The invention discloses a hull profile design method based on a convolutional neural network, and relates to the technical field of ships. The method comprises the following steps: carrying out data standardization processing on ship type geometric data of a sample ship to construct a geometric feature array of the sample ship; using the geometrical characteristic array of the sample ship as input, and using the hydrodynamic performance data as output; performing training based on a convolutional neural network to obtain a ship type performance prediction model; using the ship type performanceprediction model for carrying out hydrodynamic performance data prediction, data mining and sensitivity analysis on a target ship with known ship type geometrical data; optimizing the ship body molded line of the target ship according to the sensitivity analysis result, so that the optimization efficiency can be improved. According to the method, the ship form can be rapidly designed, an effective parameter-free ship geometric expression and analysis method is provided, data difference caused by a parameterization process in ship form line analysis is avoided, and the method is high in universality and suitable for most mainstream ship forms.
本发明公开了一种基于卷积神经网络的船体型线设计方法,涉及船舶技术领域,该方法对样本船舶的船型几何数据进行数据标准化处理构建得到样本船舶的几何特征数组,将样本船舶的几何特征数组作为输入、水动力性能数据作为输出,基于卷积神经网络训练得到船型性能预报模型,利用船型性能预报模型可以对船型几何数据已知的目标船舶进行水动力性能数据的预报、数据挖掘和敏感性分析,根据敏感性分析结果对目标船舶的船体型线优化可以提高优化效率,实现船型快速设计,而且该方法提供了一种有效的无参数船体几何表达和分析方法,避免了船体型线分析中的参数化过程带来的数据差异,通用性较高,适用于大部分主流船型。
Hull profile design method based on convolutional neural network
一种基于卷积神经网络的船体型线设计方法
04.09.2020
Patent
Elektronische Ressource
Chinesisch
Company Profile of Advanced Multi-Hull Design
British Library Conference Proceedings | 1994
|Hull profile characteristic parameter extraction method
Europäisches Patentamt | 2020
|HULL STRUCTURE MANUFACTURING METHOD, HULL STRUCTURE DESIGN METHOD, AND HULL STRUCTURE DESIGN PROGRAM
Europäisches Patentamt | 2024
|Design of FPGA Based Convolutional Neural Network Co-Processor
British Library Online Contents | 2018
|