Optimum Design of the Underwater Discharge System Based on Surrogate Modeling

Abstract

Underwater vehicles are equipped with a device for discharging projectiles, called the underwater discharge system. Because these vehicles have limited space, many studies have attempted to reduce the volume of the underwater discharge system while improving its performance. However, owing to the characteristics of the system, the occupied volume and performance conflict with each other. Therefore, minimizing the occupied volume while satisfying the required performance criteria is the most significant requirement in the design stage. To address this, an optimization problem for the underwater discharge system was mathematically formulated and solved using multi-objective genetic algorithms. Additionally, we proposed a method of using the surrogate modeling technique to find an optimum solution quickly. Linear regression, radial basis functions, and artificial neural networks were compared to generate an appropriate surrogate model. To evaluate the applicability of the proposed method, it was applied to optimize the design of an underwater discharge system. The results demonstrate that the proposed method can be effectively utilized to obtain the optimum design of an underwater discharge system within an acceptable time.