土壤污染物快速检测技术应用于农田环境保护的可行性分析

Feasibility Analysis of Rapid Detection Technology for Soil Pollutants Applied to Agricultural Environmental Protection

围绕差分光学吸收光谱技术(简称DOAS),以及太赫兹光谱技术应用于农田环境保护的可行性开展分析讨论,前者主要依靠光谱仪器,搭配粒子群算法优化,依靠深度学习自动编码器,打造基于深度神经网络的检测模型。根据实验结果显示,该方法具有较高的污染物成分检测精确性,而后者则可通过消除水对太赫兹波的影响,实现六价铬与微塑料等典型污染物的检测,根据太赫兹光谱吸收系数可以发现,该技术同样具有极高的检测可行性,极具发展潜力。

The feasibility of applying diff erential optical absorption spectroscopy(DOAS)and terahertz spectroscopy technology to agricultural environmental protection is analyzed and discussed.The former mainly relies on spectral instruments,particle swarm optimization algorithm optimization,and deep learning autoencoder to create a detection model based on deep neural networks.According to experimental results,this method has high accuracy in detecting pollutant components.The latter can achieve the detection of typical pollutants such as hexavalent chromium and microplastics by eliminating the infl uence of water on terahertz waves.According to the terahertz spectral absorption coeffi cient,it can be found that this technology also has high detection feasibility and great development potential.