Back to publications

Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation

Mack, Julian and Arcucci, Rossella and Molina-Solana, Miguel and Guo, Yike
Computer Methods in Applied Mechanics and Engineering under review, (SUBMITTED)

Abstract:

We propose a new ‘Bi-Reduced Space’ approach to solving 3D Variational Data Assimilation using Convolutional Autoencoders. We prove that our approach has the same solution as previous methods but has significantly lower computational complexity; in other words, we reduce the computational cost without affecting the data assimilation accuracy. We tested the new method with data from a real-world application: a pollution model of a site in Elephant and Castle, London and found that we could reduce the size of the background covariance matrix representation by \mathcalO(10^3) and, at the same time, increase our data assimilation accuracy with respect to existing reduced space methods.

Links:

DOI:
PDF:

Bibtex:

@article{Mack2020,
  author = {Mack, Julian and Arcucci, Rossella and Molina-Solana, Miguel and Guo, Yike},
  title = {Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation},
  journal = {Computer Methods in Applied Mechanics and Engineering},
  year = {SUBMITTED},
  volume = {under review},
  doi = {},
  comment = {},
  timestamp = {29}
}