Microseism monitoring result analysis method and device based on machine learning

Abstract

The invention discloses a microseism monitoring result analysis method and device based on machine learning. The method comprises the following steps: acquiring geological features, microseism monitoring results and fracturing engineering parameters of a to-be-researched region; establishing a geologic feature tag library, a microseism monitoring result tag library and a fracturing engineering parameter tag library; inputting the geologic feature tag library, the microseismic monitoring result tag library and the fracturing engineering parameter tag library into a machine learning program, and performing data analysis and mining to obtain a geologic feature-microseismic monitoring result correlation analysis result and a fracturing engineering parameter-microseismic monitoring result correlation analysis result; according to the geological feature-microseismic monitoring result correlation analysis result, geological factors influencing the microseismic result are determined, and according to the fracturing engineering parameter-microseismic monitoring result correlation analysis result, fracturing engineering parameters influencing the microseismic result are determined. The micro-seismic monitoring result analysis efficiency can be improved, and the accuracy of identifying the master control influence factors can be improved.

Description

Technical Field

[0001] The present invention belongs to the field of oil and gas geophysical exploration, and more specifically, relates to a method and device for analyzing microseismic monitoring results based on machine learning. Background Art

[0002] At present, the exploitation of unconventional oil and gas reservoirs represented by tight oil, shale gas, and coalbed methane has become a hot topic, and the proportion of unconventional oil and gas in the oil and gas field production has gradually increased. As a key technology to improve the recovery rate and even the success rate of production wells, the demand for hydraulic fracturing has also increased day by day. The evaluation of the hydraulic fracturing effect is generally achieved by monitoring the generated microseismic events. Microseismic monitoring is one of the most accurate, timely, and information-rich monitoring means in current reservoir fracturing. The microseismic monitoring technology is a geophysical technology that monitors the fracturing effect and the spatial distribution of underground rock fractures by observing and analyzing microseismic signals generated by operations such as water injection fracturing that cause rock fractures or dislocations.

[0003] With the development of unconventional resources represented by shale oil and gas, the microseismic monitoring technology has become a commonly used technology for real-time monitoring of fracturing operations. Technical personnel use microseismic location maps to estimate the geometric characteristics of hydraulic fractures (including volume, height, length, complexity, etc.). The microseismic monitoring results intuitively reflect the transformation effect of hydraulic fracturing on the formation. However, the occurrence and rupture energy of microseismic events are strongly correlated with the geological characteristics of the rock formations traversed by oil and gas wells, such as the development of fractures, the magnitude and direction of in-situ stress, and the brittleness index. Changes in engineering controllable parameters such as the liquid volume, displacement, construction pressure, temporary plugging, and intermediate processes during the hydraulic fracturing process often lead to changes in the transformation effect, that is, changes in the microseismic monitoring results. There are many parameters such as geological characteristics and fracturing processes, and the changes are large. It is difficult for technical personnel to identify the geological main control factors and sensitive engineering parameters that truly cause the current results from among the numerous parameters, and it is impossible to better guide the next parameter optimization and select the best well pattern deployment plan.

[0004] The current interpretation of microseismic results relies on technical personnel to manually interpret the microseismic monitoring results, which takes a long time and the analysis is not comprehensive enough. It is particularly dependent on the experience of the interpreters, and the conclusions drawn by different personnel are often inconsistent. In this case, there is an urgent need to find a convenient and fast method for data analysis and mining to find geological characteristics and fracturing engineering parameters related to the microseismic monitoring results. Summary of the Invention

[0005] The object of the present invention is to propose a method and device for analyzing microseismic monitoring results based on machine learning, so as to improve the analysis efficiency of microseismic monitoring results and improve the accuracy of identifying the main influencing factors.

