Scholarlink

<p>This paper presents a method for lithofacies classification and prediction using conventional logging data and a limited lithofacies dataset. Various deep learning regression models are employed to predict and classify lithofacies while assessing their predictive capabilities. Among them, the convolutional neural network/long short-term memory (CNN-LSTM) prediction model demonstrates good predictive performance and stronger generalizability. The introduction of this model allows for rapid and accurate identification of subsurface lithofacies and the establishment of a three-dimensional geological model based on the predicted lithofacies, which has significant implications for the exploration, development, and production of underground resources.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Lithofacies and Three-Dimensional Geological Modeling Based on CNN-LSTM Spatiotemporal Networks

  • 期刊论文

  • 发表日期：

    2025年6月6日

查看全文
  • Qianghao Liu,
  • Rong Liu,
  • Tianxin He,
  • Haoran Zhang

摘要

This paper presents a method for lithofacies classification and prediction using conventional logging data and a limited lithofacies dataset. Various deep learning regression models are employed to predict and classify lithofacies while assessing their predictive capabilities. Among them, the convolutional neural network/long short-term memory (CNN-LSTM) prediction model demonstrates good predictive performance and stronger generalizability. The introduction of this model allows for rapid and accurate identification of subsurface lithofacies and the establishment of a three-dimensional geological model based on the predicted lithofacies, which has significant implications for the exploration, development, and production of underground resources.