Multi-point aerodynamic optimization of a backward-curved impeller fan

Abstract

Centrifugal fans find widespread application in building ventilation and air-conditioning systems. Nonetheless, they face challenges in meeting the growing need for energy efficiency, better performance, and reduced noise levels. These limitations stem from the inherent constraints imposed by the circular arc blade design. In this paper, aerodynamic optimization of a backward-curved centrifugal fan was carried out. For this purpose, 3D computational fluid dynamics (CFD) simulations of the initial fan were first validated against test data. Then, using five geometric design parameters, the blade flow angle distribution, from leading edge (LE) to trailing edge (TE), was optimized within its operating range by means of a surrogate- based optimization technique. Efficiencies at different operating points of the fan were selected as objectives while the constraints of the optimization practice include flowrates as well as blade thicknesses of the initial geometry. The suggested optimum geometry generated from this process was independently simulated by CFD to assess the meta-model predictions. The computed results demonstrated up to a 1.7% increase in efficiency in its operating range compared to the initial model. Moreover, studying the features of the flow passing through the blades indicated an improved aerodynamic behavior with reduced separation zones for the optimized geometry compared with the initial one. Geometric comparison of initial and optimized geometries also revealed a curvature redistribution in the fan blades.