next generation population; s, a, c, f, e correspond to the weights of 5 behaviors, respectively; w is the inertial weight; d is the dimension of the individual position vector, r1，r2Is [0, 1 ]]Random number, Γ (x) ═ x-1! And β is a constant (here taken to be 0.5).
As shown in fig. 3, in the case of training a sample, calculating an error corresponding to each group of parameters, taking the error as a fitness function value of each dragonfly individual in the population, copying the current optimal solution to the next generation before each iteration, replacing the adaptability of the two dragonflies that are most unsuitable or worst when the two dragonflies appear with highly suitable dragonflies, and then comparing the fitness function value of each dragonfly individual calculated by each group of parameters with the error precision, if the fitness function value of each dragonfly individual is less than the error precision, ending the training and outputting the optimal parameter value; if the dragonfly individual fitness function value is larger than the error precision, iterative training is carried out, the individual, neighborhood radius and position are updated, the training is finished until the error precision is met or the maximum iteration number is reached, the optimal parameter value is output, and the optimal solution generated in the evolution process can be prevented from being damaged.
And (3) simulating an experiment selection on an MATLAB software platform, and constructing an ADA-SVR prediction model on the basis of a neural network toolbox carried by the MATLAB so as to perform experimental analysis to obtain an optimal model for predicting porosity.
The invention adopts the oil well logging data of 121 to 2430 meters in a certain area, 738 groups of data are collected from the 121 to 2430 meters depth interval for the research, and the first step is to pre-process the logging data and remove abnormal logging data before the experiment so as to accurately predict the porosity. The second step is to determine the parameters of the input and output of the model, including raw logging data and measured porosity data. Considering that some redundant data may affect the logging result in the used logging parameters, and the quality of the final prediction result of the prediction model is greatly related to the selection of the input parameters, when the network is trained, the response relationship between different types of logging parameters and the porosity is firstly analyzed, and the conventional logging parameters with the best correlation to the porosity are preferentially selected. The conventional logging parameters comprise acoustic time difference (AC), Compensated Neutron (CNL), compensated Density (DEN), natural Gamma (GR), deep induction Resistivity (RILD), microelectrode (RMN), argillaceous content (Vsh) and natural potential (SP), and the conventional logging parameters AC, CNL, DEN and GR are necessarily related to the measured porosity according to the sequence of the correlation coefficients from large to small. Therefore, four conventional logging parameters, AC, CNL, DEN, and GR, are selected as inputs to the neural network and porosity is targeted as an output in the present invention. In the sample data, a total of 739 sets of data samples are selected, and all available data sets are randomly partitioned into two different subsets, including training and testing subsets. The proportion of each of these subsets is 80% and 20% of the entire data set, respectively. In the experiment, 80% of data is randomly selected to train the artificial neural network, the rest 20% of data is used to test the network, and in order to further ensure the fairness and accuracy of the obtained result, the experiment is iterated for multiple times and the running average value is obtained. Here, a radial basis kernel function (RBF) is used to construct a Support Vector Regression (SVR) model, which depends mainly on a penalty function C and a kernel parameter G, which are preferably synchronized using an Adaptive Dragonfly Algorithm (ADA).
In order to verify the validity of the porosity SVR prediction model of the adaptive dragonfly algorithm, experiments were performed on the well log dataset and compared to the traditional classification algorithm: comparison is carried out on a Back Propagation (BP) neural network, an Extreme Learning Machine (ELM) and a standard DA-SVR model (without adding an adaptive strategy and an elite meaning), and the results are shown in fig. 4 and fig. 5 and table 1, which prove that the porosity SVR prediction model based on the adaptive dragonfly algorithm constructed by the invention has better performance in reservoir parameter prediction. The invention effectively limits the complexity of the model, and obtains a better prediction result with a prediction accuracy of 96.3% compared with other traditional neural network models. Therefore, the porosity SVR prediction method (ADA-SVR) based on the self-adaptive dragonfly algorithm has feasibility and effectiveness when used for predicting the porosity, and can be used as an effective tool for predicting other reservoir parameters.
