Petroleum exploration evaluation development intelligent decision-making system and method based on digital twinning

Abstract

The invention relates to the technical field of oil exploitation, in particular to an intelligent decision-making system and method for oil exploration, evaluation and development based on digital twinning, and the system comprises an acquisition unit, a processing unit, an analysis unit, a document unit and a sample unit, and the output end of the acquisition unit is in communication connection with the input end of the processing unit; according to the intelligent decision-making system, data can be quickly screened through the screening module in the processing unit, the data processing time is greatly shortened, the decision-making speed of the intelligent decision-making system is increased, oil exploration and development are not affected by delay, the convenience of oil exploitation is improved, and the intelligent decision-making system is suitable for popularization and application. The intelligent decision-making system has strong data analysis capability, various rules and trends hidden in the oil exploitation process can be revealed through deep mining and analysis of oil exploitation engineering data, the danger in the oil exploitation process can be reduced through risk evaluation, and the safety in the oil exploitation process is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil production, and in particular to an intelligent decision-making system and method for oil exploration, evaluation and development based on digital twins. Background Art

[0002] With the continuous growth of global energy demand, oil, as one of the indispensable energy sources, faces huge challenges in its exploration and development. During the drilling process, how to complete various tasks accurately and efficiently is not only related to the drilling cost, but also directly affects the oil production and quality.

[0003] For example, an expert decision system for oilfield development with application number CN201511005199.8 and publication date 20170707 arranges the basic data required for oilfield development into vectors and enters them into a basic vector table; normalizes the vectors in the basic vector table; determines whether the normalized vector is a reasonable vector, and if so, the determination ends; otherwise, it is an unreasonable vector and the vector classification is completed; the unreasonable vector is imported into the expert decision library, and if it is qualified, the determination ends; otherwise, the vector parameters are adjusted and the process returns to step 3. The present invention can support various types of data query, analysis and decision-making in real time to realize the networking, informatization and management of oilfield development planning, thereby comprehensively improving the level of oilfield development rules, not only improving the work efficiency of managers and decision-makers, but also formulating a fast and accurate plan for oilfield exploitation and saving costs.

[0004] Traditional drilling decision-making methods often rely on the experience and judgment of experienced engineers, lack scientificity and systematicness, and are difficult to cope with complex and changeable geological environments. In order to solve this problem, the existing technology solves this problem by building a decision-making system. However, at present, the drilling decision-making system in the existing technology needs to optimize a large amount of data when making decisions, which is time-consuming. As a result, the decision-making system cannot make decisions quickly, affecting the exploration and exploitation of oil. Therefore, it is urgent to design an intelligent decision-making system and method for oil exploration, evaluation and development based on digital twins to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent decision-making system and method for petroleum exploration, evaluation and development based on digital twins to solve the above-mentioned deficiencies in the prior art.

[0006] In order to achieve the above object, the present invention provides the following technical solutions: An intelligent decision-making system for petroleum exploration, evaluation and development based on digital twins, comprising an acquisition unit, a processing unit, an analysis unit, a copywriting unit and a sample unit, wherein an output end of the acquisition unit is communicatively connected to an input end of the processing unit, an output end of the processing unit is communicatively connected to an input end of the analysis unit, an output end of the analysis unit is communicatively connected to an input end of the copywriting unit, and an output end of the copywriting unit is communicatively connected to an input end of the sample unit; The acquisition unit is used to acquire oil production information. The acquisition unit acquires oil production information through a variety of data acquisition methods, and the various data acquisition methods are specifically sensor data acquisition, on-site monitoring data acquisition, and remote monitoring data acquisition; The processing unit integrates and unifies the data acquired by the acquisition unit through data integration technology. The processing unit includes a screening module, a unification module, and an optimization module. The screening module screens the data acquired by the acquisition unit through a constructed screening formula. The screening formula is as follows:

