机器学习替代物用于吸附到氧化物的表面复合模型
Chunhui Li1, Elijah O Adeniyi2, Piotr Zarzycki3
1Energy Geosciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA. chunhuili@lbl.gov.
Scientific reports
|March 20, 2024
概括
机器学习模型准确地预测吸附到矿物质,克服了地质核废物存储安全评估中的计算问题. 这加快了对放射性核素流动性和存储库安全性的理解.
科学领域:
- 地质化学和环境科学
- 核废物管理 核废物管理
- 计算机建模 计算建模
背景情况:
- 废核燃料的地质处置需要了解放射性核素的移动性,特别是,它在氧化状态下具有高度移动性 (U ((VI)).
- 吸附到周围的矿物表面是限制其迁移的关键过程,但传统的表面复杂化模型 (SCMs) 面临着数值融合的挑战.
- 精确的流动性建模对于核废弃物仓库的安全评估至关重要.
研究的目的:
- 开发和评估机器学习 (ML) 替代品,用于吸附的复杂表面复杂化模型 (SCM).
- 解决与传统的SCM解决方案相关的计算成本和融合问题.
- 为增强核废物迁移模型提供超快的AI/ML工具.
主要方法:
- 探索了两个ML替代品:随机森林回归器和深度神经网络 (DNN).
- 经过训练和验证的ML模型使用2-pK三层模型对氧化物表面的保留的预测.
- 将开发的ML替代模型集成到更大规模的污染物迁移模型中.
主要成果:
- 两种ML替代品都准确地复制了SCM对吸附的预测.
- 深度神经网络 (DNN) 证明了特别有效,以显著降低计算成本实现高精度.
- 机器学习替代品成功地避免了数值SCM解决器固有的趋同问题.
结论:
- 人工智能/ML替代品为模拟吸附提供了一个计算效率高,可靠的替代传统的SCM.
- 这些超快的ML替代品可以很容易地集成到更大的污染物迁移模型中,以改善存储库安全评估.
- 这项研究提出了一种基于ML的新方法,以增强对地质存储库中放射性核素迁移的理解和预测.
更多相关视频
06:45Author Spotlight: Characterizing Porous Materials for Aiding the Development of Robust Metal-Organic Frameworks with Adsorption Behavior
Published on: March 8, 2024
7.4K
09:23Removal of Trace Elements by Cupric Oxide Nanoparticles from Uranium In Situ Recovery Bleed Water and Its Effect on Cell Viability
Published on: June 21, 2015
9.8K
相关概念视频
00:43Complexometric Titration: Ligands
950
Different monodentate and polydentate ligands are used as complexing agents in complexometric titration reactions. The formation of complexes by mono- and bidentate ligands involves two or more intermediate steps, limiting their use as complexing agents. In comparison, polydentate ligands can form complexes with metal ions in a single-step process, facilitating sharper end points. This means polydentate ligands, such as amino carboxylic acid derivatives, are most commonly employed in...
950
00:49Gravimetry: Inorganic And Organic Precipitating Agents
1.3K
In gravimetry, the precipitant is chosen carefully to obtain a pure solid that can be easily filtered. Common inorganic precipitants can be used to determine several cations and anions. In some cases, the formation of the same precipitate can be used to determine the cation and the anion. For example, the reaction of barium and chromate ions to give barium chromate is used to determine both barium and chromate. However, precipitates such as hydroxides, oxalates, and metal ammonium phosphates...
1.3K
00:56Extraction: Advanced Methods
446
Metal ions can be separated from one another by complexation with organic ligands–the chelating agent– to form uncharged chelates. Here, the chelating agent must contain hydrophobic groups and behave as a weak acid, losing a proton to bind with the metal. Since most organic ligands used in this process are insoluble or undergo oxidation in the aqueous phase, the chelating agent is initially added to the organic phase and extracted into the aqueous phase. The metal-ligand complex is...
446