Model-based optimization of a conventional SBR process for biological C, N, and P removal from wastewaters

SANTA CRUZ, JUDITH A.; MORES, PATRICIA L.; SCENNA, NICOLÁS J.; MUSSATI, MIGUEL C.

Ciudad Autónoma de Buenos Aires

Congreso; 11th World Congress of Chemical Engineering (WCCE11); 2023

AAIQ Asociación Argentina de Ingenieros Químicos

The system known as sequential batch reactor (SBR) for the biological treatment of wastewater consists of a cyclic process that operates through a temporal sequence of stages: filling, reaction, sedimentation, emptying of the treated liquid, purging of sludge excess, and, eventually, idle.The goal of the work is to obtain, on the one hand, the minimum operating cycle time and, on the other hand, the minimum total annual cost, while meeting allowed discharge limits, via simultaneous mathematical optimization based on a phenomenological process model.The developed model consists of the activated sludge kinetic model ASM3 from the IWA [1] for the removal of C and N and the EAWAG Bio-P module model for P removal [2], coupled to design and dimensioning equations and to a cost model. The ordinary differential equations describing the dynamic model are discretized using the orthogonal collocation method on finite elements, resulting in a nonlinear programming (NLP) model, which is implemented in the equation-oriented modeling and optimization environment GAMS (General Algebraic Modeling System). In the cost model, the total annual cost (TAC) considers the total operating cost (TOC) and the investment cost (IC). The IC is a function of the total reaction volume and the oxygen gas-liquid transfer coefficient (kLa). The TOC includes the costs of electricity for aeration, agitation, and pumping, the cost for excess sludge treatment, and fines for pollution units discharged with the treated liquid.As mentioned, two simple-objective optimization problems are solved: (P1) minimization of the total cycle time and (P2) minimization of the total annual cost (TAC). The decision or optimization variables are the duration of each reaction stage, the solid residence time, and the kLa value at each instant of the aerobic stage. Four reactors operating in parallel are considered for treating a municipal wastewater with a volumetric flowrate and composition.For problem P1, a minimum total cycle time of 8.7 h is obtained, corresponding to a TAC of €796275/year, while for problem P2 a minimum TAC of €656566/year is obtained, corresponding to a total cycle time of 15.2 h. The minimization of the cycle time results in a total volume of each reactor of 2305 m3, while the minimization of the TAC results in a volume 90% greater, which determines that the IC of problem P2 is by 35% larger than that of problem P1. The TAC minimization solution presents better quality of the treated liquid (64%), higher total electrical energy consumption (24%), and less sludge generation (18%) with respect to the cycle time minimization solution, which determines that the TOC for TAC minimization is 31% lower. The set of operating conditions resulting from the minimization of the cycle time presents lower volume, effluent quality, and energy consumption, but higher production of sludge for disposal.