Optimization of ventilation system operation in office environment using POD model reduction and genetic algorithm

The purpose of this paper is to develop a numerical optimization methodology of ventilation system operation that help improve indoor environment quality and energy costs for space heating and cooling simultaneously. By introducing proper orthogonal decomposition (POD) based model reduction technique into computational fluid dynamics (CFD) simulation, the spatial distributions of key environmental parameters are firstly considered in optimization procedure, which enable to help improve the environment comfort of occupied zones in a office room. Four types of low-dimension parameter spaces are acquired by POD method including indoor temperature, airflow, CO2 concentration and predicted mean vote (PMV) distributions. The precisions of the reconstructed spaces are compared with those of CFD-based original ones. Because of the extremely lower dimension, resulted parameter subspaces are suitable for optimization process in this study. In the proposed genetic algorithm (GA), the objective function is constructed in a way attempting to aggregate and weight indices into one indicator, such as index of PMV, space temperature gradients (STG), indoor air quality (IAQ), energy cost of heating and cooling, etc. A simulation based case study is conducted, which indicates that the presented optimization approach is able to improve indoor comfort and energy costs of ventilation system in a balanced way.