Low-speed mudstone identification method and system based on attribute clustering
By using attribute-based clustering combined with seismic data and machine learning techniques, the problem of identifying low-velocity mudstone in medium-deep reservoirs was solved, improving identification accuracy and exploration accuracy, reducing exploration risks, and optimizing well location deployment.
Patent Information
- Application Number
- CN202411152655.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-21
AI Technical Summary
Existing technologies face difficulties in identifying low-velocity mudstone in medium-deep reservoirs, leading to decreased exploration accuracy and increased drilling risks, which in turn affects reservoir development.
An attribute-based clustering approach was adopted, combining seismic data analysis, geological modeling, and machine learning. Through in-depth analysis of seismic response characteristics and attribute optimization, low-velocity mudstone and gas-bearing sandstone were distinguished. The CLARA clustering algorithm and PAM method were used for attribute clustering analysis.
It improves the accuracy of reservoir identification in low-velocity mudstone development areas, reduces interference from bright spot traps, provides reliable exploration decision support, reduces exploration risks, and optimizes well location deployment and resource utilization efficiency.
Smart Images

Figure CN119001849A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of seismic data processing and oil and gas exploration, and in particular to a method and system for identifying low-velocity mudstone based on attribute clustering. Background Art
[0002] In recent years, with the continuous deepening of oil and gas resource exploration, bright spot recognition technology has become an important means of identifying gas-bearing reservoirs. However, although this technology has achieved remarkable success in the application of shallow reservoirs, it faces many challenges in the identification of medium and deep reservoirs. In particular, there have been repeated bright spot recognition errors in exploration practice, resulting in a decrease in exploration accuracy and affecting the effective identification and development of reservoirs.
[0003] In-depth analysis shows that one of the main reasons for these identification errors is the widespread development of low-velocity mudstone. Low-velocity mudstone is a special stratum that is usually formed below the background mudstone. Its velocity is lower and it presents a "bright spot" feature similar to that of gas-bearing reservoirs on seismic sections. This "bright spot" is easily confused with the actual gas-bearing sandstone, forming the so-called low-velocity mudstone "bright spot trap", which seriously interferes with the correct identification of gas-bearing reservoirs.
[0004] The presence of low-velocity mudstone not only makes traditional bright spot identification technology ineffective in medium-deep reservoirs, but may also lead to incorrect reservoir predictions in exploration work, thereby causing unnecessary drilling risks and economic losses. Therefore, how to accurately distinguish low-velocity mudstone from gas-bearing reservoirs has become a technical problem that needs to be solved in the exploration field.
[0005] References
[0006] [1] Zhang Mingwei, Gao Ling, Tan Jiancai, Pan Guangchao, Li Wentuo. Application of low-speed mudstone identification technology in regional exploration of Yinggehai Basin[J]. China Mining, 2018, 27(11): 145-150;
[0007] [2] Zhang Weiwei, He Min, Zhu Ming, et al. Application of AVO technology to identify non-bright spot gas reservoirs in deepwater areas, Petroleum Geophysical Exploration, 2015, 50(1): 123-128;
[0008] [3] Pei Jianxiang, Pan Guangchao, Zhu Peiyuan, Liu Feng. Identification of low-velocity mudstone traps in the overpressure zone of Yinggehai Basin and its implications for exploration. Petroleum Geophysical Exploration, 2016, 51(2): 361-370.
[0009] [4] Li Fang, Deng Yong, Liu Shiyou, et al. Effect of undercompacted low-velocity mudstone on seismic reflection and AVO [J]. Geological Science and Technology Information, 2017, 36(5): 244-248. Summary of the invention
[0010] In view of the defects of the prior art, the present invention provides a method and system for identifying low-velocity mudstone based on attribute clustering. By comprehensively applying seismic data analysis, geological modeling and machine learning technology, the effective distinction between low-velocity mudstone and gas-bearing sandstone is achieved through in-depth analysis of seismic response characteristics and attribute optimization. This intelligent prediction scheme can significantly improve the reservoir identification accuracy in low-velocity mudstone development areas, reduce the interference of bright spot traps, and thus provide more reliable technical support for exploration decisions.
