Chunjin Li ^* Hongrui Xu Tao Jiang Chengyu Liu

• Jiangsu University of Science and Technology, Zhenjiang, Jiangsu Province, China

To enhance the efficiency of the ship's propulsion system, our focus lies on the PBCF of Energy-Saving Device (ESD). We present an optimization strategy for the PBCF using a BPNN in conjunction with an intelligent optimization algorithm. The positional parameters of the PBCF relative to the propeller are expressed parametrically, and the sample space of the PBCF parameters is derived through Latin hypercube sampling. Subsequently, we establish a BPNN model with design parameters as input and hydrodynamic performance, calculated by Computational Fluid Dynamics (CFD), as output. Simultaneously, the neural network-trained approximate model serves as the objective function for the Northern Goshawk Optimization (NGO) algorithm, enhancing the iteration speed. After determining the positional parameters of the PBCF, the optimal airfoil type is selected from seven types based on its superior performance. Ultimately, we compare the flow field distributions of the optimized fins with the original propeller using visualization techniques. The results indicate that the optimized fins effectively absorb the tangential kinetic energy specifically flowing through the root section of the propeller blade, diminish the hubcap pressure difference, and suppress the generation of post-hub vortices. Under design conditions, the optimized fins contribute to a 1.5% enhancement in the propulsion system's efficiency.