PM2.5 Prediction Based on the CEEMDAN Algorithm and a Machine Learning Hybrid Model

Abstract

The current serious air pollution problem has become a closely investigated topic in people’s daily lives. If we want to provide a reasonable basis for haze prevention, then the prediction of PM2.5 concentrations becomes a crucial task. However, it is difficult to complete the task of PM2.5 concentration prediction using a single model; therefore, to address this problem, this paper proposes a fully adaptive noise ensemble empirical modal decomposition (CEEMDAN) algorithm combined with deep learning hybrid models. Firstly, the CEEMDAN algorithm was used to decompose the PM2.5 timeseries data into different modal components. Then long short-term memory (LSTM), a backpropagation (BP) neural network, a differential integrated moving average autoregressive model (ARIMA), and a support vector machine (SVM) were applied to each modal component. Lastly, the best prediction results of each component were superimposed and summed to obtain the final prediction results. The PM2.5 data of Hangzhou in recent years were substituted into the model for testing, which was compared with eight models, namely, LSTM, ARIMA, BP, SVM, CEEMDAN–ARIMA, CEEMDAN–LSTM, CEEMDAN–SVM, and CEEMDAN–BP. The results show that for the coupled CEEMDAN–LSTM–BP–ARIMA model, the prediction ability was better than all the other models, and the timeseries decomposition data of PM2.5 had their own characteristics. The data with different characteristics were predicted separately using appropriate models and the final combined model results obtained were the most satisfactory.

Deep-learning architecture for PM2.5 concentration prediction: A review

“…[ 81 ] 2021 Marrakesh, Morocco WT-CNN H/-/- 0.01 - 99.10 - Ban et al. [ 104 ] 2022 Hangzhou, China CEEMD-LSTM-BP-ARIMA D/S/T+1 4.55 3.66 - 0.79 M.A.A.A.-q et al. [ 105 ] 2021 Wuhan, China PSO-SMA-ANFIS H/S/T+1 22.39 17.50 16.83 0.51 …”

“…Ban et al. [ 104 ] incorporated CEED, BP, LSTM, and ARIMA to build a comprehensive hybrid framework that considered multiple factors and scaled for air pollutant prediction and early warning. Liu et al.…”

A Novel Hybrid Model of CEEMDAN-CNN-SAGU for Shanghai Copper Price Prediction

“…The experimental results indicated that LSTM outperformed traditional statistical forecasting methods [12]. The three gated units in LSTM can capture the dependency between time series data and improve LSTM's learning ability [13].…”

Low-altitude small target detection in sea clutter background based on improved CEEMDAN-IZOA-ELM

“…Experimental results clearly demonstrate that the predictive accuracy achieved by the mixed model significantly outperforms that of both individual and integrated benchmark models. Ban et al [ 11 ] used CEEMDAN for PM2.5 time series data and applied Long Short-Term Memory(LSTM), Back Propagation(BP), Autoregressive Integrated Moving Average Model(ARIMA), and Support Vector Machine(SVM) to each modal component to obtain a satisfactory integrated prediction model.…”