[0006] To achieve the above object, in a first aspect, the present invention proposes a method for analyzing microseismic monitoring results based on machine learning, including:

[0007] Obtain the geological characteristics, microseismic monitoring results, and fracturing engineering parameters during hydraulic fracturing in the area to be studied;

[0008] Establish a geological characteristic label library, a microseismic monitoring result label library, and a fracturing engineering parameter label library;

[0009] Input the geological characteristic label library, the microseismic monitoring result label library, and the fracturing engineering parameter label library into a machine learning program for data analysis and mining to obtain the correlation analysis results of geological characteristics - microseismic monitoring results and the correlation analysis results of fracturing engineering parameters - microseismic monitoring results;

[0010] According to the correlation analysis results of geological characteristics - microseismic monitoring results, determine the geological factors affecting the microseismic results, and according to the correlation analysis results of fracturing engineering parameters - microseismic monitoring results, determine the fracturing engineering parameters affecting the microseismic results.

[0011] Optionally, the geological characteristic labels include: reservoir characteristics, tectonic types, fracture development characteristics, bedding development characteristics, rock mechanics parameters, in-situ stress distribution, and brittleness distribution.

[0012] Optionally, the fracturing engineering parameter labels include: fracturing stage length, number of clusters, fluid volume used, displacement, wellbore pressure, proppant type, proppant quantity, and temporary plugging or mid-pump shut-off technology.

[0013] Optionally, the microseismic monitoring result labels include the number of microseismic events, event energy, geometric characteristics, fracture network complexity, and stimulation uniformity.

[0014] Optionally, the machine learning algorithm adopted by the machine learning program includes: k-means clustering algorithm.

[0015] In a second aspect, the present invention proposes an electronic device, which includes:

[0016] At least one processor; and,

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for analyzing microseismic monitoring results based on machine learning according to any one of the first aspects.

[0019] In a third aspect, the present invention provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method for analyzing microseismic monitoring results based on machine learning according to any one of the first aspect.

[0020] In a fourth aspect, the present invention provides a device for analyzing microseismic monitoring results based on machine learning, comprising:

[0021] a data acquisition module configured to acquire geological features, microseismic monitoring results, and fracturing engineering parameters during hydraulic fracturing in a region to be studied;

[0022] a tag library establishment module configured to establish a geological feature tag library, a microseismic monitoring result tag library, and a fracturing engineering parameter tag library;

[0023] a data analysis module configured to input the geological feature tag library, the microseismic monitoring result tag library, and the fracturing engineering parameter tag library into a machine learning program, perform data analysis and mining, and obtain a correlation analysis result between geological features and microseismic monitoring results and a correlation analysis result between fracturing engineering parameters and microseismic monitoring results;

[0024] a sensitive parameter identification module configured to determine geological factors affecting microseismic results according to the correlation analysis result between geological features and microseismic monitoring results, and determine fracturing engineering parameters affecting microseismic results according to the correlation analysis result between fracturing engineering parameters and microseismic monitoring results.

[0025] Optionally, the geological feature tags include: reservoir characteristics, tectonic types, fracture development characteristics, bedding development characteristics, rock mechanical parameters, in-situ stress distribution, and brittleness distribution.

[0026] Optionally, the fracturing engineering parameter tags include: fracturing stage length, number of clusters, fluid volume used, displacement, wellbore pressure, proppant type, proppant quantity, and temporary plugging or pump shutdown during operation.

[0027] The beneficial effects of the present invention are as follows:

[0028] After the method of the present invention completes microseismic monitoring of multiple multi-stage shale oil and gas fracturing wells, it collects and establishes a geological feature tag library, a fracturing engineering parameter tag library, and a microseismic monitoring result tag library, and through machine learning methods, conducts data analysis and mining, screens out sensitive geological features and fracturing engineering parameters affecting microseismic monitoring results, determines the main influencing factors, and guides the well pattern deployment and optimization of fracturing engineering parameters of shale oil and gas wells. The method of the present invention is stable and efficient. On the one hand, it reduces the workload of interpretation technicians and improves efficiency; on the other hand, the sensitive engineering parameters excavated can guide the next parameter optimization.

