摘要
采用近红外光谱检测技术测定水体中的微量持久性污染物。将卷积神经网络技术应用于近红外光谱法,与传统BP神经网络模型和PLSR模型进行对比,卷积神经网络预测模型对水体中微量污染物的含量判别较好。通过对比训练集样本数量对模型的影响,发现随着训练集样本数量的增加,模型性能显著提高,表明该模型能够适应大数据背景下水质污染物的检测需求。
The trace persistent pollutants in water was determined by near infrared spectroscopy.Applied to near infrared spectroscopy and compared with the traditional BP neural network model and PLSR model,convolution neural network prediction model could better distinguish the content of trace pollutants in water.By comparing the impact of the number of training set samples on the model,it was found that with the increase of the number of training set samples,the performance of the model was significantly improved,indicating that the model could adapt to the needs of water pollution detection in the context of big data.
作者
张林
Zhang Lin(Shangluo Artificial Intelligence Research Center,Shangluo College,Shangluo 726000,China)
出处
《化学分析计量》
CAS
2021年第8期24-27,共4页
Chemical Analysis And Meterage
基金
陕西省教育厅科学研究项目(19JK0256)。
关键词
水质
污染物
近红外光谱法
卷积神经网络
water quality
pollutant
near infrared spectroscopy
convolution neural network