Reservoir physical property parameter prediction and evaluation system for oil-gas exploration

Abstract

The invention relates to a reservoir physical property parameter prediction and evaluation system for oil-gas exploration, and aims to deal with exploration challenges under complex geological conditions. The system comprises a multi-source data acquisition module, a self-adaptive data processing module, a self-learning prediction module, a dynamic model correction module and a visual result display module. By integrating logging, core analysis, earthquake and historical exploration data, the system performs high-precision prediction on the porosity and permeability of the reservoir, and updates the model in real time through a dynamic correction technology to ensure the accuracy and reliability of a prediction result. In addition, the system also provides two-dimensional and three-dimensional visual display and uncertainty analysis functions, and assists a decision maker in identifying a high-potential reservoir area and making an optimal development scheme. The system has the interface connection capability with an external data management system, can realize automatic updating and economic analysis of data, and effectively improves the efficiency and success rate of oil-gas exploration.

Description

Technical Field

[0001] The invention relates to the technical field of oil and gas exploration, and in particular to a reservoir physical property parameter prediction and evaluation system for oil and gas exploration. Background Art

[0002] Oil and gas exploration is an extremely important part of the energy sector. With the continuous growth of global energy demand, the exploitation and exploration of oil and gas resources have become increasingly important. However, oil and gas exploration in a complex and changing geological environment is full of challenges. These challenges include not only the accurate prediction of reservoir properties, but also how to efficiently collect and process geological data from different sources to improve the success rate and economic benefits of exploration. Traditional oil and gas exploration methods often fail to provide sufficient accuracy and reliability when faced with complex geological conditions, thus affecting the effective development of resources.

[0003] 1. Limitations of Traditional Oil and Gas Exploration Methods

[0004] Traditional oil and gas exploration technologies mainly rely on geological exploration, seismic exploration, well logging analysis and other means. Although these methods have been widely used in the past few decades, their prediction results are often restricted by various factors. For example, the quality of well logging data is often affected by geological conditions, instrument accuracy and measurement environment, resulting in large uncertainties in the prediction results. In addition, traditional methods often lack effective data processing and integration methods when processing large-scale, multi-source data, resulting in low information utilization and failure to fully utilize the value of data.

[0005] 2. Difficulties in predicting reservoir properties under complex geological conditions

[0006] Under complex geological conditions, it becomes more difficult to predict reservoir physical parameters (such as porosity, permeability, etc.). Reservoir physical parameters are important factors affecting oil and gas reserves and recovery rates. Therefore, accurate prediction of these parameters is of great significance to oil and gas exploration. However, the diversity of geological structures, the complexity of reservoir lithology, and the interweaving of various geological processes make it difficult for traditional exploration methods to cope with them. These complex geological conditions may include the influence of faults, folds, volcanic activity, and other geological events, all of which may cause the reservoir physical parameters to be highly heterogeneous and uncertain.

[0007] In addition, in complex geological environments, different logging methods may produce different results. For example, acoustic logging, density logging, and resistivity logging each have their own limitations and scope of application. When these methods are applied to different types of reservoirs, inconsistent prediction results may be obtained. How to extract valuable information from these different sources of data and integrate them into a unified prediction model is an urgent problem to be solved in oil and gas exploration.

[0008] 3. Application of big data and artificial intelligence technologies in oil and gas exploration

[0009] With the development of big data and artificial intelligence technologies, the oil and gas exploration field has begun to gradually introduce these emerging technologies to cope with the limitations of traditional methods. Big data technology has greatly improved the data collection and processing capabilities in the exploration process, and can effectively process massive geological data from different sources. Through pre-processing technologies such as data cleaning, normalization, and outlier detection, the quality of data can be significantly improved and the impact of measurement errors can be reduced.

[0010] Artificial intelligence, especially the introduction of adaptive algorithms and machine learning technologies, provides a new approach to the prediction of reservoir physical parameters. These technologies can learn the relationship between geological features and reservoir physical parameters from a large amount of historical exploration data, thereby providing more accurate prediction results when new data is input. The self-learning prediction module can automatically adjust the model parameters based on the continuously updated data, making the prediction results more in line with the actual situation.

