선박의 성능 요소 추정을 위한 신경망의 실용화 연구

Abstract

In the initial ship design stage, performance predictions are generally carried out before and after the hull form design. The former is based on the main dimensions and power information, and the latter is based on the geometry of the hull form and propeller. This paper deals with the practical application of neural networks for the predi ction of a ship's performance factors before and after the hull form design. For this, the hull form parameters that affect the performance are studie d, and an optimal neural network structure based on the SSMB database is constructed. By comparing the results predicted by neural networks and the model test results, we confirmed that neural networks can be applied to practically evaluate the performance in the initial ship design stage.