| 齿轮故障诊断中的选择性集成技术 |
| |
| 引用本文： | 刘天羽,张延迟.齿轮故障诊断中的选择性集成技术[J].上海电机学院学报,2009,12(2):118-120,124. |
| |
| 作者姓名： | 刘天羽 张延迟 |
| |
| 作者单位： | 上海电机学院,电气学院,上海,200240 |
| |
| 基金项目： | 国家自然科学基金资助项目，上海市青年科技启明星计划项目(B类)，上海高校选拔培养优秀青年教师科研专项基金资助项目，上海市教育委员会重点学科 |
| |
| 摘 要： | 齿轮故障数据的建模因为数据复杂有一定的挑战性.提出将选择性集成学习技术引入到该领域,并将聚类算法用于选择性神经网络集成学习技术.对已有的选择性集成学习技术进行改进,构建了一个基于聚类的选择性神经网络.应用实际数据进行实验,结果表明新算法提高了对齿轮故障诊断的精度,同时诊断效率比基于遗传算法的选择性神经网络集成算法有了显著提高.
|
| 关 键 词： | 聚类算法 集成学习 齿轮故障诊断 |
Clustering Algorithm Based Selective Ensemble Techniques for Gear Fault Diagnosis |
| |
| LIU Tianyu,ZHANG Yanchi.Clustering Algorithm Based Selective Ensemble Techniques for Gear Fault Diagnosis[J].JOurnal of Shanghai Dianji University,2009,12(2):118-120,124. |
| |
| Authors: | LIU Tianyu ZHANG Yanchi |
| |
| Affiliation: | School of Electric;Shanghai Dianji University;Shanghai 200240;China |
| |
| Abstract: | Fault diagnosis is a critical problem in the mechanical field,here CLUSEN(CLUstering Algorithm Based Selective ENsemble) is proposed to predict the fault of gears.The procedure of CLUSEN can efficiently remove the redundancy of individual learning and help improve the diversity of ensemble methods.Experimental results on a set of gear fault data show that the CLUSEN improves the classification accuracy and is more efficiency than the GASEN method proposed before. |
| |
| Keywords: | clustering algorithm ensemble learning gear fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