[0011] 4. Necessity of real-time correction and dynamic prediction

[0012] In oil and gas exploration, the introduction of real-time correction and dynamic prediction functions has greatly improved the reliability of prediction models. Traditional prediction models are often based on static data and assumptions, and are difficult to cope with actual changes that occur during the exploration process. However, the oil and gas exploration environment is usually dynamic and may change significantly over time, such as changes in formation pressure and fluid properties. Therefore, a prediction model that can be updated and corrected in real time is of great significance for improving exploration accuracy and reducing risks.

[0013] The dynamic model correction module is designed to meet this demand. The module obtains the actual data measured during the drilling process in real time and performs online correction on the prediction model to ensure that the model can reflect the latest geological conditions. This dynamic adjustment mechanism can not only improve the accuracy of the prediction results, but also help decision makers adjust the exploration plan more quickly and reduce the risk of exploration failure.

[0014] 5. Combination of visualization technology and decision support

[0015] In the process of oil and gas exploration, decision makers often need to face a large amount of complex data and information. Therefore, visualizing the prediction results can help decision makers understand the meaning behind the data more intuitively and make scientific and reasonable decisions accordingly. Two-dimensional and three-dimensional graphic displays can not only show the spatial distribution of reservoirs, but also intuitively show high-potential reservoir areas, providing a direct reference for the formulation of development plans.

[0016] In addition, by combining risk assessment and uncertainty analysis, the system can generate comprehensive risk reports to provide multi-dimensional decision support for decision makers. These risk reports can help decision makers identify potential high-risk areas, so that they can take more cautious measures and reduce investment risks when formulating exploration and development plans.

[0017] 6. Integration of external systems and dynamic update of data

[0018] As oil and gas exploration projects expand in size and complexity, dynamic data updates and integration with external systems become increasingly important. Through interfaces with external data management systems, exploration systems can obtain the latest geological data in real time and make dynamic adjustments based on this data. This integration not only improves data utilization, but also enhances the flexibility of the system, enabling it to maintain continuous predictive capabilities in complex and changing exploration environments.

[0019] In summary, the reservoir physical property parameter prediction and evaluation system of the present invention combines big data, artificial intelligence and dynamic correction technology, which can effectively cope with the challenges of oil and gas exploration under complex geological conditions, improve prediction accuracy and decision-making efficiency, and provide strong support for the efficient development of oil and gas resources. Summary of the invention

[0020] Based on the existing technical problems, the present invention proposes a reservoir physical property parameter prediction and evaluation system for oil and gas exploration, which can predict the porosity and permeability of reservoirs under complex geological conditions, and is characterized in that the system includes:

[0021] Multi-source data acquisition module: used to collect and integrate geological data from different sources, including well logging data, core analysis data, seismic data and historical exploration data. Well logging data includes sonic logging, density logging and resistivity logging.

[0022] Adaptive data processing module: used for preprocessing the geological data, including data cleaning, outlier detection, data normalization, feature selection and multi-source data fusion;

[0023] Self-learning prediction module: used to predict reservoir physical property parameters based on the processed data based on the adaptive algorithm, the prediction parameters include porosity φ and permeability k;

[0024] Dynamic model correction module: used to correct the prediction model according to actual drilling data and update the prediction parameters to improve the accuracy and reliability of the prediction;

[0025] Visualization result display module: used to display the prediction results in two-dimensional and three-dimensional graphics, and provide multi-dimensional decision support, including the identification of high-potential reservoir areas and the recommendation of the best development plan.

[0026] As a preferred technical solution of the present invention, the adaptive data processing module includes the following steps:

[0027] Data cleaning: Remove noise and invalid data from geological data through filtering and denoising techniques;

[0028] Outlier detection: Use statistical analysis or machine learning methods to detect and process abnormal data points to ensure data reliability;

[0029] Data normalization: normalize the data from different logging methods and different geological conditions to ensure data comparability;

[0030] Feature selection: extract key features from geological data based on information gain or importance scoring algorithms, which will be used for subsequent model predictions;

[0031] Multi-source data fusion: Data from different sources are fused using weighted average or maximum information coefficient methods to generate a comprehensive feature vector.

