The customary single-arc blade applied in the multi-blade centrifugal ventilating fan for building ventilation is ordinary in shape and hard to meet the design requirements of the pertinent parameters. The article used the modified Hicks-Henne function (suitable for blade design with multi-arc and complex curve blades) to parametrically design the blades of multi-blade centrifugal fan to improve its aerodynamic performance. During the optimization process, the Audze-Eglais (AE) criterion was used to optimize the Latin hypercube test design. Based on the optimized test design method, a Kriging agent model was established. While taking total pressure efficiency and flow rate as the optimization goals, combined with the nondominated sorting genetic algorithm II (NSGA-II) to find the Pareto solution set of the agent model. Internal flow field was analyzed before and after optimization by numerical simulation of the fan, and the Computational Fluid Dynamics (CFD) results were further verified through experiments. At the maximum efficiency point, the flow rate of the fan after the optimization has increased by 1.18 m³/min, the total pressure efficiency was improved by 4.21%. The improvement of efficiency and flow rate can effectively reduce energy consumption of building ventilation.

用于建筑物通风的多叶片离心通风机中使用的常规单弧叶片形状普通，难以满足相关参数的设计要求。本文使用修改后的Hicks-Henne函数（适用于多弧和复杂曲线叶片的叶片设计）对多叶片离心风扇的叶片进行参数化设计，以改善其空气动力学性能。在优化过程中，使用Audze-Eglais（AE）准则来优化拉丁文超立方体测试设计。基于优化的测试设计方法，建立了克里格代理模型。在以总压效率和流量为最优化目标的同时，结合非支配排序遗传算法II（NSGA-II）来找到Agent模型的Pareto解集。通过对风扇的数值模拟，对优化前后的内部流场进行了分析，并通过实验进一步验证了计算流体动力学（CFD）的结果。在最高效率点，优化后的风扇流量增加了1.18 m³ / min，总压力效率提高了4.21％。效率和流量的提高可以有效降低建筑物通风的能耗。