Applying Machine Learning Techniques in Air Quality Prediction—A Bucharest City Case Study

Abstract

Air quality forecasting is very difficult to achieve in metropolitan areas due to: pollutants emission dynamics, high population density and uncertainty in defining meteorological conditions. The use of data, which contain insufficient information within the model training, and the poor selection of the model to be used limits the air quality prediction accuracy. In this study, the prediction of NO2 concentration is made for the year 2022 using a long short-term memory network (LSTM) and a gated recurrent unit (GRU). this is an improvement in terms of performance compared to traditional methods. Data used for predictive modeling are obtained from the National Air Quality Monitoring Network. The KPIs(key performance indicator) are computed based on the testing data subset when the NO2 predicted values are compared to the real known values. Further, two additional predictions were performed for two days outside the modeling dataset. The quality of the data is not as expected, and so, before building the models, the missing data had to be imputed. LSTM and GRU performance in predicting NO2 levels is similar and reasonable with respect to the case study. In terms of pure generalization capabilities, both LSTM and GRU have the maximum R2 value below 0.8. LSTM and GRU represent powerful architectures for time-series prediction. Both are highly configurable, so the probability of identifying the best suited solution for the studied problem is consequently high.

Improving Air Quality Zoning Through Deep Learning and Hyperlocal Measurements

“…Regarding the comparison with other neural networks architecture present in the state of the art, our solution improves for NO 2 , where an R 2 = 0.75 has been obtained using LSTM-GRU compared to a R 2 = 0.89 for ResNET [23]. In contrast, there are better performances for SO 2 in the literature using RNN with R 2 = 0.56 − 0 − 62 [26] .…”

“…Several neural network architectures have been proposed in the air quality dispersion problem context. For example, some studies used recurrent neural networks (RNN) and long short-term memory (LSTM) models to predict air pollution from historical time series pollutant data and meteorological data [23], [24]. For this reason, a new trend is to use the power of deep learning to improve the performance of deterministic models, which could be complemented by machine learning [25].…”