Uncertainty quantification and sensitivity analysis of an ultraviolet water disinfection photo-reactor

Abstract

In recent years, computational fluid dynamics (CFD) has been utilized as a powerful simulation tool to design and analyze the ultraviolet (UV) disinfection systems. A fact that is ignored in deterministic simulations is that most of industrial problems are associated with inevitable uncertainties in input parameters. In the present study, a non-deterministic CFD simulation of the UV disinfection process using non-intrusive polynomial chaos expansion method is performed to quantify the impact of uncertain input parameters on the performance of an annular photo-reactor. Five input variables with their probability distribution functions are determined as the uncertain inputs to evaluate their effects on the quantities of interest for different working conditions. Besides, the contribution of each uncertain variable is quantified by a sensitivity analysis using the Sobol’ indices method. The results show that the disinfection process of contaminated water with higher UV transmittances and lower flow rates is associated with more uncertainty.