[0029] The system of the present invention has other features and advantages, which will be apparent from the accompanying drawings incorporated herein and the subsequent detailed description, or will be described in detail in the accompanying drawings incorporated herein and the subsequent detailed description, and these accompanying drawings and detailed description are used together to explain the specific principles of the present invention. Description of the Drawings

[0030] By describing the exemplary embodiments of the present invention in more detail in conjunction with the accompanying drawings, the above and other objects, features, and advantages of the present invention will become more apparent. In the exemplary embodiments of the present invention, the same reference numerals generally represent the same components.

[0031] Figure 1 The flowchart showing the steps of a method for analyzing microseismic monitoring results based on machine learning according to the present invention is shown.

[0032] Figure 2 The schematic diagram showing the analysis result of the correlation between geological features (strata) and microseismic monitoring results in an embodiment of the present invention is shown.

[0033] Figure 3 The schematic diagram showing the analysis of the correlation between fracturing engineering parameters (fluid volume used) and microseismic results in an embodiment of the present invention is shown. Detailed Description of the Invention

[0034] Machine learning is a multi-disciplinary field that involves multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. Machine learning is a common research hotspot in the fields of artificial intelligence and pattern recognition, and its theories and methods have been widely applied to solve complex problems in engineering applications and scientific fields.

[0035] With the continuous increase in the demand for data analysis in various industries in the big data era, efficiently obtaining knowledge through machine learning has gradually become the main driving force for the development of current machine learning technologies. How to perform in-depth analysis on complex and diverse data based on machine learning and make more efficient use of information has become the main research direction of machine learning in the current big data environment. Therefore, machine learning is increasingly developing towards the direction of intelligent data analysis and has become an important source of intelligent data analysis technology. In addition, in the big data era, with the continuous acceleration of the data generation speed, the volume of data has grown unprecedentedly, and new types of data to be analyzed are constantly emerging. This makes intelligent computing technologies such as big data machine learning and data mining play an extremely important role in the intelligent analysis and processing applications of big data.

[0036] "Data mining" and "data analysis" are usually mentioned in the same breath. Whether it is data analysis or data mining, they both help people collect and analyze data, turn it into information, and make judgments. Data analysis and mining technologies are a combination of machine learning algorithms and data access technologies. They use the statistical analysis, knowledge discovery and other means provided by machine learning to analyze massive data, and at the same time use the data access mechanism to achieve efficient reading and writing of data. Machine learning has an irreplaceable position in the field of data analysis and mining.

[0037] At present, in this field, there is no application of the data analysis and mining method of machine learning to the interpretation of microseismic monitoring results. Therefore, the present invention proposes a method and device for analyzing microseismic monitoring results based on machine learning. Using machine learning for data analysis and mining helps to screen out the main controlling factors from geological features, screen out sensitive parameters from numerous parameters of the fracturing project, better guide the deployment of shale oil and gas well patterns, optimize fracturing project parameters, and select an economically reasonable fracturing plan.

[0038] The present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0039] Embodiment 1

[0040] This embodiment provides a method for analyzing microseismic monitoring results based on machine learning, including:

[0041] S1: Obtain the geological features, microseismic monitoring results and fracturing project parameters of the area to be studied during hydraulic fracturing;

[0042] Specifically, collect the geological features, microseismic monitoring results, and fracturing project parameters of shale oil and gas wells in the area to be studied during hydraulic fracturing. The microseismic response characteristics and identification methods of formation fracture zones and the microseismic characteristic responses and identification methods of fault activation can be used to identify and distinguish the microseismic events induced by hydraulic fracturing fractures and faults.

