Multi-objective design optimization of a high efficiency and low noise blower unit of a car air-conditioner

Car air-conditioners consist of a blower unit and a heater unit. A blower unit sends wind to a heater unit, and a heater unit adjusts the temperature inside the vehicle. Blower units of car air-conditioners are required to be smaller, lighter, noiseless, and power-saving. However, it is difficult and expensive to predict the noise directly by computational fluid dynamics simulation. Hereupon, this study employs an indirect noise prediction method based on a noise prediction theory to evaluate noise for blower units inexpensively. This method is investigated through a comparison with actual sound pressure level measurement. Then, using this method, this study moves to design optimization of a blower unit of car air-conditioners. The optimization aims to improve total pressure efficiency and sound pressure level from the current design that has been employed for a real commercial vehicle. This study employs a genetic algorithm to explore global optima in a two-objective problem. The present genetic algorithm is assisted by the Kriging surrogate model to reduce computational cost required for evaluating objective functions. The optimization results indicate that the optimized blower unit involves a multi-blade fan with the high chord-pitch ratio to decrease the loss of total pressure efficiency, which is often induced by the flow separation on the blade and the swirling flow on the meridional plane. In addition, the sound pressure level of blower unit can be reduced by decreasing the local flow velocity on the meridian plane due to a blockage factor. A blower unit, which has a scroll with a large tongue angle, shows high total pressure efficiency because the increase in eddy loss is suppressed at the tongue. They suggest the importance of the matching of multi-blade fan and scroll to achieve the good overall performance of a blower unit.