Exploration of performance optimization strategies for micro-centrifugal compressors in hydrogen fuel cells: A synergistic analysis combining one-dimensional design and three-dimensional flow fields

The design and performance optimization of micro-centrifugal compressors, serving as crucial components within hydrogen fuel cell systems , are pivotal for enhancing the overall energy conversion efficiency . Current research efforts have predominantly concentrated on the three-dimensional (3D) optimal design of blades, with comparatively limited exploration into the one-dimensional (1D) parameters that directly influence aerodynamic performance . Therefore, this study is dedicated to conducting a comprehensive and synergistic analysis of both the 1D design and 3D flow field of micro-centrifugal compressor impellers . The initial 1D parameter design was accomplished using an empirical loss model, followed by a signal-to-noise ratio (SNR) variance analysis to identify the optimal combination of 1D parameters for pressure ratio and isentropic efficiency . Furthermore, a flow field deviation analysis of impeller performance was conducted to uncover the reasons behind compressor performance limitations, particularly focusing on flow separation and blockage phenomena occurring on the blade suction surface under off-design conditions. Subsequently, the blade suction surface was optimal using the NSGA-Ⅱ algorithm. The optimization results demonstrate that, compared to the initial design, the optimal impeller exhibits an average increase in pressure ratio by 4.13 % and an improvement in isentropic efficiency by 3.11 % at the design point. At a rotational speed of 7000 rpm, the isentropic efficiency is enhanced by 3.00 %, and at 9000 rpm, the pressure ratio sees an average increase of 4.04 %. • The initial design of one-dimensional parameters is completed based on an empirical loss model. • Using variance analysis of the signal-to-noise ratio, seeking the optimal combination of one-dimensional parameters. • A flow field deviation analysis is conducted on the impeller performance to uncover the underlying causes of compressor performance limitations. • Optimization using the NSGA-II algorithm results in reduced flow separation on the leading edge and suction surface of the blades. • The algorithm-optimized model exhibits average improvements of 4.13 % in pressure ratio and 3.11 % in isentropic efficiency, respectively.