.png){kind=icon}
返回
A neural recommender system for efficient adsorbent screening
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ces.2022.117801.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Adsorption is widely used for gas separation and storage. Its performance relies heavily on the selection of suitable adsorbents. In this work, a data-driven neural recommender system is developed for a preliminary adsorbent screening. Over 32,000 single-component isotherms are retrieved from the NIST isotherm database. After a series of filtering steps, 16,470 isotherms related to gaseous adsorbates are obtained and fitted with the dual-site Langmuir sorption isotherm model. Based on these data, a 90 x 1261 sparse matrix is generated. The rows represent adsorbates at specific temperature and pressure. Each column denotes a certain adsorbent. Matrix elements are the corresponding adsorption uptakes. Based on the known entries in the matrix, the neural collaborative filtering method is employed to construct a neural recommender system. The resulting recommender system is utilized to impute the missing elements in the matrix. Finally, the completed matrix is applied to identify promising adsorbents for methane storage and ethene/ethane separation.(c) 2022 Elsevier Ltd. All rights reserved.
Keyword:
Adsorbent screening
Matrix completion
Neural recommender system
Methane storage
Ethene
ethane separation
期刊
IF:
4.3
论文数:
2.3W
被引数:
5.5W
机构
引用论文
分享收藏The Relative Influence of Lesion Length and Other Stenosis Morphologies on Procedural Success of Coronary Intervention病变长度和其他狭窄形态对冠状动脉介入手术成功的相对影响.png)
Angiology
AngiologyIF0
PREAI
分享收藏Development of a General Evaluation Metric for Rapid Screening of Adsorbent Materials for Postcombustion CO2 Capture开发用于快速筛选燃烧后CO2捕获的吸附剂材料的通用评估指标
PREAI
分享收藏Understanding the diversity of the metal-organic framework ecosystem了解金属有机框架生态系统的多样性NATURE COMMUNICATIONS
OAAI
NATURE COMMUNICATIONSIF15.7
OAAI分享收藏
查看更多内容