Abstract
PM2.5 is an important component of air pollution, its chemical composition and source are relatively complex, and it stays in the atmosphere for a long time, which has seriously endangered public health, damaged the ecosystem and even affected the climate and environment. Therefore, the prediction of PM2.5 is of great significance for public health protection and environmental management. One of the reasons why it is difficult to predict PM2.5 is the interaction between air pollutants, so a method that can not only accurately predict the change of PM2.5 concentration, but also explain the results is needed. In this paper, different surface meteorological information is used as time series data to predict the hourly PM2.5 concentration in Wuhan. The prediction can explain the influence of input variables, and use random forest to learn and predict the data. In addition, the meteorological information, air pollutant information and dust value of Zhengzhou City grouped by time series are used as input variables to reflect the air characteristics of North China. This method can learn all the specific time information of the input variables processed by time series in a balanced way, and can also explain the related effects of the input variables. In addition, it also shows that the related variables in North China have an important impact on the generation of PM2.5 in Wuhan.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call