摘要

Most of the local pressure loss coefficient (LPLC) models for duct fittings used in heating, ventilation and air conditioning (HVAC) air-side system transient simulations are simplified. The LPLCs are defined as having a constant value at the rated flow condition or as having a polynomial function of the flow ratio (or velocity ratio). To determine the influence of these simplifications, this study used a diverging tee junction as an example. First, we performed CFD calculations to generate a new LPLC dataset and trained a data-driven model using feature weighted support vector regression (FWSVR) combined with particle swarm optimization (PSO-FWSVR). Finally, we compared this new LPLC model with the two traditional models at the level of both individual junction and air-side system. The results show that the accuracy of the new model is greatly improved and the LPLC model can have a significant impact on the system operation condition.