[0032] As a preferred technical solution of the present invention, the self-learning prediction module uses the following to predict the reservoir porosity φ:

[0033]

[0034] in C adj is the correction factor generated by the correction module, where C adj is the correction factor generated by the correction module, where S ma is the rock skeleton longitudinal wave velocity, S p is the formation P-wave velocity, S s is the formation shear wave velocity.

[0035] As a preferred technical solution of the present invention, the correction factor C adj The dynamic model correction module is calculated based on the following formula:

[0036]

[0037] The actual measured value comes from the real porosity data obtained during the drilling process, and the initial predicted value is the output value of the self-learning prediction module.

[0038] As a preferred technical solution of the present invention, the self-learning prediction module further includes a prediction of the permeability k, and the prediction of the permeability uses the following formula:

[0039] k=g(X1,X2,…,X n )×C kadj

[0040] Among them, X1, X2, …, X n is the processed geological data feature, g is the permeability prediction function, C k adj is the permeability correction factor.

[0041] As a preferred technical solution of the present invention, the permeability correction factor C kadj The calculation method of is as follows: First, obtain the initial predicted value k of permeability pred , which is calculated by the self-learning prediction module based on the processed geological data and the permeability prediction function g(X1, X2, …, X n ) is calculated; secondly, the permeability measurement value k obtained during the actual drilling or logging process actual With the initial prediction value k pred The correction factor C is calculated kadj :

[0042]

[0043] Finally, the correction factor C kadj Applied to the permeability prediction model to correct and optimize the final permeability prediction results.

[0044] As a preferred technical solution of the present invention, the multi-source data fusion is achieved by the following steps:

[0045] Feature normalization: normalize the data features from different sources;

[0046] Weighted fusion: Assign weights to each data source based on its reliability and relevance, and use the weighted average method to fuse multiple data sources;

[0047] Feature Enhancement: Use the Maximum Information Coefficient (MIC) algorithm to identify and enhance the features that have the greatest impact on porosity and permeability predictions.

[0048] As a preferred technical solution of the present invention, the dynamic model correction module adopts online correction technology, so that the model can be updated in real time, and the prediction model is continuously corrected according to the latest actual measurement data to ensure the continuous effectiveness of the prediction.

[0049] As a preferred technical solution of the present invention, the visualization result display module further includes uncertainty analysis and risk assessment functions of the prediction results, which identify high-risk areas that may exist in the prediction process through a variety of statistical analysis methods and generate corresponding risk reports.

[0050] As a preferred technical solution of the present invention, the system can be connected to an external data management system interface to automatically acquire and update geological data, so that the prediction model can be dynamically adjusted based on the latest data and automatically generate an economic analysis report on reservoir development.

[0051] Beneficial effects:

[0052] First, the system integrates multi-source geological data (such as well logging data, core analysis data, seismic data, etc.) and combines it with adaptive data processing technology to achieve high-precision prediction of reservoir porosity and permeability under complex geological conditions. This multi-source data fusion and feature extraction method significantly improves the accuracy and reliability of predictions, helping users better identify high-potential reservoir areas.

[0053] Secondly, the system introduces a self-learning prediction module and a dynamic model correction module, which continuously updates and corrects model parameters so that the prediction results can reflect the latest geological conditions in real time. This online correction technology ensures the continued effectiveness and flexibility of the system in practical applications, especially in exploration environments where geological conditions are constantly changing, and can significantly reduce exploration risks.

[0054] Finally, the system's visualization results display module provides two-dimensional and three-dimensional graphic displays and uncertainty analysis functions, allowing users to intuitively understand the prediction results and formulate scientific and reasonable development plans based on the generated risk assessment report. In addition, the system can be seamlessly integrated with external data management systems to achieve automatic data updates and economic analysis of reservoir development, greatly improving the efficiency and success rate of oil and gas exploration. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a system block diagram of a reservoir physical property parameter prediction and evaluation system for oil and gas exploration proposed by the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Example 1: Multi-source data collection and preprocessing

[0058] In one embodiment of the present invention, the reservoir physical property parameter prediction and evaluation system is applied to a complex oil and gas exploration project. First, the system collects geological data from multiple data sources. These data include logging data (such as acoustic logging, density logging, resistivity logging), core analysis data, seismic data, and historical exploration data. The data acquisition module integrates data from different sources to ensure that the collected data has high timeliness and accuracy.

