| 船舶柴油机NOx排放特性神经网络预测中的试验设计 |
| |
| 引用本文： | 尹自斌,杨国豪.船舶柴油机NOx排放特性神经网络预测中的试验设计[J].舰船科学技术,2010,32(3):32-35. |
| |
| 作者姓名： | 尹自斌 杨国豪 |
| |
| 作者单位： | 集美大学,轮机工程学院,福建,厦门,361021 |
| |
| 基金项目： | 福建省青年人才创新项目(2008F3075);;福建省科技创新平台项目资助(2009H2006);;集美大学科研基金项目(ZQ2006031);;福建省教育厅科技项目(JA09153) |
| |
| 摘 要： | 研究船舶柴油机NOx排放特性神经网络预测中的学习样本选取试验设计方法。根据用于主机的船舶柴油机可能持续运行范围的工况变化特点,提出采用功率因素变边界的均匀设计法进行试验设计选取样本,并验证了其可行性。研究结果表明,变边界均匀设计法选取的样本用于神经网络训练,预测精度明显高于随机样本选取法。4位级变边界均匀设计法选取的样本训练得到的神经网络模型,NOx排放浓度预测误差小于3.8%,NOx比排放预测误差小4.5%。
|
| 关 键 词： | 船舶柴油机 NOx排放 神经网络 均匀设计 学习样本 |
Test design on prediction of NOx-emission from marine diesel engine using neural network |
| |
| YIN Zi-bin,YANG Guo-hao.Test design on prediction of NOx-emission from marine diesel engine using neural network[J].Ship Science and Technology,2010,32(3):32-35. |
| |
| Authors: | YIN Zi-bin YANG Guo-hao |
| |
| Institution: | Marine Engineering College;Jimei University;Xiamen361021;China |
| |
| Abstract: | Test design on selection of the learning samples for prediction of NOx-emission from marine diesel engine using neural network was studied. According to the variation characteristics of the operating condition of marine propulsion diesel engines in service,an improved uniform design method which considered the changes on power factor boundary was proposed to design test and select learning samples of neural networks. The results show that the forecasting precision of neural networks model trained with sampl... |
| |
| Keywords: | marine diesel engine NOx-emission neural networks uniform design learning samples |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
