WO2024060427A1 - 河道与断层同步检测方法及装置 - Google Patents

河道与断层同步检测方法及装置 Download PDF

Info

Publication number
WO2024060427A1
WO2024060427A1 PCT/CN2022/138860 CN2022138860W WO2024060427A1 WO 2024060427 A1 WO2024060427 A1 WO 2024060427A1 CN 2022138860 W CN2022138860 W CN 2022138860W WO 2024060427 A1 WO2024060427 A1 WO 2024060427A1
Authority
WO
WIPO (PCT)
Prior art keywords
amplitude value
data
seismic data
stack seismic
dimensional post
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2022/138860
Other languages
English (en)
French (fr)
Inventor
李海山
杨午阳
魏新建
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Petrochina Co Ltd
Original Assignee
Petrochina Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Petrochina Co Ltd filed Critical Petrochina Co Ltd
Publication of WO2024060427A1 publication Critical patent/WO2024060427A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
    • G01V1/30—Analysis

Definitions

  • the invention relates to the field of seismic data interpretation, and specifically to a method for synchronous detection of river channels and faults, a device for synchronous detection of river channels and faults, an electronic device and a machine-readable storage medium.
  • River channels are one of the important geological targets in oil and gas exploration. It is crucial to accurately detect the location of river channels and reveal their development processes. In terms of basin and prospect scale analysis, mastering the channel characteristics is helpful for quantitative analysis of landforms and accurate understanding of the deposition process, and helps to understand the depositional environment and the main directions of sediments over time; in terms of reservoir scale analysis, channel sandstone It is crucial to determine the volume of the river and the properties of the fluid filling it; in the oil and gas field development stage, accurately depicting the macroscopic distribution characteristics, sedimentation and change patterns of the river channel can provide guidance for the selection of development target blocks, preparation of oil and gas development plans, well locations and wells Provide important basis for trajectory design, etc.
  • the purpose of the embodiments of the present invention is to provide a method and device for synchronous detection of river channels and faults to solve the above problem of manually interpreting river channels, which is time-consuming, inaccurate, and subject to human subjectivity; using conventional convolutional neural networks River detection is highly dependent on labels and label generation methods, resulting in poor generalization ability in other work areas, and only considering the impact of folds and ignoring the impact of faults, resulting in low detection accuracy.
  • embodiments of the present invention provide a method for synchronous detection of river channels and faults, including:
  • the three-dimensional post-stack seismic data is processed, including:
  • processed three-dimensional post-stack seismic data is obtained.
  • determine the normalized amplitude value based on the minimum amplitude value and the maximum amplitude value including:
  • ⁇ A is the normalized amplitude value
  • a max is the maximum amplitude value
  • a min is the minimum amplitude value.
  • transform the amplitude value of each data in the three-dimensional post-stack seismic data based on the minimum amplitude value to obtain the first transformed data including:
  • A′ i A i -A min ;
  • A′ i is the first transformed data; A i is the maximum amplitude value; A min is the minimum amplitude value.
  • the processed three-dimensional post-stack seismic data is obtained, including:
  • the processed three-dimensional post-stack seismic data is calculated using the following calculation formula:
  • A′′ i is the processed three-dimensional post-stack seismic data
  • A′ i is the first transformed data
  • ⁇ A is the normalized amplitude value
  • the method also includes:
  • the training sample set is used as the input of the deep learning neural network, and the river channel and fault detection model is trained.
  • the invention also provides a device for synchronizing detection of river channels and faults, including:
  • Data acquisition module used to acquire three-dimensional post-stack seismic data
  • a data processing module used for processing the three-dimensional post-stack seismic data based on the minimum amplitude value and the maximum amplitude value in the three-dimensional post-stack seismic data;
  • the result output module is used to use the processed three-dimensional post-stack seismic data as the input of the river channel and fault detection model to obtain the river channel and fault detection results;
  • the data processing module specifically includes:
  • a data amplitude determination module used to determine the minimum amplitude value and the maximum amplitude value of the three-dimensional post-stack seismic data
  • a normalized amplitude value determination module configured to determine a normalized amplitude value based on the minimum amplitude value and the maximum amplitude value
  • An amplitude value transformation module configured to transform the amplitude value of each data in the three-dimensional post-stack seismic data based on the minimum amplitude value to obtain the first transformed data
  • a data output module is used to obtain processed three-dimensional post-stack seismic data based on the normalized amplitude value and the first transformed data.
  • ⁇ A is the normalized amplitude value
  • a max is the maximum amplitude value
  • a min is the minimum amplitude value.
  • A′ i A i ⁇ A min ;
  • A′ i is the first transformed data; A i is the maximum amplitude value; A min is the minimum amplitude value.
  • the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the processor executes the computer program, the above-mentioned steps are implemented. Synchronous detection method of river channels and faults.
  • the present invention also provides a machine-readable storage medium. Instructions are stored on the machine-readable storage medium. The instructions are used to cause the machine to execute the above-mentioned synchronized detection method of river channels and faults.
  • This technical solution performs data processing on the 3D post-stack seismic data by obtaining the minimum amplitude value and maximum amplitude value of the 3D post-stack seismic data, and obtains the processed 3D post-stack seismic data, and uses the processed 3D post-stack seismic data to Input the pre-trained detection model of river channels and faults to obtain the detection results of river channels and faults.
  • This technical solution can be adapted to the detection of different lithological river channel positions formed in different sedimentary environments. In addition, it can eliminate the problems of accurate detection of river channels caused by faults. Impact, the detection of complex rivers is realized, and the detection process is short, efficient, accurate, and has strong generalization ability.
  • Figure 1 is a flow chart of the method for synchronous detection of river channels and faults provided by the present invention
  • Figure 2 is a flow chart of data processing in the method for synchronous detection of river channels and faults provided by the present invention
  • Figure 3 is a schematic structural diagram of the device for synchronous detection of river channels and faults provided by the present invention.
