Estimation of gas hydrate saturation based on seismic attributes and deep feedforward neural network

被引：0
|
作者：
Meng, Dajiang [1 ,2 ]
Chen, Xi [1 ,2 ]
Lu, Yunqian [1 ,2 ]
Gu, Yuan [1 ,2 ]
Wen, Pengfei [1 ,2 ]
机构：
[1] National Engineering Research Center of Gas Hydrate Exploration and Development, Guangdong, Guangzhou,511458, China
[2] Key Laboratory of Marine Mineral Resources, Ministry of Natural Resources, Guangzhou Marine Geological Survey, China Geological Survey, Guangdong, Guangzhou,511458, China
来源：
Dizhi Xuebao/Acta Geologica Sinica | 2024年 / 98卷 / 09期
关键词：
Gas hydrate saturation is an important parameter to evaluate the potential of gas hydrate resources. Conventional seismic inversion methods for reservoir prediction; however; exhibit limitations in accuracy and efficiency; mainly due to their inability to adequately address the inherent nonlinear relationship between seismic data and reservoir parameters. The rapid development of artificial intelligence (AI) has led to the exploration of novel approaches for seismic exploration. Among these; artificial neural networks (ANNs) have emerged as a promising tool. ANNs possess the capability to learn complex patterns from large datasets; enabling them to approximate nonlinear functions with high accuracy. This inherent ability to model complex relationships makes ANNs particularly suitable for inverting subsurface reservoir characteristics; including gas hydrate saturation. Therefore; this paper proposes a novel method for predicting gas hydrate saturation based on seismic attributes and deep feedforward neural networks. This approach consider the advantages and disadvantages of conventional linear formula and petrophysical model methods. The proposed method involves a three-step process. First; different types of seismic attribute bodies with a strong correlation to gas hydrate saturation are identified based on logging and seismic data. These attributes are then used to construct multi-dimensional sample label data. Second; a strategy combining seismic inversion with end-to-end (seismic data-reservoir physical data) inversion is used to test and train the parameters of the fully connected neural network. This includes optimizing the number of hidden layers; the number of neurons; and the number of iterations. Finally; the trained neural network is applied to the seismic data body to obtain predictions of hydrate saturation. Actual data application results show that the saturation predictions based on seismic attributes and deep feedforward neural networks exhibit high accuracy and low multi-solution; showing good agreement with the logging data; which proves that the method has good application value. Furthermore; the predicted spatial distribution characteristics of hydrate indicate that the hydrate accumulation pattern in the study area is a flat-lying transitional free gas to gas hydrate system. © 2024 Geological Society of China. All rights reserved;
D O I：
10.19762/j.cnki.dizhixuebao.2024133
中图分类号：
学科分类号：
摘要：
引用
收藏
页码：2723 / 2736
相关论文
共 50 条
  • [41] Prediction of sand thickness using seismic attributes based on BP-neural network
    Gao, Feng
    NATURAL RESOURCES AND SUSTAINABLE DEVELOPMENT II, PTS 1-4, 2012, 524-527 : 180 - 183
  • [42] A Convolutional Neural Network based Deep Learning Technique for Identifying Road Attributes
    Jan, Zohaib
    Verma, Brijesh
    Affum, Joseph
    Atabak, Sam
    Moir, Lachlan
    2018 INTERNATIONAL CONFERENCE ON IMAGE AND VISION COMPUTING NEW ZEALAND (IVCNZ), 2018,
  • [43] BatTS: a hybrid method for optimizing deep feedforward neural network
    Pan, Sichen
    Gupta, Tarun Kumar
    Raza, Khalid
    PEERJ COMPUTER SCIENCE, 2023, 9
  • [44] BowTie - A Deep Learning Feedforward Neural Network for Sentiment Analysis
    Vassilev, Apostol
    MACHINE LEARNING, OPTIMIZATION, AND DATA SCIENCE, 2019, 11943 : 360 - 371
  • [45] Improved weight initialization for deep and narrow feedforward neural network
    Lee, Hyunwoo
    Kim, Yunho
    Yang, Seung Yeop
    Choi, Hayoung
    NEURAL NETWORKS, 2024, 176
  • [46] Automatic NMO correction and velocity estimation by a feedforward neural network
    Calderon-Macias, C
    Sen, MK
    Stoffa, PL
    GEOPHYSICS, 1998, 63 (05) : 1696 - 1707
  • [47] Comparison of Ensemble Simple Feedforward Neural Network and Deep Learning Neural Network on Phishing Detection
    Soon, Gan Kim
    Chiang, Liew Chean
    On, Chin Kim
    Rusli, Nordaliela Mohd
    Fun, Tan Soo
    COMPUTATIONAL SCIENCE AND TECHNOLOGY (ICCST 2019), 2020, 603 : 595 - 604
  • [48] Gas chimney and hydrocarbon detection using combined BBO and artificial neural network with hybrid seismic attributes
    Ramya, J.
    Somasundareswari, D.
    Vijayalakshmi, P.
    SOFT COMPUTING, 2020, 24 (03) : 2341 - 2354
  • [49] A seismic facies classification method based on the convolutional neural network and the probabilistic framework for seismic attributes and spatial classification
    Liu, Zhege
    Cao, Junxing
    Lu, Yujia
    Chen, Shuna
    Liu, Jianli
    INTERPRETATION-A JOURNAL OF SUBSURFACE CHARACTERIZATION, 2019, 7 (03): : SE225 - SE236
  • [50] Super resolution DOA estimation based on deep neural network
    Liu, Wanli
    SCIENTIFIC REPORTS, 2020, 10 (01)
12345