Energy efficiency and operational stability optimization of mixed-flow pumps based on hydrodynamic parameter control

The fluid in a mixed-flow pump exhibits complex flow behavior due to the coupled effects of radial and axial forces, which poses a significant challenge to the further improvement of its performance. This study proposes an expert-independent design method for mixed-flow pumps, which enhances operational efficiency and stability by regulating flow-field hydrodynamic parameters. During the optimization process, three key measures are integrated: (1) a steady-state index at 0.5Qdes is introduced to ensure that the optimization duration aligns with engineering requirements; (2) the entropy weight method is incorporated to enhance optimization robustness; and (3) a neural network-coupled evolutionary algorithm is employed to decouple the optimization process from expert experience. The results show that the weighted efficiency of the optimized model increased by 3.92%, while the pressure pulsation amplitude at the main frequency decreased by an average of 42.5%, reaching a maximum reduction of 72.8%. A further analysis of energy losses revealed that the redistribution of hydrodynamic parameters on the blade surface effectively suppressed the energy loss induced by jet-wake evolution at the blade trailing edge and reduced the intensity of rotor–stator interactions. This study offers reliable guidance for improving the performance of turbomachinery.