论文部分内容阅读
激光诱导击穿光谱(LIBS)技术具有快速、非接触、无需制样等特点,适合应用于转炉钢水成分的在线分析。由于转炉终点可由Si、Mn含量和温度来判定,因此提出了钢水成分中Si和Mn的LIBS定性分析方法。通过光谱仪采集激光激发的光谱,经过预处理和寻峰等操作后,以原子光谱数据库(NIST)为参考标准,找出Si和Mn对应的特征谱线波长和光谱强度,利用支持向量机(SVM)强大的分类功能和采集到的245组数据中的210组学习得到支持向量分类(SVC)模型,利用SVC模型预测这245组数据,结果证明该模型的准确率为98%以上,将其应用在相同实验条件的情况下,会大大减少LIBS定性分析时间。
Laser Induced Breakdown Spectroscopy (LIBS) has the characteristics of fast, non-contact, sample-free and so on. It is suitable for on-line analysis of molten steel components in converter. Since the end of the converter can be judged by Si, Mn content and temperature, a qualitative analysis method of LIBS of Si and Mn in molten steel is proposed. The spectrum of laser excitation was collected by spectrometer. After preprocessing and peak-finding operation, the spectral line and spectral intensity of Si and Mn were found by using the atomic spectroscopy database (NIST) as a reference standard. Support Vector Machine (SVM) ) Strong classification function and the collected 245 groups of data 210 groups to learn the support vector classification (SVC) model, the use of SVC model to predict these 245 sets of data, the results show that the accuracy of the model was 98%, its application In the case of the same experimental conditions, LIBS qualitative analysis will be greatly reduced.