Design And Multi-objective Optimization Of Mixed-flow Compressor

With the rapid development of aviation science and technology,micro turbojet engines want to impact higher performance,further achieve miniaturization,and low cost and high performance.Micro turbojet engines based on mixed-flow compressors have become a research hotspot.The mixed-flow compressor is a form between the centrifugal compressor and the axial compressor,which includes the advantages of both.It has attracted more and more attention to realize large flow rate,high pressure ratio and low cost under limited size.This paper takes the compressor of a micro turbojet engine proposed by a key domestic science and technology project in 2022 as the research objective.In order to meet the project requirements: the thrust of the micro turbojet engine reaches 100 da N,and the maximum radial size does not exceed 250 mm,design and optimization are carried out for the mixed-flow compressor of the micro turbojet engine under the indicators.(1)In this paper,for the micro turbojet engine with a thrust of 100 da N,the overall performance of the turbojet engine is estimated,and then the performance is allocated to the compressor by a professional soft.The preliminary aerodynamic design of the mixed-flow compressor is completed through one-dimensional design and three-dimensional modeling of the impeller and diffuser.The modeling is completed using Solid Works.(2)The simulation analysis is carried out for the mixed-flow compressor,and the working characteristic curve of the compressor is drawn.It’s found that the mixed-flow compressor has a wide surge margin and a narrow blockage margin.The low speed working state is relatively stable,and it’s easy to lose stability at high speed.The efficiency at the design point reaches 82.17%,and the pressure ratio reaches 4.771,meeting the design requirements.The flow field of the compressor impeller is stable,and the gas state of the flow field in the vaneless section and diffuser section is disordered.The main characteristics are the leakage vortex at the vaneless section,the shock wave at the inlet of the diffuser,and the low-speed flow cluster on the suction surface at the outlet of the blade.(3)According to the simulation results,the hub diameter,the number of impeller blades and diffuser blades,the axial length of the diffuser,the control points of the vaneless section and leading edge,and the curvature of the diffuser channel are taken as the optimization parameters,and the efficiency and pressure ratio of the mixed-flow compressor are taken as the optimization objectives.The sample library is established through orthogonal experiments,and the SOM(self-organization maps)correlation analysis is carried out for the optimization parameters in the sample library,so that the hidden features behind the high-dimensional data are displayed on the two-dimensional plane to understand the relationship between the parameters and the optimization objectives.Meanwhile,optimization range of the parameters is redefined,which provides a reference for the further research of the diagonal flow compressor.Finally,taking the compressor performance,including pressure ratio and efficiency,as the optimization objectives,a coupling platform of neural network-genetic algorithm is built for multi-objective optimization.(4)The performance and flow field of the mixed-flow compressor before and after optimization are compared and analyzed.The results show that the pressure ratio at the design point is increased by 1.6% and 5.5% after the first and second optimization,and the efficiency is increased by 4% and 4.2% respectively.The flow field of the mixed-flow compressor is improved to different degrees after the two optimizations.