WO2020082824A1 - 模型生成方法及装置、岩性识别方法及装置 - Google Patents
模型生成方法及装置、岩性识别方法及装置 Download PDFInfo
- Publication number
- WO2020082824A1 WO2020082824A1 PCT/CN2019/097226 CN2019097226W WO2020082824A1 WO 2020082824 A1 WO2020082824 A1 WO 2020082824A1 CN 2019097226 W CN2019097226 W CN 2019097226W WO 2020082824 A1 WO2020082824 A1 WO 2020082824A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- model
- lithology
- logging
- training
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Definitions
- the present application relates to the field of data processing technology, and in particular, to a model generation method and device and a lithology identification method and device.
- lithological identification and reservoir division of wellbore formations are the basis for identification and evaluation of oil and gas layers, and also an important basis for studying reservoir depositional patterns.
- lithology identification there are three methods to realize lithology identification: cuttings logging, core drilling and logging interpretation.
- cuttings logging For debris logging, due to the late time and some artificial factors, there will be certain errors.
- Drilling and coring is currently the most reliable and intuitive method for lithology identification.
- the current commonly used lithology identification method is based on the high-resolution underground petrophysical response information of the whole well section collected by the logging data, and the log data interpretation and analysis of the underground petrophysical response information is carried out to realize the formation of the whole well section Lithology identification.
- the underground rock physical response information can specifically include spontaneous potential curve, natural gamma curve, resistivity and other logging curves.
- the method of manual interpretation is generally used to identify the lithology of strata in the whole well section by using the underground rock physical response information.
- the structure of the actual stratum is complex and has heterogeneous characteristics, which requires high technical level of the interpreter, which leads to the problems of strong randomness, poor interpretation accuracy and effect, and inaccurate depth of the layer interface. It will consume a lot of labor time of interpreters, and the efficiency of lithology identification is not high.
- the present application is proposed in order to provide a model generation method and device and a lithology identification method and device that overcome the above problems or at least partially solve the above problems.
- the training data set includes multiple sets of instances, and each set of the examples includes at least one log data and one data annotation, the at least one log data includes a natural gamma curve and a natural potential curve of the target area , Any one or more of resistivity and well diameter, the data annotation is used to mark the lithology of the at least one logging data corresponding to different formation depths;
- the trained model is used to identify lithology at different formation depths in the area to be measured.
- the at least one log data includes a natural gamma curve and a natural potential curve of the target area.
- the obtaining the training data set specifically includes:
- the log interpretation includes conclusions of lithology interpretation at different formation depths
- the at least one log data and the obtained data tag are used as a set of the examples.
- acquiring the at least one log data and then including:
- the preprocessing includes any one or more of unit conversion, depth alignment, curve modification and splicing, curve smoothing, environmental correction, well deviation correction, and logging data standardization.
- the model is a decision tree, and the method further includes:
- pruning is performed on the trained model.
- the lithology of different formation depths in the area to be measured is obtained
- the identification model is obtained according to any one of the model generation methods provided in the above embodiments; the target logging data input to the identification model is the same as the logging data included in each group of instances in the training data set of the identification model.
- a model generation device provided by an embodiment of the present application includes: an acquisition unit and a training unit;
- the acquisition unit is used to obtain a training data set;
- the training data set includes multiple sets of instances, and each set of the instances includes at least one log data and one data label, and the at least one log data includes the naturalness of the target area Any one or more of a gamma curve, a spontaneous potential curve, a resistivity and a bore diameter, the data annotation is used to mark the lithology of the at least one logging data corresponding to different formation depths;
- the training unit is used to train the pre-built model using the training data set to obtain a trained model; the trained model is used to identify lithology at different formation depths in the area to be measured.
- the at least one log data includes a natural gamma curve and a natural potential curve of the target area.
- the acquiring unit specifically includes: acquiring a subunit and a labeling subunit;
- the obtaining subunit is used to obtain the at least one logging data and its corresponding logging interpretation; the logging interpretation includes lithological interpretation conclusions of different formation depths;
- the labeling subunit is configured to, after obtaining a data label corresponding to the at least one log data according to the logging interpretation, label the at least one log data and the obtained data as a group of the examples.
- the acquiring unit further includes: a processing subunit;
- the processing subunit is used to perform preprocessing and / or data normalization processing on the at least one logging data
- the preprocessing includes any one or more of unit conversion, depth alignment, curve modification and splicing, curve smoothing, environmental correction, well deviation correction, and logging data standardization.
- the model is a decision tree
- the device further includes: a processing unit;
- the processing unit is configured to perform pruning processing on the trained model according to a preset rule.
- a lithology identification device provided by an embodiment of the present application includes: a data acquisition unit, a model input unit, and a lithology determination unit;
- the data obtaining unit is used to obtain at least one target logging data of the area to be measured;
- the model input unit is used to input the at least one target logging data into a recognition model
- the lithology determining unit is used to obtain lithology at different formation depths in the area to be measured according to the output of the identification model;
- the identification model is obtained according to any one of the model generation methods provided in the above embodiments; the target logging data input to the identification model is the same as the logging data included in each group of instances in the training data set of the identification model.
- An embodiment of the present application further provides a storage medium, which is characterized in that a program is stored thereon, and when the program is executed by a processor, any one of the model generation methods provided in the foregoing embodiments is implemented, or the implementation is implemented as described above Any one of the lithology identification methods provided in the example.
- An embodiment of the present application further provides a processor, wherein the processor is used to run a program, and when the program is run, any one of the model generation methods provided in the foregoing embodiments is executed, or Any one of the lithology identification methods provided in the above embodiments.
- a model generation method and device utilize multiple sets of examples composed of at least one log data obtained by detecting the target area in advance and corresponding lithology data corresponding to different formation depths.
- the constructed model is trained to obtain lithology for different formation depths in the area to be tested.
- This application makes full use of the lithology information contained in the logging curve to study the different characteristics and rules under the sand and mudstone profiles, to realize the automatic determination and division of lithology, to achieve a high degree of automation of data processing, intelligent lithology identification, and improve rock
- the speed and accuracy of sex recognition can reduce the labor intensity of interpreters and reduce the requirements for interpreters' practical experience and professional technical knowledge. It is accurate and efficient without artificial randomness.
- FIG. 1 shows a schematic flowchart of a model generation method provided by an embodiment of the present application
- FIG. 2 shows multiple logging data of a target area provided by a specific embodiment of the present application
- FIG. 3 shows a set of examples provided by specific embodiments of the present application.
- FIG. 4 is a schematic flowchart of a lithology identification method provided by an embodiment of the present application.
- FIG. 5 shows a schematic structural diagram of a model generation device provided by an embodiment of the present application
- FIG. 6 shows a schematic flow structure diagram of a lithology identification device provided by an embodiment of the present application.
- the interpretation and analysis of logging data is generally carried out manually.
- the lithology of different formation depths in the area to be measured is carried out. Identify.
- the lithology and interface can be divided by the natural gamma curve, natural potential curve of the area to be measured, combined with the resistivity and the lithology changes reflected by other logging curves (such as well diameter, etc.).
- the natural potential curve that is, the curve of the natural potential with the depth of the well
- the common curve is "half amplitude point" (that is, the amplitude is half of the highest amplitude in the curve).
- the thickness of the formation is different, and the shape of the curve and the position of the "half width point" of the actual layer interface will change.
- the natural gamma curve that is, the total natural gamma ray intensity curve measured by the gamma ray detector
- the sandstone shows a low value, with the mud content in the rock layer.
- Increase the value of the curve gradually increases, which can be used as the basis for the division of sand and mudstone profiles.
- the technical level of the interpreter is relatively high, resulting in artificial stratification with strong randomness, poor interpretation accuracy and effect, and inaccurate depth of the layer interface.
- the interpretation work will also be It takes a lot of labor time for interpreters.
- the embodiments of the present application provide a method and device for model generation and a method and device for lithology identification, making full use of the lithology information contained in the logging curve, and using the logging data obtained by logging to train the model
- the trained model can be used to realize the automatic determination and division of lithology. It can realize highly automated data processing and intelligent identification of lithology, improve the speed and accuracy of lithology identification, and reduce the labor intensity of interpreters.
- FIG. 1 is a schematic flowchart of a model generation method provided by an embodiment of the present application.
- the model generation method provided by the embodiment of the present application includes the following steps S101-S102.
