Application of a generalized hybrid machine learning model for the prediction of H2S and VOCs removal in a compact trickle bed bioreactor (CTBB)

Krzysztof Barbusiński; Bartosz Szeląg; Anita Parzentna-Gabor; Damian Kasperczyk; Eldon R Rene

doi:10.1016/j.chemosphere.2024.142181

Application of a generalized hybrid machine learning model for the prediction of H₂S and VOCs removal in a compact trickle bed bioreactor (CTBB)

Chemosphere. 2024 Apr 27:360:142181. doi: 10.1016/j.chemosphere.2024.142181. Online ahead of print.

Authors

Krzysztof Barbusiński¹, Bartosz Szeląg², Anita Parzentna-Gabor³, Damian Kasperczyk³, Eldon R Rene⁴

Affiliations

¹ Department of Water and Wastewater Engineering, Silesian University of Technology, Konarskiego 18, 44-100, Gliwice, Poland.
² Warsaw University of Life Sciences, Nowoursynowska 166, 02-787, Warsaw, Poland. Electronic address: bszelag@tu.kielce.pl.
³ Ekoinwentyka Ltd., Szyb Walenty 26, 41-700, Ruda Śląska, Poland.
⁴ Department of Water Supply, Sanitation and Environmental Engineering, IHE Delft Institute for Water Education, P.O. Box 3015, 2601DA Delft, Netherlands.

PMID: 38685329
DOI: 10.1016/j.chemosphere.2024.142181

Abstract

This study presents a generalized hybrid model for predicting H₂S and VOCs removal efficiency using a machine learning model: K-NN (K - nearest neighbors) and RF (random forest). The approach adopted in this study enabled the (i) identification of odor removal efficiency (K) using a classification model, and (ii) prediction of K <100%, based on inlet concentration, time of day, pH and retention time. Global sensitivity analysis (GSA) was used to test the relationships between the inputs and outputs of the K-NN model. The results from classification model simulation showed high goodness of fit for the classification models to predict the removal of H₂S and VOCs (SPEC = 0.94-0.99, SENS = 0.96-0.99). It was shown that the hybrid K-NN model applied for the "Klimzowiec" WWTP, including the pilot plant, can also be applied to the "Urbanowice" WWTP. The hybrid machine learning model enables the development of a universal system for monitoring the removal of H₂S and VOCs from WWTP facilities.

Keywords: Compact trickle bed bioreactor (CTBB); H(2)S; Machine learning; Odors; VOCs.