Multi-Surrogate Model Aided Bow Optimization of River and Coastal Connection Ship

Abstract In the simulation-based ship design, performance evaluation plays a crucial role. However, the process of evaluating the hydrodynamic performance of a ship can be both time-consuming and expensive, especially when relying on high-fidelity computational fluid dynamics (CFD) methods. To address this issue, this paper constructs an approach to hull form optimization design based on a multi-surrogate model. The study focuses on the bow optimization of River and Coastal Connection Ship. Two classic surrogate models are utilized to compare the multi-surrogate model’s approximation capabilities in four test functions. The numerical experimental results demonstrate that the multi-surrogate model exhibits superior forecast accuracy and robustness compared to the other models. Furthermore, a simulation-based optimization design model for the resistance performance of river and coastal connection ships is established, utilizing the aforementioned surrogate model and solved through an evolutionary algorithm. The result indicates that the multi-surrogate model facilitates the attainment of improved optimization design results. The optimization of the bow shape of the river and coastal connection ship through the multi-surrogate model effectively enhances the resistance performance of the target vessel. The proposed approach offers a more efficient and cost-effective method for evaluating the hydrodynamic performance of ships and optimizing their hull forms. The use of a multi-surrogate model significantly reduces the computational cost of the optimization process, while also improving the accuracy and robustness of the results.