| 基于引力球结构支持向量机多类算法的涡轮泵故障诊断 |
| |
| 引用本文： | 袁胜发,褚福磊.基于引力球结构支持向量机多类算法的涡轮泵故障诊断[J].宇航学报,2006,27(4):635-639. |
| |
| 作者姓名： | 袁胜发 褚福磊 |
| |
| 作者单位： | 1. 清华大学精密仪器与机械学系,北京,100084;江西理工大学机电工程学院,赣州,341000
2. 清华大学精密仪器与机械学系,北京,100084 |
| |
| 基金项目： | 国家自然科学基金;教育部跨世纪优秀人才培养计划 |
| |
| 摘 要： | 在涡轮泵故障诊断中，多类故障诊断是经常出现的问题。为提高多类故障诊断速度，在球结构支持向量机的基础上，提出一种引力球结构的支持向量机多类算法，该算法充分考虑样本分布疏松程度，经过试验优化分析得到最佳分类引力公式。用该算法和其他常用算法对涡轮泵仿真故障进行分类比较，结果表明基于引力球结构的支持向量机故障诊断算法学习速度快，诊断效果好。
|
| 关 键 词： | 涡轮泵 故障诊断 支持向量机 多类算法 |
| 文章编号： | 1000-1328（2006）04-0635-05 |
| 收稿时间： | 02 24 2005 12:00AM |
| 修稿时间： | 2005-02-242005-06-01 |
Fault Diagnosis of Turbo-pump Based on Gravitational Sphere Multi-Category Support Vector Machines |
| |
| YUAN Sheng-fa,CHU Fu-lei.Fault Diagnosis of Turbo-pump Based on Gravitational Sphere Multi-Category Support Vector Machines[J].Journal of Astronautics,2006,27(4):635-639. |
| |
| Authors: | YUAN Sheng-fa CHU Fu-lei |
| |
| Institution: | 1. Department of Precision Instruments and Mechanology, Tsinghua University, Beijing 100084, China; 2. School of Machinery and Power-generating Equipment Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China |
| |
| Abstract: | Multi-category faults are common in turbo-pump. In order to speed up the multi-category faults diagnosis, a gravitational sphere multi-category support vector machines based on sphere-structured support vector machines is presented in this paper. The distribution density of samples is fully taken into account in the new method, and the gravitational classification formulas are optimized by many analyses and experiments. This new method and several other normal methods are used in the diagnosis of simulated multi-category faults in turbo-pump, and result shows that the diagnosis time is shorter and the diagnosis correct rate is higher with the new method. |
| |
| Keywords: | Turbo-pump Fault diagnosis Support vector machines Multi-category |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
