Application of box-behnken design with response surface for optimizing ventilation system of underground shelter

Ventilation shaft is one of the effective elements in natural ventilation for ensuring acceptable Indoor Air Quality (IAQ) and thermal comfort. It has been found that the opening of ventilation shaft plays a significant role in the ventilation efficiency of an underground shelter. In this study, we aim to develop a predictive ventilation rate model for a naturally-ventilated underground shelter. Computational Fluid Dynamics (CFD) was employed as a simulation tool, where the result was validated with experimental data obtained from the previous literature. Goal Driven Optimization (GDO) was used for the optimization process by considering three geometrical factors and their effects on the objective function. From this study, it is found that the predicted response surface values agree well with the CFD values and hence the predictive model is reliable.