Deep-AIR: A Hybrid CNN-LSTM Framework for Fine-Grained Air Pollution Estimation and Forecast in Metropolitan Cities
نویسندگان
چکیده
Air pollution presents a serious health challenge in urban metropolises. While accurately monitoring and forecasting air are highly crucial, existing data-driven models have yet fully captured the complex interactions between temporal characteristics of spatial dynamics. Our proposed Deep-AIR fills this gap to provide fine-grained city-wide estimation station-wide forecast, by exploiting domain-specific features (including Pollution, Weather, Urban Morphology, Transport, Time-sensitive features), with hybrid CNN-LSTM structure capture spatio-temporal features, 1x1 convolution layers enhance learning interaction. outperforms compatible baselines higher accuracy 1.5%, 2.7%, 2.3% for Hong Kong 1.4%, 1.4% 3.3% Beijing 1-hr estimation, 24-hr forecasts, respectively. Saliency analysis reveals that Kong, including street canyon road density, best predictors NO2, while historical pollutants weather, PM2.5. For Beijing, pollutant data, traffic congestion, wind direction seasonal indicator all pollutants. PM10 is achieving forecast accuracy, whilst CO results.
منابع مشابه
Forecasting Air Pollution Concentrations in Iran, Using a Hybrid Model
The present study aims at developing a forecasting model to predict the next year’s air pollution concentrations in the atmosphere of Iran. In this regard, it proposes the use of ARIMA, SVR, and TSVR, as well as hybrid ARIMA-SVR and ARIMA-TSVR models, which combined the autoregressive part of the autoregressive integrated moving average (ARIMA) model with the support vector regression technique...
متن کاملForecasting Air Pollution Concentrations in Iran, Using a Hybrid Model
The present study aims at developing a forecasting model to predict the next year’s air pollution concentrations in the atmosphere of Iran. In this regard, it proposes the use of ARIMA, SVR, and TSVR, as well as hybrid ARIMA-SVR and ARIMA-TSVR models, which combined the autoregressive part of the autoregressive integrated moving average (ARIMA) model with the support vector regression technique...
متن کاملDeep Air Learning: Interpolation, Prediction, and Feature Analysis of Fine-grained Air Quality
The interpolation, prediction, and feature analysis of fine-gained air quality are three important topics in the area of urban air computing. The solutions to these topics can provide extremely useful information to support air pollution control, and consequently generate great societal and technical impacts. Most of the existing work solves the three problems separately by different models. In...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Access
سال: 2022
ISSN: ['2169-3536']
DOI: https://doi.org/10.1109/access.2022.3174853