基于PLS-SVM解析光度法多组分同时测定数据的比较  

Comparison of Simultaneous Determination Data of Multicomponent by Analytical Spectrophotometry Based on PLS-SVM Method

作　　者：潘国锋[1] 

机构地区：[1]江南大学轻工过程先进控制教育部重点实验室

出　　处：《光谱实验室》2013年第1期98-104,共7页Chinese Journal of Spectroscopy Laboratory

基　　金：江苏省高技术研究项目(工业)(BG2006010);江南大学博士研究生科学研究基金(JUDCF12030)

摘　　要：针对含有一定浓度氮、磷水样的紫外吸收光谱,利用支持向量机与偏最小二乘回归相结合的方法对水样中的氮磷浓度进行回归建模,所建模型可以很好地检测出水样中的氮磷含量。仿真结果表明,用这种方法建立的模型,氮麟浓度检测的相对误差都在10%以内,完全满足这方面的精度要求。Aim at UV absorption spectra of water containing certain concentration of nitrogen and phosphorus,the concentrations of nitrogen and phosphorus in water sample were treated by a regression model according to support vector machine（SVM） and partial least-squares regression method（PLS）,and established model can well detect the contents of nitrogen and phosphorus in water sample.The simulation results showed that the relative errors of nitrogen and phosphorus detection were less than 10%through this established model,and the model can fully meet the requirements of precision.

关 键 词：支持向量机 偏最小二乘回归 吸收光谱 精度 

分 类 号：O433.51[机械工程—光学工程] O657.32[理学—光学]