Intelligent design method of hull form based on knowledge-driven approach

In the realm of ship design, the hull form plays a crucial role in determining a ship's hydrodynamic performance. However, existing hull form design methods have significant drawbacks, such as lengthy design cycles and poor reusability of design knowledge. This not only increases design costs but also hinders innovation in ship design. Therefore, there is an urgent need to develop a rapid and intelligent hull form design method.

To address these issues, this research focuses on the Yangtze River cargo ships and leverages the knowledge engineering principles. A comprehensive knowledge base is constructed, including a knowledge graph of hull form parameters, parent ship cases, and design rules; the knowledge graph depicts the mapping relationships between various hull form parameters and total resistance performance under different conditions; the parent ship cases provide practical references, while the design rules summarize how geometric parameters affect ship resistance. Based on this knowledge base, a hybrid reasoning model for hull form design has been developed. This model integrates case-based reasoning (CBR), model-based reasoning (MBR), and rule-based reasoning (RBR) to fully utilize the knowledge base and derive more accurate hull form design parameters.

This research uses the hull form design of a 13 000 DWT bulk carrier operating on the Yangtze River as a case study. A set of hull form geometric parameters is obtained through intelligent reasoning based on the knowledge base. Taking into account spatial layout constraints and parameter interactions, a parametric hull form design scheme is generated. By inputting this scheme into the parametric geometric model, a geometric model of the ship is created. The total resistance of both the initial ship type and the optimized hull form is calculated using STAR-CCM +, a CFD software based on fully viscous flow. The results show that, compared with the initial ship type, the optimized hull form achieves a 3.06% reduction in total resistance. Furthermore, analysis reveals that the reduction is primarily attributed to a decrease in residual resistance. The optimized bow and stern shapes contribute to a more uniform pressure distribution, thereby reducing pressure differentials and overall resistance.

In conclusion, the knowledge-driven intelligent method for hull form design proposed in this study can significantly improve the efficiency and quality of hull form design. It can quickly generate hull form schemes derived through reasoning that meet design requirements, demonstrating strong potential for engineering applications.