Modeling air pollution under uncertainty by using deep generative models

TOUTOUH, JAMAL; ROSSIT, DIEGO GABRIEL; NESMACHNOW, SERGIO

Irkutsk

Congreso; 1st International Workshop on Advanced Information and Computation Technologies and Systems (AICTS); 2020

Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the Russian Academy of Sciences

Modeling and forecasting ambient air pollution is a relevant problem because it is helpful for decision-makers and urban city planners to understand one of the main health problems in the urban area, the degradation of air quality due to motorized road mobility. This article addresses data-driven outdoor pollution modeling given limited data about road traffic density and pollution concentration. In order to deal with such a lack of data, we propose a deep generative model able to synthesize synthetic nitrogen dioxide concentration given road traffic density. Specically, we train conditional generative adversarial networks. The experimentalanalysis indicates that the proposed approach generates accurate and diverse pollution data, while requiring reduced computational time.