[0043] S2: Establish a geological feature tag library, a microseismic monitoring result tag library, and a fracturing project parameter tag library;

[0044] Specifically, establish libraries of geological feature tags, microseismic monitoring result tags, and fracturing engineering parameter tags by classification. Among them, geological feature tags include reservoir characteristics, tectonic types, fracture development characteristics, bedding development characteristics, rock mechanics parameters, in-situ stress distribution, brittleness distribution, etc. Fracturing engineering parameter tags include fracture section length, number of clusters, fluid volume used, displacement, pressure, proppant type and quantity, temporary plugging or mid-pump stoppage process, etc. Microseismic monitoring result tags include the number of microseismic events, event energy, geometric characteristics (propagation length, width, height, volume, etc.), fracture network complexity, stimulation balance, etc. By sorting out the microseismic, geological feature, and engineering parameter tags, establish a sample library.

[0045] S3: Input the libraries of geological feature tags, microseismic monitoring result tags, and fracturing engineering parameter tags into a machine learning program for data analysis and mining to obtain the correlation analysis results between geological features and microseismic monitoring results and the correlation analysis results between fracturing engineering parameters and microseismic monitoring results.

[0046] Specifically, input the sample library into a machine learning program for data analysis and mining to obtain the correlation analysis results. Among them, the machine learning algorithms used in the machine learning program include: the k-means clustering algorithm, which is an iterative clustering analysis algorithm, and its steps are as follows:

[0047] a. First, randomly select K points in the sample as the clustering centers.

[0048] b. Calculate the distances between other samples in the sample and these K clustering centers respectively, and classify these samples into the category of their nearest clustering center.

[0049] c. Calculate the average value for each category of the classified samples above to obtain new clustering centroids.

[0050] d. Compare with the K clustering centroids obtained in the previous calculation. If the clustering centroids change, go to process b; otherwise, go to process e.

[0051] e. When the centroids do not change (when a centroid is found and the samples assigned to this centroid are the same in each iteration, that is, the newly generated clusters are the same in each iteration and all sample points will no longer transfer from one cluster to another, the centroid will not change), stop and output the clustering result.

[0052] S4: Determine the geological factors affecting the microseismic results based on the correlation analysis results between geological features and microseismic monitoring results, and determine the fracturing engineering parameters affecting the microseismic results based on the correlation analysis results between fracturing engineering parameters and microseismic monitoring results.

[0053] Specifically, based on the correlation analysis results of geological features - microseismic monitoring results, determine the geological factors (sensitive geological features) affecting microseismic results, providing guidance for the next-step oil and gas development well pattern deployment; based on the correlation analysis results of fracturing engineering parameters - microseismic monitoring results, determine the fracturing engineering parameters (sensitive parameters) affecting microseismic results, and optimize the relevant parameters, such as selecting reasonable liquid usage, proppant dosage, and displacement, etc. After identifying the sensitive geological and engineering parameters, the process ends.

[0054] This method has carried out relevant research and tests in the microseismic monitoring of multiple shale gas and shale oil wells. This method can stably and efficiently identify the main geological and pressure engineering control factors affecting microseismic results. On the one hand, it reduces the workload of interpretation technicians and improves efficiency; on the other hand, the sensitive engineering parameters excavated can guide the next-step parameter optimization.

[0055] Example 2

[0056] Seven microseismic monitoring wells were implemented in a certain shale oil block, with a designed well depth of 5000 - 6000 meters, a vertical depth of 3500 - 4200 meters, a to-be-fractured long horizontal section of 1500 - 1800 meters, and a fracturing engineering design of 20 - 30 stages.

[0057] According to the actual situation of the study area, using the microseismic response characteristics and identification methods of formation fracture zones and the microseismic characteristic responses and identification methods of fault activation, identify and distinguish the microseismic events caused by hydraulic fracturing fractures and fault induction, providing a basis for the correct use of microseismic processing and interpretation data. Collect the geological features, microseismic monitoring results, and fracturing engineering parameters of the area to be studied, and establish geological feature tags, microseismic monitoring result tags, and fracturing engineering parameter tag libraries in categories.

