| 基于光滑支持向量回归的汽轮机振动分析 |
| |
| 引用本文： | 张超,韩璞.基于光滑支持向量回归的汽轮机振动分析[J].电力系统及其自动化学报,2008,20(2):65-69. |
| |
| 作者姓名： | 张超 韩璞 |
| |
| 作者单位： | 1. 华北电力大学机械工程系暨电站设备状态监测与控制教育部重点实验室,保定,071003
2. 华北电力大学控制科学与工程学院,保定,071003 |
| |
| 基金项目： | 华北电力大学校科研和教改项目 |
| |
| 摘 要： | 传统的时间序列分析方法如统计和神经网络并不适用于复杂的非线性系统.为此,引入了支持向量回归(SVR)算法用以解决复杂时间序列的预测问题,在Matlab6.5环境下编程实现SVR算法,对汽轮机振动数据进行拟合和预测,并与神经网络方法进行比较.为了降低计算的复杂度,在SVR的基础上,采用光滑化算法,对汽轮机振动时间序列进行预测分析.仿真结果表明,光滑支持向量回归(SSVR)在预测性能方面明显优于神经网络.与SVR相比,SSVR具有更好的收敛速度和精度,适于进行复杂动态系统的时间序列分析.
|
| 关 键 词： | 机器学习 时间序列分析 支持向量回归 汽轮机状态监测 光滑化方法 |
| 文章编号： | 1003-8930(2008)02-0065-04 |
| 修稿时间： | 2007年4月19日 |
Analysis of Turbine's Vibration Based on Smooth Support Vector Regression |
| |
| ZHANG Chao,HAN Pu.Analysis of Turbine''''s Vibration Based on Smooth Support Vector Regression[J].Proceedings of the CSU-EPSA,2008,20(2):65-69. |
| |
| Authors: | ZHANG Chao HAN Pu |
| |
| Affiliation: | ZHANG Chao1,HAN Pu2(1.Department of Mechanical Engineering , the Key Laboratory ofCondition Monitoring & Control for Power Plant Equipments,Ministry ofEducation,North China Electricity Power University,Baoding 071003,China,2.School of Control Science , Engineering,North China ElectricityPower University,China) |
| |
| Abstract: | Traditional time series analysis methods such as statistics and neural network are not fit for complicated non-linear system.Therfore,the method of suppot vector regression(SVR) is introduced to resolve the prediction problem of complicated time series.The SVR method is achieved under the environment of Matlab 6.5,which has fitted and predicted the vibration data of turbline with the comparison with neural network.For the purpose of reducing calculation complexity,smooth arithmetic is imported based on SVR ... |
| |
| Keywords: | machine learning time series analysis support vector regression trubine state monitoring smooth method |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