TABLE 1
The method optimizes the dragonfly algorithm by using a self-adaptive strategy and an elite, and then optimizes the parameters supporting vector regression by using the improved dragonfly algorithm, so that the solving speed and the solution precision are improved, the reservoir parameter prediction work is more convenient and efficient, and a favorable tool is provided for geologists. According to the method, a Support Vector Regression (SVR) prediction model is selected, so that less parameter adjustment and higher convergence rate can be realized when small sample data are processed; the training data is input into the model, and the RBF kernel function is selected in the invention, so that the model fully learns the internal rules of the training data sample, thereby improving the accuracy and precision of the model, and having better generalization capability than other methods.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A reservoir porosity prediction method, comprising the steps of:
s1, preprocessing the collected original logging data, eliminating redundant data in the original logging data, then performing parameter use correlation analysis, and preferably selecting a parameter with high porosity correlation as a training sample input;
s2, optimizing the step length of the dragonfly algorithm by adopting a self-adaptive strategy, and selecting dragonfly individuals with high adaptive values by adopting an elite strategy during sample training; and then optimizing the support vector regression parameters by using the optimized dragonfly algorithm, and finally establishing an ADA-SVR prediction model by using the optimized support vector regression parameters to realize the prediction of the porosity.
2. The method for predicting the porosity of the reservoir according to claim 1, wherein the original well logging data are converted into dimensionless index values, each index value is in the same number level, and then comprehensive evaluation analysis is performed to obtain the relevance between each index value and the porosity, preferably the first N parameters with high relevance to the porosity are used as training sample input.
3. A reservoir porosity prediction method according to claim 1, wherein the optimized regression function f (x) is:
wherein w is a weight vector; b is an offset;
is the mapping of the input feature space to the high-dimensional feature space, and epsilon is the loss function.
4. The reservoir porosity prediction method of claim 1, wherein the support vector regression is implemented from support vector machine extended regression, and a radial basis kernel function is adopted as a kernel function of the support vector machine.
5. The reservoir porosity prediction method of claim 1, wherein the dragonfly algorithm is made adaptive to the search space, and the adaptive strategy step size is as follows:
where fit (t) is the function fitness value, bestf (t) and worst fitness value, respectively, and t represents the current number of iterations.
6. The method as claimed in claim 1, wherein when training the sample, calculating the error corresponding to each set of parameters, and using the error as the fitness function value of each dragonfly individual in the population.
7. The method of claim 6, wherein before each iteration, the current optimal solution is copied into the next generation, when two dragonflies that are the least suitable or worst appear, the adaptability of the current optimal solution is replaced by the dragonflies that are highly suitable, after that, the individual fitness function value of each dragonflies calculated by each group of parameters is compared with the error precision, if the individual fitness function value of the dragonflies is less than the error precision, the training is finished, and the optimal parameter value is output; if the dragonfly individual fitness function value is larger than the error precision, iterative training is carried out, the individual, neighborhood radius and position are updated, the training is finished until the error precision is met or the maximum iteration number is reached, the optimal parameter value is output, and the optimal solution generated in the evolution process can be prevented from being damaged.
8. A reservoir porosity prediction system is characterized by comprising a prediction module and a training module, wherein the training module is used for carrying out model training according to a stored prediction model and logging training data and storing the trained model to the prediction module, the prediction model is established according to a support vector regression method after optimization is carried out on support vector regression parameters by a dragonfly algorithm after self-adaptive strategy optimization is adopted, and the prediction module is used for predicting porosity according to collected data.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the computer program is executed by the processor.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110211673.1A CN113064216A (en) | 2021-02-25 | 2021-02-25 | Reservoir porosity prediction method, system, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110211673.1A CN113064216A (en) | 2021-02-25 | 2021-02-25 | Reservoir porosity prediction method, system, equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113064216A true CN113064216A (en) | 2021-07-02 |
Family