[0058] Input the above multiple tag libraries into a machine learning program, conduct data analysis and mining, and obtain the correlation analysis results. Determine that the main geological factors affecting microseismic results are the development conditions of faults and fracture zones. Identify the main fracturing engineering parameters as controllable engineering parameters such as liquid usage, displacement, and wellbore pressure, providing a basis for the next-step engineering parameter optimization and formulating a reasonable and economic fracturing plan.

[0059] In this embodiment, Figure 2 It is a schematic diagram of the correlation analysis between geological features (strata) and microseismic results. It can be seen that there are significant differences in the fracture development of each small layer between the Ⅳ and Ⅴ subsections of the study area, which have a greater impact on the number of microseismic events. The average number of events in each small layer of the Ⅳ subsection is significantly higher than that of the Ⅴ subsection. Figure 3It is a schematic diagram for the correlation analysis of fracturing engineering parameters (fluid volume used) and microseismic results. It can be seen that the fluid volume used in different fracturing wells in the study area is a sensitive parameter, and it has a strong correlation with the number of microseismic events. The higher the fluid volume used, the higher the cost. Therefore, a fluid volume of 4000 - 5000 cubic meters is the most economical and reasonable solution.

[0060] Example 3

[0061] This embodiment provides a device for analyzing microseismic monitoring results based on machine learning, including:

[0062] A data acquisition module, configured to acquire geological features, microseismic monitoring results, and fracturing engineering parameters during hydraulic fracturing in the area to be studied;

[0063] A tag library establishment module, configured to establish a geological feature tag library, a microseismic monitoring result tag library, and a fracturing engineering parameter tag library; among them, the geological feature tags include: reservoir characteristics, tectonic types, fracture development characteristics, bedding development characteristics, rock mechanics parameters, in-situ stress distribution, and brittleness distribution. The fracturing engineering parameter tags include: fracturing stage length, number of clusters, fluid volume used, displacement, wellbore pressure, proppant type, proppant quantity, and temporary plugging or mid-pump stoppage technology. The microseismic monitoring result tags include the number of microseismic events, event energy, geometric characteristics, fracture network complexity, and stimulation uniformity.

[0064] A data analysis module, configured to input the geological feature tag library, the microseismic monitoring result tag library, and the fracturing engineering parameter tag library into a machine learning program, perform data analysis and mining, and obtain the correlation analysis results between geological features and microseismic monitoring results and the correlation analysis results between fracturing engineering parameters and microseismic monitoring results;

[0065] A sensitive parameter identification module, configured to determine geological factors affecting microseismic results according to the correlation analysis results between geological features and microseismic monitoring results, and determine fracturing engineering parameters affecting microseismic results according to the correlation analysis results between fracturing engineering parameters and microseismic monitoring results.

[0066] Example 4

[0067] This embodiment provides an electronic device, and the electronic device includes:

[0068] At least one processor; and,

[0069] A memory communicatively connected to the at least one processor; wherein,

[0070] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for analyzing microseismic monitoring results based on machine learning described in the above embodiment.

[0071] An electronic device according to an embodiment of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0072] The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In an embodiment of the present disclosure, the processor is used to run the computer-readable instructions stored in the memory.

[0073] Those skilled in the art should understand that, in order to solve the technical problem of how to obtain good user experience effects, this embodiment may also include well-known structures such as communication buses, interfaces, etc., and these well-known structures should also be included in the protection scope of the present disclosure.

[0074] For a detailed description of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

[0075] Embodiment 5

[0076] This embodiment provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a computer to execute the method for analyzing microseismic monitoring results based on machine learning described in the foregoing embodiments.

[0077] A computer-readable storage medium according to an embodiment of the present disclosure stores non-transitory computer-readable instructions thereon. When the non-transitory computer-readable instructions are run by a processor, all or part of the steps of the methods of the foregoing embodiments of the present disclosure are executed.

