Carbon and nutrient removal from domestic wastewaters in a modified 5-stage Bardenpho process via fuzzy modeling approach

Gradual increase in the generation of wastewater results from the increasing global population. Thus, new treatment techniques and systems for controlling the treatment process depending on wastewater characteristics are desirable. This paper presents the use of a pilot-scale modified five-stage Bardenpho process with a 10 m³/day capacity for the treatment of real municipal wastewater. The process was developed for this study, and the steady-state removal efficiencies for COD (chemical oxygen demand), TKN (total Kjeldahl nitrogen), NH₄⁺-N (ammonium nitrogen), PO₄^3--P (phosphate phosphoms), SS (suspended solids), and VSS (volatile suspended solids) were 87±5%, 86±12%, 93±14%, 89±9%, 88±8%, 94±4%, and 94±4%, respectively. In the study, the effluent COD, TKN, and TP concentrations were also estimated using a fuzzy logic approach. The results showed that coefficients of determination are higher than 0.80 suggesting that the presented fuzzy logic approach may confidently be used for the estimation of the treatment performance.