[0011] In order to achieve the above invention object, the technical solution adopted by the present invention is as follows:
[0012] A low-speed mudstone identification method based on attribute clustering includes the following steps:
[0013] Step 1: Low-velocity mudstone model study:
[0014] Based on the logging data and seismic data of the exploration wells in the work area, the development characteristics of low-velocity mudstone are studied and analyzed.
[0015] The low-velocity mudstone is divided into bright spot type low-velocity mudstone and non-bright spot type low-velocity mudstone, and its performance including well logging curve, cross plot, seismic section and AVO characteristics is analyzed to obtain the well analysis results.
[0016] Step 2: Model construction and earthquake forward modeling:
[0017] According to the well analysis results, a three-layer horizontal layered medium model is established.
[0018] Model 1 is constructed, which includes a thick layer of low-velocity mudstone and a thin layer of gas-bearing sandstone. The seismic forward modeling response is low-frequency and medium-amplitude.
[0019] Model 2 was constructed, which included a thin layer of low-velocity mudstone and a thin layer of gas-bearing sandstone, and the seismic forward modeling response was low frequency and high amplitude.
[0020] Models 3 and 4 were constructed, which are bright spot type low-velocity mudstones, and the seismic forward modeling response is low frequency and high amplitude.
[0021] Model 5 is constructed, which is a non-bright spot low-velocity mudstone model, and the seismic forward modeling response is low frequency and low amplitude.
[0022] Step 3: Extraction and optimization of seismic attributes:
[0023] Extract multiple seismic attributes including amplitude, instantaneous and spectral.
[0024] Insensitive attributes are eliminated and preferred seismic attributes are retained for subsequent analysis.
[0025] The selected seismic attributes are standardized and the correlation coefficients are calculated.
[0026] Based on the correlation coefficient value, redundant attributes are eliminated and attributes with lower correlation are retained for cluster analysis.
[0027] Step 4: Attribute cluster analysis:
[0028] The CLARA clustering algorithm is used to perform attribute clustering analysis on the selected seismic attributes.
[0029] The PAM method is used to find the K best center points in the selected sample set, and these center points are applied to the entire database for attribute clustering.
[0030] Through multiple cycles, the total clustering cost is calculated and optimized, and finally the best set of representative objects is determined.
[0031] Step 5: Classification of recognition results:
[0032] According to the attribute clustering results, the identification results are divided into five categories: normal compacted mudstone, bright spot type low-velocity mudstone, non-bright spot type low-velocity mudstone and two types of gas-bearing sandstone with different seismic responses.
[0033] The attribute clustering results were compared and analyzed with the well data to confirm the accuracy of identification of low-velocity mudstone and gas-bearing sandstone.
[0034] Finally, based on the clustering results, the reservoir and low-velocity mudstone are identified and differentiated to guide exploration and well deployment.
[0035] Furthermore, the sub-steps of step three are as follows:
[0036] Step 3.1: Extract seismic attributes;
[0037] Collect forward modeling result data: obtain seismic data obtained through seismic forward modeling.
[0038] Select the attribute type, including amplitude attributes, instantaneous attributes and spectrum attributes;
[0039] Amplitude attributes include: peak amplitude, root mean square amplitude, and amplitude ratio.
[0040] Instantaneous attributes include: instantaneous phase, instantaneous frequency, and instantaneous amplitude.
[0041] Spectrum attributes include: power spectrum density, frequency peak, and bandwidth.
[0042] Calculate seismic attributes and calculate each attribute. Extract the specific values of the selected amplitude, instantaneous and spectrum attributes from the forward modeling data.
[0043] Step 3.2: Attribute sensitivity analysis;
[0044] Screening sensitive attributes: Based on the analysis of the seismic response characteristics of low-velocity mudstone and gas-bearing sandstone, sensitive attributes are screened out. Sensitive attributes can effectively distinguish different types of reservoirs.
[0045] Attributes that have no significant contribution to target identification are eliminated, leaving only attributes that have identification effects on low-velocity mudstone and gas sandstone.