- the obtained training data set includes multiple sets of instances, and each set of instances includes at least one log data and one data annotation.
- at least one log data includes any one or more of the natural gamma curve, spontaneous potential curve, resistivity and bore diameter of the target area, and the data annotation is used to mark at least one log data corresponding to the lithology of different formation depths .
- At least one logging data included in each group of examples is obtained by logging experiments on the same target area. In practical applications, you can use different logging tools for the same target area to collect and record the numerical results of formation information measurement with depth changes. Any one or more of the multiple logging data obtained is a group of examples.
- the natural gamma curve is the curve of the ray intensity with the depth obtained by measuring the total natural gamma ray intensity of the rock with the gamma ray detector
- the natural potential curve is the curve of the natural potential with the well depth measured by the potential measurement equipment .
- Fig. 2 illustrates a schematic diagram of various logging data in a target area, including natural gamma curve, natural potential curve and well diameter.
- each set of examples includes multiple different logging data of the same target area for model training to more accurately describe the lithology characteristics of different formation depths
- improve Model prediction is the accuracy of lithology identification.
- each group of examples includes at least the natural gamma curve and the natural potential curve of the target area.
- a step of preprocessing and / or data normalization processing may be included on the at least one log data.
- the preprocessing step edits and corrects the originally collected raw data, which can specifically include any of unit conversion, depth alignment, curve modification and splicing, curve smoothing, environmental correction, well deviation correction, and log data standardization. One or more, the specific processing is not repeated here.
- the normalization process is to scale the fluctuation range of the command curve to between 0 and 1.
- the maximum value on the curve corresponds to 1
- the minimum value corresponds to 0, and then all values on the curve are linearly converted to between 0 and 1 to achieve the normalization of the curve.
- the normalization process not only eliminates the difference in dimensionality of different logging curves, but also avoids the tendency to focus on which curves with smaller dimensions and larger values when training the model, which is helpful to improve the accuracy of the model and unify the data.
- the range can make the trained model more general.
- the above describes how to obtain at least one logging data in each group of examples.
- the following examples illustrate how to obtain the data annotations included in each group of examples.
- step S101 may specifically include:
- S1011 Acquire at least one log data of the target area and its corresponding log interpretation.
- logging interpretation includes lithological interpretation conclusions for different formation depths in the target area, which can be carried out by experienced interpreters based on drilling, logging, core, logging data and subsequent production test data
- Reference data interpretation such as verified accurate horizon data can be obtained, and the specific conclusion of logging interpretation (such as whether the formation depth is sandstone or mudstone, volcanic rock or metamorphic rock, etc.) can be set according to actual needs.
- the use of log interpretation by experienced interpreters for model training can effectively transform and reuse the knowledge and experience of the personnel, and reduce the work of the experienced interpreters on the basis of ensuring the accuracy of lithology identification
- For strength and logging data acquisition method please refer to the relevant description above, which will not be repeated here.
- the data annotation is obtained according to the log interpretation.
- the data annotation can be an annotation of whether the formation depth is mudstone or sandstone at different strata in the target area. Because the sandstone and mudstone reservoir type is mainly mudstone and sandstone, the sandstone is the reservoir. Using this data labeling, the reservoirs in the stratum in the area to be tested can be divided and determined initially. Reservoir refers to a rock layer with connected pores, allowing oil and gas to be stored and infiltrated therein; the better the porosity and permeability of the reservoir, the more favorable it is for storing oil and gas.
- the concept of the reservoir only illustrates the reservoir With the ability to store oil and gas, not all reservoirs have already stored oil and gas.
- the data annotation may also be an annotation of whether the different formation depths of the target area are volcanic or metamorphic rocks, or an acidic rock and a basic rock.
- Fig. 3 combines the example of Fig. 2 with sandstone and mudstone as examples, and illustrates the logging data and corresponding data annotations in a group of examples.
- S102 Use the training data set to train the pre-built model to obtain the trained model.
- the trained model can be used to identify lithology at different formation depths in the area to be tested, and the recognition result is related to the data annotation included in the training data set.
- any kind of machine learning can be used
- the algorithm performs model training.
- the pre-built model may be a neural network model, any kind of classifier, any kind of decision tree, etc., which is not limited in the embodiments of the present application.
- decision tree As an example, it represents a mapping relationship between object attributes and object values.
- decision trees There are several methods for generating decision trees: classification trees, regression trees, classification regression (Classification And Regression Trees, CART) trees, and Chi-Square Automatic Interaction Detection (CHAID), etc., the examples of this application are not limited Decision tree generation method.
- the method may further include:
- pruning the trained model to reduce the complexity of the model and enhance the generalization ability of the model.
- Pruning is one of the methods to stop branches of decision trees. There are two types of pruning: pre-pruning and post-pruning.
- Pre-pruning is to set an index during the growth of the tree. When the index is reached, it will stop growing. This is easy to produce "view limitation", that is, once the branch is stopped, making node N become a leaf node, its successor is cut off Any possibility of a node performing a "good” branch operation. It is not strictly said that this will stop the branch will mislead the learning algorithm, resulting in the largest difference in tree impurity purity is too close to the root node.
- post-pruning the tree must first fully grow until the leaf nodes have the smallest impurity value, so that they can overcome the "vision limitation".
- multiple sets of instances composed of at least one log data obtained by detecting the target area in advance and data corresponding to lithology corresponding to different formation depths are used to train the pre-built model to obtain the data to be measured. Recognize the lithology of different stratum depths in the region.
- This application makes full use of the lithology information contained in the logging curve to study the different characteristics and rules under the sand and mudstone profiles, to realize the automatic determination and division of lithology, to achieve a high degree of automation of data processing, intelligent lithology identification, and improve rock
- the speed and accuracy of sexual recognition reduce the labor intensity of interpreters and reduce the requirements for interpreters' practical experience and professional technical knowledge.
- the recognition results are accurate, without artificial randomness, and the recognition process is efficient.
- an embodiment of the present application also provides a method for using the generated model to perform lithology identification.
- FIG. 4 is a schematic flowchart of a lithology identification method provided by an embodiment of the present application.
- the lithology identification method provided by the embodiment of the present application includes the following steps S401-S403.
- S401 Obtain at least one target logging data of the area to be measured.
- S402 Input at least one target logging data into the recognition model.
- the input recognition model is obtained in advance according to any one of the model generation methods provided in the above embodiments.
- the method for obtaining the recognition model please refer to the relevant description above, which will not be repeated here.
- at least one target logging data input to the model needs to be the same as the logging data included in each group of examples in the training data set used for the identification model training.
- each group of examples includes a natural gamma curve and a natural potential curve
- at least one target logging curve of the input recognition model in step S402 is the natural gamma curve and the natural potential curve of the area to be measured.
- the identification model and the target logging curve of the area to be tested to identify the lithology of different stratum depths in the area not only can it improve The speed and accuracy of lithology identification, the identification results are accurate, without artificial randomness, and the identification process is efficient. It can also reduce the labor intensity of the interpreter and reduce the requirements for the interpreter's practical experience and professional technical knowledge.
- an embodiment of the present application further provides a model generation method.
- FIG. 5 is a schematic structural diagram of a model generation apparatus provided by an embodiment of the present application.
- the model generating apparatus includes: an obtaining unit 501 and a training unit 502;
- the obtaining unit 501 is used to obtain a training data set;
- the training data set includes multiple sets of instances, and each set of instances includes at least one log data and one data label, and at least one log data includes a natural gamma curve and a natural potential curve of the target area , Any one or more of resistivity and bore diameter, data annotation is used to mark at least one log data corresponding to the lithology of different formation depths;
- the training unit 502 is used to train the pre-built model using the training data set to obtain the trained model; the trained model is used to identify the lithology of different formation depths in the area to be measured.
- At least one log data includes a natural gamma curve and a natural potential curve of the target area.
- the obtaining unit 501 may specifically include: obtaining a subunit and a labeling subunit;
- Obtaining subunit used to obtain at least one log data and its corresponding logging interpretation; logging interpretation includes lithological interpretation conclusions of different formation depths;
- the labeling subunit is used to obtain at least one log data corresponding to at least one log data according to the log interpretation, and label at least one log data and the obtained data as a group of examples.
- the obtaining unit 501 may further include: a processing subunit;
- a processing subunit used for preprocessing and / or data normalization processing of at least one logging data
- Preprocessing includes any one or more of unit conversion, depth alignment, curve modification and splicing, curve smoothing, environmental correction, well deviation correction, and log data standardization.
