摘要
为了更好地对钢液进行定量分析,利用激光诱导击穿技术(LIBS)建立支持向量机模型,使用遗传算法优化支持向量机的参数。以钢液中锰元素的质量分数进行验证性试验,通过与传统方法作比较,将两种方法的试验结果进行对比来达到试验目的,最终的试验结果通过以下3个参考量可以看出,即均方根误差、相对标准误差和相关系数,分别为0.612%、9.37%、0.948。结果表明,使用遗传算法的支持向量机模型对分析性能有一定的提高。
In order to obtain better quantitative analysis for liquid steel,the technology of laser induced breakdown（LIBS）is used to establish the support vector machine model,whose parameters are optimized by the genetic algorithm.The experimental verification is performed on Mn element concentration in the liquid steel,which is comparing the support vector machine model with the traditional method.And the three reference quantities,namely,root mean square error,relative standard error and the correlation coefficient,which are obtained from the experimental results,are 0.612%,9.37% and 0.948,respectively.Results show that the genetic algorithm of support vector machine model can have certain enhancement in the analytical performance.
出处
《中国冶金》
CAS
2016年第11期30-33,40,共5页
China Metallurgy
基金
国家自然科学基金资助项目(61271402)
关键词
遗传算法
支持向量机
激光诱导击穿光谱
定量分析
genetic algorithm
support vector machine
laser induced breakdown spectroscopy
quantitative
