相关实验视频

Updated: May 10, 2026

Fabrication of Gate-tunable Graphene Devices for Scanning Tunneling Microscopy Studies with Coulomb Impurities

Fabrication of Gate-tunable Graphene Devices for Scanning Tunneling Microscopy Studies with Coulomb Impurities

Published on: July 24, 2015

机器学习方法的比较研究,用于模拟水处理中的基于石墨烯的吸附

Thu Thao Thi Tran¹, Saeedeh Babaee¹, Deneyelle Wilson¹

¹Department of Chemistry and Chemical Engineering, Florida Institute of Technology, Melbourne, Florida, USA.

Water environment research : a research publication of the Water Environment Federation

|December 23, 2025

概括

多项式回归 (PR) 在污染物去除预测的小数据集上扎. 支持矢量机 (SVM) 和人工深度学习神经网络 (ANNDL) 为优化水处理吸附系统提供了卓越的性能和灵活性.

科学领域:

环境科学环境科学
材料科学材料科学材料科学
数据科学数据科学数据科学

背景情况:

多项式回归 (PR) 被广泛用于预测污染物去除效率,但缺乏灵活性.
像支持向量机 (SVM) 和人工深度学习神经网络 (ANNDL) 这样的机器学习方法显示出希望,但需要对小数据集进行优化.
对于水污染物处理的吸附研究通常涉及小数据集,需要强大的建模方法.

研究的目的:

评估多项式回归 (PR),支持向量机 (SVM) 和人工深度学习神经网络 (ANNDL) 的性能,以使用小数据集预测水污染物去除效率.
为了比较这些方法在数据大小和参数限制方面的灵活性和容忍度.
为优化吸附系统和水处理中的实验设计提供建议.

主要方法:

利用了四个已发表的数据集 (石墨烯相关的纳米材料),数据大小小小 (20-30个样本).
应用多项式回归 (PR),支持向量机 (SVM) 具有辐射基函数内核,以及使用TensorFlow框架的人工深度学习神经网络 (ANNDL).
在交叉验证下评估模型性能和可靠性,重点关注数据大小和参数灵敏度.

主要成果:

多项式回归 (PR) 显示可靠性,但对数据大小和参数限制高度敏感.
与PR相比,带有辐射基函数内核的支持矢量机器 (SVM) 和人工深度学习神经网络 (ANNDL) 显示出更高的性能和更大的耐受性对数据大小和参数的变化.
无论是SVM还是ANNDL,都被证明在预测有限数据的污染物去除效率方面更强大.

结论:

SVM和ANNDL比PR更适合在吸附研究中常见的小型数据集中建模污染物去除效率.
选择建模技术显著影响水处理系统的可靠性和优化.
未来的研究应该专注于利用先进的机器学习技术,以实现强大的实验设计和有效的水污染物处理.

关键词:

在 RSM RSM 中.威尼斯人百叶窗威尼斯人百叶窗吸附吸附是一种吸附.进行交叉验证.机器学习是机器学习.

更多相关视频

Preparation of Biomass-based Mesoporous Carbon with Higher Nitrogen-/Oxygen-chelating Adsorption for CuII Through Microwave Pre-Pyrolysis

Preparation of Biomass-based Mesoporous Carbon with Higher Nitrogen-/Oxygen-chelating Adsorption for CuII Through Microwave Pre-Pyrolysis

Published on: February 12, 2019

Development and Functionalization of Electrolyte-Gated Graphene Field-Effect Transistor for Biomarker Detection

Development and Functionalization of Electrolyte-Gated Graphene Field-Effect Transistor for Biomarker Detection

Published on: February 1, 2022

相关概念视频

Adsorption of Gases on Solids

Adsorption of Gases on Solids

Adsorption is a process where molecules, known as the adsorbates, accumulate on a surface, which is referred to as the adsorbent or substrate. Occurring at the solid-gas interface, this phenomenon is crucial in various scientific and industrial contexts. The reverse of adsorption is desorption.Two types of adsorptions exist: physical (physisorption) and chemical (chemisorption). Physisorption involves gas molecules held to the solid's surface by relatively weak intermolecular van der Waals...

Adsorption Isotherms I

Adsorption Isotherms I

Adsorption isotherms are mathematical models that describe how molecules in a gas or liquid phase interact with surfaces. Two of the most common isotherm models are the Langmuir and Freundlich isotherms, which relate to Type I monolayer chemisorption. The Langmuir model is based on four key assumptions:• Adsorption cannot exceed monolayer coverage.• All surface sites are equivalent.• Molecules adsorb only at vacant sites.• There are no interactions between adsorbed molecules.Consider the...

Adsorption Isotherms II

Adsorption Isotherms II

Brunauer, Emmett, and Teller (BET) introduced a theory in 1938 that modified Langmuir's assumptions to explain multilayer physical adsorption. This theory is applicable to Type II isotherms and provides a more realistic picture of adsorption processes. The BET theory assumes a uniform solid surface with localized adsorption sites, where adsorption at one site doesn't affect adsorption at neighboring sites. This theory also allows for the possibility of additional molecules being adsorbed on top...