[0007] in, To obtain the data set actually obtained by the unit, To obtain the data set that the unit ideally obtains, if If the actual data obtained by the acquisition unit meets the requirements, the data can be saved and called directly. Not true, the actual data acquired by the acquisition unit does not meet the requirements, the data is deleted and the data is acquired again for screening, the unification module unifies the data screened by the screening module through a unified method of spatiotemporal granularity, and the optimization module optimizes the data unified by the unification module through three methods: abnormal value processing, missing value processing and data standardization; The analysis unit analyzes and evaluates the data processed by the processing unit through a big data analysis algorithm, and analyzes and predicts the data processed by the processing unit through a fusion model technology. The analysis unit includes an evaluation module and a model module. The evaluation module combines the data processed by the processing unit to perform risk assessment and oil exploration assessment through quantitative risk assessment and artificial neural network algorithm technology, and the specific steps of risk assessment are as follows: S1-1. Information collection: Obtain relevant information on the risk assessment of oil extraction projects, and after sorting and screening, use it as the necessary input data for quantitative risk assessment; S1-2. Hazard identification: Search historical data and identify possible hazard factors by combining oil extraction geographic information and construction information; S1-3. Consequence analysis and calculation: Based on the various dangerous events identified, simulate and analyze the consequences and determine the danger level; S1-4. Risk calculation and evaluation: PhastRisk is used to calculate and analyze various risk results, establish risk acceptance standards, determine whether the risks faced are acceptable, propose protective measures, and prepare quantitative risk evaluation reports; The specific steps of oil exploration and evaluation are as follows: S2-1. Index classification and data processing: Obtain geological data of the oil area to be mined, and classify and normalize the data; S2-2. Select a mode to train the network: select a three-layer BP network to train the data, specify the number of learning times as 1000 times, and require the system error to reach 0.0001; S2-3. Network performance test: select one of the above modes to test the network and use the rotation return test method; S2-4. Regression analysis and evaluation: Perform regression analysis on the tested data and output evaluation results based on the regression analysis.

[0008] In another embodiment provided by the present invention, the model module constructs a fusion model based on the data processed by the processing unit and the evaluation result of the evaluation module, and the specific steps of constructing the fusion model are as follows: S3-1. Data integration and preprocessing: Integrate big data from multiple sources, including geological data, drilling engineering parameters, real-time sensor data, etc., clean, convert and standardize the data to ensure data accuracy and consistency; S3-2. Model architecture design: Based on the requirements and characteristics of oil production engineering, design the architecture of the fusion model; S3-3. Feature extraction and selection: Using big data analysis technology, extract features that have a significant impact on drilling decisions from raw data; S3-4. Model training and optimization: Use the training data set to train the fusion model and optimize the model performance by adjusting the model parameters and algorithms; S3-5. Model evaluation and verification: Use the test data set to evaluate and verify the trained model to ensure the accuracy and reliability of the model in practical applications. The evaluation indicators include accuracy, recall rate, and F1 score; S3-6. Model fusion and application: Integrate the trained model into the drilling intelligent decision-making support system to achieve real-time data analysis and decision support; The copywriting unit calls the data to generate a decision plan according to the analysis result of the analysis unit. The copywriting unit includes a search module, a generation module, and a judgment module. The search module can search the Internet and the data in the sample unit through keyword search technology. The generation module can summarize the data retrieved by the search module through the copywriting summary algorithm and generate exclusive copywriting. The judgment module will input the plan generated by the generation model into the model module in the analysis unit for verification; The sample unit is used to collect relevant data on oil production. The sample unit is established through distributed database technology and stores past oil production information.

