B. VIVEKANANDAN AND A. SESHAGIRI RAO
Abstract
In this paper, the control strategies such as PI and model predictive control (MPC) are applied to benchmark simulation model no.1 (BSM1) of an activated sludge process. The main objective is to compare the PI and MPC control strategies based on the performance with respect to effluent concentration under constraints and operating costs. In this study, nitrate and nitrite concentration (SNO) is controlled in the second anoxic reactor using PI control strategy and the dissolved oxygen concentration (SO) is controlled in the last aerobic reactor using either MPC or PI control strategy of the activated sludge process. No case studies of BSM1 activated sludge process have been described in the literature for the municipal wastewater treatment plant located in India. In this paper, simulations are carried out using the influent data collected from the municipal wastewater treatment plant, located in India. The influent variables data are takenfrom the municipal wastewater treatment plant during the dry weather period. The influent fractionation is evaluated using activated sludge model no.1 (ASM1). The simulation response shows that the advanced MPC controller gives better performance with respect to constraints than simple PI control especially effluent ammonia concentration. This study also indicates that dissolved oxygen MPC and nitrate and nitrite PI (MPC-PI) control strategies have achieved nearly similar operating costs as with dissolved oxygen PI and nitrate and nitrite PI (PI-PI) control strategies.