基于光信息处理的污水检测分析方法研究

Research on sewage detection and analysis method based on optical information processing

本次实验设计了一款水中持久性污染物富集装置,通过该装置对水中的低浓度污染物进行富集,然后采用卷积神经网络技术建立污染物含量的预测模型,对污染物的含量进行定量分析,实验过程中以苯酚类和重金属类持久性污染物作为研究对象,研究表明采用卷积神经网络技术建立的水中持久性污染物含量预测模型预测相关性最优,且系统具有较好的稳定性。

A device for enriching persistent pollutants in water was designed in this experiment,it was used to enrich lowconcentration pollutants in water,then convolutional neural network technology was used to establish a predictive model of pollutant content.Quantitative analysis of the content of phenols and heavy metal persistent pollutants during the experiment shows that the prediction model of persistent pollutants in water established by convolutional neural network technology has the best predictive correlation,and the best stability.