Application of agent model in the optimal design of AIG for SCR-DeNOx system

Abstract

Purpose Ammonia injection grid (AIG) is used as an input device for ammonia which reacts with NOx in the selective catalytic reduction (SCR) reactor. However, non-uniform concentration distribution of ammonia could produce partially poisoning or deposits of the catalyst. In this work, for making ammonia widely distributed throughout the flue gas and fully mixed, an optimization method of AIG is proposed. Design/methodology/approach Depending on the complexity of fluid flow, the relation between the concentration distributions of ammonia and the geometric parameters of AIG is nonlinear. Based on a certain amount of AIG samples, the computational fluid dynamics (CFD) simulations are applied to propose the agent model which describes the functional relation of the deviation of ammonia concentration and the geometric parameters of AIG. The optimization model of AIG based on the agent model is established. The optimized AIG based on the agent model can be used to produce uniform concentration distributions of ammonia, especially in the case that velocity distribution of flue gas is non-uniform. Findings For qualitatively confirming this optimization method, the three-dimensional CFD simulation of the optimized AIG is carried out. The results reveal that the diffusion process of ammonia gas is consistent with the development of the local vortices, which have a certain relation with the velocity distribution of the flue gas. The unequal ammonia injection designed by the optimization based on the agent model promotes a better mixing of ammonia and flue gas. Originality/value In this work, first, the method for optimizing AIG based on the agent model is proposed. Second, the three-dimensional CFD modeling and simulation of the optimized AIG is carried out, and the mixing effects of ammonia and flue gas are presented.