Study on design performance evaluation model for ship based on genetic-RBF neural network

摘要：

Design performance evaluation for ship has a great importance to improve the quality of ship design and manufacture ship with high quality.In the paper,we employ a novel hybrid method including RBF neural network and genetic algorithm to evaluate design performance of ship.Genetic algorithm is applied to choose the suitable parameters of RBF neural network.The main affection parameters of design performance evaluation for ship are introduced.And then,genetic-RBF neural network evaluation model is introduced.Finally,the experimental data of design performance evaluation for ship is applied to study the evaluation ability of G-RBF neural network.The experimental results show that G-RBF neural network can gain better design performance evaluation results for ship than traditional RBF neural network.

关键词： genetic-RBF neural network design evaluation evaluation parameters ship genetic algorithm

作者: Yao Jing-zheng Han Duan-feng

作者单位: Multihull Ship Technology Key Laboratory of Fundamental Science for National Defence Harbin Engineer College of Shipbuilding Engineering Harbin Engineering University Harbin 150001,China

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 44-47

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）