| 大型回转机械故障的组合网络识别方法 |
| |
| 引用本文： | 韩震宇,屈梁生.大型回转机械故障的组合网络识别方法[J].西安交通大学学报,1996,30(10):70-76,82. |
| |
| 作者姓名： | 韩震宇 屈梁生 |
| |
| 作者单位： | 西安交通大学 |
| |
| 摘 要： | 神经网络通过对故障样本的学习后，对未知的故障的样本具有较高的正确识别率，从神经网络对故障的识别检验结果中发现，神经网络对单一故障的分类与组合故障的分类相差较大，文中分析了产生这现象的原因，并利用组合网络来克服单一网络对组合故障分类精度不够高的缺陷。
|
| 关 键 词： | 故障诊断 转子机械 模式识别 神经网络 |
THE METHOD OF MULTI NETWORK RECOGNITION OF LARGE SIZED ROTATING MACHINERY FAULTS |
| |
| Han,Zhenyu,Qu,Liangsheng.THE METHOD OF MULTI NETWORK RECOGNITION OF LARGE SIZED ROTATING MACHINERY FAULTS[J].Journal of Xi'an Jiaotong University,1996,30(10):70-76,82. |
| |
| Authors: | Han Zhenyu Qu Liangsheng |
| |
| Abstract: | Neural networks can correctly recognize an unknown fault after it has been trained with a group of faults examples. From the test results of faults recognition,it is found out that classification effect of neural networks has apparent difference between single fault and combining fault. Theoretical analysis to the above phenomena is given and a type of network called multi network is built up to improve classification results of neural networks to combining fault. |
| |
| Keywords: | fault diagnosis rotating machinery pattern recognition neural networks |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