[0046] Preserve preferred attributes: From the originally extracted attributes, retain the seismic attributes that are useful for distinguishing low-velocity mudstone and gas sandstone.
[0047] Step 3.3: Standardize the retained earthquake attributes to have the same scale and mean.
[0048] Step 3.4: Calculate the correlation coefficient between all seismic attributes. The correlation coefficient is used to measure the linear relationship between attributes, ranging from -1 to 1.
[0049] The calculated correlation coefficients are statistically analyzed to generate a correlation coefficient matrix.
[0050] Attribute pairs with correlation coefficient values higher than 0.8 were identified, and it was considered that there was a strong correlation between them, and redundant attributes were considered for removal.
[0051] Attribute pairs with correlation coefficient values lower than 0.2 are identified as having weak correlations and are not suitable for attribute clustering.
[0052] Step 3.5: Based on the correlation coefficient analysis results, select attributes with correlation coefficients between 0.25 and 0.75 for subsequent attribute clustering analysis.
[0053] The screened earthquake attributes are organized into a data format suitable for cluster analysis to ensure that the data set is complete and meets the requirements of cluster analysis.
[0054] Furthermore, the sub-steps of step 4 are as follows:
[0055] Step 4.1: Select the seismic attributes selected in step 3 as the analysis object and organize the data to ensure that the format and structure meet the requirements of the CLARA clustering algorithm.
[0056] Step 4.2: Set the parameters of the CLARA clustering algorithm, including sample size and number of samplings, and randomly select a subsample set from the entire database. The sample size is determined by the parameter settings.
[0057] Step 4.3: Initialization: select K initial center points in the selected sample set using the PAM method; assignment step: assign each data point in the sample set to the nearest center point to form K clusters; update: calculate the total cost of the points within each cluster, find and update the center point of each cluster to reduce the total cost within the cluster; iteration: repeat the assignment and update steps until the center point no longer changes significantly or the upper limit of the number of iterations is reached.
[0058] Step 4.4: Expand the center points, apply the K center points obtained by the PAM method to the entire database, assign each data point to the nearest center point, and form the final K clusters.
[0059] Step 4.5: Calculate the final clustering cost, including the total distance of the points within the cluster and the total distance of the center point; adjust the K value or other parameters, and re-execute the CLARA algorithm to optimize the clustering results; use clustering evaluation indicators (such as silhouette coefficient, clustering effectiveness index, etc.) to evaluate the quality of the clustering results, compare the consistency of the clustering results with the actual data, and verify the effectiveness of the clustering.
[0060] Step 4.6: Output the results, record the final attribute clustering results, including the center point of each cluster and the data points within the cluster; generate a report, prepare an attribute clustering analysis report, including a detailed description and analysis of the clustering results.
[0061] The present invention also discloses a low-speed mudstone identification system based on attribute clustering, which can be used to implement the above-mentioned low-speed mudstone identification method based on attribute clustering. The system includes:
[0062] Data acquisition module: responsible for collecting and organizing logging data and seismic data of exploration wells, including data acquisition instruments, data storage devices and data transmission interfaces.
[0063] Data preprocessing module: cleans, denoises and converts the collected data, including data cleaning algorithms, denoising filters and format conversion tools.
[0064] Low-velocity mudstone pattern research module: Analyze the pattern of low-velocity mudstone based on exploration well data, including low-velocity mudstone classification analysis, feature extraction and pattern research tools, divide low-velocity mudstone into bright spot type and non-bright spot type, and analyze its performance in logging curves, crossplots, seismic profiles and AVO characteristics.
[0065] Model building and seismic forward modeling module: Based on the well analysis results, a three-layer horizontal layered medium model is established and seismic forward modeling is performed, including model building tools, forward simulation software and model verification functions, and multiple models (such as model one to model five) are constructed to simulate different seismic response characteristics.
[0066] Seismic attribute extraction module: extracts a variety of seismic attributes from the forward modeling results, including amplitude, instantaneous and spectrum attributes, and performs attribute extraction, screening and analysis.
