LAUSR

Abstract The aim of the study was to supplement the information missing in the literature on the influence of organic compounds speciation (percentage of chemical oxygen demand (COD) fractions, readily biodegradable and non–biodegradable dissolved organic matter; slowly biodegradable and non‒biodegradable insoluble organic matter) in wastewater on chemically assisted primary sedimentation process and its potential impact on the energy and economic balance of the plant. The statistical modelling of the chemically assisted primary sedimentation process using artificial neural networks (ANN) was a novelty. It was also important to check the differences resulting from the use of coagulants, polyelectolites and their doses. For both large wastewater treatment plants (>500,000 people equivalent) located in the various regions of the country in Central Europe only ferric chloride had a positive balance. The addition of polyelectrolyte in the Wroclaw wastewater did not affect the chemically assisted primary sedimentation process but in the Bialystok wastewater it strengthened the process effect. In both plants only doses from 10 to 40 mgFe/dm3 are cost-effective. The results indicate that the addition of polyelectrolyte should be taken into account in this application and was correlated with the amount of colloidal COD fractions in wastewater. The developed ANN models reflected well the effectiveness of COD removal in both large WWTP while the sensitivity analysis of individual ANN models showed differences in the variables affecting the calculation process of both models, despite similar values of parameters characterizing wastewater in both wastewater treatment plants.