Centrifugal Compressor Stage Efficiency and Rotor Stiffness Augmentation via Artificial Neural Networks

TitleCentrifugal Compressor Stage Efficiency and Rotor Stiffness Augmentation via Artificial Neural Networks
Publication TypeConference Paper
Year of Publication2021
AuthorsAgnolucci A, Marconcini M, Arnone A, Toni L, Grimaldi A, Giachi M
Conference NameASME Turbo Expo 2021 Turbomachinery Technical Conference and Exposition
PublisherASME
Conference LocationVirtual Event, June 7-11, 2021
Abstract

Centrifugal compressor stages with high rotor stiffness (i.e.impeller hub-to-outer-diameter ratio) may represent a crucial element  to  cope  with  tight  rotordynamic  requirements  and  constraints that are needed for certain applications.  On the otherhand, high stiffness has a detrimental effect on the aerodynamicperformance.   Thus,  an  accurate  design  and  optimization  arerequired to minimize the performance gap with respect to low-stiffness stages.  This paper shows a redesign and optimizationprocedure of a centrifugal compressor stage aimed at increasing the impeller stiffness while keeping high aerodynamic performance. Two different optimization steps are employed to con-sider a wide design space while keeping the computational costas low as possible.  At first the attention is focused on the im-peller only,  then the diffuser and the return channel are takeninto account.  The multi-objective and multi-operating point op-timization makes use of artificial neural networks (ANNs) as asurrogate model to obtain the response surfaces.  RANS calculations are carried out using the TRAF code and are employedto create the training dataset.  Once the ANN has been trained,an optimization strategy is used to find the constrained optimum geometries for the impeller and the static components. The opti-mized high stiffness stage is finally compared to the low stiffnessone to assess its applicability

Notes

ASME paper GT2021-59998

Refereed DesignationRefereed