- the trained model is a decision tree
- the device may further include: a processing unit;
- the processing unit is used for pruning the trained model according to preset rules.
- multiple sets of instances composed of at least one log data obtained by detecting the target area in advance and data corresponding to lithology corresponding to different formation depths are used to train the pre-built model to obtain the data to be measured. Recognize the lithology of different stratum depths in the region.
- This application makes full use of the lithology information contained in the logging curve to study the different characteristics and rules under the sand and mudstone profiles, to realize the automatic determination and division of lithology, to achieve a high degree of automation of data processing, intelligent lithology identification, and improve rock
- the speed and accuracy of sexual recognition reduce the labor intensity of interpreters and reduce the requirements for interpreters' practical experience and professional technical knowledge.
- the recognition results are accurate, without artificial randomness, and the recognition process is efficient.
- an embodiment of the present application further provides a lithology identification device.
- FIG. 6 is a schematic structural diagram of a lithology identification device provided by an embodiment of the present application.
- the lithology identification device includes: a data acquisition unit 601, a model input unit 602, and a lithology determination unit 603;
- the data obtaining unit 601 is used to obtain at least one target logging data of the area to be measured;
- the model input unit 602 is used to input at least one target logging data into the recognition model
- the lithology determination unit 603 is used to obtain the lithology of different formation depths in the area to be measured according to the output of the identification model;
- the recognition model is obtained according to any one of the model generation methods provided in the above embodiments; the target logging data input to the recognition model is the same as the logging data included in each group of instances in the training data set of the recognition model.
- the identification model and the target logging curve of the area to be tested to identify the lithology of different stratum depths in the area not only can it improve The speed and accuracy of lithology identification, the identification results are accurate, without artificial randomness, and the identification process is efficient. It can also reduce the labor intensity of the interpreter and reduce the requirements for the interpreter's practical experience and professional technical knowledge.
- the model generation device and the lithology identification device both include a processor and a memory.
- the above acquisition unit, training unit, data acquisition unit, model input unit, and lithology determination unit are all stored as program units in the corresponding memory
- the processor executes the above program units stored in the memory to realize the corresponding functions.
- the processor contains a core, and the core retrieves the corresponding program unit from the memory.
- One or more kernels can be set. By adjusting the kernel parameters, high automation of data processing and intelligentization of lithology identification can be achieved. The speed and accuracy of lithology identification can be improved. The requirements of professional technical knowledge are accurate and efficient without artificial randomness.
- the memory may include non-permanent memory, random access memory (RAM) and / or non-volatile memory in a computer-readable medium, such as read only memory (ROM) or flash memory (flash RAM), and the memory includes at least one Memory chip.
- RAM random access memory
- ROM read only memory
- flash RAM flash memory
- An embodiment of the present application provides a storage medium on which a program is stored, and when the program is executed by a processor, the model generation method or the lithology identification method is implemented.
- An embodiment of the present application provides a processor for running a program, wherein the model generation method or the lithology identification method is executed when the program is run.
- An embodiment of the present application provides a device.
- the device includes a processor, a memory, and a program stored on the memory and executable on the processor.
- the processor executes the program, the following steps are implemented:
- the training data set includes multiple sets of instances, and each set of the examples includes at least one log data and one data annotation, the at least one log data includes a natural gamma curve and a natural potential curve of the target area , Any one or more of resistivity and well diameter, the data annotation is used to mark the lithology of the at least one logging data corresponding to different formation depths;
- the trained model is used to identify lithology at different formation depths in the area to be measured.
- the at least one log data includes a natural gamma curve and a natural potential curve of the target area.
- the obtaining the training data set specifically includes:
- the log interpretation includes conclusions of lithology interpretation at different formation depths
- the at least one log data and the obtained data tag are used as a set of the examples.
- acquiring the at least one log data and then including:
- the preprocessing includes any one or more of unit conversion, depth alignment, curve modification and splicing, curve smoothing, environmental correction, well deviation correction, and logging data standardization.
- the model is a decision tree, and the method further includes:
- pruning is performed on the trained model.
- the lithology of different formation depths in the area to be measured is obtained
- the recognition model is obtained according to any one of the model generation methods provided in the above embodiments; the target logging data input to the recognition model is the same as the logging data included in each group of instances in the training data set of the recognition model .
- the devices in this article can be servers, PCs, PADs, mobile phones, etc.
- the present application also provides a computer program product, which when executed on a data processing device, is suitable for executing a program initialized with the following method steps: obtaining a training data set; the training data set includes multiple sets of instances, each of which Examples include at least one log data and one data annotation, the at least one log data includes any one or more of the natural gamma curve, natural potential curve, resistivity and bore diameter of the target area, the data annotation is used To mark the lithology of the at least one logging data corresponding to different formation depths;
- the trained model is used to identify lithology at different formation depths in the area to be measured.
- the at least one log data includes a natural gamma curve and a natural potential curve of the target area.
- the obtaining the training data set specifically includes:
- the log interpretation includes conclusions of lithology interpretation at different formation depths
- the at least one log data and the obtained data tag are used as a set of the examples.
- acquiring the at least one log data and then including:
- the preprocessing includes any one or more of unit conversion, depth alignment, curve modification and splicing, curve smoothing, environmental correction, well deviation correction, and logging data standardization.
- the model is a decision tree, and the method further includes:
- pruning is performed on the trained model.
- the lithology of different formation depths in the area to be measured is obtained
- the recognition model is obtained according to any one of the model generation methods provided in the above embodiments; the target logging data input to the recognition model is the same as the logging data included in each group of instances in the training data set of the recognition model .
- the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, the present application 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.
- computer usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
- These computer program instructions may also be stored in a computer readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, the instructions The device implements the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and / or block diagrams.
- 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 performed on the computer or other programmable device to produce computer-implemented processing, which is executed on the computer or other programmable device
- the instructions provide steps for implementing the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and / or block diagrams.
- the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.
- processors CPUs
- input / output interfaces output interfaces
- network interfaces network interfaces
- memory volatile and non-volatile memory
- the memory may include non-permanent memory, random access memory (RAM) and / or non-volatile memory in a computer-readable medium, such as read only memory (ROM) or flash memory (flash RAM).
- RAM random access memory
- ROM read only memory
- flash RAM flash memory
- Computer-readable media including permanent and non-permanent, removable and non-removable media, can store information by any method or technology.
- the information may be computer readable instructions, data structures, modules of programs, or other data.
- Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM)), digital versatile disc (DVD) or other optical storage , Magnetic tape cassette, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media, can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.
- the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, the present application 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.