[0059] Data cleaning: During data processing, the data cleaning step is used to remove noise and invalid data. Noise can be removed by filtering techniques, while invalid data is handled by detecting and removing outliers.

[0060] Outlier detection: Detect and handle outliers in data using statistical analysis or machine learning algorithms. For example, use methods such as box plots or Z scores to identify outliers in well logging data and handle them appropriately.

[0061] Data normalization: Since logging data and geological data come from different sources, the measurement units and dimensions may be different, so the data needs to be normalized to unify the scales of various data sources. Normalization methods can include maximum and minimum normalization, Z-score normalization, etc.

[0062] Feature selection and multi-source data fusion: After data processing, the system uses information gain or importance scoring algorithms to extract key features from a large amount of geological data, and fuses data from different sources to generate a comprehensive feature vector through weighted average or maximum information coefficient method. This process ensures the diversity and representativeness of the data.

[0063] Example 2: Prediction of reservoir physical property parameters based on self-learning prediction module

[0064] Based on the processed geological data, the self-learning prediction module applies an adaptive algorithm to predict the porosity φ and permeability k of the reservoir. Specifically, the prediction of porosity is based on the formula:

[0065]

[0066] in C adj is the correction factor generated by the correction module, where C adj is the correction factor generated by the correction module, where S ma is the rock skeleton longitudinal wave velocity, S p is the formation P-wave velocity, S s is the formation shear wave velocity.

[0067] The permeability is predicted using the following formula:

[0068] k=g(X1,X2,…,X n )×C kadj

[0069] Among them, X1, X2, …, X n is the processed geological data feature, g is the permeability prediction function, C k adj is the permeability correction factor.

[0070] Permeability correction factor C kadj The calculation method of is as follows: First, obtain the initial predicted value k of permeability pred , which is calculated by the self-learning prediction module based on the processed geological data and the permeability prediction function g(X1, X2, …, X n ) is calculated; secondly, the permeability measurement value k obtained during the actual drilling or logging process actual With the initial prediction value k pred The correction factor C is calculated kadj :

[0071]

[0072] Finally, the correction factor C kadj Applied to the permeability prediction model to correct and optimize the final permeability prediction results.

[0073] In actual operation, the system continuously adjusts and optimizes the adaptive algorithm to adapt to data input under different geological conditions, thereby improving the accuracy of prediction results.

[0074] Example 3: Dynamic model correction and real-time prediction update

[0075] In actual application, the system uses the dynamic model correction module to perform online correction on the prediction model. For example, during the drilling process, the system obtains the actual measurement values ​​(such as porosity, permeability, etc.) in real time, compares them with the initial prediction values, and calculates new correction factors Cadj and Ckadj. The system automatically applies these correction factors to the model and updates the prediction results to ensure the prediction accuracy and reliability of the model.

[0076] The module’s online correction function allows the system to adjust the prediction model at any time under complex and changing geological conditions and ensure the continued effectiveness of the prediction. This is especially important in deep well drilling or multi-layer reservoir exploration projects.

[0077] Example 4: Visualization results display and decision support

[0078] After the prediction is completed, the system will present the prediction results to the user in the form of two-dimensional and three-dimensional graphics. Through the visual result display module, users can intuitively identify high-potential reservoir areas and obtain suggestions for the best development plan. In addition, the system also provides uncertainty analysis and risk assessment functions to identify high-risk areas in the prediction process and generate corresponding risk reports. These reports will help users consider potential risks and benefits more comprehensively when making exploration decisions.

[0079] For example, in an oil and gas field development project, the system identified several high-potential reservoir areas and recommended a development plan with better economic efficiency based on historical data and prediction results. Users used visualization tools to conduct in-depth analysis of the prediction results and combined them with risk assessment reports to ultimately make the best development decision.

[0080] Example 5: Integration and dynamic adjustment with external data management systems

[0081] In another embodiment of the present invention, the reservoir physical property parameter prediction and evaluation system can be seamlessly integrated with an external data management system. Through the interface connection with the external data management system, the system can automatically obtain and update the latest geological data, and dynamically adjust the prediction model based on these data. The system can not only adjust the model parameters in real time, but also automatically generate an economic analysis report on reservoir development, providing data support for the economic benefit evaluation of the project.

