Multi-objective optimization design of axial flow fan rotor blades and guide vanes based on coordinated radial twist

Abstract

To address the issue of flow interference and energy loss caused by insufficient aerodynamic matching between rotor blades and guide vanes in low-pressure industrial axial fans, this study proposes a coordinated radial twist blade design method for rotor blades and guide vanes. A parameterized model of blade radial twist is constructed based on Non-Uniform Rational B-Spline (NURBS) curves, in which the control point coordinates (k₁, k₂, k₃, c₁, c₂, c₃) are extracted as design variables to generate three-dimensional twisted blade profiles that adapt to local inflow characteristics. Then, samples are selected using the Optimal Latin Hypercube Sampling (OLHS) method, and a collaborative optimization of fan total pressure and efficiency is achieved by integrating a Hippopotamus Optimization-enhanced Backpropagation Neural Network surrogate model (HO-BP) with a multi-objective evolutionary algorithm. The optimization results show that, compared to the original design, the optimized model achieves improvements of 11.89% in total pressure and 6.50% in efficiency under design conditions, and up to 18.45% and 8.48%, respectively, under off-design conditions, significantly expanding the operational range of the fan. Further flow field analysis indicates that the coordinated radial twist design reconstructs the radial load distribution on the blades, effectively alleviates hub corner separation and tip leakage, suppresses flow separation on the suction surface, and improves flow matching between rotor blades and guide vanes. These improvements lead to reduced flow interference and energy loss, thereby enhancing the fan’s aerodynamic performance and operational stability.