用脉冲分布分析发射光谱技术快速表征夹杂物—最新研究进展

Rapid inclusion characterisation by pulse distribution analysis optical emission spectroscopy-recent development

本文旨在进一步开发和验证脉冲分布分析发射光谱法快速用于钢铁生产中夹杂物的表征。本实验主要对一些操作参数如:火花能量,火花频率和门控时间(TRS)进行研究。结果显示低火花能量有利于检出限的提高,但是由于样品量较小不利于测量统计。研究证明测量频率对分析结果没有显著影响,但是一些现有仪器由于电子器件限制不能处理脉冲分布分析模式中最高频率300Hz以上的火花。门控时间对于金属含量以上硅离群值的检出限有微幅改善。本研究还发现因为强度的不对称分布使得测定"假"离群值有风险,这就导致了检测离群值更先进算法的发展,提高了方法的准确性。另一个限制是颗粒物数量密度不可超过约10 000夹杂物/mm3,以便能够有效的检测单个夹杂物。本文提出了可以克服这一限制的方法,但该方法还没有被评估。通过参考方法验证了夹杂物中铝含量定量测定的准确性,结果证明现代化脉冲分布分析是快速表征钢中夹杂物的强大工具。本法分析速度对过程反馈足够快,且可以控制。

The purpose of the work is to further develop and validate the optical emission method pulse distribution analysis（PDA） for rapid inclusion characterisation in steel production.The experimental work was focused on investigation of several operational parameters： spark energy,spark frequency and time gating of signal acquisition.The results showed that a low spark energy improves the detection limit,but at the expense of measuring statistics due to a smaller sample volume.The measuring frequency proved to have no significant influence on the analytical results,but several existing instruments cannot handle the highest spark frequencies above about 300 Hz in PDA mode due to limitations in the electronics.Investigation of time gating（TRS） gave only a marginal improvement in the detection limit of Si outliers above the metallic content.The work also revealed that there is a risk to detect ＂false＂ outliers due to asymmetric intensity distributions.This has resulted in the development of more advanced algorithms for outlier detection,increasing the accuracy of the method.Another limitation found is that the particle number density must not exceed about 10 000 inclusions/mm3 for the method to effectively detect single inclusions.A method to overcome this limitation has been suggested,but not yet evaluated.The accuracy of quantitative determination of the Al content in inclusions has been verified by reference methods.In conclusion,it has been demonstrated that state-of-the-art PDA is a very powerful technique for rapid inclusion characterisation in steels.Furthermore,the speed of analysis is sufficiently high for process feedback and controllable.