水体透射光谱结合主成分分析(PCA)改进化学需氧量(COD)含量估算研究  

Estimation of COD Content by TransmissionSpectroscopy Combined with PCA

作　　者：王彩玲[1] 位欣欣 WANG Cailing;WEI Xinxin(School of Computer Science,Xi′an Shiyou University,Xi′an,Shaanxi 710065,China)

机构地区：[1]西安石油大学计算机学院,西安710065

出　　处：《中国无机分析化学》2024年第4期410-417,共8页Chinese Journal of Inorganic Analytical Chemistry

基　　金：陕西省重点研发计划项目(2023-YBSF-437);国家自然科学基金资助项目(31160475,61401439)。

摘　　要：为了解决传统的化学需氧量(Chemical Oxygen Demand,COD)测量方法耗时较长,不利于快速、实时地获取水体中COD的信息等问题。通过采集100组COD水体光谱信息,分别使用3种不同的高光谱数据预处理方法对光谱数据进行预处理,并基于不同的预处理方法分别建立高斯过程回归模型(Gaussian Process Regression,GPR)和BP神经网络模型,分析不同预处理方法对模型精度的影响,建立了基于透射光谱测量结合主成分分析(Principal Component Analysis,PCA)改进水体COD含量估算模型。对各模型结合PCA数据降维方法进行模型的改进,通过比较模型的精度选择最优模型进行水体COD含量的检测。结果显示:相比于原始光谱数据建立的GPR模型和BP神经网络模型,数据预处理后的模型精度明显提升;且结合PCA对预处理后的数据进一步降维处理后,模型精度得到了进一步的提升。其中,基于标准正态变量变换特征结合PCA改进BP神经网络模型R 2高达0.9940,均方根误差RMSE为0.022540。证明了基于PCA数据降维方法对预处理后的光谱数据进行降维处理,有利于去除光谱中的冗余信息,提取特征信息,可以实现COD含量估算模型的优化,从而为传统COD测量方法存在的问题提出了一种新的解决思路。COD is an important indicator of organic pollution in water,and the higher the COD,the more serious the degree of water pollution.To solve the traditional method of COD determination is time-consuming,not conducive to rapid,real-time access to COD information in the water and other issues,in this paper,an improved model for COD determination in water on the basis of transmission spectroscopy measurement combined with principal component analysis(PCA)was proposed.Specifically,100 groups of COD water body spectral information were collected,and three different hyperspectral data preprocessing methods were used to preprocess the spectral data,and Gaussian process regression(GPR)and BP neural network models were constructed based on different preprocessing methods to analyze the effects of different preprocessing methods on the accuracy of the models.Compared with GPR model and BP neural network model constructed from original spectrum data,it was found that after data pre-processing,there was a significant improvement in model accuracy,and after further dimension reduction of pre-processing data combined with PCA,there was a further improvement in model accuracy.Among them,the R 2 of the improved BP neural network model based on standard normal variable transformed features combined with PCA was as high as 0.9940,and the RMSE was 0.022540.This proved that the dimensionality reduction of the preprocessed spectral data based on the PCA data dimensionality reduction method was helpful to remove the redundant information in the spectral data and extract the feature information,and optimize the COD content estimation model,thereby solving the problems of traditional COD measurement methods.Thus,a new idea for the solution of the problems that exist in the traditional method of COD measurement is proposed.

关 键 词：透射光谱法 COD含量预测 PCA 高斯过程回归 BP神经网络 

分 类 号：O657.39[理学—分析化学] O433.4[理学—化学]