[0067] Attribute optimization module: standardizes the extracted seismic attributes and performs correlation coefficient statistics, removes insensitive attributes, and retains the preferred attributes for subsequent attribute clustering analysis, including standardization tools, correlation coefficient calculation tools and redundant attribute elimination functions.
[0068] Attribute clustering analysis module: Use the CLARA clustering algorithm and PAM method to perform attribute clustering analysis on the selected seismic attributes, including the CLARA clustering algorithm tool, the PAM method implementation module and the clustering result optimization function.
[0069] Identification result classification module: classifies and analyzes the attribute clustering results, including identification result classification tools, data comparison and analysis functions and result verification modules, and identifies and distinguishes reservoirs and low-velocity mudstones based on the attribute clustering results.
[0070] The present invention also discloses a computer device, including a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned low-speed mudstone identification method based on attribute clustering when executing the program.
[0071] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned low-speed mudstone identification method based on attribute clustering is implemented.
[0072] Compared with the prior art, the advantages of the present invention are:
[0073] 1. Improve recognition accuracy: Through the attribute clustering-based method, the present invention can effectively distinguish between bright spot type and non-bright spot type low-velocity mudstone, thereby improving the recognition accuracy of low-velocity mudstone and reducing recognition errors.
[0074] 2. Optimizing attribute selection: By extracting and optimizing seismic attributes, the present invention eliminates insensitive or redundant attributes and retains the most valuable attributes for low-velocity mudstone identification, thereby improving the effect and efficiency of cluster analysis.
[0075] 3. Enhanced model adaptability: By using different seismic models (such as gas-bearing sandstone and low-velocity mudstone models) for forward simulation, the present invention can meet the needs of low-velocity mudstone identification under various geological conditions and has strong adaptability and versatility.
[0076] 4. Reduced manual intervention: Through automated data processing, model building, attribute extraction, and cluster analysis, the present invention significantly reduces the need for manual intervention and improves work efficiency and consistency.
[0077] 5. Improve exploration accuracy: The present invention provides reliable technical support for exploration and development through accurate identification of reservoirs and low-velocity mudstone, helping to reduce exploration risks and improve exploration success rates.
[0078] 6. Support decision optimization: The identification results generated by the present invention can provide a scientific basis for development decisions of oil and gas fields, optimize well site deployment and reservoir development strategies, and improve resource utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] Figure 1 It is a flow chart of a low-velocity mudstone identification method based on attribute clustering according to an embodiment of the present invention. DETAILED DESCRIPTION
[0080] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples.
[0081] like Figure 1 As shown, the present invention provides a low-speed mudstone identification method based on attribute clustering, comprising the following steps:
[0082] Step 1: Conduct model research on low-velocity mudstone based on the exploration wells in the work area;
[0083] Zhang Mingwei et al. (2018) divided low-velocity mudstone into bright spot type low-velocity mudstone and non-bright spot type low-velocity mudstone. [1] , this embodiment will classify low-speed mudstone on this basis. Bright spot type low-speed mudstone is developed in Well A. On the intersection diagram and logging curve, it can be seen that the velocity at 4000m is obviously reversed. Low-speed mudstone is developed below 4000m, and it appears as a bright spot on the seismic profile, which affects the identification of gas-bearing sandstone. It appears as Class I on the AVO feature, which is confused with the AVO feature of gas-bearing sandstone. Therefore, the low-speed mudstone cannot be completely identified by the AVO feature. Non-bright spot type low-speed mudstone is developed in Well B. The combination pattern of low-speed mudstone and gas-bearing sandstone can be obtained on the logging intersection diagram and logging curve. Low-speed mudstone is developed directly on the sandstone layer in this well. On the pre-stack AVO angle gather and AVO feature, the sandstone layer (Class I-II) and low-speed mudstone (Class IV) can be clearly divided. The sandstone layer appears as a bright spot on the seismic profile. Well C has developed non-bright spot type low-velocity mudstone. The velocity of the mudstone in this well has obvious velocity reversal at 3994 meters. A set of thin sandstone layers is developed above 3987 meters, and a set of low-velocity mudstone with a thickness of 50 meters is developed below. The underlying stratum of the mudstone is gas-bearing sandstone. The well shows low-frequency weak amplitude in both forward logs and seismic profiles. It is Class IV in AVO characteristics, which can be distinguished from gas-bearing sandstone. Based on the analysis of the above three wells, the bright spot type low-velocity mudstone is a false "bright spot" that appears on the seismic profile. The low-velocity mudstone developed on the sandstone layer will not affect the identification of the sandstone.
