激光诱导击穿光谱的重叠峰分辨的方法
于海斌; 张博; 孙兰香; 杨志家; 辛勇; 丛智博
2013-05-01
专利国别中国
专利号CN103076308B
专利类型发明授权
产权排序1
权利人中国科学院沈阳自动化研究所
其他题名Laser-induced breakdown spectroscopy overlapped peak resolution method
中文摘要本发明涉及激光诱导击穿光谱的数据预处理分析领域,具体是一种基于合理的谱峰数学模型,通过无约束最优化算法计算得到重叠峰分辨之后的谱峰相关信息的方法。本发明基于谱峰数学模型,通过无约束最优化算法计算重叠峰分辨之后的谱线相关信息,进而分离重叠谱线。本发明不仅可以根据实际谱线的情况进行选择计算,而且仅需要很少的参数就可以确定其谱线信息,提高了样品组分分析的性能和准确性;只需要通过直观的选择谱峰位置就可以确定所有的最优化算法所需要的初始化参数,而且方法计算速度快,易于实施;直观的比较来选择确定最终的处理结果;提高物质组分量化分析的准确度。
是否PCT专利
英文摘要The invention relates to the field of laser-induced breakdown spectroscopy data pretreatment analysis. Specifically, with the method, based on a reasonable spectral peak mathematical model, with an unconstrained optimization algorithm, spectral peak related information after overlapped peak resolution is obtained by calculation. According to the method, based on a spectral peak mathematical model, and with an unconstrained optimization algorithm, spectral line related information after overlapped peak resolution is obtained by calculation, and overlapped spectral lines are separated. According to the invention, selective calculation can be carried out according to actual spectral line situations, and spectral line information can be determined with a small amount of parameters. Therefore, sample component analysis performance and accuracy are improved. Spectral peak position is visually selected, and initialization parameters needed by all optimization algorithms can be determined. The method also has the advantages of high calculation speed and easy implementation. With the method, a final processing result is selected by visual comparison, such that substance component quantitative analysis accuracy is improved.

[en;CN103076308A]

公开日期2014-09-17
申请日期2011-10-25
语种中文
专利申请号CN201110328277.3
专利代理沈阳科苑专利商标代理有限公司 21002
内容类型专利
源URL[http://ir.sia.ac.cn/handle/173321/15513]  
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
于海斌,张博,孙兰香,等. 激光诱导击穿光谱的重叠峰分辨的方法. CN103076308B. 2013-05-01.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

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。


©版权所有 ©2017 CSpace - Powered by CSpace