Abstract
In this paper, we propose a source term estimation approach for air pollution monitoring based on a physics-informed machine learning approach using radial basis function-generated finite differences (RBF-FD) approximations, rather than using neural network-based approximations. This approach looks promising for detecting a static pollution source, at a particularly low computing cost and based on a network of fixed or mobile sensors. A 3D case study demonstrates the effectiveness of the approach.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have