Abstract

With the development of modern industry, it can be said that from industrial production to the basic necessities of life, plastic products can be seen everywhere. What follows is the disposal of plastic waste. How to deal with plastic waste correctly is closely related to people’s health and the earth’s environment. In this paper, laser-induced breakdown spectroscopy is applied for the online in situ detection of plastic and its combustion smoke. According to the spectrum obtained, the main components of plastics are Ca, Mg, Na, K, C, H and O. In the spectrum of smoke, the characteristic peaks of the elements C, H, O and N are observed. In addition, the online detection of heavy metal pollution is simulated. Finally, principal component analysis and error back propagation artificial neural networks were applied for the identification of different plastics and their combustion smoke, and the identification accuracy reaches 92.73% and 77.42% respectively. All the results indicate that laser-induced breakdown spectroscopy technology has great potential in the identification of similar objects and the monitoring of air pollutants.

摘要

随着现代工业的发展，可以说从工业生​​产到衣食住行，塑料制品的身影无处不在。接下来是塑料垃圾的处理。如何正确处理塑料垃圾，关系到人们的身体健康和地球环境。在本文中，激光诱导击穿光谱应用于在线原位塑料及其燃烧烟雾的检测。根据得到的光谱，塑料的主要成分为Ca、Mg、Na、K、C、H和O。在烟雾光谱中，观察到元素C、H、O和N的特征峰。此外，还模拟了重金属污染的在线检测。最后，应用主成分分析和误差反向传播人工神经网络对不同塑料及其燃烧烟气进行识别，识别准确率分别达到92.73%和77.42%。这些结果表明，激光诱导击穿光谱技术在相似物体的识别和空气污染物的监测方面具有巨大的潜力。