  • Figure 4 is a schematic structural diagram of the data processing module in the device for synchronous detection of river channels and faults provided by the present invention
  • Figure 5 is a schematic diagram of the first synthetic three-dimensional post-stack seismic data provided by the present invention.
  • Figure 6 is a schematic diagram of three-dimensional river channel data provided by the present invention based on the synthesized three-dimensional post-stack seismic data in Figure 5;
  • Figure 7 is a schematic diagram of three-dimensional fault data obtained based on the synthesized three-dimensional post-stack seismic data in Figure 5 provided by the present invention.
  • Figure 8 is a schematic diagram of the second type of synthetic three-dimensional post-stack seismic data provided by the present invention.
  • Figure 9 is a schematic diagram of three-dimensional river channel data obtained based on the synthesized three-dimensional post-stack seismic data in Figure 8 provided by the present invention.
  • Figure 10 is a schematic diagram of three-dimensional fault data obtained based on the synthesized three-dimensional post-stack seismic data in Figure 8 provided by the present invention.
  • Figure 11 is a schematic diagram of three-dimensional post-stack seismic data of a certain work area provided by the present invention.
  • Figure 12 is a schematic diagram of the overlay display based on the three-dimensional post-stack seismic data and corresponding river channel detection results provided by the present invention in Figure 11;
  • Figure 13 is a schematic diagram of a superposition display based on the three-dimensional post-stack seismic data and corresponding fault detection results in Figure 11 provided by the present invention.
  • Figure 1 is a flow chart of the method for synchronous detection of river channels and faults provided by the present invention
  • Figure 3 is a schematic structural diagram of the device for synchronous detection of river channels and faults provided by the present invention
  • Figure 4 is a diagram of data processing in the device for synchronous detection of river channels and faults provided by the present invention.
  • FIG. 5 is a schematic diagram of the first synthesized three-dimensional post-stack seismic data provided by the present invention
  • Figure 6 is a schematic diagram of the three-dimensional river channel data obtained based on the synthesized three-dimensional post-stack seismic data in Figure 5 provided by the present invention
  • Figure 7 is a schematic diagram of the three-dimensional fault data obtained based on the synthesized three-dimensional post-stack seismic data in Figure 5 provided by the present invention
  • Figure 8 is a schematic diagram of the second type of synthesized three-dimensional post-stack seismic data provided by the present invention
  • Figure 9 is a schematic diagram of the second synthesized three-dimensional post-stack seismic data provided by the present invention based on Figure 5
  • Figure 10 is a schematic diagram of the three-dimensional fault data obtained by synthesizing the three-dimensional post-stack seismic data in Figure 8 provided by the present invention
  • an embodiment of the present invention provides a method for synchronous detection of river channels and faults, including:
  • Step 1 Obtain three-dimensional post-stack seismic data
  • Step 2 Process the three-dimensional post-stack seismic data based on the minimum amplitude value and the maximum amplitude value in the three-dimensional post-stack seismic data;
  • step two the specific steps of step two include:
  • Step 201 determining the minimum amplitude and the maximum amplitude of the three-dimensional post-stack seismic data
  • Step 202 Determine a normalized amplitude value based on the minimum amplitude value and the maximum amplitude value
  • Step 203 Transform the amplitude value of each data in the three-dimensional post-stack seismic data based on the minimum amplitude value to obtain first transformed data, where the first transformed data is a three-dimensional post-stack seismic with a minimum amplitude value of 0. data;
  • Step 204 Obtain processed three-dimensional post-stack seismic data based on the normalized amplitude value and the first transformation data, and the amplitude range of the processed three-dimensional post-stack seismic data is between 0 and 1.
  • determining a normalized amplitude value includes:
  • ⁇ A is the normalized amplitude value
  • a max is the maximum amplitude value
  • a min is the minimum amplitude value.
  • the amplitude value of each data in the three-dimensional post-stack seismic data is transformed to obtain first transformed data, including:
  • A′ i A i -A min ;
  • A′ i is the first transformed data; A i is the maximum amplitude value; A min is the minimum amplitude value.
  • processed three-dimensional post-stack seismic data including:
  • the processed three-dimensional post-stack seismic data is calculated using the following calculation formula:
  • A′′ i is the processed three-dimensional post-stack seismic data
  • A′ i is the first transformed data
  • ⁇ A is the normalized amplitude value
  • step two can also be described as:
  • the minimum amplitude value and the maximum amplitude value of the obtained three-dimensional post-stack seismic data are compared with the amplitude minimum value of the three-dimensional post-stack seismic data to obtain three-dimensional post-stack seismic data with a minimum amplitude value of 0 (i.e. first transformed data); make a difference between the maximum amplitude value and the minimum amplitude value to obtain a normalized amplitude value; compare the three-dimensional post-stack seismic data with the minimum amplitude value of 0 (i.e., the first transformed data) and The normalized amplitude value is used to obtain the processed three-dimensional post-stack seismic data, and the amplitude range of the data is between 0 and 1. Specifically, the following calculation formula is used to express:
  • A′′ i is the processed 3D post-stack seismic data; Ai is the maximum amplitude; Amax is the maximum amplitude; Amin is the minimum amplitude.
  • Step 3 Use the processed three-dimensional post-stack seismic data as the input of the river channel and fault detection model to obtain the river channel and fault detection results.
  • the data size is 700*1600*400
  • the spatial sampling interval is 12.5m
  • the time sampling interval is 2ms
  • the vertical coordinate is the longitudinal measurement line number
  • the horizontal coordinate is the contact measurement Line number
  • vertical coordinate is time.
  • the method also includes:
  • the training sample set is used as the input of the deep learning neural network, and the river channel and fault detection model is trained.
  • the training sample set for training the river channel and fault detection model is also processed (normalized), so that each sample is composed of a numerical range between 0 and 1. It consists of synthetic 3D post-stack seismic data and corresponding accurately labeled 3D channel data and accurately labeled 3D fault data. Therefore, in order to ensure the consistency of the input of the pre-trained river channel and fault detection models, the input three-dimensional post-stack seismic data was also normalized.
  • the river channel and fault detection model has two task outputs, that is, the processed three-dimensional post-stack seismic data is input into the pre-trained river channel and fault detection model for prediction, and the river channel and fault detection model can be used for prediction.
