Structural sensitivity analysis and multi-objective optimization of liquid nozzle performances with the internal cavitation flow

Abstract

The transient cavitation flow within the nozzle of a high-pressure micro-size injector is characterized by high turbulence and multi-phase. The inception and development of this flow are markedly affected by the nozzle geometries, which exert a direct influence on the injector nozzle reliability and the spray characteristics. This study presents a comprehensive evaluation of nozzle performance in terms of flow coefficient, spray patterns and cavitation erosion, with a particular focus on the sensitivity of the nozzle parameters using a combination of optical experiment, numerical simulation, and machine learning. According to the sensitivity analysis, the parameters of hole taper coefficient, needle lift and the combined needle lift and hole height have the most significant influence on the nozzle performances. Furthermore, a multi-objective optimization of the parameters of a nine-hole marine nozzle was conducted based on the Radial Basis Function neural network and the Non-dominated Sorting Genetic Algorithms-II genetic algorithm. Results present that the flow coefficient of the fitness-optimal nozzle shows an approximately 25% increase, while the near-nozzle liquid phase diffusion angle simultaneously demonstrates a marked increase exceeding 70%, and the cavitation erosion risk is sharply reduced.