Application of knowledge-based engineering for ship optimisation design

Abstract

This paper presents a knowledge-based engineering (KBE) application for ship structural optimisation design. Ship structural design variables are discrete, involving a variety of design requirements and constraints; this leads to highly non-linear, multimodal, and convergence problems in structural optimisation. Ship optimisation development has ignored the application of design specifications, expert experience and other knowledge accumulated in the design process. The best solution is to combine an advanced optimisation technique with KBE, which can capture rich information to apply in optimisation loops. During the optimisation process, there are three steps for applying the KBE technique. The first step is to select the design variable values and set the design optimisation according to the knowledge base. The second step is to select the proper design process to support the optimisation design. The final step is to check the best optimisation solutions according to the KBE techniques. This approach was applied to the example of a watertight transverse bulkhead. The optimisation design of the bulkhead shows that the optimised weight of the structure is lower than that for the design without KBE, while meeting the required constraints. The design process with KBE therefore ensures reasonable performance of this structure while achieving the minimum weight objective. In addition, it combines a structural parameterised model with mathematical optimisation, which provides a shortcut for designers with less experience in structural design. In conclusion, this proposed methodology combines knowledge from different sources and applies it to the structural optimisation design of ships, which can improve the design level and reduce designer dependence on experience.

Prior knowledge‐infused neural network for efficient performance assessment of structures through few‐shot incremental learning

“…KBAI has been used in engineering applications as diverse as product design, and industrial equipment fault diagnosis (Sakaguchi & Matsumoto, 1983;Sapuan, 2001). Yang et al (2014) collected previous shipbuilding experience and design standards, and specifications as a candidate library to facilitate the optimization of a watertight bulkhead in a ship. Li et al (2020) established a knowledge-based neural network for protecting the power transformer by letting a neural network learn plenty of cases judged by power experts.…”

“…Yang et al. (2014) collected previous shipbuilding experience and design standards, and specifications as a candidate library to facilitate the optimization of a watertight bulkhead in a ship. Li et al.…”

Development of Knowledge-Based Engineering System for Structural Size Optimization of External Fixation Device

“…The integration of this software, together with skeleton-based generative modeling methods [1] and algorithms for optimization, as well as systems for knowledge-based engineering (KBE), can be developed. KBE systems have integrated tools for parametric CAD modeling, numerical analysis using finite element modeling (FEM) and structural optimization of various designs [2,3].…”

Knowledge-Based Engineering–based method for containership lashing bridge optimization design and structural improvement with functionally graded thickness plates

“…The optimization model was established complying with the result of knowledge discovery, and the robustness optimization parameters were attained. Yang et al 28 proposed a KBE methodology for the ship hull's structure optimization design. Their methodology can achieve the specification-based optimization and help engineering designers learn the rules and regulations during the ship optimization process.…”