Abstract: Objectives This study addresses the limitation of existing deep learning-based ship cross-section topology optimization methods, which can only be applied to single structural sections. A fast topology optimization method for general ship cross-sections is proposed, which enables the direct application of neural network prediction results in engineering analysis. Methods The proposed methodology employs an automated parametric modeling and computation technique to construct a large-scale, structurally diverse dataset of ship cross-sectional static analysis and topology optimization results for deep supervised learning. This enables the trained neural network to rapidly derive rational topology optimization configurations for various ship structural cross-sections. Furthermore, addressing the challenge of directly applying neural network predictions to engineering analysis, an algorithm was developed for the automatic reconstruction of finite element models from the binarized density tensor output by the neural network. This overcomes the limitations inherent in model reconstruction using the element removal method, thereby providing a foundation for verifying the consistency of mechanical performance between network predictions and traditional iterative calculation results. Results Experimental results demonstrate that applying the proposed method to predict the topological configuration of ship cross-sections with various structures reduces computation time by two orders of magnitude, with an average prediction accuracy exceeding 90%. Sampling inspection results indicate that the finite element models reconstructed based on the network predictions avoid stress concentration effects, and the deviation in mechanical performance compared to traditional iterative calculations is less than 3%, further verifying the reliability of the proposed method. Conclusions The proposed method provides a general solution for the rapid topology optimization of ship cross-sections, reduces ship design costs, and possesses significant engineering value.