| 基于粗糙集和信息熵的BP神经网络故障诊断 |
| |
| 引用本文： | 王发兴,沈永红.基于粗糙集和信息熵的BP神经网络故障诊断[J].计算机辅助工程,2009,18(3):78-82. |
| |
| 作者姓名： | 王发兴 沈永红 |
| |
| 作者单位： | 1. 南京邮电大学,通达学院,南京,210046
2. 天水师范学院,数学与统计学院,甘肃,天水741000 |
| |
| 摘 要： | 针对粗糙集只能处理量化数据,容错和推广能力较差的缺点以及BP神经网络的维数灾难问题,提出1种基于信息熵的粗糙集属性离散化方法. 该方法利用粗糙集对属性进行约简,解决BP神经网络的维数灾难问题,并将BP神经网络用于模式分类补偿粗糙集属性约简用于模式分类时的不足. 实例分析表明该方法具有较好的故障诊断效果.
|
| 关 键 词： | 粗糙集 信息熵 离散化 神经网络 故障诊断 |
| 收稿时间： | 4/6/2009 4:26:30 PM |
| 修稿时间： | 8/28/2009 3:07:39 PM |
Fault Diagnosis of BP Neural Network Based on Rough Set and Information Entropy |
| |
| wangfaxing and shenyonghong.Fault Diagnosis of BP Neural Network Based on Rough Set and Information Entropy[J].Computer Aided Engineering,2009,18(3):78-82. |
| |
| Authors: | wangfaxing and shenyonghong |
| |
| Affiliation: | 1. Tongda College;Nanjing Univ. of Post & Telecommunications;Nanjing 210046;China;2. School of Mathmatics & Statistics;Tianshui Normal Univ.;Tianshui Gansu 741000;China |
| |
| Abstract: | Rough set can only process quantization data,and the ability of fault-tolerant and generalization is weak,meanwhile,BP neural network has the dimension disaster problem. So a rough set attribute discretization method based on information entropy is proposed. The attribute is reduced to solve the dimension disaster problem of BP neural network. BP neural network is used to deal with the pattern classification to make up for the shortcoming brought by attribute reduction. The example result shows that the met... |
| |
| Keywords: | rough set information entropy discretization neural network fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机辅助工程》浏览原始摘要信息 |
|
点击此处可从《计算机辅助工程》下载全文 |