[0084] Step 2: Model construction and earthquake forward simulation;
[0085] Comprehensive well analysis, a three-layer horizontal layered medium model was established and seismic forward simulation was performed. Model 1 and Model 2 are both gas-bearing sandstones, where Model 1 is a thick layer of low-velocity mudstone plus a thin layer of gas-bearing sandstone, and the forward response is low frequency and medium amplitude; Model 2 is a thin low-velocity mudstone plus a thin gas-bearing sandstone layer, and the forward response is low frequency and high amplitude. Models 3 and 4 are bright spot type low-velocity mudstones, and their forward responses are both low frequency and high amplitude. Model 5 is a non-bright spot type low-velocity mudstone model, and the forward response is low frequency and low amplitude.
[0086] Step 3: Extraction and optimization of seismic attributes;
[0087] According to the forward modeling results, a variety of seismic attributes were extracted, including amplitude, instantaneous, and spectrum. Insensitive seismic attributes were eliminated, and the remaining 13 seismic attributes were used for subsequent attribute clustering analysis. The forward modeling results show that low-velocity mudstone and gas-bearing sandstone both have abnormal amplitudes in seismic response, but they can be distinguished in seismic attributes. Then, the 13 selected seismic attributes were standardized and the correlation coefficients were statistically analyzed. The higher the correlation coefficient value in the statistical results, the better the correlation between the two attributes. According to the statistical results, when the correlation coefficient is greater than 0.8, it indicates that one of the two attributes is redundant and can be removed; when the correlation coefficient is less than 0.2, it indicates that the two attributes are basically uncorrelated and are not suitable for attribute clustering. According to the statistical results, the correlation coefficients between the four seismic attributes of instantaneous true amplitude*instantaneous phase cosine, amplitude peak state, maximum peak amplitude, and average instantaneous phase are all between 0.25 and 0.75, which can be used for attribute clustering analysis.
[0088] Step 4: attribute clustering and result analysis;
[0089] Seismic attribute technology has been developed to a mature level. With the widespread application of neural networks in the field of oil and gas exploration, the technology of seismic attribute clustering using unsupervised neural networks has also emerged. This method uses the clustering method in large applications (Clustering Large Applications, CLARA), performs seismic attribute optimization, and then uses attribute clustering to identify reservoirs and low-velocity mudstones. The CLARA clustering algorithm does not consider the entire data set, but selects a small part of the data as a sample, uses the Partitioning Around Medoid (PAM) clustering method around the center point for each sample set, and uses the best cluster as the output. The steps of the CLARA algorithm are as follows: ① Repeat steps ② to ④ in a loop with a step length equal to the number of sampling times; ② Randomly extract a sample of N objects from the entire database, and call the PAM method to find the K best center points of the sample from the sample; ③ Apply these K center points to the entire database, and for each non-representative object X, determine which representative object selected from the sample is closest to it; ④ Calculate the total clustering cost obtained in the previous step. If the value is less than the current minimum value, replace the current minimum value with this value, and retain the K representative objects obtained in this sampling as the best representative object collection so far; ⑤ Return to step ① and start the next loop. After the algorithm ends, the best representative object collection is obtained.
[0090] According to the results of well analysis and attribute analysis, the attribute clustering results are divided into 5 categories, namely normal compacted mudstone, bright spot type low-velocity mudstone, non-bright spot type low-velocity mudstone and two gas-bearing sandstones with different seismic responses. Attribute clustering analysis was performed at the bottom of the H layer. According to the well data, the bottom of the H layer in the three wells is mudstone, well B is normal compacted mudstone, well A is bright spot type low-velocity mudstone, and well C is non-bright spot type low-velocity mudstone. In the clustering results, well B is pointed out as normal compacted mudstone; well A is bright spot type low-velocity mudstone; but well C is non-bright spot type low-velocity mudstone in the clustering results. The clustering results are consistent with the well understanding.
