Improving AUV Thruster Performance: Optimizing Hydrodynamic Nose-Cone Section with Automated Genetic Algorithm and Computational Fluid Dynamics Integration

Abstract

To maximize the speed of Autonomous Underwater Vehicles (AUVs), it is crucial to carefully consider the design of the conical nose, situated at the forefront of the underwater thruster. The study introduces a method that significantly increases the thruster-generated output velocity by optimizing the curvature of the cone's spline and the length of the conical nose geometry using a Genetic Algorithm (GA). This optimization process facilitates real-time interactions between the GA and Computational Fluid Dynamics (CFD)-based simulations, permitting dynamic adjustments to the design parameters. The GA utilizes velocity metrics derived from fluid flow analysis in CFD simulations as a criterion for fitness. Informed by this criterion, the GA generates new design iterations, which undergo further CFD simulations for continued evaluation and refinement. This iterative process identifies promising combinations of conical nose shape and length for improving thruster output velocity. Through this analysis two intermediate nose cone prototypes were generated. The study found that modifications to the nose cone geometry increased the thruster output velocity from a baseline of approximately 5.05 m/s to a maximum value of 5.41 m/s among the evaluated configurations, representing a 7.13% improvement. This configuration was identified as the best-performing nose cone geometry in terms of output velocity within the investigated design space.