- computer usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Geophysics And Detection Of Objects (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
一种模型生成方法及装置和一种岩性识别方法及装置,该模型生成方法包括:获得训练数据集(S101);训练数据集包括多组实例,每组实例包括至少一个测井数据和一个数据标注,至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,数据标注用于标注至少一个测井数据对应不同地层深度的岩性;利用训练数据集对预先构建的模型进行训练,得到训练后的模型(S102);训练后的模型用于对待测区域不同地层深度的岩性进行识别。能够实现数据处理的高度自动化、岩性识别的智能化,提高岩性识别的速度和精度,减轻解释人员的劳动强度,降低对解释人员实践经验和专业技术知识的要求。
Description
本申请要求于2018年10月25日提交中国专利局、申请号为201811251307.3、发明名称为“模型生成方法及装置、岩性识别方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及数据处理技术领域,尤其涉及一种模型生成方法及装置和一种岩性识别方法及装置。
在油气勘探中,对井眼地层进行岩性识别、储层划分是对油气层识别和评价的基础,也是研究油藏沉积模式的重要依据。通常,岩性识别的实现方法有三种:岩屑录井、钻井取心和测井解释。针对岩屑录井,由于迟到时间的存在,以及一些人为的因素,会存在着一定的误差。而钻井取心是目前进行岩性识别最可靠最直观的方法,但油田开发取心井一般较少,取心的井段有限,无法获取井眼地层中全井段的岩性信息。因此,目前常用的岩性识别手段是根据测井资料采集到的全井段高分辨率的地下岩石物理响应信息,对地下岩石物理响应信息进行测井数据的解释分析,从而实现全井段地层的岩性识别。
地下岩石物理响应信息,具体可以包括自然电位曲线、自然伽马曲线、电阻率及其他测井曲线等。目前,一般采用人工解释的方法利用地下岩石物理响应信息对全井段地层的岩性进行识别。然而,实际地层的结构复杂、具有非均质性的特性,对解释人员的技术水平要求较高,导致人工分层存在随机性较强、解释精度和效果较差、层界面深度不精确的问题,又会耗费解释人员大量的劳动时间,岩性识别的效率不高。
发明内容
鉴于上述问题,提出了本申请以便提供一种克服上述问题或者至少部分地解决上述问题的一种模型生成方法及装置和一种岩性识别方法及装置。
本申请实施例提供的一种模型生成方法,包括:
获得训练数据集;所述训练数据集包括多组实例,每组所述实例包括至少一个测井数据和一个数据标注,所述至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,所述数据标注用于标注所述至少一个测井数据对应不同地层深度的岩性;
利用所述训练数据集对预先构建的模型进行训练,得到训练后的模型;所述训练后的模型用于对待测区域不同地层深度的岩性进行识别。
可选的,所述至少一个测井数据包括所述目标区域的自然伽马曲线和自然电位曲线。
可选的,所述获得训练数据集,具体包括:
获取所述至少一个测井数据及其对应的测井解释;所述测井解释包括不同地层深度的岩性解释结论;
根据所述测井解释,得到所述至少一个测井数据对应的一个数据标注后,将所述至少一个测井数据和得到的数据标注作为一组所述实例。
可选的,获取所述至少一个测井数据,之后还包括:
对所述至少一个测井数据进行预处理和/或数据归一化处理;
所述预处理包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个。
可选的,所述模型为决策树,所述方法还包括:
按照预设规则,对所述训练后的模型的进行剪枝处理。
本申请实施例提供的一种岩性识别方法,包括:
获得待测区域的至少一个目标测井数据;
将所述至少一个目标测井数据输入识别模型;
根据所述识别模型的输出,得到所述待测区域不同地层深度的岩性;
其中,所述识别模型根据上述实施例提供的模型生成方法中任意一种得到;输入所述识别模型的目标测井数据与所述识别模型的训练数据集中每组实例包括的测井数据相同。
本申请实施例提供的一种模型生成装置,包括:获取单元和训练单元;
所述获取单元,用于获得训练数据集;所述训练数据集包括多组实例,每组所述实例包括至少一个测井数据和一个数据标注,所述至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,所述数据标注用于标注所述至少一个测井数据对应不同地层深度的岩性;
所述训练单元,用于利用所述训练数据集对预先构建的模型进行训练,得到训练后的模型;所述训练后的模型用于对待测区域不同地层深度的岩性进行识别。
可选的,所述至少一个测井数据包括所述目标区域的自然伽马曲线和自然电位曲线。
可选的,所述获取单元,具体包括:获取子单元和标注子单元;
所述获取子单元,用于获取所述至少一个测井数据及其对应的测井解释;所述测井解释包括不同地层深度的岩性解释结论;
所述标注子单元,用于根据所述测井解释,得到所述至少一个测井数据对应的一个数据标注后,将所述至少一个测井数据和得到的数据标注作为一组所述实例。
可选的,所述获取单元,还包括:处理子单元;
所述处理子单元,用于对所述至少一个测井数据进行预处理和/或数据归一化处理;
所述预处理包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个。
可选的,所述模型为决策树,所述装置还包括:处理单元;
所述处理单元,用于按照预设规则,对所述训练后的模型的进行剪枝处理。
本申请实施例提供的一种岩性识别装置,包括:数据获取单元、模型输入单元和岩性确定单元;
所述数据获取单元,用于获得待测区域的至少一个目标测井数据;
所述模型输入单元,用于将所述至少一个目标测井数据输入识别模型;
所述岩性确定单元,用于根据所述识别模型的输出,得到所述待测区域不同地层深度的岩性;
其中,所述识别模型根据上述实施例提供的模型生成方法中任意一种得到;输入所述识别模型的目标测井数据与所述识别模型的训练数据集中每组实例包括的测井数据相同。
本申请实施例还提供了一种存储介质,其特征在于,其上存储有程序,该程序被处理器执行时实现如上述实施例提供的模型生成方法中任意一种,或者,实现如上述实施例提供的岩性识别方法中任意一种。
本申请实施例还提供了一种处理器,其特征在于,所述处理器用于运行程序,其中,所述程序运行时执行如上述实施例提供的模型生成方法中任意一种,或者,执行如上述实施例提供的岩性识别方法中任意一种。
借由上述技术方案,本申请提供的一种模型生成方法及装置,利用预先对目标区域检测得到的至少一个测井数据及其对应不同地层深度岩性的数据标注组成的多组实例,对预先构建的模型进行训练,得到用于对待测区域不同地层深度的岩性进行识别。本申请充分利用测井曲线所蕴含的岩性信息,研究砂泥岩剖面下不同的特征规律,实现岩性的自动判断及划分,能够实现数据处理的高度自动化、岩性识别的智能化,提高岩性识别的速度和精度,减轻解释人 员的劳动强度,降低对解释人员实践经验和专业技术知识的要求,准确高效,无人工随机性。
上述说明仅是本申请技术方案的概述,为了能够更清楚了解本申请的技术手段,而可依照说明书的内容予以实施,并且为了让本申请的上述和其它目的、特征和优点能够更明显易懂,以下特举本申请的具体实施方式。
通过阅读下文优选实施方式的详细描述,各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的,而并不认为是对本申请的限制。而且在整个附图中,用相同的参考符号表示相同的部件。在附图中:
图1示出了本申请实施例提供的一种模型生成方法的流程示意图;
图2示出了本申请具体实施例提供的一个目标区域的多个测井数据;
图3示出了本申请具体实施例提供的一组实例;
图4示出了本申请实施例提供的一种岩性识别方法的流程示意图;
图5示出了本申请实施例提供的一种模型生成装置的结构示意图;
图6示出了本申请实施例提供的一种岩性识别装置的流结构示意图。
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例,然而应当理解,可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反,提供这些实施例是为了能够更透彻地理解本公开,并且能够将本公开的范围完整的传达给本领域的技术人员。
目前,对测井数据的解释分析一般采用人工方法进行,借助测井解释专业应用软件,参考已有的录井、岩心、试油及生产测试等数据,对待测区域不同 地层深度的岩性进行识别。通常,可以用待测区域的自然伽马曲线、自然电位曲线结合电阻率及其他测井曲线(如井径等)反映的岩性变化来划分岩性及界面。
其中,自然电位曲线,即自然电位随井深变化的曲线,对划分砂泥岩剖面及砂岩渗透层直观而有效,常用曲线“半幅点”(即幅值为曲线中最高幅值一半的采用点)划分渗透层的顶底界,地层厚度的不同,曲线的形态及实际层位界面“半幅点”的位置会有变化。类似的,自然伽马曲线,即利用伽马射线探测器测量岩石总的自然伽马射线强度曲线,在砂泥岩剖面中会有不同的数值显示,砂岩显示低值,随着岩层中泥质含量的增加,曲线的数值逐渐增大,可作为砂泥岩剖面划分的依据。但是,由于实际地层的非均质性,对解释人员的技术水平要求较高,导致人工分层存在随机性较强,解释精度和效果较差,层界面深度不精确的问题,解释工作同样会耗费解释人员大量的劳动时间。
为此,本申请实施例提供了一种模型生成方法及装置和一种岩性识别方法及装置,充分利用测井曲线所蕴含的岩性信息,利用测井得到的测井数据进行模型的训练,训练得到的模型可以用于实现岩性的自动判断及划分,能够实现数据处理的高度自动化、岩性识别的智能化,提高岩性识别的速度和精度,减轻解释人员的劳动强度。
基于上述思想,为使本申请的上述目的、特征和优点能够更加明显易懂,下面结合附图对本申请的具体实施方式做详细的说明。