[0078] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (such as CD-ROM and DVD), magneto-optical storage media (such as MO), magnetic storage media (such as magnetic tape or removable hard disk), media with built-in rewritable non-volatile memory (such as memory card), and media with built-in ROM (such as ROM cartridge).

[0079] The embodiments of the present invention have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for analyzing microseismic monitoring results based on machine learning, characterized in that, it includes: Obtain the geological characteristics, microseismic monitoring results, and fracturing engineering parameters during hydraulic fracturing in the area to be studied; Establish a geological characteristics label library, a microseismic monitoring results label library, and a fracturing engineering parameters label library; Input the geological characteristics label library, the microseismic monitoring results label library, and the fracturing engineering parameters label library into a machine learning program for data analysis and mining to obtain the correlation analysis results between geological characteristics and microseismic monitoring results and the correlation analysis results between fracturing engineering parameters and microseismic monitoring results; According to the correlation analysis results between geological characteristics and microseismic monitoring results, determine the geological factors affecting microseismic results, and according to the correlation analysis results between fracturing engineering parameters and microseismic monitoring results, determine the fracturing engineering parameters affecting microseismic results.

2. The method for analyzing microseismic monitoring results based on machine learning according to claim 1, characterized in that, Geological characteristics labels include: reservoir characteristics, tectonic type, fracture development characteristics, bedding development characteristics, rock mechanics parameters, in-situ stress distribution, and brittleness distribution.

3. The method for analyzing microseismic monitoring results based on machine learning according to claim 1, characterized in that, Fracturing engineering parameter labels include: fracturing stage length, number of clusters, liquid consumption, displacement, wellbore pressure, proppant type, proppant quantity, and temporary plugging or mid-pump shut-off technology.

4. The method for analyzing microseismic monitoring results based on machine learning according to claim 1, characterized in that, Microseismic monitoring results labels include the number of microseismic events, event energy, geometric characteristics, fracture network complexity, and transformation balance.

5. The method for analyzing microseismic monitoring results based on machine learning according to claim 1, characterized in that, The machine learning algorithm adopted by the machine learning program includes: k-means clustering algorithm.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for analyzing microseismic monitoring results based on machine learning according to any one of claims 1-5.

7. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to execute the method for analyzing microseismic monitoring results based on machine learning according to any one of claims 1-5.

8. A device for analyzing microseismic monitoring results based on machine learning, characterized in that, it includes: A data acquisition module for obtaining the geological characteristics, microseismic monitoring results, and fracturing engineering parameters during hydraulic fracturing in the area to be studied; A label library establishment module for establishing a geological characteristics label library, a microseismic monitoring results label library, and a fracturing engineering parameters label library; A data analysis module, which is used to input the geological feature tag library, the microseismic monitoring result tag library, and the fracturing engineering parameter tag library into a machine learning program for data analysis and mining, so as to obtain the correlation analysis result between geological features and microseismic monitoring results and the correlation analysis result between fracturing engineering parameters and microseismic monitoring results; A sensitive parameter identification module, which is used to determine the geological factors affecting the microseismic results according to the correlation analysis result between geological features and microseismic monitoring results, and determine the fracturing engineering parameters affecting the microseismic results according to the correlation analysis result between fracturing engineering parameters and microseismic monitoring results.

9. The microseismic monitoring result analysis device based on machine learning according to claim 8, characterized in that, The geological feature tags include: reservoir characteristics, tectonic type, fracture development characteristics, bedding development characteristics, rock mechanical parameters, in-situ stress distribution, and brittleness distribution.

10. The microseismic monitoring result analysis device based on machine learning according to claim 8, characterized in that, The fracturing engineering parameter tags include: fracturing stage length, number of clusters, fluid volume used, displacement, wellbore pressure, proppant type, proppant quantity, and temporary plugging or mid-pump shutdown technology.

Images