Performance and model of a full-scale up-flow anaerobic sludge blanket (UASB) to treat the pharmaceutical wastewater containing 6-APA and amoxicillin

A full-scale test was conducted with an up-flow anaerobic sludge blanket (UASB) pre-treating pharmaceutical wastewater containing 6-aminopenicillanic acid (6-APA) and amoxicillin. The aim of the study is to investigate the performance of UASB in the condition of a high chemical oxygen demand (COD) loading rate from 12.57 to 21.02 kg m−3 d−1 and a wide pH from 5.57 to 8.26, in order to provide a reference for treating the similar chemical synthetic pharmaceutical wastewater containing 6-APA and amoxicillin. The results demonstrated that the UASB average percentage reduction in COD, 6-APA and amoxicillin were 52.2%, 26.3% and 21.6%, respectively. In addition, three models, built on the back propagation neural network (BPNN) theory and linear regression techniques were developed for the simulation of the UASB system performance in the biodegradation of pharmaceutical wastewater containing 6-APA and amoxicillin. The average error of COD, 6-APA and amoxicillin were −0.63%, 2.19% and 5.40%, respectively. The results indicated that these models built on the BPNN theory were well-fitted to the detected data, and were able to simulate and predict the removal of COD, 6-APA and amoxicillin by UASB.