基于XGBoost算法的水质CODmn预测模型研究

Research on CODmn Prediction Model Based on XGBoost Algorithm

为解决传统化学方法测定水质高锰酸盐指数时效性差、操作繁琐、易产生二次污染的问题,本文提出一种基于XGBoost算法预测水质高锰酸盐指数的模型。首先采用紫外-可见光谱仪获取水样的光谱特征,然后使用KNN对光谱数据进行异常值预处理,搭建XGBoost算法模型实现预测水质高锰酸盐指数。实验结果表明,基于本文的数据,相对于SVR算法,利用XGBoost算法搭建的模型在预测结果上有较好的拟合优度和模型精度。

In order to solve the problems of poor timeliness,cumbersome operation and secondary pollution caused by traditional chemical methods,a prediction model of permanganate index based on XGBoost algorithm is proposed in this paper.Firstly,the spectral characteristics of water samples were obtained by UV visible spectrometer,and then the abnormal values of spectral data were preprocessed by KNN,and XGBoost algorithm model was established to predict the permanganate index of water quality.The experimental results show that,compared with SVR algorithm,the model built by XGBoost algorithm has better goodness of fit and model accuracy.