Optimization of low-temperature multi-stage submersible pump based on blade load

The multi-stage submersible pump is a power conveyor for low-temperature media, which is conveyed by the rotating of centrifugal impellers. In this study, the impellers of a multi-stage submersible pump were optimized to improve the efficiency under the premise of the constant total blade load and head. Based on the analysis of performance and flow for each stage of the 18-stage submersible pump, the optimization scheme composed of the first stage, the middle stage, and the last stage was determined. The blade outlet angle, average blade wrapping angle, and blade wrapping angle difference were selected as optimization parameters through Plackett–Burman experimental design and significance analysis, and the blade profile was redrawn by changing the blade load distribution. The performance prediction model was built based on the Kriging response surface model, and then, the global optimal blade profile was found by non-dominated Sorting Genetic Algorithm II. The efficiencies of the 3-stage submersible pump and 18-stage submersible pump with optimized impeller increased by 2.35% and 2.01%, respectively. Under the design condition, the flow rate loss and pressure pulsation at the impeller outlet decreased and the stator–rotor interaction between the impeller and guide vane was weakened. This will lead to a reduction in unstable flow such as secondary flow and vortices, and an improvement of flow stability at the impeller outlet.