Objective  As creating finite element models for the finite element analysis (FEA) of ship structures is time-consuming and highly dependent on engineers' experience, this paper proposes a knowledge-based fast finite element modeling method to solve this problem.

  Methods  The proposed method makes full use of the existing model data of the CAD system and shares model data between the CAD and CAE systems through model features. First, based on expert knowledge, the CAD geometric features are marked with engineering semantics to generate CAD information models. Second, the feature reduction of the CAD information model is carried out based on the constructed knowledge rules. Finally, the neutral file XML is used to construct a knowledge-based data transfer and matching method which allows the geometric features and attribute information to be matched with the pre-constructed CAE parameterized template, thereby speeding up the finite element modeling of the ship structure.

  Results  Taking the structural strength analysis of a bulk carrier of 66000 tons under typical loading conditions as an example, this approach can realize the automatic extraction, labeling and model reduction of the geometric model, boundary conditions and other information, and quickly complete the finite element modeling and analysis of the structure through the knowledge-based feature matching rules, effectively reducing data loss in the transmission process and realizing the reuse of knowledge.

  Conclusions  The proposed fast finite element modeling method with embedded knowledge reuse rules improves the conversion efficiency of data, information and knowledge between the CAD and CAE systems, and has good universality and expansibility in practical engineering.