参见图1,该图为本申请实施例提供的一种模型生成方法的流程示意图。
本申请实施例提供的模型生成方法,包括如下步骤S101-S102。
S101:获得训练数据集。
在本申请实施例中,得到的训练数据集包括多组实例,每组实例包括至少一个测井数据和一个数据标注。其中,至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,数据标注用于 标注至少一个测井数据对应不同地层深度的岩性。
需要说明的是,每组实例包括的至少一个测井数据是对同一目标区域进行测井实验采集得到的。实际应用中,可以通过对同一目标区域使用不同的测井仪器,采集记录随着深度变化的地层信息测量数值结果,得到的多个测井数据中的任意一个或多个为一组实例中所包括的至少一个测井数据。例如,自然伽马曲线是利用伽马射线探测器测量岩石总的自然伽马射线强度得到的射线强度随深度变化的曲线,自然电位曲线是利用电位测量设备测得的自然电位随井深变化的曲线。图2举例示出了一个目标区域的多种测井数据的示意图,包括自然伽马曲线、自然电位曲线和井径。
这里还需要说明的是,本申请发明人在研究中发现,由于实际地层的非均质性,单一的测井曲线很难表征地层的真实性,因此,在本申请实施例一些可能的实现方式中,为了进一步提高对岩性识别的准确性,在每组实例中包括同一目标区域的多个不同的测井数据用于模型的训练,以更加准确的描述不同地层深度的岩性特征,提高模型预测即岩性识别的准确性。在一个例子中,每组实例至少包括目标区域的自然伽马曲线和自然电位曲线。
在本申请实施例一些可能的实现方式中,为了提高数据的精度,在步骤S101之后还可以包括对至少一个测井数据进行预处理和/或数据归一化处理的步骤。
其中,预处理步骤对初始采集到的原始数据进行编辑和校正等处理,具体可以包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个,具体处理过程这里不再一一赘述。
归一化处理是指令曲线的波动范围缩放到0至1之间。在一个例子中,可以取曲线上最大值对应于1、最小值对应于0,然后线性地将该曲线上所有值换算到0至1之间,以实现曲线的归一化。归一化处理不仅消除了不同测井曲 线在量纲上的差异,避免模型训练时倾向于重点考虑哪些量纲较小、数值较大的曲线,有利于提高模型的准确性,还统一了数据的范围可以使得训练得到的模型更加通用。
这里需要说明的是,本申请实施例对预处理和归一化处理的具体实现方式不进行限定,以上仅为示例性说明,不应视作对本申请的限制。
以上内容对如何得到每组实例中的至少一个测井数据进行了说明,下面举例说明如何得到每组实例所包括的数据标注。
在本申请实施例一些可能的实现方式中,步骤S101具体可以包括:
S1011:获取目标区域的至少一个测井数据及其对应的测井解释。
在本申请实施例中,测井解释包括对目标区域内不同地层深度的岩性解释结论,可以是通过有经验的解释人员根据钻井、录井、岩心、测井资料以及后续的生产测试数据进行验证的准确层位数据等参考数据解释得到,测井解释的具体结论(如该地层深度是砂岩还是泥岩、是火山岩还是变质岩等)可以根据实际需求具体设定。由于利用有经验的解释人员的测井解释进行模型的训练,可以使得该人员的知识经验可以得到高效的转化和复用,在保证岩性识别的准确度的基础上减轻经验的解释人员的工作强度,测井数据的获得方式参见上面的相关说明即可,这里不再赘述。
S1012:根据测井解释,得到至少一个测井数据对应的一个数据标注后,将至少一个测井数据和得到的数据标注作为一组实例。
可以理解的是,数据标注根据测井解释得到。在一个具体的场景中,数据标注可以是对目标区域不同地层深度是泥岩还是砂岩进行的标注,由于砂泥岩油气层储层类型主要为泥岩、砂岩两种岩性,其中砂岩为储集层,利用这种数据标注可以对待测区域内地层中的储集层进行初步的划分和确定。储集层指的是具有连通孔隙、允许油气在其中储存和渗滤的岩层;储集层的孔渗性越好则有越有利于储集油气,储集层的概念仅说明了储集层具备了储集油气的能力, 并非所有的储集层都已经储集了油气。在另一个具体的场景中,数据标注还可以是对目标区域不同地层深度是火山岩还是变质岩进行标注,或者酸性岩和基性岩进行标注。
图3结合图2的例子以砂岩和泥岩为例,举例示出了一组实例中的测井数据和对应的数据标注。为了方便编程处理,可以在数据标注中用0表示泥岩,用1表示砂岩,本申请对此不进行限定。
S102:利用训练数据集对预先构建的模型进行训练,得到训练后的模型。
在本申请实施例中,训练后的模型可以用于对待测区域不同地层深度的岩性进行识别,识别的结果与训练数据集中包括的数据标注相关,实际应用中,可以利用任意一种机器学习算法进行模型的训练,预先构建的模型可以是神经网络模型、任意一种分类器、任意一种决策树等,本申请实施例不进行限定。
以决策树(decision tree)为例,其代表的是对象属性与对象值之间的一种映射关系。决策树有几种生成方法:分类树、回归树、分类回归(Classification And Regression Trees,CART)树和卡方自动交互检测法(Chi-Square Automatic Interaction Detector,CHAID)等,本申请实施例不限定决策树的生成方法。
当训练后的模型为决策树时,在本申请实施例一些可能的实现方式中,步骤S102之后还可以包括:
按照预设规则,对训练后的模型的进行剪枝处理,以降低模型的复杂度、同时增强模型泛化能力。
剪枝是决策树停止分支的方法之一,剪枝有分预先剪枝和后剪枝两种。预先剪枝是在树的生长过程中设定一个指标,当达到该指标时就停止生长,这样做容易产生“视界局限”,就是一旦停止分支,使得节点N成为叶节点,就断绝了其后继节点进行“好”的分支操作的任何可能性。不严格的说这会已停止的分支会误导学习算法,导致产生的树不纯度降差最大的地方过分靠近根节点。后剪枝中树首先要充分生长,直到叶节点都有最小的不纯度值为止,因而 可以克服“视界局限”。然后对所有相邻的成对叶节点考虑是否消去它们,如果消去能引起令人满意的不纯度增长,那么执行消去,并令它们的公共父节点成为新的叶节点。这种“合并”叶节点的做法和节点分支的过程恰好相反,经过剪枝后叶节点常常会分布在很宽的层次上,树也变得非平衡。后剪枝技术的优点是克服了“视界局限”效应,而且无需保留部分样本用于交叉验证,所以可以充分利用全部训练集的信息。在具体实施时,可以根据训练数据集中数据的分布情况以及对于测试集预测效果的分析,对决策树进行适当的剪枝处理,这里不再赘述。
还需要说明的是,在实际应用中,随着测井数据的丰富以及测井解释的更新,可以根据新得到的测井数据和其对应的测井解释,或根据测井数据和新的测井解释得到新的实例作为训练数据集对训练后的模型进行更新,实现模型的迭代更新优化,可以使训练得到的模型更贴近地质构造的演变进程,从而使得模型对岩性识别的预测精度能够始终保持在高水平。
在本申请实施例中,利用预先对目标区域检测得到的至少一个测井数据及其对应不同地层深度岩性的数据标注组成的多组实例,对预先构建的模型进行训练,得到用于对待测区域不同地层深度的岩性进行识别。本申请充分利用测井曲线所蕴含的岩性信息,研究砂泥岩剖面下不同的特征规律,实现岩性的自动判断及划分,能够实现数据处理的高度自动化、岩性识别的智能化,提高岩性识别的速度和精度,减轻解释人员的劳动强度,降低对解释人员实践经验和专业技术知识的要求,识别结果准确、无人工随机性,识别过程高效。
基于上述实施例提供的模型生成方法,本申请实施例还提供了一种利用生成的模型进行岩性识别的方法。
参见图4,该图为本申请实施例提供的一种岩性识别方法的流程示意图。
本申请实施例提供的岩性识别方法,包括如下步骤S401-S403。
S401:获得待测区域的至少一个目标测井数据。
S402:将至少一个目标测井数据输入识别模型。
S403:根据识别模型的输出,得到待测区域不同地层深度的岩性。
在本申请实施例中,输入的识别模型预先是根据上述实施例提供的模型生成方法中的任意一种得到的,识别模型的得到方法具体参见上面的相关说明即可,这里不再赘述。这里需要说明的是,为了得到准确的岩性识别结果,输入模型的至少一个目标测井数据需要与识别模型训练得到是所用的训练数据集中每组实例包括的测井数据相同。例如,每组实例包括自然伽马曲线和自然电位曲线,则步骤S402中输入识别模型的至少一个目标测井曲线为待测区域的自然伽马曲线和自然电位曲线。
在本申请实施例中,预先利用测井曲线所蕴含的岩性信息得到识别模型后,利用该识别模型和待测区域的目标测井曲线对待区域不同地层深度的岩性进行识别,不仅能够提高岩性识别的速度和精度,识别结果准确、无人工随机性,识别过程高效,还能够减轻解释人员的劳动强度,降低对解释人员实践经验和专业技术知识的要求。
基于上述实施例提供的模型生成方法,本申请实施例还提供了一种模型生成方法。
参见图5,该图为本申请实施例提供的一种模型生成装置的结构示意图。
本申请实施例提供的模型生成装置,包括:获取单元501和训练单元502;
获取单元501,用于获得训练数据集;训练数据集包括多组实例,每组实例包括至少一个测井数据和一个数据标注,至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,数据标注用于标注至少一个测井数据对应不同地层深度的岩性;
训练单元502,用于利用训练数据集对预先构建的模型进行训练,得到训 练后的模型;训练后的模型用于对待测区域不同地层深度的岩性进行识别。
可选的,至少一个测井数据包括目标区域的自然伽马曲线和自然电位曲线。
在本申请实施例一些可能的实现方式中,获取单元501,具体可以包括:获取子单元和标注子单元;
获取子单元,用于获取至少一个测井数据及其对应的测井解释;测井解释包括不同地层深度的岩性解释结论;
标注子单元,用于根据测井解释,得到至少一个测井数据对应的一个数据标注后,将至少一个测井数据和得到的数据标注作为一组实例。
在本申请实施例一些可能的实现方式中,获取单元501,还可以包括:处理子单元;
处理子单元,用于对至少一个测井数据进行预处理和/或数据归一化处理;
预处理包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个。