[0091] Currently, AVO types and AVO anomalies are used to distinguish low-velocity mudstone and sandstone reservoirs. [2-4] This technology has high requirements for pre-stack seismic data, and requires high-quality pre-stack seismic data to realize AVO extraction and analysis. Compared with AVO analysis technology, using post-stack seismic for attribute clustering can avoid the influence of pre-stack seismic data on AVO.
[0092] In another embodiment of the present invention, a low-velocity mudstone identification system based on attribute clustering is provided. The system can be used to implement the above-mentioned low-velocity mudstone identification method based on attribute clustering, and specifically includes:
[0093] Data acquisition module: responsible for collecting and organizing logging data and seismic data of exploration wells, including data acquisition instruments, data storage devices and data transmission interfaces.
[0094] Data preprocessing module: cleans, denoises and converts the collected data, including data cleaning algorithms, denoising filters and format conversion tools.
[0095] Low-velocity mudstone pattern research module: Analyze the pattern of low-velocity mudstone based on exploration well data, including low-velocity mudstone classification analysis, feature extraction and pattern research tools, divide low-velocity mudstone into bright spot type and non-bright spot type, and analyze its performance in logging curves, crossplots, seismic profiles and AVO characteristics.
[0096] Model building and seismic forward modeling module: Based on the well analysis results, a three-layer horizontal layered medium model is established and seismic forward modeling is performed, including model building tools, forward simulation software and model verification functions, and multiple models (such as model one to model five) are constructed to simulate different seismic response characteristics.
[0097] Seismic attribute extraction module: extracts a variety of seismic attributes from the forward modeling results, including amplitude, instantaneous and spectrum attributes, and performs attribute extraction, screening and analysis.
[0098] Attribute optimization module: standardizes the extracted seismic attributes and performs correlation coefficient statistics, removes insensitive attributes, and retains the preferred attributes for subsequent attribute clustering analysis, including standardization tools, correlation coefficient calculation tools and redundant attribute elimination functions.
[0099] Attribute clustering analysis module: Use the CLARA clustering algorithm and PAM method to perform attribute clustering analysis on the selected seismic attributes, including the CLARA clustering algorithm tool, the PAM method implementation module and the clustering result optimization function.
[0100] Identification result classification module: classifies and analyzes the attribute clustering results, including identification result classification tools, data comparison and analysis functions and result verification modules, and identifies and distinguishes reservoirs and low-velocity mudstones based on the attribute clustering results.
[0101] In another embodiment of the present invention, a terminal device is provided, the terminal device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used for the operation of the low-speed mudstone identification method based on attribute clustering in the above embodiment.
[0102] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk storage.
[0103] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the low-speed mudstone identification method based on attribute clustering in the above embodiment; one or more instructions in the computer-readable storage medium are loaded and executed by the processor.
[0104] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0105] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0106] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0108] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the implementation methods of the present invention, and should be understood that the protection scope of the present invention is not limited to such special statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.