  • One branch of the model outputs the river channel detection result
  • another branch of the network model outputs the fault detection result.
  • the river channel and fault detection model in this invention sequentially builds an encoder part, a decoder part and an output part based on the principle of deep learning:
  • the encoder part consists of five stages and is used to extract features at different scales from the input 3D post-stack seismic data.
  • Each stage contains one or two convolutional layers and learns a residual function that adds the input of each stage to the output of the last convolutional layer of that stage.
  • the introduction of the residual learning mechanism helps overcome performance degradation and improve accuracy.
  • the convolution kernel size of the convolution layer is 3*3*3, and pooling is implemented through a convolution operation with a convolution kernel size of 2*2*2 and a stride of 2. It can be noticed that the number of features at each stage is doubled and the resolution is halved.
  • batch normalization Batch Nom1alization
  • ReLU Rectified Linear Unit
  • the decoder part consists of four stages for systematically aggregating multi-scale information. Similar to the encoder part, each stage also contains one or two convolutional layers, also using a residual learning mechanism. The size of the convolution kernel of the convolutional layer at each stage is also 3*3*3, and upsampling is achieved through a deconvolution operation with a convolution kernel size of 2*2*2 and a stride of 2. Through layer-hopping connections, the features extracted from the encoder part are connected to the corresponding stages of the decoder part for spatial resolution compensation.
  • the output part consists of two branches sharing the same decoding features.
  • Each branch consists of two convolution layers with a convolution kernel size of 3*3*3 and a convolution layer with a convolution kernel size of l*l*l.
  • batch normalization Batch Nom1alization
  • sigmoid activation function The first branch is used for river detection tasks, and the second branch is used for fault detection tasks.
  • the normalized synthetic 3D post-stack seismic data is input into the channel and fault detection model in batches, and the batch size is set to 4.
  • the adaptive moment estimation optimization algorithm is used to optimize the network, the learning rate is set to 0.0001, the total training period is set to 100, and all samples in the training sample set are traversed in one period.
  • the longitudinal coordinate in Figures 5, 6 and 7 is the longitudinal survey line number
  • the horizontal coordinate is the contact survey line number
  • the vertical coordinate is time.
  • the channel position and The fault locations have been accurately marked; among them, the river channel location in Figure 6 is marked as 1, and other locations are marked as 0; among them, the fault location in Figure 7 is marked as 1, and other locations are marked as 0.
  • the vertical coordinate in Figures 8, 9 and 10 is the longitudinal survey line number
  • the horizontal coordinate is the contact survey line number
  • the vertical coordinate is time.
  • the fold and fault structure information in the second synthetic three-dimensional post-stack seismic data generated by the present invention is rich and diverse, and can highly approximate the actual data, and the positions of river channels and faults are obtained.
  • Accurate labeling where the river channel position in Figure 9 is labeled as 1 and other positions are labeled as 0; among them, the fault position in Figure 10 is labeled as 1 and other positions are labeled as 0.
  • an embodiment of the present invention also provides a device for synchronizing detection of river channels and faults, including:
  • Data acquisition module 10 used to acquire three-dimensional post-stack seismic data
  • the data processing module 20 is used to process the three-dimensional post-stack seismic data based on the minimum amplitude value and the maximum amplitude value in the three-dimensional post-stack seismic data;
  • the result output module 30 is used to use the processed three-dimensional post-stack seismic data as the input of the river channel and fault detection model to obtain the river channel and fault detection results.
  • the data processing module 20 specifically includes:
  • the data amplitude determination module 21 is used to determine the minimum amplitude value and the maximum amplitude value of the three-dimensional post-stack seismic data;
  • the normalized amplitude value determination module 22 is used to determine the normalized amplitude value based on the minimum amplitude value and the maximum amplitude value;
  • the amplitude value transformation module 23 is used to transform the amplitude value of each data in the three-dimensional post-stack seismic data based on the minimum amplitude value to obtain the first transformed data;
  • the data output module 24 is used to obtain processed three-dimensional post-stack seismic data based on the normalized amplitude value and the first transformed data.
  • ⁇ A is the normalized amplitude value
  • a max is the maximum amplitude value
  • a min is the minimum amplitude value.
  • the first transformation data is obtained using the following calculation formula:
  • A′ i A i -A min ;
  • A′ i is the first transformed data; A i is the maximum amplitude value; A min is the minimum amplitude value.
  • An embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the processor executes the computer program, the above-mentioned river channel and Fault synchronization detection method.
  • Embodiments of the present invention provide a machine-readable storage medium. Instructions are stored on the machine-readable storage medium. The instructions are used to cause the machine to execute the above-mentioned synchronized detection method of river channels and faults.
  • river channels of different shapes are added to the model in the form of relative impedance, the river channel and fault detection model can be adapted to different lithological channels formed in different depositional environments. Location detection.
  • the river channel and fault detection model can adapt to the detection of complex river channel locations such as closely spaced or intertwined ones.
  • the river channel and fault detection model can simultaneously detect complex structural backgrounds.
  • the underlying river channels and faults are effectively eliminated, thereby effectively eliminating the impact of faults on accurate detection of river channels;
  • the program is stored in a storage medium and includes several instructions to cause the microcontroller, chip or processor to (processor) executes all or part of the steps of the methods described in various embodiments of this application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code. .
  • any combination of different implementation modes of the embodiments of the present invention can also be performed. As long as they do not violate the ideas of the embodiments of the present invention, they should also be regarded as the content disclosed in the embodiments of the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Acoustics & Sound (AREA)
  • Environmental & Geological Engineering (AREA)
  • Geology (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics (AREA)
  • Geophysics And Detection Of Objects (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