可选的,训练得到的模型为决策树,该装置还可以包括:处理单元;
处理单元,用于按照预设规则,对训练后的模型的进行剪枝处理。
在本申请实施例中,利用预先对目标区域检测得到的至少一个测井数据及其对应不同地层深度岩性的数据标注组成的多组实例,对预先构建的模型进行训练,得到用于对待测区域不同地层深度的岩性进行识别。本申请充分利用测井曲线所蕴含的岩性信息,研究砂泥岩剖面下不同的特征规律,实现岩性的自动判断及划分,能够实现数据处理的高度自动化、岩性识别的智能化,提高岩性识别的速度和精度,减轻解释人员的劳动强度,降低对解释人员实践经验和专业技术知识的要求,识别结果准确、无人工随机性,识别过程高效。
基于上述实施例提供的模型生成方法和岩性识别方法,本申请实施例还提 供了一种岩性识别装置。
参见图6,该图为本申请实施例提供的一种岩性识别装置的结构示意图。
本申请实施例提供的岩性识别装置,包括:数据获取单元601、模型输入单元602和岩性确定单元603;
数据获取单元601,用于获得待测区域的至少一个目标测井数据;
模型输入单元602,用于将至少一个目标测井数据输入识别模型;
岩性确定单元603,用于根据识别模型的输出,得到待测区域不同地层深度的岩性;
其中,识别模型根据上述实施例提供的模型生成方法中的任意一种得到;输入识别模型的目标测井数据与识别模型的训练数据集中每组实例包括的测井数据相同。
在本申请实施例中,预先利用测井曲线所蕴含的岩性信息得到识别模型后,利用该识别模型和待测区域的目标测井曲线对待区域不同地层深度的岩性进行识别,不仅能够提高岩性识别的速度和精度,识别结果准确、无人工随机性,识别过程高效,还能够减轻解释人员的劳动强度,降低对解释人员实践经验和专业技术知识的要求。
所述模型生成装置和岩性识别装置均包括处理器和存储器,上述获取单元、训练单元、数据获取单元、模型输入单元和岩性确定单元等均作为程序单元存储在对应的存储器中,由对应的处理器执行存储在存储器中的上述程序单元来实现相应的功能。
处理器中包含内核,由内核去存储器中调取相应的程序单元。内核可以设置一个或以上,通过调整内核参数来能够实现数据处理的高度自动化、岩性识别的智能化,提高岩性识别的速度和精度,减轻解释人员的劳动强度,降低对解释人员实践经验和专业技术知识的要求,准确高效,无人工随机性。
存储器可能包括计算机可读介质中的非永久性存储器,随机存取存储器 (RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM),存储器包括至少一个存储芯片。
本申请实施例提供了一种存储介质,其上存储有程序,该程序被处理器执行时实现所述模型生成方法或所述岩性识别方法。
本申请实施例提供了一种处理器,所述处理器用于运行程序,其中,所述程序运行时执行所述模型生成方法或所述岩性识别方法。
本申请实施例提供了一种设备,设备包括处理器、存储器及存储在存储器上并可在处理器上运行的程序,处理器执行程序时实现以下步骤:
获得训练数据集;所述训练数据集包括多组实例,每组所述实例包括至少一个测井数据和一个数据标注,所述至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,所述数据标注用于标注所述至少一个测井数据对应不同地层深度的岩性;
利用所述训练数据集对预先构建的模型进行训练,得到训练后的模型;所述训练后的模型用于对待测区域不同地层深度的岩性进行识别。
可选的,所述至少一个测井数据包括所述目标区域的自然伽马曲线和自然电位曲线。
可选的,所述获得训练数据集,具体包括:
获取所述至少一个测井数据及其对应的测井解释;所述测井解释包括不同地层深度的岩性解释结论;
根据所述测井解释,得到所述至少一个测井数据对应的一个数据标注后,将所述至少一个测井数据和得到的数据标注作为一组所述实例。
可选的,获取所述至少一个测井数据,之后还包括:
对所述至少一个测井数据进行预处理和/或数据归一化处理;
所述预处理包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个。
可选的,所述模型为决策树,所述方法还包括:
按照预设规则,对所述训练后的模型的进行剪枝处理。
或者,实现以下步骤:
获得待测区域的至少一个目标测井数据;
将所述至少一个目标测井数据输入识别模型;
根据所述识别模型的输出,得到所述待测区域不同地层深度的岩性;
其中,所述识别模型根据上述实施例提供的模型生成方法中的任意一种得到;输入所述识别模型的目标测井数据与所述识别模型的训练数据集中每组实例包括的测井数据相同。
本文中的设备可以是服务器、PC、PAD、手机等。
本申请还提供了一种计算机程序产品,当在数据处理设备上执行时,适于执行初始化有如下方法步骤的程序:获得训练数据集;所述训练数据集包括多组实例,每组所述实例包括至少一个测井数据和一个数据标注,所述至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,所述数据标注用于标注所述至少一个测井数据对应不同地层深度的岩性;
利用所述训练数据集对预先构建的模型进行训练,得到训练后的模型;所述训练后的模型用于对待测区域不同地层深度的岩性进行识别。
可选的,所述至少一个测井数据包括所述目标区域的自然伽马曲线和自然电位曲线。
可选的,所述获得训练数据集,具体包括:
获取所述至少一个测井数据及其对应的测井解释;所述测井解释包括不同地层深度的岩性解释结论;
根据所述测井解释,得到所述至少一个测井数据对应的一个数据标注后,将所述至少一个测井数据和得到的数据标注作为一组所述实例。
可选的,获取所述至少一个测井数据,之后还包括:
对所述至少一个测井数据进行预处理和/或数据归一化处理;
所述预处理包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个。
可选的,所述模型为决策树,所述方法还包括:
按照预设规则,对所述训练后的模型的进行剪枝处理。
或者,有如下方法步骤的程序:
获得待测区域的至少一个目标测井数据;
将所述至少一个目标测井数据输入识别模型;
根据所述识别模型的输出,得到所述待测区域不同地层深度的岩性;
其中,所述识别模型根据上述实施例提供的模型生成方法中的任意一种得到;输入所述识别模型的目标测井数据与所述识别模型的训练数据集中每组实例包括的测井数据相同。
本领域内的技术人员应明白,本申请的实施例可提供为方法、系统、或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中,计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
存储器可能包括计算机可读介质中的非永久性存储器,随机存取存储器(RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM)。存储器是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM))、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包 括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括要素的过程、方法、商品或者设备中还存在另外的相同要素。
本领域技术人员应明白,本申请的实施例可提供为方法、系统或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
以上仅为本申请的实施例而已,并不用于限制本申请。对于本领域技术人员来说,本申请可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在本申请的权利要求范围之内。
Claims (10)
- 一种模型生成方法,其特征在于,所述方法包括:获得训练数据集;所述训练数据集包括多组实例,每组所述实例包括至少一个测井数据和一个数据标注,所述至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,所述数据标注用于标注所述至少一个测井数据对应不同地层深度的岩性;利用所述训练数据集对预先构建的模型进行训练,得到训练后的模型;所述训练后的模型用于对待测区域不同地层深度的岩性进行识别。
- 根据权利要求1所述的方法,其特征在于,所述至少一个测井数据包括所述目标区域的自然伽马曲线和自然电位曲线。
- 根据权利要求1所述的方法,其特征在于,所述获得训练数据集,具体包括:获取所述至少一个测井数据及其对应的测井解释;所述测井解释包括不同地层深度的岩性解释结论;根据所述测井解释,得到所述至少一个测井数据对应的一个数据标注后,将所述至少一个测井数据和得到的数据标注作为一组所述实例。
- 根据权利要求3所述的方法,其特征在于,获取所述至少一个测井数据,之后还包括:对所述至少一个测井数据进行预处理和/或数据归一化处理;所述预处理包括单位转换、深度对齐、曲线修改和拼接、曲线平滑、环境校正、井斜校正和测井数据标准化中的任意一个或多个。
- 一种岩性识别方法,其特征在于,所述方法包括:获得待测区域的至少一个目标测井数据;将所述至少一个目标测井数据输入识别模型;根据所述识别模型的输出,得到所述待测区域不同地层深度的岩性;其中,所述识别模型根据权利要求1-4任意一项所述的模型生成方法得到;输入所述识别模型的目标测井数据与所述识别模型的训练数据集中每组实例包括的测井数据相同。
- 一种模型生成装置,其特征在于,所述装置包括:获取单元和训练单元;所述获取单元,用于获得训练数据集;所述训练数据集包括多组实例,每组所述实例包括至少一个测井数据和一个数据标注,所述至少一个测井数据包括目标区域的自然伽马曲线、自然电位曲线、电阻率和井径中的任意一个或多个,所述数据标注用于标注所述至少一个测井数据对应不同地层深度的岩性;所述训练单元,用于利用所述训练数据集对预先构建的模型进行训练,得到训练后的模型;所述训练后的模型用于对待测区域不同地层深度的岩性进行识别。
