The application of machine learning methods for prediction of metal sorption onto biochars

Xinzhe Zhu, Xiaonan Wang, Yong Sik Ok

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The adsorption of six heavy metals (lead, cadmium, nickel, arsenic, copper, and zinc) on 44 biochars were modeled using artificial neural network (ANN) and random forest (RF) based on 353 dataset of adsorption experiments from literatures. The regression models were trained and optimized to predict the adsorption capacity according to biochar characteristics, metal sources, environmental conditions (e.g. temperature and pH), and the initial concentration ratio of metals to biochars. The RF model showed better accuracy and predictive performance for adsorption efficiency (R2 = 0.973) than ANN model (R2 = 0.948). The biochar characteristics were most significant for adsorption efficiency, in which the contribution of cation exchange capacity (CEC) and pHH2O of biochars accounted for 66% in the biochar characteristics. However, surface area of the biochars provided only 2% of adsorption efficiency. Meanwhile, the models developed by RF had better generalization ability than ANN model. The accurate predicted ability of developed models could significantly reduce experiment workload such as predicting the removal efficiency of biochars for target metal according to biochar characteristics, so as to select more efficient biochar without increasing experimental times. The relative importance of variables could provide a right direction for better treatments of heavy metals in the real water and wastewater.

Original languageEnglish
Article number120727
JournalJournal of hazardous materials
DOIs
Publication statusPublished - 2019 Oct 15

Fingerprint

Learning systems
Sorption
sorption
Metals
adsorption
Adsorption
metal
prediction
artificial neural network
Neural networks
Heavy metals
Neural Networks (Computer)
heavy metal
Heavy Metals
Arsenic
cation exchange capacity
Cadmium
method
machine learning
biochar

Keywords

  • Artificial intelligence
  • Charcoal
  • Machine learning
  • Pyrolysis
  • Sorption model

ASJC Scopus subject areas

  • Environmental Engineering
  • Environmental Chemistry
  • Waste Management and Disposal
  • Pollution
  • Health, Toxicology and Mutagenesis

Cite this

The application of machine learning methods for prediction of metal sorption onto biochars. / Zhu, Xinzhe; Wang, Xiaonan; Ok, Yong Sik.

In: Journal of hazardous materials, 15.10.2019.

Research output: Contribution to journalArticle

@article{b1da57afb99d4bf785a76a1890edd295,
title = "The application of machine learning methods for prediction of metal sorption onto biochars",
abstract = "The adsorption of six heavy metals (lead, cadmium, nickel, arsenic, copper, and zinc) on 44 biochars were modeled using artificial neural network (ANN) and random forest (RF) based on 353 dataset of adsorption experiments from literatures. The regression models were trained and optimized to predict the adsorption capacity according to biochar characteristics, metal sources, environmental conditions (e.g. temperature and pH), and the initial concentration ratio of metals to biochars. The RF model showed better accuracy and predictive performance for adsorption efficiency (R2 = 0.973) than ANN model (R2 = 0.948). The biochar characteristics were most significant for adsorption efficiency, in which the contribution of cation exchange capacity (CEC) and pHH2O of biochars accounted for 66{\%} in the biochar characteristics. However, surface area of the biochars provided only 2{\%} of adsorption efficiency. Meanwhile, the models developed by RF had better generalization ability than ANN model. The accurate predicted ability of developed models could significantly reduce experiment workload such as predicting the removal efficiency of biochars for target metal according to biochar characteristics, so as to select more efficient biochar without increasing experimental times. The relative importance of variables could provide a right direction for better treatments of heavy metals in the real water and wastewater.",
keywords = "Artificial intelligence, Charcoal, Machine learning, Pyrolysis, Sorption model",
author = "Xinzhe Zhu and Xiaonan Wang and Ok, {Yong Sik}",
year = "2019",
month = "10",
day = "15",
doi = "10.1016/j.jhazmat.2019.06.004",
language = "English",
journal = "Journal of Hazardous Materials",
issn = "0304-3894",
publisher = "Elsevier",

}

TY - JOUR

T1 - The application of machine learning methods for prediction of metal sorption onto biochars

AU - Zhu, Xinzhe

AU - Wang, Xiaonan

AU - Ok, Yong Sik

PY - 2019/10/15

Y1 - 2019/10/15

N2 - The adsorption of six heavy metals (lead, cadmium, nickel, arsenic, copper, and zinc) on 44 biochars were modeled using artificial neural network (ANN) and random forest (RF) based on 353 dataset of adsorption experiments from literatures. The regression models were trained and optimized to predict the adsorption capacity according to biochar characteristics, metal sources, environmental conditions (e.g. temperature and pH), and the initial concentration ratio of metals to biochars. The RF model showed better accuracy and predictive performance for adsorption efficiency (R2 = 0.973) than ANN model (R2 = 0.948). The biochar characteristics were most significant for adsorption efficiency, in which the contribution of cation exchange capacity (CEC) and pHH2O of biochars accounted for 66% in the biochar characteristics. However, surface area of the biochars provided only 2% of adsorption efficiency. Meanwhile, the models developed by RF had better generalization ability than ANN model. The accurate predicted ability of developed models could significantly reduce experiment workload such as predicting the removal efficiency of biochars for target metal according to biochar characteristics, so as to select more efficient biochar without increasing experimental times. The relative importance of variables could provide a right direction for better treatments of heavy metals in the real water and wastewater.

AB - The adsorption of six heavy metals (lead, cadmium, nickel, arsenic, copper, and zinc) on 44 biochars were modeled using artificial neural network (ANN) and random forest (RF) based on 353 dataset of adsorption experiments from literatures. The regression models were trained and optimized to predict the adsorption capacity according to biochar characteristics, metal sources, environmental conditions (e.g. temperature and pH), and the initial concentration ratio of metals to biochars. The RF model showed better accuracy and predictive performance for adsorption efficiency (R2 = 0.973) than ANN model (R2 = 0.948). The biochar characteristics were most significant for adsorption efficiency, in which the contribution of cation exchange capacity (CEC) and pHH2O of biochars accounted for 66% in the biochar characteristics. However, surface area of the biochars provided only 2% of adsorption efficiency. Meanwhile, the models developed by RF had better generalization ability than ANN model. The accurate predicted ability of developed models could significantly reduce experiment workload such as predicting the removal efficiency of biochars for target metal according to biochar characteristics, so as to select more efficient biochar without increasing experimental times. The relative importance of variables could provide a right direction for better treatments of heavy metals in the real water and wastewater.

KW - Artificial intelligence

KW - Charcoal

KW - Machine learning

KW - Pyrolysis

KW - Sorption model

UR - http://www.scopus.com/inward/record.url?scp=85066994123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066994123&partnerID=8YFLogxK

U2 - 10.1016/j.jhazmat.2019.06.004

DO - 10.1016/j.jhazmat.2019.06.004

M3 - Article

JO - Journal of Hazardous Materials

JF - Journal of Hazardous Materials

SN - 0304-3894

M1 - 120727

ER -