一种河道与断层同步检测方法，属于地震资料解释技术领域，包括：获取三维叠后地震数据；基于三维叠后地震数据中的振幅最小值和振幅最大值，对三维叠后地震数据进行处理；将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果。还涉及一种检测装置、电子设备和机器可读存储介质。

Description

河道与断层同步检测方法及装置 技术领域
本发明涉及地震资料解释领域，具体地涉及一种河道与断层同步检测方法、一种河道与断层同步检测装置、一种电子设备及一种机器可读存储介质。
背景技术
河道是油气勘探中重要的地质目标之一，准确检测河道位置并揭示其发育过程至关重要。在盆地和远景规模分析方面，掌握河道特征有助于地貌的定量分析和沉积过程的准确认识，有助于了解沉积环境和沉积物随时间推移的主要方向；在储层规模分析方面，河道砂岩的体积及其中填充流体属性的确定至关重要；在油气田开发阶段，准确刻画河道平面宏观展布特征、沉积及其变化规律，可为开发目标区块优选、油气开发方案编制、井位和井轨迹设计等提供重要依据。
现有技术中，三维叠后地震数据被广泛用于河道解释。但是，采用手动解释河道，耗费时间长、解释结果不准确且受人为主观性影响；采用常规的卷积神经网络进行河道检测，存在高度依赖于标签和标签生成方式，导致在其他工区中的泛化能力很差，以及只考虑褶皱的影响，忽略断层的影响，导致检测精度大大降低。
发明内容
本发明实施例的目的是提供一种河道与断层同步检测方法及装置，用以解决上述采用手动解释河道，耗费时间长、解释结果不准确且受人为主观性影响；采用常规的卷积神经网络进行河道检测，存在高度依赖于标签和标签生成方式，导致在其他工区中的泛化能力很差，以及只考虑褶皱的影响，忽略断层的影响，导致检测精度低的问题。
为了实现上述目的，本发明实施例提供一种河道与断层同步检测方法，包括：
获取三维叠后地震数据；
基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理；
将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果；
基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理，包括：
确定所述三维叠后地震数据的振幅最小值和振幅最大值；
基于所述振幅最小值和所述振幅最大值，确定归一化振幅值；
基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据；
基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据。
可选的，基于所述振幅最小值和所述振幅最大值，确定归一化振幅值，包括：
利用以下计算公式计算得到归一化振幅值：
ΔA＝A max-A min；
其中，ΔA为归一化振幅值；A max为振幅最大值；A min为振幅最小值。
可选的，基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据，包括：
利用以下计算公式得到第一变换数据：
A′ i＝A i-A min；
其中，A′ i为第一变换数据；A i为振幅最大值；A min为振幅最小值。
可选的，基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据，包括：
利用以下计算公式计算得到处理后的三维叠后地震数据：
Figure PCTCN2022138860-appb-000001
其中，A″ i为处理后的三维叠后地震数据；A′ i为第一变换数据；ΔA为归一化振幅值。
可选的，所述方法还包括：
基于地质和地球物理理论合成多个样本数据，并构建训练样本集；
将所述训练样本集作为深度学习神经网络的输入，训练得到所述河道与断层检测模型。
本发明还提供一种河道与断层同步检测装置，包括：
数据获取模块，用于获取三维叠后地震数据；
数据处理模块，用于基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理；
结果输出模块，用于将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果；
所述数据处理模块具体包括：
数据振幅确定模块，用于确定所述三维叠后地震数据的振幅最小值和振幅最大值；
归一化振幅值确定模块，用于基于所述振幅最小值和所述振幅最大值，确定归一化振幅值；
振幅值变换模块，用于基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据；
数据输出模块，用于基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据。
可选的，利用以下计算公式计算得到归一化振幅值：
ΔA＝A max-A min；
其中，ΔA为归一化振幅值；A max为振幅最大值；A min为振幅最小值。
可选的，利用以下计算公式得到第一变换数据：
A′ i＝A i-A min；
其中，A′ i为第一变换数据；A i为振幅最大值；A min为振幅最小值。
另一方面，本发明提供一种电子设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述的河道与断层同步检测方法。
另一方面，本发明还提供一种机器可读存储介质，该机器可读存储介质上存储有指令，该指令用于使得机器执行上述的河道与断层同步检测方法。
本技术方案通过获取的三维叠后地震数据的振幅最小值和振幅最大值，对三维叠后地震数据进行数据处理，得到处理后的三维叠后地震数据，并将处理后的三维叠后地震数据输入预先训练完成的检测河道与断层检测模型中，得到河道与断层检测结果，本技术方案能够适应在不同沉积环境中形成的不同岩性河道位置的检测，另外，能够消除断层对河道准确检测的影响，实现复杂河道的检测，且检测过程耗时短、效率高、检测结果精确、泛化能力强。
本发明实施例的其它特征和优点将在随后的具体实施方式部分予以详细说明。
附图说明
附图是用来提供对本发明实施例的进一步理解，并且构成说明书的一部分，与下面的具体实施方式一起用于解释本发明实施例，但并不构成对本发明实施例的限制。在附图中：
图1是本发明提供的河道与断层同步检测方法的流程图；
图2是本发明提供的河道与断层同步检测方法中数据处理的流程图；
图3是本发明提供的河道与断层同步检测装置的结构示意图；
图4是本发明提供的河道与断层同步检测装置中数据处理模块的结构示意图；
图5是本发明提供的第一种合成三维叠后地震数据的示意图；
图6是本发明提供的基于图5中合成三维叠后地震数据得到的三维河道数据示意图；
图7是本发明提供的基于图5中合成三维叠后地震数据得到的三维断层数据示意图；
图8是本发明提供的第二种合成三维叠后地震数据的示意图；