[0009] The intelligent decision-making method for petroleum exploration and evaluation development based on digital twins includes the following steps: Step S1. Obtaining oil extraction information: Using various sensors, instruments and equipment, various data in the oil extraction process are collected in real time, and logging on to relevant websites to search for previous oil extraction information; Step S2. Data analysis and mining: Optimize the acquired data, and then use machine learning, deep learning and other algorithms to analyze and mine the acquired data to extract valuable information. Then, build a fusion model through digital twin technology, and optimize the model through learning algorithms; Step S3. Decision support and visualization: Based on the analysis results, provide intelligent auxiliary decision support for various problems in oil extraction engineering, and build a visualization interface to intuitively display the decision.

[0010] In the above technical scheme, the digital twin-based petroleum exploration, evaluation and development intelligent decision-making system and method provided by the present invention have the following beneficial effects: (1) The intelligent decision-making system designed by the present invention can quickly screen data through the screening module in the processing unit, which greatly reduces the time for data processing and improves the speed at which the intelligent decision-making system makes decisions, so that oil exploration and development will not be affected by delays, thereby improving the convenience of oil extraction.

[0011] (2) The intelligent decision-making system designed by the present invention has powerful data analysis capabilities. Through in-depth mining and analysis of oil production engineering data, it can reveal various laws and trends hidden in the oil production process. Through risk assessment, it can reduce the dangers in the oil production process and improve the safety of the oil production process.

[0012] (3) The intelligent decision-making system and method designed by the present invention can process massive amounts of data, provide real-time analysis results, and help decision makers make accurate and wise decisions, which helps to reduce decision-making risks and improve exploration efficiency, which is crucial in the rapidly changing oilfield exploration environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0014] Figure 1 A system flow chart of an embodiment of an intelligent decision-making system and method for petroleum exploration, evaluation and development based on digital twins provided by the present invention.

[0015] Figure 2 A schematic diagram of the processing unit structure provided for an embodiment of an intelligent decision-making system and method for petroleum exploration, evaluation and development based on digital twins of the present invention.

[0016] Figure 3 A schematic diagram of the analysis unit structure provided for an embodiment of an intelligent decision-making system and method for petroleum exploration, evaluation and development based on digital twins of the present invention.

[0017] Figure 4 A schematic diagram of the copywriting unit structure provided for an embodiment of an intelligent decision-making system and method for petroleum exploration, evaluation and development based on digital twins of the present invention.

[0018] Figure 5 A method flow chart is provided for an embodiment of an intelligent decision-making system and method for petroleum exploration and evaluation development based on digital twins of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0020] like Figure 1-4 As shown, the petroleum exploration, evaluation and development intelligent decision-making system based on digital twin provided by the embodiment of the present invention includes an acquisition unit, a processing unit, an analysis unit, a copywriting unit, and a sample unit. The output end of the acquisition unit is communicatively connected to the input end of the processing unit, the output end of the processing unit is communicatively connected to the input end of the analysis unit, the output end of the analysis unit is communicatively connected to the input end of the copywriting unit, and the output end of the copywriting unit is communicatively connected to the input end of the sample unit; The acquisition unit is used to acquire oil production information. The acquisition unit acquires oil production information through a variety of data acquisition methods, and the multiple data acquisition methods are specifically sensor data acquisition, on-site monitoring data acquisition, and remote monitoring data acquisition; It should be noted that oil extraction information mainly includes the following aspects: ‌Exploration information‌: The first step in oil extraction is exploration, which uses physical methods such as seismic wave detection technology, magnetic field, gravity field, and chemical methods such as soil gas measurement and surface chemical composition analysis to determine the location and reserves of oil;‌ ‌Drilling information‌: Drilling is a key step in oil extraction. Various drilling equipment and technologies are required during the drilling process, such as oil drilling rigs, drill bits and screw drilling tools. The types of drill bits include scraper bits, roller bits, fixed cutter bits, etc., which are suitable for different formation conditions;‌ Oil production technology: Oil production technology includes self-flowing oil production, artificial lifting oil production, water injection, fracturing and acidizing, etc. These technical means are aimed at improving the efficiency and economic benefits of oil production; ‌Oilfield development and oil refining and processing information‌: The oil industry includes oilfield geological exploration, oilfield development and oil extraction, transportation, refining and processing, which together constitute a complete oil extraction and processing process;‌ Drilling technology: Modern drilling technologies such as rotary steering technology and full-process casing drilling technology can improve drilling efficiency and reduce costs. The rotary steering technology has the characteristics of low friction and torsional resistance during drilling, high drilling speed and low cost. The full-process casing drilling technology simplifies the conventional drilling operation process and greatly shortens the well construction cycle; The processing unit integrates and unifies the data acquired by the acquisition unit through data integration technology. The processing unit includes a screening module, a unification module, and an optimization module. The screening module screens the data acquired by the acquisition unit through a constructed screening formula. The screening formula is as follows:

