Constructing a PM2.5 concentration prediction model by combining auto-encoder with Bi-LSTM neural networks

Exposure to indoor particulate matter less than 2.5 µm in diameter (PM2.5) is a critical health risk factor. Therefore, measuring indoor PM2.5 concentrations is important for assessing their health risks and further investigating the sources and influential factors. However, installing monitoring instruments to collect indoor PM2.5 data is difficult and expensive. Therefore, several indoor PM2.5 concentration prediction models have been developed. However, these prediction models only assess the daily average PM2.5 concentrations in cold or temperate regions. The factors that influence PM2.5 concentration differ according to climatic conditions. In this study, we developed a prediction model for hourly indoor PM2.5 concentrations in Taiwan (tropical and subtropical region) by using a multiple linear regression model and investigated the impact factor. The sample comprised 93 study cases (1979 measurements) and 25 potential predictor variables. Cross-validation was performed to assess performance. The prediction model explained 74% of the variation, and outdoor PM2.5 concentrations, the difference between indoor and outdoor CO2 levels, building type, building floor level, bed sheet cleaning, bed sheet replacement, and mosquito coil burning were included in the prediction model. Cross-validation explained 75% of variation on average. The results also confirm that the prediction model can be used to estimate indoor PM2.5 concentrations across seasons and areas. In summary, we developed a prediction model of hourly indoor PM2.5 concentrations and suggested that outdoor PM2.5 concentrations, ventilation, building characteristics, and human activities should be considered. Moreover, it is important to consider outdoor air quality while occupants open or close windows or doors for regulating ventilation rate and human activities changing also can reduce indoor PM2.5 concentrations.

Download Full-text

A PM2.5 concentration prediction model based on multi-task deep learning for intensive air quality monitoring stations

Journal of Cleaner Production ◽

10.1016/j.jclepro.2020.122722 ◽

2020 ◽

Vol 275 ◽

pp. 122722

Author(s):

Qiang Zhang ◽

Shun Wu ◽

Xiangwen Wang ◽

Binzhen Sun ◽

Haimeng Liu

Keyword(s):

Deep Learning ◽

Air Quality ◽

Prediction Model ◽

Quality Monitoring ◽

Air Quality Monitoring ◽

Model Based ◽

Air Quality Monitoring Stations ◽

Concentration Prediction ◽

Pm2.5 Concentration ◽

Monitoring Stations

Download Full-text

PM2.5 Concentration Prediction Model Based on Chaotic Genetic Neural Network

2019 Chinese Control And Decision Conference (CCDC) ◽

10.1109/ccdc.2019.8832724 ◽

2019 ◽

Cited By ~ 1

Author(s):

Wang Xiao-min ◽

Zheng Yu-tong ◽

Lei Ting

Keyword(s):

Neural Network ◽

Prediction Model ◽

Model Based ◽

Genetic Neural Network ◽

Concentration Prediction ◽

Pm2.5 Concentration

Download Full-text

A novel hourly PM2.5 concentration prediction model based on feature selection, training set screening, and mode decomposition-reorganization

Sustainable Cities and Society ◽

10.1016/j.scs.2021.103348 ◽

2021 ◽

pp. 103348

Author(s):

Wei Sun ◽

Zhiwei Xu

Keyword(s):

Feature Selection ◽

Prediction Model ◽

Training Set ◽

Model Based ◽

Mode Decomposition ◽

Concentration Prediction ◽

Pm2.5 Concentration

Download Full-text

PM2.5 Concentration Prediction for Air Pollution using Machine Learning Algorithms

Menoufia Journal of Electronic Engineering Research ◽

10.21608/mjeer.2019.67375 ◽

2019 ◽

Vol 28 (1) ◽

pp. 349-354 ◽

Cited By ~ 1

Author(s):

Ahmed Samy Abd El Aziz Moursi ◽

Marwa Shouman ◽

Ezz El-din Hemdan ◽

Nawal El-Fishawy

Keyword(s):

Machine Learning ◽

Air Pollution ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Concentration Prediction ◽

Pm2.5 Concentration

Download Full-text

Prediction Modeling of PM-10 in Chiangmai City Moat by Using Artificial Neural Networks

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.781.628 ◽

2015 ◽

Vol 781 ◽

pp. 628-631 ◽

Cited By ~ 1

Author(s):

Rati Wongsathan ◽

Issaravuth Seedadan ◽

Metawat Kavilkrue

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Prediction Model ◽

Chiang Mai ◽

Adverse Health Effects ◽

High Pollution ◽

Artificial Neural ◽

Input Variables ◽

Pm 10 ◽

Respiratory Conditions

A mathematical prediction model has been developed in order to detect particles with a diameter of 10 micrometers or less (PM-10) that are responsible for adverse health effects because of their ability to cause serious respiratory conditions in areas of high pollution such as Chiang Mai City moat area. The prediction model is based on 3 types of Artificial Neural Networks (ANNs), including Multi-layer perceptron (MLP-NN), Radial basis function (RBF-NN), and hybrid of RBF and Genetic algorithm (RBF-NN-GA). The model uses 8 input variables to predict PM-10, consisting of 4 air pollution substances ( CO, O3, NO2 and SO2) and 4 meteorological variables related PM-10 (wind speed, temperature, atmospheric pressure and relative humidity). These 3 types of ANN have proved efficient instrument in predicting the PM-10. However, the performance of RBF-NN was superior in comparison with MLP-NN and RBF-NN-GA respectively.

Download Full-text

Atmospheric PM2.5 Prediction Based on Multiple Model Adaptive Unscented Kalman Filter

Atmosphere ◽

10.3390/atmos12050607 ◽

2021 ◽

Vol 12 (5) ◽

pp. 607

Author(s):

Jihan Li ◽

Xiaoli Li ◽

Kang Wang ◽

Guimei Cui

Keyword(s):

Kalman Filter ◽

Unscented Kalman Filter ◽

Support Vector ◽

Multiple Model ◽

Single Model ◽

Adaptive Unscented Kalman Filter ◽

Time Periods ◽

Atmospheric Pm2.5 ◽

Concentration Prediction ◽

Pm2.5 Concentration

The PM2.5 concentration model is the key to predict PM2.5 concentration. During the prediction of atmospheric PM2.5 concentration based on prediction model, the prediction model of PM2.5 concentration cannot be usually accurately described. For the PM2.5 concentration model in the same period, the dynamic characteristics of the model will change under the influence of many factors. Similarly, for different time periods, the corresponding models of PM2.5 concentration may be different, and the single model cannot play the corresponding ability to predict PM2.5 concentration. The single model leads to the decline of prediction accuracy. To improve the accuracy of PM2.5 concentration prediction in this solution, a multiple model adaptive unscented Kalman filter (MMAUKF) method is proposed in this paper. Firstly, the PM2.5 concentration data in three time periods of the day are taken as the research object, the nonlinear state space model frame of a support vector regression (SVR) method is established. Secondly, the frame of the SVR model in three time periods is combined with an adaptive unscented Kalman filter (AUKF) to predict PM2.5 concentration in the next hour, respectively. Then, the predicted value of three time periods is fused into the final predicted PM2.5 concentration by Bayesian weighting method. Finally, the proposed method is compared with the single support vector regression-adaptive unscented Kalman filter (SVR-AUKF), autoregressive model-Kalman (AR-Kalman), autoregressive model (AR) and back propagation neural network (BP). The prediction results show that the accuracy of PM2.5 concentration prediction is improved in whole time period.

Download Full-text