图9是本发明提供的基于图8中合成三维叠后地震数据得到的三维河道数据示意图；
图10是本发明提供的基于图8中合成三维叠后地震数据得到的三维断层数据示意图；
图11是本发明提供的某一工区三维叠后地震数据的示意图；
图12是本发明提供的基于图11中三维叠后地震数据与相应的河道检测结果的叠合展示示意图；
图13是本发明提供的基于图11中三维叠后地震数据与相应的断层检测结果的叠合展示示意图。
附图标记说明
10-数据获取模块；                 20-数据处理模块；
30-结果输出模块；                 21-数据振幅确定模块；
22-归一化振幅值确定模块；         23-振幅值变换模块；
24-数据输出模块。
具体实施方式
以下结合附图对本发明实施例的具体实施方式进行详细说明。应当理解的是，此处所描述的具体实施方式仅用于说明和解释本发明实施例，并不用于限制本发明实施例。
术语“第一”、“第二”、“第三”等仅用于区分描述，而不能理解为指示或暗示相对重要性。
此外，“大致”、“基本”等用语旨在说明相关内容并不是要求绝对的精确，而是可以有一定的偏差。例如：“大致相等”并不仅仅表示绝对的相等，由于实际生产、操作过程中，难以做到绝对的“相等”，一般都存在一定的偏差。因此，除了绝对相等之外，“大致等于”还包括上述的存在一定偏差的情况。以此为例，其他情况下，除非有特别说明，“大致”、“基本”等用语均为与上述类似的含义。
图1是本发明提供的河道与断层同步检测方法的流程图；图3是本发明提供的河道与断层同步检测装置的结构示意图；图4是本发明提供的河道与断层同步检测装置中数据处理模块的结构示意图；图5是本发明提供的第一种合成三维叠后地震数据的示意图；图6是本发明提供的基于图5中合成三维叠后地震数据得到的三维河道数据示意图；图7是本发明提供的基于图5中合成三维叠后地震数据得到的三维断层数据示意图；图8是本发明提供的第二种合成三维叠后地震数据的示意图；图9是本发明提供的基于图8中合成三维叠后地震数据得到的三维河道数据示意图；图10是本发明提供的基于图8中合成三维叠后地震数据得到的三维断层数据示意图；图11是本发明提供的某一工区三维叠后地震数据的示意图；图12是本发明提供的基于图11中三维叠后地震数据与相应的河道检测结果的叠合展示示意图；图13是本发明提供的基于图11中三维叠后地震数据与相应的断层检测结果的叠合展示示意图。
实施例1
如图1所示，本发明实施例提供一种河道与断层同步检测方法，包括：
步骤一、获取三维叠后地震数据；
具体地，基于现有地震数据得到三维叠后地震数据的相关步骤时本领域技术人员熟知的，此处不再赘述。
步骤二、基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理；
具体地，如图2所示，步骤二的具体步骤包括：
步骤201、确定所述三维叠后地震数据的振幅最小值和振幅最大值；
步骤202、基于所述振幅最小值和所述振幅最大值，确定归一化振幅值；
步骤203、基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据，其中，第一变换数据为振幅最小值为0的三维叠后地震数据；
步骤204、基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据，并且处理后的三维叠后地震数据的振幅的范围在0与1之间。
进一步地，基于所述振幅最小值和所述振幅最大值，确定归一化振幅值，包括：
利用以下计算公式计算得到归一化振幅值：
ΔA＝A max-A min；
其中，ΔA为归一化振幅值；A max为振幅最大值；A min为振幅最小值。
进一步地，基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据，包括：
利用以下计算公式得到第一变换数据：
A′ i＝A i-A min；
其中，A′ i为第一变换数据；A i为振幅最大值；A min为振幅最小值。
进一步地，基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据，包括：
利用以下计算公式计算得到处理后的三维叠后地震数据：
Figure PCTCN2022138860-appb-000002
其中，A″ i为处理后的三维叠后地震数据；A′ i为第一变换数据；ΔA为归一化振幅值。
因此，步骤二的具体步骤还可以描述为：
获取的三维叠后地震数据的振幅最小值和振幅最大值，对将所述三维叠后地震数据的振幅与所述振幅最小值作差，得到振幅最小值为0的三维叠后地震数据(即第一变换数据)；将所述振幅最大值与所述振幅最小值作差，得到归一化振幅值；将所述振幅最小值为0的三维叠后地震数据(即第一变换数据)与所述归一化振幅值做商，得到处理后的三维叠后地震数据，且数据的振幅的范围在0与1之间，具体采用以下计算公式进行表示：
Figure PCTCN2022138860-appb-000003
其中，A″ i为处理后的三维叠后地震数据；A i为振幅最大值；A max为振幅最大值；A min为振幅最小值。
步骤三、将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果。
具体地，如图11、图12和图13所示，数据大小为700*1600*400，空间采样间隔为12.5m，时间采样间隔为2ms，纵向坐标为纵测线号，横向坐标为联络测线号，垂向坐标为时间。采用上述的方法进行检测，得到检测结果，如图12和图13所示，具有复杂构造背景的三维叠后地震数据中所有的河道位置(深色位置)和断层位置(深色位置)都得到了准确的检测。
进一步地，所述方法还包括：
基于地质和地球物理理论合成多个样本数据，并构建训练样本集；
将所述训练样本集作为深度学习神经网络的输入，训练得到所述河道与断层检测模型。
具体地，为了保证河道与断层检测模型的泛化能力，同样对训练河道与断层检测模型的训练样本集进行处理(归一化处理)，使得每个样本由数值范围在0与1之间的合成三维叠后地震数据及相应的准确标注三维河道数据和准确标注三维断层数据组成。因此为了保证预先训练好的河道与断层检测模型输入的一致性，对输入的三维叠后地震数据也进行了归一化处理。考虑到人工标注三维地震数据耗时、不准确且主观性强，为了克服这个问题，基于地质和地球物理理论，自动生成大小为96*96*96的9600对具有多样和复杂构造模式的合成三维叠后地震数据及相应的准确标注三维河道数据和准确标注三维断层数据，构成训练样本集。
本发明实施例中，河道与断层检测模型有两个任务输出，即将处理后的三维叠后地震数据输入到所述预先训练好的河道与断层检测模型中进行预测，可以由该河道与断层检测模型的其中一个分支输出河道检测结果，由该网络模型的另外一个分支输出断层检测结果。需要说明的是，本发明实施例中，对河道与断层检测模型的卷积层数、卷积核的大小、步长、激活函数等要素均根据本发明需求做了精细的可行性和必要性设计。本发明中的河道与断层检测模型根据深度学习原理依次搭建了编码器部分、解码器部分和输出部分：
