A Case Study On The Prediction Of PM2.5 And PM10 In Lanzhou Based On Variational Mode Decomposition

The atmosphere is one of the essential elements of human activities.However,the rapid development of modern industry and the growth of human population caused a large amount of burning of coal and oil fuel,which produced a lot of chemical waste through air exhaust,soot and many other forms.All these emissions to the atmosphere have exceeded the capacity of the atmospheric environment.The pollution has very negative impacts to our everyday life and our own health.In this thesis,a novel hybrid model based on variational mode decomposition(VMD)was developed for forcasting and analysis.The historical data of atmospheric PM2.5 and PM10 in 2016 in Lanzhou,Gansu Province were collected for analysis and prediction.The VMD was initially used in the preprocessing stage to reconstruct data.Besides,we used hybrid model for searching the optimal weight and threshold of the back propagation neural networks.Numerical results indicate that the developed hybrid model has higher accuracy on predicting PM2.5 and PM10 concentrations when compared to other previously reported individual models and traditional models.