Reducing Quantitative Fluctuation of Laser-Induced Breakdown Spectroscopy by Kalman Filtering
文献类型：会议论文
| 作者 | Sun LX(孙兰香) ; Cong ZB(丛智博); Xin Y(辛勇); Qi LF(齐立峰); Li Y(李洋)
|
| 出版日期 | 2013 |
| 会议名称 | 2013 2nd International Conference on Measurement, Instrumentation and Automation |
| 会议日期 | April 23-24, 2013 |
| 会议地点 | Guilin, China |
| 关键词 | Fluctuation Kalman Filter (KF) Laser-Induced Breakdown Spectroscopy Quantitative Analysis |
| 页码 | 243-247 |
| 中文摘要 | Laser-induced breakdown spectroscopy (LIBS) is excellent for its potential of online compositional analysis. Large signal fluctuation is the major obstacle of LIBS for quantitative analysis application. A kalman filtering method is proposed to estimate the elemental concentration and smooth the quantitative results. The system state model and the measurement model are deduced. The relation matrix between the measured values and system state is estimated based on calibration curve built on some standard samples, and the measurement noise matrix is estimated by the variance of multiple measurements of the spectral intensity. In order to make Kalman filter follow the changes of elemental concentration, the initial value of the covariance matrix of estimation error is reset as a certain rule. The experimental results show that the Kalman filtering method can greatly reduce the fluctuation of quantitative results and improve the measurement accuracy. |
| 收录类别 | EI ; CPCI（ISTP） |
| 产权排序 | 1 |
| 会议录 | Applied Mechanics and Materials
 |
| 会议录出版者 | Trans Tech Publications Ltd |
| 会议录出版地 | Zurich-Durnten, Switzerland |
| 语种 | 英语 |
| WOS记录号 | WOS:000329080300044 |
| 源URL | [http://ir.sia.cn/handle/173321/12430]  |
| 专题 | 沈阳自动化研究所_工业控制网络与系统研究室 |
| 推荐引用方式 GB/T 7714 | Sun LX,Cong ZB,Xin Y,et al. Reducing Quantitative Fluctuation of Laser-Induced Breakdown Spectroscopy by Kalman Filtering[C]. 见:2013 2nd International Conference on Measurement, Instrumentation and Automation. Guilin, China. April 23-24, 2013. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。