编码器部分由五个阶段组成，用于从输入的三维叠后地震数据中提取不同尺度的特征。每个阶段包含一个或两个卷积层，并且学习一个残差函数，即将每个阶段的输入与该阶段最后一个卷积层的输出相加。残差学习机制的引入，有助于克服性能下降问题并提高准确性。每个阶段中，卷积层的卷积核大小为3*3*3，池化是通过卷积核大小为2*2*2和步长为2的卷积操作实现的。可以注意到，每个阶段的特征数量翻倍而分辨率减半。在每个卷积层后面，应用了批标准化 (Batch Nom1alization)和ReLU(Rectified Linear Unit)激活函数，用以提高网络的稳定性和非线性逼近能力。
解码器部分由四个阶段组成，用于系统地聚合多尺度信息。类似于编码器部分，每个阶段也包含一个或两个卷积层，同样采用残差学习机制。每个阶段卷积层卷积核的大小也是3*3*3，通过卷积核大小为2*2*2和步长为2的反卷积操作来实现上采样。通过跃层连接，将从编码器部分提取的特征与解码器部分相应的阶段进行连接，用以进行空间分辨率补偿。
输出部分由共享相同解码特征的两个分支组成，每个分支由两个卷积核大小为3*3*3的卷积层和一个卷积核大小为l*l*l的卷积层组成，然后是批标准化(Batch Nom1alization)和sigmoid激活函数。第一个分支用于河道检测任务，第二个分支用于断层检测任务。
在进行训练时，将进行归一化处理的合成三维叠后地震数据分批次输入河道与断层检测模型，批次大小设置为4。使用自适应矩估计优化算法来最优化网络，学习率设置为0.0001，总训练周期设置为100，一个周期遍历训练样本集中的所有样本。
更具体地，图5、图6和图7中的纵向坐标为纵测线号，横向坐标为联络测线号，垂向坐标为时间。如图5、图6和图7所示，本发明生成的第一种该样本中经过复杂构造变形(包括一系列交替的褶皱变形和断裂变形)的合成三维叠后地震数据中，河道位置和断层位置都得到了准确标注；其中，图6中河道位置标注为1，其他位置标注为0；其中，图7中断层位置标注为1，其他位置标注为0。
更具体地，图8、图9和图10的纵向坐标为纵测线号，横向坐标为联络测线号，垂向坐标为时间。如图8、图9和图10所示，本发明生成的第二种合成三维叠后地震数据中的褶皱和断层构造信息丰富多样，能够高度逼近实际资料，且河道位置和断层位置都得到了准确标注，其中，图9中河道位置标注为1，其他位置标注为0；其中，图10中断层位置标注为1，其他位置标注为0。
以上述的第一种合成三维叠后地震数据和第二种合成三维叠后地震数据为例，为河道与断层检测模型的训练提供了准确标注样本。训练样本集中足够数目的标签数据的存在，为本发明基于深度学习的河道和断层同步检测方法的快速准确实现奠定了坚实的大数据基础。
实施例2
如图3所示，本发明实施例还提供一种河道与断层同步检测装置，包括：
数据获取模块10，用于获取三维叠后地震数据；
数据处理模块20，用于基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理；
结果输出模块30，用于将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果.
进一步地，如图4所示，所述数据处理模块20具体包括：
数据振幅确定模块21，用于确定所述三维叠后地震数据的振幅最小值和振幅最大值；
归一化振幅值确定模块22，用于基于所述振幅最小值和所述振幅最大值，确定归一化振幅值；
振幅值变换模块23，用于基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据；
数据输出模块24，用于基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据。
进一步地，利用以下计算公式计算得到归一化振幅值：
ΔA＝A max-A min；
其中，ΔA为归一化振幅值；A max为振幅最大值；A min为振幅最小值。
进一步地，利用以下计算公式得到第一变换数据：
A′ i＝A i-A min；
其中，A′ i为第一变换数据；A i为振幅最大值；A min为振幅最小值。
实施例3
本发明实施例提供一种电子设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述的河道与断层同步检测方法。
实施例4
本发明实施例提供一种机器可读存储介质，该机器可读存储介质上存储有指令，该指令用于使得机器执行上述的河道与断层同步检测方法。
通过上述的技术方案，针对河道和断层的检测无论是检测精度还是检测效率都得到显著提升：
1、由于在构建双任务准确标注数据集的过程中，不同形态的河道是以相对阻抗的形式添加到模型中的，因此河道与断层检测模型可以适应在不同沉积环境中形成的不同岩性河道位置的检测。
2、由于在构建双任务准确标注数据集的过程中，模型中模拟的不同形态河道的数量是随机设置的，并且河道的方向、宽度、长度、厚度和相对阻抗值随着 空间位置而变化，因此河道与断层检测模型可以适应诸如紧密相间或者相互交织的复杂河道位置的检测。
3、由于在构建双任务准确标注数据集的过程中，进行了多样的褶皱和断裂变形，再加上数据增广和任务之间的知识共享，河道与断层检测模型能够同步检测出复杂构造背景下的河道和断层，进而有效地消除断层对河道准确检测的影响；
4、由于本发明基于深度学习算法，且在GPU上进行运算，因此具有较高的计算效率，对于4.56GB的三维地震数据体，使用NVIDIA V100 GPU仅需4.4分钟即可完成整个三维地震数据体中所有河道和断层的检测，而人工解释需要数天且精度不能得到保证。
以上结合附图详细描述了本发明实施例的可选实施方式，但是，本发明实施例并不限于上述实施方式中的具体细节，在本发明实施例的技术构思范围内，可以对本发明实施例的技术方案进行多种简单变型，这些简单变型均属于本发明实施例的保护范围。
另外需要说明的是，在上述具体实施方式中所描述的各个具体技术特征，在不矛盾的情况下，可以通过任何合适的方式进行组合。为了避免不必要的重复，本发明实施例对各种可能的组合方式不再另行说明。
本领域技术人员可以理解实现上述实施例方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，该程序存储在一个存储介质中，包括若干指令用以使得单片机、芯片或处理器(processor)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
此外，本发明实施例的各种不同的实施方式之间也可以进行任意组合，只要其不违背本发明实施例的思想，其同样应当视为本发明实施例所公开的内容。

