The profile and axial designs of a volute tongue have great influences on the internal flow conditions and aerodynamic performance of a forward-curved blade centrifugal fan. In order to achieve a high quality of internal flow fields and fan aerodynamic performance, we propose an inclined bionic volute tongue (IBVT) described with rotation angle and scale coefficient at the front and back walls of the volute, which varies the profile of the volute tongue and the distance between the impeller and the volute tongue in the axial direction. Meanwhile, an adaptive pseudo expected improvement matrix (APEIM) parallel multi-objective efficient global optimization (EGO) algorithm is developed by introducing a flexible number of added points for each iteration of optimization on account of the optimization problem itself. On this basis, IBVT is optimized to maximize the fan aerodynamic performance in terms of total pressure efficiency and static pressure rise by virtue of numerical simulation. The results show that the developed APEIM parallel multi-objective EGO algorithm is capable of accurately and efficiently finding the Pareto front solutions. IBVT is considered to have the effectiveness of aerodynamic performance improvement for a forward-curved blade centrifugal fan. Within the fan with IBVT, the flow fields in the axial direction become more uniform with the dimension reduction of the inactive zone around the front span. The impeller-volute tongue interaction is greatly improved with the decreases of reversed and recirculated flow coefficients. IBVT also benefits the reduction of flow losses in the volute. By adopting the analysis of variance (ANOVA), the aerodynamic performance of the fan is highly affected by the rotation angle and scale coefficient of IBVT at the front wall of the volute. This work is helpful to guide the design of the volute tongue for a forward-curved blade centrifugal fan.

蜗壳舌的轮廓和轴向设计对前弯叶片离心式风扇的内部流动条件和空气动力性能有很大影响。为了获得高质量的内部流场和风扇空气动力学性能，我们提出了一种倾斜的仿生蜗壳舌（IBVT），其在蜗壳的前后壁上均具有旋转角度和比例系数，从而改变了蜗壳的轮廓舌和叶轮与蜗壳舌之间的轴向距离。同时，由于优化问题本身，通过为每次优化迭代引入灵活的加分点数，开发了自适应伪预期改进矩阵（APEIM）并行多目标高效全局优化（EGO）算法。在此基础上，通过数值模拟，IBVT经过优化，可在总压效率和静压上升方面最大化风扇的空气动力学性能。结果表明，所开发的APEIM并行多目标EGO算法能够准确有效地找到Pareto前沿解。IBVT被认为具有改善前弯叶片离心式风扇空气动力学性能的功效。在带有IBVT的风扇内，随着前跨度周围非活动区域的尺寸减小，轴向流场变得更加均匀。随着反向流动系数和再循环流动系数的减小，叶轮与蜗壳舌的相互作用大大改善。IBVT还有利于减少蜗壳中的流量损失。通过采用方差分析（ANOVA），风扇的空气动力性能受蜗壳前壁IBVT的旋转角度和比例系数的很大影响。这项工作有助于指导向前弯曲的叶片离心风扇的蜗壳舌的设计。