Claims
1. A low-velocity mudstone identification method based on attribute clustering, characterized in that: The following steps are involved: Step 1: Low-velocity mudstone model study: Based on the logging data and seismic data of the exploration wells in the work area, the development characteristics of low-velocity mudstone are studied and analyzed; The low-velocity mudstone is divided into bright-spot type low-velocity mudstone and non-bright-spot type low-velocity mudstone, and its performance in well logging curves, crossplots, seismic sections and AVO characteristics is analyzed to obtain the well analysis results; Step 2: Model construction and earthquake forward modeling: Based on the well analysis results, a three-layer horizontal layered medium model was established; Model 1 was constructed, which included thick low-velocity mudstone and thin gas-bearing sandstone, and the seismic forward modeling response was low-frequency and medium-amplitude; Model 2 was constructed, which included a thin layer of low-velocity mudstone and a thin layer of gas-bearing sandstone, and the seismic forward modeling response was low frequency and high amplitude; Models 3 and 4 were constructed, which are bright spot type low-velocity mudstones, and the seismic forward modeling response is low frequency and high amplitude; Model 5 is constructed, which is a non-bright spot type low-velocity mudstone model, and the seismic forward modeling response is low frequency and low amplitude; Step 3: Seismic attribute extraction and optimization: Extract multiple seismic attributes, including amplitude, instantaneous and spectrum; Eliminate insensitive attributes and retain preferred seismic attributes for subsequent analysis; The selected seismic attributes are standardized and the correlation coefficients are calculated; According to the correlation coefficient value, redundant attributes are eliminated and attributes with lower correlation are retained for cluster analysis; Step 4: Attribute cluster analysis: Use the CLARA clustering algorithm to perform attribute clustering analysis on the selected seismic attributes; The PAM method is used to find the K best center points in the selected sample set, and these center points are applied to the entire database for attribute clustering; Through multiple cycles, the total clustering cost is calculated and optimized, and finally the best representative object set is determined; Step 5: Classification of recognition results: According to the attribute clustering results, the identification results are divided into five categories: normal compacted mudstone, bright spot type low-velocity mudstone, non-bright spot type low-velocity mudstone and two types of gas-bearing sandstone with different seismic responses; The attribute clustering results were compared and analyzed with the well data to confirm the accuracy of identification of low-velocity mudstone and gas-bearing sandstone; Finally, based on the clustering results, the reservoir and low-velocity mudstone are identified and differentiated to guide exploration and well deployment.
2. The low-velocity mudstone identification method based on attribute clustering according to claim 1 is characterized in that: The sub-steps of step three are as follows: Step 3.1: Extract seismic attributes; Collect forward modeling result data: obtain seismic data obtained through seismic forward modeling; Select the attribute type, including amplitude attributes, instantaneous attributes and spectrum attributes; Amplitude attributes include: peak amplitude, root mean square amplitude, amplitude ratio; Instantaneous attributes include: instantaneous phase, instantaneous frequency, instantaneous amplitude; Spectrum attributes include: power spectrum density, frequency peak, and bandwidth; Calculate seismic attributes and calculate each attribute; extract the specific values of the selected amplitude, instantaneous and spectrum attributes from the forward modeling data; Step 3.2: Attribute sensitivity analysis; Screening sensitive attributes: Screen out sensitive attributes based on the analysis of the seismic response characteristics of low-velocity mudstone and gas-bearing sandstone; sensitive attributes can effectively distinguish different types of reservoirs; Eliminate the attributes that have no significant contribution to target identification, leaving the attributes that have identification effects on low-velocity mudstone and gas sandstone; Preserve preferred attributes: From the originally extracted attributes, retain the seismic attributes useful for distinguishing low-velocity mudstone and gas-bearing sandstone; Step 3.3: Standardize the retained earthquake attributes to have the same scale and mean; Step 3.4: Calculate the correlation coefficient between all seismic attributes; the correlation coefficient is used to measure the linear relationship between attributes, ranging from -1 to 1; Perform statistical analysis on the calculated correlation coefficients to generate a correlation coefficient matrix; Determine the attribute pairs with correlation coefficient values higher than 0.8, consider them to have strong correlation, and consider removing redundant attributes; Attribute pairs with correlation coefficient values lower than 0.2 were identified, and the correlation between them was considered weak and unsuitable for attribute clustering; Step 3.5: According to the correlation coefficient analysis results, select attributes with correlation coefficients between 0.25 and 0.75 for subsequent attribute clustering analysis; The screened earthquake attributes are organized into a data format suitable for cluster analysis to ensure that the data set is complete and meets the requirements of cluster analysis.