Claims (10)

  1. 一种河道与断层同步检测方法，其特征在于，包括：
    获取三维叠后地震数据；
    基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理；
    将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果；
    基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理，包括：
    确定所述三维叠后地震数据的振幅最小值和振幅最大值；
    基于所述振幅最小值和所述振幅最大值，确定归一化振幅值；
    基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据；
    基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据。
  2. 根据权利要求1所述的方法，其特征在于，基于所述振幅最小值和所述振幅最大值，确定归一化振幅值，包括：
    利用以下计算公式计算得到归一化振幅值：
    ΔA＝A max-A min；
    其中，ΔA为归一化振幅值；A max为振幅最大值；A min为振幅最小值。
  3. 根据权利要求1所述的方法，其特征在于，基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据，包括：
    利用以下计算公式得到第一变换数据：
    A′ i＝A i-A min；
    其中，A′ i为第一变换数据；A i为振幅最大值；A min为振幅最小值。
  4. 根据权利要求1所述的方法，其特征在于，基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据，包括：
    利用以下计算公式计算得到处理后的三维叠后地震数据：
    Figure PCTCN2022138860-appb-100001
    其中，A″ i为处理后的三维叠后地震数据；A′ i为第一变换数据；ΔA为归一化振幅值。
  5. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    基于地质和地球物理理论合成多个样本数据，并构建训练样本集；
    将所述训练样本集作为深度学习神经网络的输入，训练得到所述河道与断层检测模型。
  6. 一种河道与断层同步检测装置，其特征在于，包括：
    数据获取模块，用于获取三维叠后地震数据；
    数据处理模块，用于基于三维叠后地震数据中的振幅最小值和振幅最大值，对所述三维叠后地震数据进行处理；
    结果输出模块，用于将处理后的三维叠后地震数据作为河道与断层检测模型的输入，得到河道与断层检测结果；
    所述数据处理模块具体包括：
    数据振幅确定模块，用于确定所述三维叠后地震数据的振幅最小值和振幅最大值；
    归一化振幅值确定模块，用于基于所述振幅最小值和所述振幅最大值，确定归一化振幅值；
    振幅值变换模块，用于基于所述振幅最小值对所述三维叠后地震数据中每一数据的振幅值进行变换，得到第一变换数据；
    数据输出模块，用于基于所述归一化振幅值和所述第一变换数据，得到处理后的三维叠后地震数据。
  7. 根据权利要求6所述的装置，其特征在于，利用以下计算公式计算得到归一化振幅值：
    ΔA＝A max-A min；
    其中，ΔA为归一化振幅值；A max为振幅最大值；A min为振幅最小值。
  8. 根据权利要求6所述的装置，其特征在于，利用以下计算公式得到第一变换数据：
    A′ i＝A i-A min；
    其中，A′ i为第一变换数据；A i为振幅最大值；A min为振幅最小值。
  9. 一种电子设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求1-5中任一项所述的河道与断层同步检测方法。
  10. 一种机器可读存储介质，该机器可读存储介质上存储有指令，该指令用于使得机器执行权利要求1-5中任一项所述的河道与断层同步检测方法。
PCT/CN2022/138860 2022-09-22 2022-12-14 河道与断层同步检测方法及装置 Ceased WO2024060427A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211169806.4 2022-09-22
CN202211169806.4A CN117784225B (zh) 2022-09-22 2022-09-22 河道与断层同步检测方法及装置

Publications (1)

Publication Number Publication Date
WO2024060427A1 true WO2024060427A1 (zh) 2024-03-28