- 根据权利要求6所述的装置,其特征在于,所述获取单元,具体包括:获取子单元和标注子单元;所述获取子单元,用于获取所述至少一个测井数据及其对应的测井解释;所述测井解释包括不同地层深度的岩性解释结论;所述标注子单元,用于根据所述测井解释,得到所述至少一个测井数据对应的一个数据标注后,将所述至少一个测井数据和得到的数据标注作为一组所述实例。
- 一种岩性识别装置,其特征在于,所述装置包括:数据获取单元、模型输入单元和岩性确定单元;所述数据获取单元,用于获得待测区域的至少一个目标测井数据;所述模型输入单元,用于将所述至少一个目标测井数据输入识别模型;所述岩性确定单元,用于根据所述识别模型的输出,得到所述待测区域不同地层深度的岩性;其中,所述识别模型根据权利要求1-4任意一项所述的模型生成方法得到;输入所述识别模型的目标测井数据与所述识别模型的训练数据集中每组实例包括的测井数据相同。
- 一种存储介质,其特征在于,其上存储有程序,该程序被处理器执行时实现如权利要求1-4任一项所述的模型生成方法,或者,实现如权利要求5所述的岩性识别方法。
- 一种处理器,其特征在于,所述处理器用于运行程序,其中,所述程序运行时执行如权利要求1-4任一项所述的模型生成方法,或者,执行如权利要求5所述的岩性识别方法。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811251307.3A CN111104819A (zh) | 2018-10-25 | 2018-10-25 | 模型生成方法及装置、岩性识别方法及装置 |
CN201811251307.3 | 2018-10-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020082824A1 true WO2020082824A1 (zh) | 2020-04-30 |
Family
ID=70330793
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/097226 WO2020082824A1 (zh) | 2018-10-25 | 2019-07-23 | 模型生成方法及装置、岩性识别方法及装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111104819A (zh) |
WO (1) | WO2020082824A1 (zh) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111598444A (zh) * | 2020-05-15 | 2020-08-28 | 中国石油天然气集团有限公司 | 基于卷积神经网络的测井岩性识别方法及系统 |
CN111767674A (zh) * | 2020-05-29 | 2020-10-13 | 中国科学技术大学 | 一种基于主动域适应的测井岩性识别方法 |
CN111862778A (zh) * | 2020-06-04 | 2020-10-30 | 中国地质科学院 | 一种浅层岩性地质图生成方法、装置、储存介质及设备 |
CN112348922A (zh) * | 2020-11-06 | 2021-02-09 | 长江大学 | 一种测井曲线自动绘制方法、系统、装置及存储介质 |
CN112541523A (zh) * | 2020-11-17 | 2021-03-23 | 中海油田服务股份有限公司 | 一种泥质含量计算方法和装置 |
CN112784980A (zh) * | 2021-01-05 | 2021-05-11 | 中国石油天然气集团有限公司 | 一种智能化测井层位划分方法 |
CN113344050A (zh) * | 2021-05-28 | 2021-09-03 | 中国石油天然气股份有限公司 | 一种基于深度学习的岩性智能化识别方法及系统 |
CN114280689A (zh) * | 2021-12-20 | 2022-04-05 | 中国石油大学(北京) | 基于岩石物理知识确定储层孔隙度的方法、装置及设备 |
CN114492521A (zh) * | 2022-01-21 | 2022-05-13 | 成都理工大学 | 一种基于声振信号的随钻岩性智能识别方法与系统 |
CN114764423A (zh) * | 2022-05-23 | 2022-07-19 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | 一种测井智能解释系统 |
CN115576028A (zh) * | 2022-12-01 | 2023-01-06 | 武汉盛华伟业科技股份有限公司 | 基于支持向量机的地质特征层预测方法及系统 |
CN115877464A (zh) * | 2022-12-30 | 2023-03-31 | 中海石油(中国)有限公司深圳分公司 | 一种岩性识别方法、装置、计算机设备及存储介质 |
CN116070089A (zh) * | 2023-02-21 | 2023-05-05 | 北京金阳普泰石油技术股份有限公司 | 一种基于ResNet回归模型的地层划分方法、装置和计算机设备 |
CN116227308A (zh) * | 2023-05-09 | 2023-06-06 | 广东石油化工学院 | 一种浅层测井自然电场的数值模拟方法及系统 |
CN117872494A (zh) * | 2024-03-12 | 2024-04-12 | 北京科技大学 | 深部金属矿花岗岩岩石内部节理信息获取方法、预测方法 |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111753958A (zh) * | 2020-06-22 | 2020-10-09 | 成都理工大学 | 基于测井数据深度学习的灯影组微生物岩微相识别方法 |
CN112144594B (zh) * | 2020-08-10 | 2022-03-18 | 中交第二航务工程局有限公司 | 一种基于tpot的铣槽机施工地层识别方法 |
CN111980688B (zh) * | 2020-09-01 | 2021-11-23 | 中国石油集团渤海钻探工程有限公司 | 一种基于集成学习算法的井斜角度预测方法 |
CN112903607A (zh) * | 2021-02-23 | 2021-06-04 | 谢跃红 | 地下地质勘探方法、装置、设备及存储介质 |
CN112801035B (zh) * | 2021-02-24 | 2023-04-07 | 山东大学 | 基于知识与数据双驱动的搭载式岩性智能识别方法及系统 |
CN112990320A (zh) * | 2021-03-19 | 2021-06-18 | 中国矿业大学(北京) | 一种岩性的分类方法、装置、电子设备及存储介质 |
CN117633658B (zh) * | 2024-01-25 | 2024-04-19 | 北京大学 | 岩石储层岩性识别方法及系统 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050206378A1 (en) * | 2004-03-18 | 2005-09-22 | Baker Hughes Incorporated | Rock properties prediction, categorization, and recognition from NMR echo-trains using linear and nonlinear regression |
CN105388531A (zh) * | 2015-10-19 | 2016-03-09 | 成都理工大学 | 一种基于支持向量回归机和核fisher分析的岩性识别方法 |
CN105488248A (zh) * | 2015-11-18 | 2016-04-13 | 山东科技大学 | 一种深部矿层和岩层的判定方法 |
CN105697002A (zh) * | 2014-11-24 | 2016-06-22 | 中国石油化工股份有限公司 | 一种用于识别煤系地层岩性的方法 |
CN106291713A (zh) * | 2016-08-24 | 2017-01-04 | 中国石油天然气股份有限公司 | 一种确定地层岩性的处理方法及装置 |
CN107133670A (zh) * | 2017-04-21 | 2017-09-05 | 中国科学院大学 | 一种基于决策树数据挖掘算法的复杂岩性识别方法及系统 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011146734A2 (en) * | 2010-05-19 | 2011-11-24 | Schlumberger Canada Limited | Pulse neutron formation gas identification with lwd measurements |
CN108122066B (zh) * | 2017-11-13 | 2021-08-03 | 中国石油天然气股份有限公司 | 储层岩性的确定方法和装置 |
-
2018
- 2018-10-25 CN CN201811251307.3A patent/CN111104819A/zh active Pending
-
2019
- 2019-07-23 WO PCT/CN2019/097226 patent/WO2020082824A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050206378A1 (en) * | 2004-03-18 | 2005-09-22 | Baker Hughes Incorporated | Rock properties prediction, categorization, and recognition from NMR echo-trains using linear and nonlinear regression |
