| 对ANN技术表达船体曲线算法的改进 |
| |
| 引用本文： | 谭家华,包丛喜,Bao Cong-xi,TAN Jia-hua.对ANN技术表达船体曲线算法的改进[J].船舶工程,2001,3(1):18-20. |
| |
| 作者姓名： | 谭家华 包丛喜 Bao Cong-xi TAN Jia-hua |
| |
| 作者单位： | 上海交通大学船舶与海洋工程学院 |
| |
| 摘 要： | 应用人工神经元网络（ANN）技术表达船体曲线。根据问题性质，选用小波基作为前向单层神经网络的神经元激励函数，结合逐层学习（OHLO）算法对一艘3．6万吨散货船的后半体进行了表达。编程运算结果表明，该方法速度较传统的BP算法有较大提高。
|
| 关 键 词： | 人工神经网络 BP算法 改进 激励函数 船体曲线 表达 |
Improvement on Algorithm of Ship Lines with ANN Technology |
| |
| TAN Jia-hua.Improvement on Algorithm of Ship Lines with ANN Technology[J].Ship Engineering,2001,3(1):18-20. |
| |
| Authors: | TAN Jia-hua |
| |
| Abstract: | By Applying technology of artificial neural networks(ANN)to express hull lines,a wavelet function is chosen as the activation function of the neurons of the progressive single-layer network in combination with the algorithm of optimizing hidden layer output(OHLO).The method is used to express the aft part of the hull of a 36 000 t bulk carrier.The results show that the calculation speed of this improved algorithm is much quicker than that of the conventional BP Algorithm. |
| |
| Keywords: | Artificial neural networks(ANN) BP algorithm improvement Activation functions Hull lines |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《船舶工程》浏览原始摘要信息 |
| 点击此处可从《船舶工程》下载免费的PDF全文 |
