Zobrazit v EDS

Analysis of Response Surface and Artificial Neural Network for Cr(Ⅵ) Removal Column Experiment.

Uloženo v:
Podrobná bibliografie
Název: Analysis of Response Surface and Artificial Neural Network for Cr(Ⅵ) Removal Column Experiment.
Autoři: Ren, Zhongyu, Li, Zhicong, Tang, Haokai, Yang, Lin, Zhu, Jinrun, Jing, Qi
Zdroj: Water (20734441); Apr2025, Vol. 17 Issue 8, p1211, 21p
Témata: ARTIFICIAL neural networks, FOURIER transform infrared spectroscopy, ZERO-valent iron, RESPONSE surfaces (Statistics), FILTER paper
Abstrakt: In this study, inexpensive, environmentally friendly, and biodegradable cellulose filter paper was used to load nano zero-valent iron (nZVI), effectively improving the dispersibility of nZVI and successfully preparing the supported modified cellulose filter paper (FP-nZVI). Subsequently, the capacity of FP-nZVI to remove Cr(VI) in a flow system was explored. FP-nZVI was characterized by scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), and X-ray diffraction (XRD). Traditional single-factor experiments often require a large number of repeated experiments when analyzing the interactions among multiple variables, resulting in a long experimental cycle and high consumption of experimental materials. This research used the Response Surface Methodology (RSM) based on the Box-Behnken Design (BBD) and the Artificial Neural Network (ANN) to optimize and predict the removal process of Cr(VI). This RSM investigated the interactions between the response variable (Cr(VI) removal rate) and the independent variables (Cr(VI) concentration, pH value, and flow rate). A highly significant quadratic regression model was constructed, which was proven by a high F value (93.92), an extremely low p-value (<0.0001), and a high determination coefficient (R2 = 0.9918). An ANN model was established to forecast the correlation between independent variables and the removal rate of Cr(VI). Both models demonstrate remarkable consistency with the experimental data; however, from the perspective of statistical parameters, the ANN model has more significant advantages; the coefficient of determination R2 reaches 0.9937, which is higher than that of RSM (0.9918); the values of indicators such as MSE, RMSE, MAE, MAPE, AAD, and SEP are all smaller than those of RSM. The ANN exhibits greater excellence in prediction error, value fluctuation, and closeness to the actual value and has a more excellent prediction ability. The experiment for treating Cr(VI) with FP-nZVI was optimized, achieving good results. Meanwhile, it also provides a valuable reference for similar experimental studies. [ABSTRACT FROM AUTHOR]
Copyright of Water (20734441) is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Databáze: Biomedical Index
Popis
Abstrakt:In this study, inexpensive, environmentally friendly, and biodegradable cellulose filter paper was used to load nano zero-valent iron (nZVI), effectively improving the dispersibility of nZVI and successfully preparing the supported modified cellulose filter paper (FP-nZVI). Subsequently, the capacity of FP-nZVI to remove Cr(VI) in a flow system was explored. FP-nZVI was characterized by scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), and X-ray diffraction (XRD). Traditional single-factor experiments often require a large number of repeated experiments when analyzing the interactions among multiple variables, resulting in a long experimental cycle and high consumption of experimental materials. This research used the Response Surface Methodology (RSM) based on the Box-Behnken Design (BBD) and the Artificial Neural Network (ANN) to optimize and predict the removal process of Cr(VI). This RSM investigated the interactions between the response variable (Cr(VI) removal rate) and the independent variables (Cr(VI) concentration, pH value, and flow rate). A highly significant quadratic regression model was constructed, which was proven by a high F value (93.92), an extremely low p-value (<0.0001), and a high determination coefficient (R<sup>2</sup> = 0.9918). An ANN model was established to forecast the correlation between independent variables and the removal rate of Cr(VI). Both models demonstrate remarkable consistency with the experimental data; however, from the perspective of statistical parameters, the ANN model has more significant advantages; the coefficient of determination R<sup>2</sup> reaches 0.9937, which is higher than that of RSM (0.9918); the values of indicators such as MSE, RMSE, MAE, MAPE, AAD, and SEP are all smaller than those of RSM. The ANN exhibits greater excellence in prediction error, value fluctuation, and closeness to the actual value and has a more excellent prediction ability. The experiment for treating Cr(VI) with FP-nZVI was optimized, achieving good results. Meanwhile, it also provides a valuable reference for similar experimental studies. [ABSTRACT FROM AUTHOR]
ISSN:20734441
DOI:10.3390/w17081211