Learned 1-D advection solver to accelerate air quality modeling

TL;DR

This paper introduces a neural network-based advection solver that accelerates air quality modeling by up to 12.5 times, demonstrating promising accuracy and generalization in realistic wind datasets.

Contribution

It presents the first learned surrogate model for advection in air quality models, using CNNs to significantly speed up computations with good accuracy.

Findings

Achieved up to 12.5x acceleration in advection calculations.

Maintained good overall accuracy with some instability cases.

Showed promising generalization to different wind datasets.

Abstract

Accelerating the numerical integration of partial differential equations by learned surrogate model is a promising area of inquiry in the field of air pollution modeling. Most previous efforts in this field have been made on learned chemical operators though machine-learned fluid dynamics has been a more blooming area in machine learning community. Here we show the first trial on accelerating advection operator in the domain of air quality model using a realistic wind velocity dataset. We designed a convolutional neural network-based solver giving coefficients to integrate the advection equation. We generated a training dataset using a 2nd order Van Leer type scheme with the 10-day east-west components of wind data on 39$^{\circ}$N within North America. The trained model with coarse-graining showed good accuracy overall, but instability occurred in a few cases. Our approach achieved up to 12.5$\times$ acceleration. The learned schemes also showed fair results in generalization tests.