论文部分内容阅读
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 analyze the molten steel better, a support vector machine (SVM) model is established by laser induced breakdown technique (LIBS) and the parameters of SVM are optimized by genetic algorithm. In the liquid steel manganese content of the mass fraction of the confirmatory test, compared with the traditional method, the two methods of test results were compared to achieve the purpose of the experiment, the final test results can be seen by the following three reference quantities, namely Root mean square error, relative standard error and correlation coefficient were 0.612%, 9.37%, 0.948 respectively. The results show that the support vector machine model using genetic algorithm can improve the performance of the analysis.