A 3D convolutional neural network model with multiple outputs for simultaneously estimating the reactive transport parameters of sandstone from its CT images

作     者：Haiying Fu Shuai Wang Guicheng He Zhonghua Zhu Qing Yu Dexin Ding 

作者机构：Key Discipline Laboratory for National Defense for Biotechnology in Uranium Mining and HydrometallurgyUniversity of South ChinaHengyang421001PR China School of Resource Environment and Safety EngineeringUniversity of South ChinaHengyang421001PR China 

出 版 物：《Artificial Intelligence in Geosciences》 (地学人工智能（英文）)

年 卷 期：2024年第5卷第1期

页      面：310-319页

核心收录：

学科分类：081803[工学-地质工程] 08[工学] 0818[工学-地质资源与地质工程] 

基　　金：supported by the National Natural Science Foundation of China (12105139 and 42277264) National Key Research and Development Program of China (2021YFC2902104) Education Department of Hunan Province (21B0446) 

主　　题：Reactive transport CNN model with multiple outputs Sandstone Tortuosity Permeability 

摘      要：Porosity,tortuosity,specific surface area(SSA),and permeability are four key parameters of reactive transport modeling in sandstone,which are important for understanding solute transport and geochemical reaction pro-cesses in sandstone *** four parameters reflect the characteristics of pore structure of sandstone from different perspectives,and the traditional empirical formulas cannot make accurate predictions of them due to their complexity and *** this paper,eleven types of sandstone CT images were firstly segmented into numerous subsample images,the porosity,tortuosity,SSA,and permeability of the subsamples were calculated,and the dataset was *** 3D convolutional neural network(CNN)models were subse-quently established and trained to predict the key reactive transport parameters based on subsample CT images of *** results demonstrated that the 3D CNN model with multiple outputs exhibited excellent prediction ability for the four parameters compared to the traditional empirical *** particular,for the prediction of tortuosity and permeability,the 3D CNN model with multiple outputs even showed slightly better prediction ability than its single-output variant ***,it demonstrated good generalization per-formance on sandstone CT images not included in the training *** study showed that the 3D CNN model with multiple outputs has the advantages of simplifying operation and saving computational resources,which has the prospect of popularization and application.