CN105697002A (zh) * | 2014-11-24 | 2016-06-22 | 中国石油化工股份有限公司 | 一种用于识别煤系地层岩性的方法 |
CN105388531A (zh) * | 2015-10-19 | 2016-03-09 | 成都理工大学 | 一种基于支持向量回归机和核fisher分析的岩性识别方法 |
CN105488248A (zh) * | 2015-11-18 | 2016-04-13 | 山东科技大学 | 一种深部矿层和岩层的判定方法 |
CN106291713A (zh) * | 2016-08-24 | 2017-01-04 | 中国石油天然气股份有限公司 | 一种确定地层岩性的处理方法及装置 |
CN107133670A (zh) * | 2017-04-21 | 2017-09-05 | 中国科学院大学 | 一种基于决策树数据挖掘算法的复杂岩性识别方法及系统 |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111598444A (zh) * | 2020-05-15 | 2020-08-28 | 中国石油天然气集团有限公司 | 基于卷积神经网络的测井岩性识别方法及系统 |
CN111767674A (zh) * | 2020-05-29 | 2020-10-13 | 中国科学技术大学 | 一种基于主动域适应的测井岩性识别方法 |
CN111767674B (zh) * | 2020-05-29 | 2024-03-29 | 中国科学技术大学 | 一种基于主动域适应的测井岩性识别方法 |
CN111862778A (zh) * | 2020-06-04 | 2020-10-30 | 中国地质科学院 | 一种浅层岩性地质图生成方法、装置、储存介质及设备 |
CN112348922A (zh) * | 2020-11-06 | 2021-02-09 | 长江大学 | 一种测井曲线自动绘制方法、系统、装置及存储介质 |
CN112348922B (zh) * | 2020-11-06 | 2023-04-07 | 长江大学 | 一种测井曲线自动绘制方法、系统、装置及存储介质 |
CN112541523B (zh) * | 2020-11-17 | 2023-02-28 | 中海油田服务股份有限公司 | 一种泥质含量计算方法和装置 |
CN112541523A (zh) * | 2020-11-17 | 2021-03-23 | 中海油田服务股份有限公司 | 一种泥质含量计算方法和装置 |
CN112784980A (zh) * | 2021-01-05 | 2021-05-11 | 中国石油天然气集团有限公司 | 一种智能化测井层位划分方法 |
CN112784980B (zh) * | 2021-01-05 | 2024-05-28 | 中国石油天然气集团有限公司 | 一种智能化测井层位划分方法 |
CN113344050B (zh) * | 2021-05-28 | 2024-03-26 | 中国石油天然气股份有限公司 | 一种基于深度学习的岩性智能化识别方法及系统 |
CN113344050A (zh) * | 2021-05-28 | 2021-09-03 | 中国石油天然气股份有限公司 | 一种基于深度学习的岩性智能化识别方法及系统 |
CN114280689A (zh) * | 2021-12-20 | 2022-04-05 | 中国石油大学(北京) | 基于岩石物理知识确定储层孔隙度的方法、装置及设备 |
CN114492521A (zh) * | 2022-01-21 | 2022-05-13 | 成都理工大学 | 一种基于声振信号的随钻岩性智能识别方法与系统 |
CN114764423A (zh) * | 2022-05-23 | 2022-07-19 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | 一种测井智能解释系统 |
CN114764423B (zh) * | 2022-05-23 | 2024-05-07 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | 一种测井智能解释系统 |
CN115576028A (zh) * | 2022-12-01 | 2023-01-06 | 武汉盛华伟业科技股份有限公司 | 基于支持向量机的地质特征层预测方法及系统 |
CN115877464A (zh) * | 2022-12-30 | 2023-03-31 | 中海石油(中国)有限公司深圳分公司 | 一种岩性识别方法、装置、计算机设备及存储介质 |
CN115877464B (zh) * | 2022-12-30 | 2024-02-13 | 中海石油(中国)有限公司深圳分公司 | 一种岩性识别方法、装置、计算机设备及存储介质 |
CN116070089A (zh) * | 2023-02-21 | 2023-05-05 | 北京金阳普泰石油技术股份有限公司 | 一种基于ResNet回归模型的地层划分方法、装置和计算机设备 |
CN116227308B (zh) * | 2023-05-09 | 2023-07-18 | 广东石油化工学院 | 一种浅层测井自然电场的数值模拟方法及系统 |
CN116227308A (zh) * | 2023-05-09 | 2023-06-06 | 广东石油化工学院 | 一种浅层测井自然电场的数值模拟方法及系统 |
CN117872494A (zh) * | 2024-03-12 | 2024-04-12 | 北京科技大学 | 深部金属矿花岗岩岩石内部节理信息获取方法、预测方法 |
CN117872494B (zh) * | 2024-03-12 | 2024-06-07 | 北京科技大学 | 深部金属矿花岗岩岩石内部节理信息获取方法、预测方法 |
Also Published As
Publication number | Publication date |
---|---|
CN111104819A (zh) | 2020-05-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020082824A1 (zh) | 模型生成方法及装置、岩性识别方法及装置 | |
CN109800863B (zh) | 一种基于模糊理论和神经网络的测井相识别方法 | |
CN111665560B (zh) | 油气层识别方法、装置、计算机设备及可读存储介质 | |
CN111472751B (zh) | 测井解释方法、知识图谱构建方法及相关装置 | |
US9176255B2 (en) | Permeability prediction systems and methods using quadratic discriminant analysis | |
CN104318109A (zh) | 基于支持向量机的页岩气储层识别方法 | |
CN109597129B (zh) | 基于目标检测的缝洞型油藏串珠状反射特征识别方法 | |
CN112784980B (zh) | 一种智能化测井层位划分方法 | |
CN104977613A (zh) | 基于多信息的碳酸盐岩岩相古地理重构方法及装置 | |
SHAN et al. | Identification of complex lithology for tight sandstone gas reservoirs sase on BP neural net | |
CN108952699A (zh) | 一种复杂地质钻进过程地层岩性智能识别方法 | |
CN107942404A (zh) | 一种确定裂缝的方法及装置 | |
CN107956465A (zh) | 基于关联井的全区多井测井曲线标准化方法及装置 | |
CN107688037A (zh) | 一种利用核磁测井t2分布确定井下岩石粒度曲线的方法 | |
CN107345481A (zh) | 煤田测井曲线标准化方法 | |
CN105986819B (zh) | 用于测井资料自动处理与综合解释的方法和装置 | |
CN108035709A (zh) | 一种页岩储层质量的确定方法及装置 | |
CN105003257A (zh) | 一种定性识别高温高压甲烷气层与二氧化碳气层的方法 | |
CA3227514A1 (en) | Characterizing effects of co2 chemical reaction with rock minerals during carbon capture and sequestration | |
CN107448195A (zh) | 识别地层中轻质油层与凝析气层的方法及应用 | |
AU2017279838B1 (en) | Method for classifying deep rock geofacies based on data mining | |
Zhonghua | Seismic data attribute extraction based on Hadoop platform | |
CN116201535B (zh) | 油气藏目标井标志地层自动划分方法、装置及设备 | |
CN114086887B (zh) | 一种基于人工智能的待钻井眼轨道井下规划方法 | |
Ma et al. | Design and development of intelligent well logging interpretation system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19876259 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19876259 Country of ref document: EP Kind code of ref document: A1 |