[0021] in, To obtain the data set actually obtained by the unit, To obtain the data set that the unit ideally obtains, if If the actual data obtained by the acquisition unit meets the requirements, the data can be saved and called directly. Not true, the actual data acquired by the acquisition unit does not meet the requirements, the data is deleted and the data is acquired again for screening, the unification module unifies the data screened by the screening module through a unified method of spatiotemporal granularity, and the optimization module optimizes the data unified by the unification module through three methods: abnormal value processing, missing value processing and data standardization; It should be noted that outlier processing: through statistical analysis of the data, the maximum value, minimum value, lower quartile value, upper quartile value and median value are calculated. For each type of data, outliers are those that exceed the range corresponding to the physical meaning. The processing methods adopted by the author include: ① elimination; ② substitution with the average value; ③ substitution with the median; ④ substitution with the boundary value; ⑤ ignoring and not processing; Missing value processing: Missing value processing usually uses interpolation, which is divided into univariate interpolation and multivariate interpolation. Univariate interpolation is a simpler processing method, including four processing methods: using mean value substitution, median substitution, highest frequency value substitution and given constant value substitution. Multivariate interpolation is to model the characteristics of each missing value as a function of other characteristics and use the estimated value for estimation; Data standardization mainly includes: ① Ignoring the shape of feature distribution, directly removing the mean of a feature, and then scaling it by dividing it with the standard deviation of the non-constant feature; ② Scaling the feature to a given minimum and maximum value (0-1), or converting the maximum absolute value of each feature to unit size; The analysis unit analyzes and evaluates the data processed by the processing unit through a big data analysis algorithm, and analyzes and predicts the data processed by the processing unit through a fusion model technology. The analysis unit includes an evaluation module and a model module. The evaluation module combines the data processed by the processing unit to perform risk assessment and oil exploration assessment through quantitative risk assessment and artificial neural network algorithm technology, and the specific steps of risk assessment are as follows: S1-1. Information collection: Obtain relevant information on the risk assessment of oil extraction projects, and after sorting and screening, use it as the necessary input data for quantitative risk assessment; S1-2. Hazard identification: Search historical data and identify possible hazard factors by combining oil extraction geographic information and construction information; S1-3. Consequence analysis and calculation: Based on the various dangerous events identified, simulate and analyze the consequences and determine the danger level; S1-4. Risk calculation and evaluation: PhastRisk is used to calculate and analyze various risk results, establish risk acceptance standards, determine whether the risks faced are acceptable, propose protective measures, and prepare quantitative risk evaluation reports; The specific steps of oil exploration and evaluation are as follows: S2-1. Index classification and data processing: Obtain geological data of the oil area to be mined, and classify and normalize the data; S2-2. Select a mode to train the network: select a three-layer BP network to train the data, specify the number of learning times as 1000 times, and require the system error to reach 0.0001; S2-3. Network performance test: select one of the above modes to test the network and use the rotation return test method; S2-4. Regression analysis and evaluation: Perform regression analysis on the tested data and output evaluation results based on the regression analysis; The model module constructs a fusion model based on the data processed by the processing unit and the results of the evaluation module, and the specific steps of constructing the fusion model are as follows: S3-1. Data integration and preprocessing: Integrate big data from multiple sources, including geological data, drilling engineering parameters, real-time sensor data, etc., clean, convert and standardize the data to ensure data accuracy and consistency; S3-2. Model architecture design: Based on the requirements and characteristics of oil production engineering, design the architecture of the fusion model; S3-3. Feature extraction and selection: Using big data analysis technology, extract features that have a significant impact on drilling decisions from raw data; S3-4. Model training and optimization: Use the training data set to train the fusion model and optimize the model performance by adjusting the model parameters and algorithms; S3-5. Model evaluation and verification: Use the test data set to evaluate and verify the trained model to ensure the accuracy and reliability of the model in practical applications. The evaluation indicators include accuracy, recall rate, and F1 score; S3-6. Model fusion and application: Integrate the trained model into the drilling intelligent decision-making support system to achieve real-time data analysis and decision support; The copywriting unit calls the data to generate a decision plan according to the analysis result of the analysis unit. The copywriting unit includes a search module, a generation module, and a judgment module. The search module can search the Internet and the data in the sample unit through keyword search technology. The generation module can summarize the data retrieved by the search module through the copywriting summary algorithm and generate exclusive copywriting. The judgment module will input the plan generated by the generation model into the model module in the analysis unit for verification; The sample unit is used to collect relevant data on oil production. The sample unit is established through distributed database technology and stores past oil production information.