Family

ID=90382434

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/138860 Ceased WO2024060427A1 (zh) 2022-09-22 2022-12-14 河道与断层同步检测方法及装置

Country Status (2)

Country Link
CN (1) CN117784225B (zh)
WO (1) WO2024060427A1 (zh)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140188392A1 (en) * 2012-12-27 2014-07-03 Schlumberger Technology Corporation Normalization seismic attribute
CN109655894A (zh) * 2017-10-12 2019-04-19 中国石油化工股份有限公司 碳酸盐岩古河道地震反演低频模型构建方法及系统
CN113296152A (zh) * 2020-02-21 2021-08-24 中国石油天然气集团有限公司 断层检测方法及装置
CN113703044A (zh) * 2020-05-20 2021-11-26 中国石油化工股份有限公司 古河道宽度的校正方法、装置、电子设备及存储介质
CN113900141A (zh) * 2020-07-06 2022-01-07 中国石油天然气股份有限公司 油气分布预测方法及装置
US20220027126A1 (en) * 2020-07-22 2022-01-27 Shanghai Cambricon Information Technology Co., Ltd. Data normalization processing method, storage medium and computer equipment
CN114462113A (zh) * 2020-10-21 2022-05-10 中国石油化工股份有限公司 基于振幅比的隐蔽河道刻画方法、装置、电子设备及介质
CN114755721A (zh) * 2021-01-12 2022-07-15 中国石油化工股份有限公司 基于Unet的河道砂体顶面预测方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103792576A (zh) * 2014-01-28 2014-05-14 中国石油天然气股份有限公司 一种基于梯度结构张量的储层非均质检测方法及设备
CN103869358A (zh) * 2014-02-18 2014-06-18 中国石油天然气股份有限公司 一种基于直方图均衡的断层识别方法及设备
CN107632320A (zh) * 2017-08-21 2018-01-26 西安交通大学 基于同步提取s变换的地震资料时频分析方法
CN113805235B (zh) * 2020-06-11 2023-12-26 中国石油天然气股份有限公司 基于卷积神经网络的三维地震相识别方法及装置
CN113902769B (zh) * 2021-08-18 2024-07-05 南方海洋科学与工程广东省实验室(广州) 一种基于深度学习语义分割的地震断层识别方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140188392A1 (en) * 2012-12-27 2014-07-03 Schlumberger Technology Corporation Normalization seismic attribute
CN109655894A (zh) * 2017-10-12 2019-04-19 中国石油化工股份有限公司 碳酸盐岩古河道地震反演低频模型构建方法及系统
CN113296152A (zh) * 2020-02-21 2021-08-24 中国石油天然气集团有限公司 断层检测方法及装置
CN113703044A (zh) * 2020-05-20 2021-11-26 中国石油化工股份有限公司 古河道宽度的校正方法、装置、电子设备及存储介质
CN113900141A (zh) * 2020-07-06 2022-01-07 中国石油天然气股份有限公司 油气分布预测方法及装置
US20220027126A1 (en) * 2020-07-22 2022-01-27 Shanghai Cambricon Information Technology Co., Ltd. Data normalization processing method, storage medium and computer equipment
CN114462113A (zh) * 2020-10-21 2022-05-10 中国石油化工股份有限公司 基于振幅比的隐蔽河道刻画方法、装置、电子设备及介质
CN114755721A (zh) * 2021-01-12 2022-07-15 中国石油化工股份有限公司 基于Unet的河道砂体顶面预测方法

Also Published As

Publication number Publication date
CN117784225B (zh) 2025-04-08
CN117784225A (zh) 2024-03-29

Similar Documents

Publication Publication Date Title
An et al. Deep convolutional neural network for automatic fault recognition from 3D seismic datasets
CN112731522B (zh) 地震地层智能识别方法、装置、设备及存储介质
US9804282B2 (en) Computer-assisted fault interpretation of seismic data
CN114549589B (zh) 一种基于轻量化神经网络的旋转体振动位移测量方法及系统
CN106405651B (zh) 一种基于测井匹配的全波形反演初始速度模型构建方法
CN110361778A (zh) 一种基于生成对抗网络的地震数据重建方法
CN107807387A (zh) 基于神经网络的地震初至波走时获取方法
CN109272029B (zh) 井控稀疏表征大规模谱聚类地震相划分方法
CN105954802A (zh) 一种岩性数据体的转换方法及装置
CN109444879A (zh) 一种DInSAR近断层同震形变场提取方法
Wang et al. Seismic stratum segmentation using an encoder–decoder convolutional neural network
CN113065467A (zh) 一种基于深度学习的卫星图像低相干区域识别方法及装置
Alfarhan et al. Concurrent detection of salt domes and faults using resnet with u-net
Chen et al. CO 2 Seg: Automatic CO 2 segmentation from 4-D seismic image using convolutional vision transformer
US11880008B2 (en) Velocity model construction
Jiang et al. Deep learning-based optimization framework for full-waveform inversion in tunnels
Yang et al. Unsupervised learning inversion of seismic velocity models based on a multi‐scale strategy
CN112464547B (zh) 地震倾角和方位角的确定方法及装置
WO2024060427A1 (zh) 河道与断层同步检测方法及装置
CN116413783A (zh) 地震数据解释方法及装置
Zhang et al. Automatic salt dome detection using U-net
CN113111928A (zh) 一种基于地学数据库的半监督学习矿产资源定量预测方法
CN119251613A (zh) 一种基于稀疏标注的半监督3d断层识别方法
Devarakota et al. Deep learning in salt interpretation from R&D to deployment: Challenges and lessons learned
Zhang et al. 3D UXSE-Net for Seismic Channel Detection Based on Satellite Image Enhanced Synthetic Datasets

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: 22959426

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: 22959426

Country of ref document: EP

Kind code of ref document: A1