Deep learning modeling for removal of cadmium (Cd) (II) ions from water using kenaf fiber biochar as an adsorbent

Heavy metals in the water bodies can be detrimental to human health and also the environment. Removal of heavy metal can be done using adsorbent derived from plant materials. In this study, ANN modeling was performed for the removal of cadmium (Cd) (II) ions from water using Kenaf fiber biochar as a...

Full description

Saved in:
Bibliographic Details
Main Authors: Arumugasamy, Senthil Kumar, Selvarajoo, Anurita
Format: Conference Proceeding
Language:English
Subjects:
Adsorbents
Adsorption
Algorithms
Cadmium
Errors
Heavy metals
Kenaf
Machine learning
Modelling
Pyrolysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Heavy metals in the water bodies can be detrimental to human health and also the environment. Removal of heavy metal can be done using adsorbent derived from plant materials. In this study, ANN modeling was performed for the removal of cadmium (Cd) (II) ions from water using Kenaf fiber biochar as an adsorbent. Operating conditions of the process: chemical impregnation (NaOH: KF), pyrolysis temperature and pyrolysis time were used as the inputs and biochar yield, cadmium removal, adsorption capacity, and specific surface area of the process were used as the outputs of the ANN model. The model was trained using 12 Training algorithms (TA) of ANN to optimize the TA that provides the best performance in terms of prediction. The results show that no TA’s performance has been consistent in terms of lowest mean squared error (MSE), mean absolute error (MAE), mean absolute performance error (MAPE) for all the 4 outputs. Levenberg Marquart (LM) has been successful in providing lowest MSE, MAE, MAPE of 5.23, 2.23, 0.07 respectively for biochar yield and 2.69, 1.14, 0.07 respectively for adsorption capacity. Also, since highest accuracy of 83.33 was obtained for adsorption capacity with LM, it is optimized to be best performing TA.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0195568