[0022] Intelligent decision-making methods for petroleum exploration, evaluation and development based on digital twins, such as Figure 5 As shown, the following steps are included: Step S1. Obtaining oil extraction information: Using various sensors, instruments and equipment, various data in the oil extraction process are collected in real time, and logging on to relevant websites to search for previous oil extraction information; Step S2. Data analysis and mining: Optimize the acquired data, and then use machine learning, deep learning and other algorithms to analyze and mine the acquired data to extract valuable information. Then, build a fusion model through digital twin technology, and optimize the model through learning algorithms; Step S3. Decision support and visualization: Based on the analysis results, provide intelligent auxiliary decision support for various problems in oil extraction engineering, and build a visualization interface to intuitively display the decision.

[0023] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An intelligent decision-making system for petroleum exploration, evaluation and development based on digital twins, including an acquisition unit, a processing unit, an analysis unit, a document unit and a sample unit, characterized in that: The output end of the acquisition unit is in communication connection with the input end of the processing unit, the output end of the processing unit is in communication connection with the input end of the analysis unit, the output end of the analysis unit is in communication connection with the input end of the copywriting unit, and the output end of the copywriting unit is in communication connection with the input end of the sample unit; The acquisition unit is used to acquire oil production information; The processing unit integrates and unifies the data acquired by the acquisition unit through data integration technology; The analysis unit analyzes and evaluates the data processed by the processing unit through a big data analysis algorithm, and analyzes and predicts the data processed by the processing unit through a fusion model technology; The copywriting unit calls data to generate a decision plan according to the analysis result of the analysis unit; The sample unit is used to collect relevant data on oil production.

2. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 1 is characterized in that: The acquisition unit collects the information of oil production through a variety of data collection methods, and the various data collection methods are specifically sensor data collection, on-site monitoring data collection and remote monitoring data collection.

3. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 1 is characterized in that: The processing unit includes a screening module, a unification module, and an optimization module. The screening module screens the data acquired by the acquisition unit through a constructed screening formula. The screening formula is as follows: in, To obtain the data set actually obtained by the unit, To obtain the data set that the unit ideally obtains, if If the actual data obtained by the acquisition unit meets the requirements, the data can be saved and called directly. Not true. The actual data obtained by the acquisition unit does not meet the requirements. Delete the data and obtain the data again for screening.

4. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 3 is characterized in that: The unification module unifies the data screened by the screening module through a unified method of spatiotemporal granularity, and the optimization module optimizes the data unified by the unification module through three methods: abnormal value processing, missing value processing and data standardization.

5. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 1 is characterized in that: The analysis unit includes an evaluation module and a model module. The evaluation module combines the data processed by the processing unit to perform risk assessment and oil exploration assessment through quantitative risk assessment and artificial neural network algorithm technology. The specific steps of risk assessment are as follows: S1-1. Information collection: Obtain relevant information on the risk assessment of oil extraction projects, and after sorting and screening, use it as the necessary input data for quantitative risk assessment; S1-2. Hazard identification: Search historical data and identify possible hazard factors by combining oil extraction geographic information and construction information; S1-3. Consequence analysis and calculation: Based on the various dangerous events identified, simulate and analyze the consequences and determine the danger level; S1-4. Risk calculation and evaluation: PhastRisk is used to calculate and analyze various risk results, establish risk acceptability standards, determine whether the risks faced are acceptable, propose protective measures, and prepare quantitative risk evaluation reports.

6. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 5 is characterized in that: The specific steps of oil exploration and evaluation are as follows: S2-1. Index classification and data processing: Obtain geological data of the oil area to be mined, and classify and normalize the data; S2-2. Select a mode to train the network: select a three-layer BP network to train the data, specify the number of learning times as 1000 times, and require the system error to reach 0.0001; S2-3. Network performance test: select one of the above modes to test the network and use the rotation return test method; S2-4. Regression analysis and evaluation: Perform regression analysis on the tested data and output evaluation results based on the regression analysis.

7. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 5 is characterized in that: The model module constructs a fusion model based on the data processed by the processing unit and the results of the evaluation module, and the specific steps of constructing the fusion model are as follows: S3-1. Data integration and preprocessing: Integrate big data from multiple sources, including geological data, drilling engineering parameters, real-time sensor data, etc., clean, convert and standardize the data to ensure data accuracy and consistency; S3-2. Model architecture design: Based on the requirements and characteristics of oil production engineering, design the architecture of the fusion model; S3-3. Feature extraction and selection: Using big data analysis technology, extract features that have a significant impact on drilling decisions from raw data; S3-4. Model training and optimization: Use the training data set to train the fusion model and optimize the model performance by adjusting the model parameters and algorithms; S3-5. Model evaluation and verification: Use the test data set to evaluate and verify the trained model to ensure the accuracy and reliability of the model in practical applications. The evaluation indicators include accuracy, recall rate, and F1 score; S3-6. Model fusion and application: Integrate the trained model into the drilling intelligent decision-making support system to achieve real-time data analysis and decision support.

8. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 1 is characterized in that: The copywriting unit includes a search module, a generation module, and a judgment module. The search module can search the Internet and data in the sample unit through keyword search technology. The generation module can summarize and process the data retrieved by the search module through a copywriting summary algorithm and generate exclusive copywriting. The judgment module will input the solution generated by the generation model into the model module in the analysis unit for verification.

9. The petroleum exploration, evaluation and development intelligent decision-making system based on digital twin according to claim 1 is characterized in that: The sample unit is established by using a distributed database technology, and the sample unit stores past oil production information.

10. An intelligent decision-making method for petroleum exploration, evaluation and development based on digital twins, characterized in that: The following steps are involved: Step S1. Obtaining oil extraction information: Using various sensors, instruments and equipment, various data in the oil extraction process are collected in real time, and logging on to relevant websites to search for previous oil extraction information; Step S2. Data analysis and mining: Optimize the acquired data, and then use machine learning, deep learning and other algorithms to analyze and mine the acquired data to extract valuable information. Then, build a fusion model through digital twin technology, and optimize the model through learning algorithms; Step S3. Decision support and visualization: Based on the analysis results, provide intelligent auxiliary decision support for various problems in oil extraction engineering, and build a visualization interface to intuitively display the decision.

Images