论文部分内容阅读
应用近红外光谱技术结合连续投影算法(SPA)实现了油中含水量的分析。对57个油样进行光谱扫描,通过比较不同预处理方法,以相关系数(R)和均方根误差(RMSE)作为模型评价指标,建立油中含水量预测的全波段偏最小二乘法(PLS)模型。同时应用SPA提取有效波长,作为PLS的输入变量,建立了SPA-PLS模型。结果表明经连续投影算法提取24个特征波长建立的模型,所用变量数仅占全波段的4.68%,SPA-PLS优于全波段的PLS模型,其对验证集样本进行预测的相关系数和均方根误差分别为0.994 4和5.455 1×10-5,获得了满意的预测精度。说明应用光谱技术检测油中含水量是可行的,并能获得满意的预测精度,为进一步应用光谱技术进行油中其他污染物的在线监测提供了新的方法。
Near infrared spectroscopy combined with continuous projection algorithm (SPA) to achieve the analysis of water content in oil. Spectral scanning of 57 oil samples was carried out. By comparing different pretreatment methods, correlation coefficient (R) and root mean square error (RMSE) were used as model evaluation indices to establish the full-band PLS )model. At the same time, SPA was used to extract the effective wavelength. As the input variable of PLS, the SPA-PLS model was established. The results show that the 24 variables extracted by continuous projection algorithm use only 4.68% of the total number of variables, and the SPA-PLS is better than the full-band PLS model. The correlation coefficient and the mean square The root errors were 0.994 4 and 5.455 1 × 10-5, respectively, with satisfactory prediction accuracy. It is feasible to detect the water content in the oil by using the spectroscopic technique, and the satisfactory prediction accuracy can be obtained. It provides a new method for the further application of the spectroscopic technique to the on-line monitoring of other pollutants in the oil.