Journal
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
Volume 36, Issue 5, Pages 1255-1276Publisher
SPRINGER
DOI: 10.1007/s00477-021-02153-3
Keywords
Spatio-temporal forecasting; Particulate matter; Autoencoder; Generative adversarial network; Inverse problem; CNN-LSTM
Categories
Funding
- All-India Council for Technical Education (AICTE)-NDF Scheme
Ask authors/readers for more resources
Accurately forecasting PM2.5 data is beneficial to society but challenging due to non-linear spatio-temporal correlations and incomplete training data. This paper proposes a generative modeling approach called Air-GAN to address these challenges and demonstrates superior performance in predicting real-time PM2.5 data. Air-GAN combines CNN-LSTM Encoder with conditional WGAN to capture non-linear correlations and generate accurate forecast estimates.
Accurate forecasting of air pollutant PM2.5(particulate matter with diameter less than 2.5 mu m) is beneficial to society. However, the non-linear spatio-temporal correlations, multi-feasible forecast values and incomplete training data due to stochasticity make it challenging for discriminative deep learning approaches to forecasting PM2.5 data. In this paper, a generative modeling approach is proposed to overcome the challenges in forecasting PM2.5 data by considering it as an ill-posed inverse problem. To strengthen its applicability, the proposed approach is theoretically validated. Furthermore, based on the proposed generative modeling, an Autoencoder-based generative adversarial network (GAN) named Air-GAN is developed. Air-GAN combines a convolutional neural network- long short-term memory (CNN-LSTM) based Encoder with a conditional Wasserstein GAN (WGAN) to capture non-linear correlations in the data distribution via inverse mapping from the forecast distribution. The condition vector to conditional WGAN is the novelty in Air-GAN, which employs this inverse learning and allows the WGAN's Generator to generate accurate forecast estimates from noise distribution. The condition vector is composed of two elements: (1) the category label of the best correlated meteorological parameter with the PM2.5 data, assigned using an efficient classifier and (2) the output of the CNN-LSTM-based Encoder which is the latent representation of the forecast. The extensive evaluation of Air-GAN for predicting the real-time PM2.5 data of Delhi demonstrates its superior performance with an average inference error of 5.3 mu g/m(3), which achieves 31.7% improvement over the baseline approaches. The improved performance of Air-GAN demonstrates its efficiency to forecast stochastic PM2.5 data by generalizing to out-of-distribution data.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available