Multi-objective optimization of the flush-type intake duct for a waterjet propulsion system

Abstract

The intake duct is an important component of waterjet propulsion system and its hydraulic performance is directly connected with the capability of the waterjet pump and even the propulsion performance of the system. In order to optimize the flush-type intake duct, the present paper proposes a multi-objective optimization system including Design of Experiments (DOE), Computational Fluid Dynamics (CFD), three-dimensional parametric design, approximate model, the modified Non-dominated Sorting Genetic Algorithm-II (NSGA-II) and a Technique for Ordering Preferences by Similarity to Ideal Solution (TOPSIS). The nonuniformity and perpendicularity of outflow, and hydraulic efficiency for the intake duct are treated as the optimization objectives with four geometrical parameters as the design variables. Local sensitivity analyses indicate that the optimization objectives are significantly affected by the inclination angle and slightly affected by the lip vertical distance of the intake duct. During IVR = 0.3–0.8, the outflow quality and hydraulic efficiency of the optimized intake duct are greatly improved. The nonuniformity after optimization decreases by 27.8% and perpendicularity increases by 3.07% at the design condition of IVR = 0.7 and V s = 19.49 m/s. Based on the flow at the outlet plane of intake duct, the pressure distribution after optimization is very uniform and the tangential velocity is small without obvious secondary flow. During the application to a mixed-flow waterjet propulsion system at various navigation speeds, the nonuniformity of the optimized intake duct decreases by 20.4% and perpendicularity increases by 4.11% on average, demonstrating that the optimized intake duct performs a better outflow quality. It is also noted that the non-uniform flow from the intake duct mainly affects pressure distribution on the suction surface of impeller blade and causes different shaft power. • A multi-objective optimization system is proposed by incorporating DOE, CFD, three-dimensional parametric design, approximate model, the modified NSGA-II and TOPSIS. • The three optimization objectives are significantly affected by the inclination angle and slightly affected by the lip vertical distance. • Non-uniform suction flows mainly affect the flow field on the impeller suction surfaces and cause the difference in the shaft power.