Effect of Parametric Uncertainty of Selected Classification Models and Simulations of Wastewater Quality Indicators on Predicting the Sludge Volume Index
More details
Hide details
University of Agriculture in Cracow, Department of Sanitary Engineering and Water Management, Kraków, Poland
University of Agriculture in Cracow, Department of Ecology Climatology and Air Protection, Kraków, Poland
Slovak University of Agriculture in Nitra, Department of Water Resources and Environmental Engineering, Nitra, Slovakia
Krzysztof Chmielowski   

University of Agriculture in Cracow, Department of Sanitary Engineering and Water Management, al. Mickiewicza 24/28, 30-059 Kraków, Poland, al. Mickiewicza 24/28, 30-059 Kraków Kraków, Poland
Submission date: 2018-05-21
Final revision date: 2018-11-06
Acceptance date: 2018-11-21
Online publication date: 2019-10-30
Publication date: 2020-01-16
Pol. J. Environ. Stud. 2020;29(2):1101–1110
This article presents a method for assessing the impact of the predictive uncertainty of selected wastewater quality indicators and the parametric uncertainty of classification models on the forecast results of simulating activated sludge sedimentation using classification models. The data for the calculations were obtained from monitoring carried out at a municipal wastewater treatment plant with a capacity of 72,000 m3/d1, located in the Sitkówka-Nowiny commune. The treatment plant receives wastewater, mostly from Kielce city. In the article the possibility of modeling the sedimentation of activated sludge at a wastewater treatment plant using logistic regression and Gompertz models was presented. The included values of the variables (i.e., sewage quality indicators) have been predicted by black-box methods (support vectors and k-nearest neighbor). This approach can be used to improve the operational efficiency of the bioreactor when continuous measurements of sewage quality are not available.