ID=76559163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110211673.1A Pending CN113064216A (en) | 2021-02-25 | 2021-02-25 | Reservoir porosity prediction method, system, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113064216A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117807859A (en) * | 2022-10-13 | 2024-04-02 | 任丘市智科石油技术有限公司 | Porosity and permeability calculation method and system based on machine learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130080066A1 (en) * | 2011-09-28 | 2013-03-28 | Saudi Arabian Oil Company | Reservoir properties prediction with least square support vector machine |
| CN106556863A (en) * | 2015-09-29 | 2017-04-05 | 中国石油天然气股份有限公司 | Porosity prediction method based on Depth Domain prestack angle gathers |
| US20180018561A1 (en) * | 2016-07-14 | 2018-01-18 | King Fahd University Of Petroleum And Minerals | Apparatuses, systems, and methodologies for permeability prediction |
| CN110084435A (en) * | 2019-05-07 | 2019-08-02 | 合肥工业大学 | A kind of gas reservoir means of interpretation and system |
| CN112230283A (en) * | 2020-10-12 | 2021-01-15 | 北京中恒利华石油技术研究所 | Seismic porosity prediction method based on logging curve support vector machine modeling |
-
2021
- 2021-02-25 CN CN202110211673.1A patent/CN113064216A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130080066A1 (en) * | 2011-09-28 | 2013-03-28 | Saudi Arabian Oil Company | Reservoir properties prediction with least square support vector machine |
| CN106556863A (en) * | 2015-09-29 | 2017-04-05 | 中国石油天然气股份有限公司 | Porosity prediction method based on Depth Domain prestack angle gathers |
| US20180018561A1 (en) * | 2016-07-14 | 2018-01-18 | King Fahd University Of Petroleum And Minerals | Apparatuses, systems, and methodologies for permeability prediction |
| CN110084435A (en) * | 2019-05-07 | 2019-08-02 | 合肥工业大学 | A kind of gas reservoir means of interpretation and system |
| CN112230283A (en) * | 2020-10-12 | 2021-01-15 | 北京中恒利华石油技术研究所 | Seismic porosity prediction method based on logging curve support vector machine modeling |
Non-Patent Citations (2)
| Title |
|---|
| ZHONGWEI LI ET AL.: "Prediction and application of porosity based on support vector regression model optimized by adaptive dragonfly algorithm", 《ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS》 * |
| 邓社根等: "砂泥岩储层孔隙度预测的改进模型", 《黑龙江科技大学学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117807859A (en) * | 2022-10-13 | 2024-04-02 | 任丘市智科石油技术有限公司 | Porosity and permeability calculation method and system based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109611087B (en) | Volcanic oil reservoir parameter intelligent prediction method and system | |
| CN110717249B (en) | Shale gas reservoir logging porosity rapid prediction method and system | |
| US20040254734A1 (en) | Method for optimizing production of an oil reservoir in the presence of uncertainties | |
| CN108563837B (en) | Method and system for correcting model parameters of alluvial river water sand model in real time | |
| CN114723095A (en) | Missing well logging curve prediction method and device | |
| CN113009553A (en) | Interpretable seismic motion parameter probability density distribution prediction method based on NGboost and SHAP values | |
| CN105205239A (en) | Method and device for modeling reservoir physical property parameter | |
| CN112861890B (en) | Reservoir evaluation model construction method and reservoir identification method | |
| CN112253087A (en) | Biological disturbance reservoir physical property calculation method based on multi-source logging data | |
| CN110988997A (en) | Hydrocarbon source rock three-dimensional space distribution quantitative prediction technology based on machine learning | |
| CN113064216A (en) | Reservoir porosity prediction method, system, equipment and readable storage medium | |
| CN111751878B (en) | Method and device for predicting transverse wave speed | |
| CN117251802A (en) | Heterogeneous reservoir parameter prediction method and system based on transfer learning | |
| Sarkheil et al. | The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements | |
| US20230141334A1 (en) | Systems and methods of modeling geological facies for well development | |
| CN115238860B (en) | Leakage pressure prediction model generation method and device | |
| CN111273349A (en) | Transverse wave velocity extraction method and processing terminal for seabed shallow sediment layer | |
| Anifowose et al. | A functional networks-type-2 fuzzy logic hybrid model for the prediction of porosity and permeability of oil and gas reservoirs | |
| CN113240161B (en) | Method and device for establishing net present value prediction model, storage medium and electronic equipment | |
| CN111594156B (en) | Natural gas hydrate saturation calculation method and system | |
| CN110032777B (en) | Drilling fluid density safety window estimation method based on uncertainty analysis | |
| CN114764546A (en) | Method for predicting residual oil of fault block oil reservoir based on generation of antagonistic neural network | |
| CN112946754B (en) | Reservoir porosity prediction method and device | |
| CN117407841B (en) | Shale layer seam prediction method based on optimization integration algorithm | |
| CN108132486A (en) | A kind of heavy magnetic gradient combines the Optimized Simulated annealing method of Interface Inversion with seismic data |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210702 |
|
| RJ01 | Rejection of invention patent application after publication |