Investigation, biokinetic calculation, and modelling of a real combined industrial wastewater biological treatment process by activated sludge models
Bijan Bina,Nasim Nikzad,Soudabeh Ghodsi,Seyed Alireza Momeni,Hossein Movahedian Attar,Mahsa Janati,Farzaneh Mohammadi
DOI: https://doi.org/10.34172/ehem.2022.35
2022-10-21
Environmental Health Engineering and Management
Abstract:Background: Treatment of combined industrial wastewater from industrial parks is one of the most complex and difficult wastewater treatment processes. Also, the accuracy of biological models for the prediction of the performance of these processes has not been sufficiently evaluated. Therefore, in this study, the International Association on Water Quality (IAWQ(-Activated Sludge Model No. 1 (ASM1) was implemented for the Jey industrial park in Isfahan province, Iran. Methods: The Jey IPWWTP process is a combination of anaerobic and aerobic biological processes. To evaluate the overall performance of IPWWTP, organic compounds, suspended solids, nutrients, attached biomass, and some operating parameters were measured during 6 months. Then, the biokinetic coefficients of aerobic processes were determined using Monod equations. Finally, the aerobic processes were modeled using ASM1 implemented in STOAT software. Results: The values of the biokinetic coefficients K, Y, Ks, Kd, and μmax were calculated as 2.7d- , 0.34 mg VSS/mg COD, 133.36 mg/L COD, 0.03d- , and 0.93d- , respectively. Based on the default coefficients and conditions of the ASM model, the difference between the experiments and model prediction was about 2 to 98%. After calibrating the ASM model, the difference between the experiments and prediction in all parameters was reduced to less than 10%. Conclusion: Investigations showed that the default coefficients and operation conditions of the ASM1 model do not have good predictability for complex industrial wastewaters and the outputs show a low accuracy compared to the experiments. After calibrating the kinetic coefficients and operating conditions, the model performance is acceptable and the predictions show a good agreement with the experiments.