Multi-parameter coupled optimization design and hydrodynamic performance analysis of a full tubular pump impeller

To enhance the hydraulic performance and operational stability of a full tubular pump (FTP), this study conducts a systematic investigation and optimization of the impeller structural parameters based on computational fluid dynamics (CFD) simulations and multi-parameter optimization techniques. The number of blades and hub ratio were selected as key geometric variables, and their effects on the head and efficiency characteristics were analyzed using a single-factor approach. The results indicate that increasing the blade number improves the head but shifts the best-efficiency point toward a lower flow rate. In contrast, decreasing the hub ratio deteriorates the low-flow performance while slightly improving the high-flow performance, exhibiting a nonlinear hydraulic response. Subsequently, a surrogate model was constructed using a radial basis function (RBF) neural network (NN) within the Isight platform, and a multi-objective optimization was performed employing the multi-island genetic algorithm (MIGA). Under the constraint of an approximately constant specific speed, the blade outlet angle and blade cascade solidity were identified as the dominant factors influencing head and efficiency. The optimized impeller demonstrated a remarkable improvement in overall efficiency, a more uniform distribution of the blade boundary vorticity flux (BVF), and a significantly improved internal flow structure.