Comparing three mathematical models using different substrates for prediction of partial nitrification.

Modelling of partial nitrification process is affected by several factors such as selection of true substrates, FA and FNA inhibition, and pH effect on growth rate. Among these factors, the selection of true substrates is very critical as it affects the structure of the model. In the present work, a new model adopting free ammonia (FA) and free nitrous acids (FNA) as the true substrate for ammonia oxidizing bacteria (AOB) and nitrite oxidizing bacteria (NOB) was proposed. Then the proposed model was compared with two reported models which adopted ammonium and nitrite, and FA and nitrite as the true substrate for AOB and NOB, respectively. The three mathematical models were compared in terms of predicted minimum dissolved oxygen (DO) in response to varied solids retention time (SRT) (10-30 d), pH (7-8.5), and temperature (10-35 °C). The input kinetic values were justified and updated based on statistical analysis of literature data. Adopting FA as the true substrate increased the minimum DO for AOB. Further, experimental data from different literature studies were taken for model simulation and comparison. Inconsistency was observed between the model prediction and literature data for all three models. The model that adopted ammonium and nitrite as the true substrate for AOB and NOB had better consistency with literature data than other two models. The affecting factors for the model prediction was classified into three levels and discussed in detail. Future work was proposed. The results of this study provide valuable information for the design and modelling of partial nitrification process.