3. The low-speed mudstone identification method based on attribute clustering according to claim 1 is characterized in that: The sub-steps of step 4 are as follows: Step 4.1: Select the earthquake attributes selected in step 3 as the analysis object, organize the data to ensure that the format and structure meet the requirements of the CLARA clustering algorithm; Step 4.2: Set the parameters of the CLARA clustering algorithm, including the sample size and number of samplings, and randomly select a subsample set from the entire database. The sample size is determined by the parameter settings; Step 4.3: Initialization: In the selected sample set, select K initial center points using the PAM method; Assignment step: assign each data point in the sample set to the nearest center point to form K clusters; update, calculate the total cost of the points in each cluster, find and update the center point of each cluster to reduce the total cost within the cluster; Iterate, repeat the allocation and update steps until the center point no longer changes significantly or the upper limit of the iteration number is reached; Step 4.4: Expand the center points, apply the K center points obtained by the PAM method to the entire database, assign each data point to the nearest center point, and form the final K clusters; Step 4.5: Calculate the final clustering cost, including the total distance of the points within the cluster and the total distance of the center point; adjust the K value or other parameters, and re-execute the CLARA algorithm to optimize the clustering results; Use clustering evaluation indicators to evaluate the quality of clustering results, compare the consistency of clustering results with actual data, and verify the effectiveness of clustering; Step 4.6: Output the results, record the final attribute clustering results, including the center point of each cluster and the data points within the cluster; Generate reports and prepare attribute clustering analysis reports, including detailed descriptions and analysis of clustering results.
4. A low-speed mudstone identification system based on attribute clustering, characterized by: The system can be used to implement the low-speed mudstone identification method based on attribute clustering according to any one of claims 1 to 3, and the system comprises: Data acquisition module: responsible for collecting and organizing well logging data and seismic data of exploration wells, including data acquisition instruments, data storage devices and data transmission interfaces; Data preprocessing module: cleans, denoises and converts the collected data, including data cleaning algorithms, denoising filters and format conversion tools; Low-velocity mudstone pattern research module: Analyze the pattern of low-velocity mudstone based on exploration well data, including low-velocity mudstone classification analysis, feature extraction and pattern research tools, divide low-velocity mudstone into bright spot type and non-bright spot type, and analyze its performance in logging curves, crossplots, seismic sections and AVO characteristics; Model building and seismic forward modeling module: Based on the well analysis results, a three-layer horizontal layered medium model is established and seismic forward modeling is performed, including model building tools, forward modeling software and model verification functions, and models one to five are constructed to simulate different seismic response characteristics; Seismic attribute extraction module: extracts a variety of seismic attributes from the forward modeling results, including amplitude, instantaneous and spectrum attributes, and performs attribute extraction, screening and analysis; Attribute optimization module: standardizes the extracted seismic attributes and performs correlation coefficient statistics, removes insensitive attributes, and retains the preferred attributes for subsequent attribute clustering analysis, including standardization tools, correlation coefficient calculation tools, and redundant attribute elimination functions; Attribute clustering analysis module: Use the CLARA clustering algorithm and PAM method to perform attribute clustering analysis on the selected seismic attributes, including the CLARA clustering algorithm tool, the PAM method implementation module and the clustering result optimization function; Identification result classification module: classifies and analyzes the attribute clustering results, including identification result classification tools, data comparison and analysis functions and result verification modules, and identifies and distinguishes reservoirs and low-velocity mudstones based on the attribute clustering results.
5. A computer device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for identifying low-speed mudstone based on attribute clustering as claimed in any one of claims 1 to 3 is implemented.
6. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the method for identifying low-speed mudstone based on attribute clustering as claimed in any one of claims 1 to 3 is implemented.
Citation Information
Patent Citations
Horizontal well track optimization method based on RS three-dimensional sensitivity seismic attribution analysis
CN103472484A
Sandstone reservoir grey matter removal prediction method of grey matter probability body
CN117148448A
Tight sandstone reservoir prediction method and device based on pre-stack elastic factor and medium
CN117890967A
Rock Reservoir Structure Characterization Method, Device, Computer-Readable Storage Medium and Electronic Equipment
US20210356623A1
Reservoir parameter prediction method and apparatus based on geological characteristic constraint, and storage medium
US20230314649A1