[0082] For example, an oil and gas company uses this system in the exploration process. Through integration with external data management systems, it can obtain the latest geological exploration data in real time and adjust development strategies based on these data. The economic analysis report generated by the system shows that the development of a certain reservoir has a high economic value, which ultimately helps the company make wise investment decisions.

[0083] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0084] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration, the system can predict the porosity and permeability of reservoirs under complex geological conditions, characterized in that: The system comprises: Multi-source data acquisition module: used to collect and integrate geological data from different sources, including well logging data, core analysis data, seismic data and historical exploration data. Well logging data includes sonic logging, density logging and resistivity logging. Adaptive data processing module: used for preprocessing the geological data, including data cleaning, outlier detection, data normalization, feature selection and multi-source data fusion; Self-learning prediction module: used to predict reservoir physical parameters based on the processed data based on the adaptive algorithm. The prediction parameters include porosity φ and permeability k; Dynamic model correction module: used to correct the prediction model according to actual drilling data and update the prediction parameters; Visualization result display module: used to display the prediction results in two-dimensional and three-dimensional graphical form, including the identification of high-potential reservoir areas and the recommendation of the best development plan.

2. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The adaptive data processing module comprises the following steps: Data cleaning: Remove noise and invalid data from geological data through filtering and denoising techniques; Outlier detection: Detect and process anomalous data points using statistical analysis or machine learning methods; Data normalization: normalize the data from different logging methods and different geological conditions; Feature selection: extracting key features from geological data based on information gain or importance scoring algorithms; Multi-source data fusion: Data from different sources are fused using weighted average or maximum information coefficient methods to generate a comprehensive feature vector.

3. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The self-learning prediction module predicts reservoir porosity φ using the following: in C adj is the correction factor generated by the correction module, where S ma is the rock skeleton longitudinal wave velocity, S p is the formation P-wave velocity, S s is the formation shear wave velocity.

4. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 3, characterized in that: The correction factor C adj The dynamic model correction module is calculated based on the following formula: The actual measured value comes from the real porosity data obtained during the drilling process, and the initial predicted value is the output value of the self-learning prediction module.

5. The reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The self-learning prediction module further includes the prediction of the permeability k, and the prediction of the permeability uses the following formula: k=g(X1,X2,…,X n )×C kadj Among them, X1, X2, …, X n is the processed geological data feature, g is the permeability prediction function, C k adj is the permeability correction factor.

6. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 5, characterized in that: The permeability correction factor C kadj The calculation method is as follows: First, obtain the initial predicted value k of permeability pred , which is calculated by the self-learning prediction module based on the processed geological data and the permeability prediction function g(X1, X2, …, X n ) is calculated; Secondly, the permeability measurement value k obtained during actual drilling or logging actual With the initial prediction value k pred The correction factor C is calculated kadj : Finally, the correction factor Ckadj is applied to the permeability prediction model to update the model parameters and optimize the final permeability prediction results.

7. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The multi-source data fusion is achieved by the following steps: Feature normalization: normalize the data features from different sources; Weighted fusion: Assign weights to each data source based on its reliability and relevance, and use the weighted average method to fuse multiple data sources; Feature Enhancement: Use the Maximum Information Coefficient algorithm to identify and enhance the features that have the greatest impact on porosity and permeability predictions.

8. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The dynamic model correction module adopts online correction technology to enable the model to be updated in real time, and the prediction model is continuously corrected according to the latest actual measurement data to ensure the continued effectiveness of the prediction.

9. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The visualization result display module further includes uncertainty analysis and risk assessment functions of the prediction results, which identify high-risk areas that may exist in the prediction process through a variety of statistical analysis methods and generate corresponding risk reports.

10. A reservoir physical property parameter prediction and evaluation system for oil and gas exploration according to claim 1, characterized in that: The system can be connected to an external data management system interface to automatically acquire and update geological data, so that the prediction model can be dynamically adjusted based on the latest data and automatically generate an economic analysis report for reservoir development.

Images