EnKF data-driven reduced order assimilation system

Liu, C.; Fu, Rui; Xiao, Dunhui; Ştefanescu, Răzvan; Sharma, Prakhar; Zhu, Chuanhua; Sun, Shuchao; Wang, C.

doi:10.1016/j.enganabound.2022.02.016

Cited by 28 publications

(15 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much research effort 29,30,45,46 has been given to learn the underlying dynamics in the reduced-order space (also known as the latent space). Among them, the RNNs 47 take a sequence of inputs and are generally used to handle time-series prediction.…”

Section: Related Work and Our Contributionsmentioning

confidence: 99%

“…42,43 More precisely, CAE have shown its strength in model reduction of laminar flows 44 and recently a CAE encoded data-driven model is developed for estimating indoor airflow and ventilation from the numerical simulation results, using DA technique to perform corrections with sensor data. 30 Much research effort 29,30,45,46 has been given to learn the underlying dynamics in the reduced-order space (also known as the latent space). Among them, the RNNs 47 take a sequence of inputs and are generally used to handle timeseries prediction.…”

Section: Related Work and Our Contributionsmentioning

confidence: 99%

See 1 more Smart Citation

Ensemble latent assimilation with deep learning surrogate model: application to drop interaction in a microfluidics device

et al. 2022

View full text Add to dashboard Cite

show abstract

Section: Related Work and Our Contributionsmentioning

confidence: 99%

Section: Related Work and Our Contributionsmentioning

confidence: 99%

Ensemble latent assimilation with deep learning surrogate model: application to drop interaction in a microfluidics device

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In order to incorporate unseen real-time observation data efficiently, the recent works of [7,8,25] introduce the concept of LA where an AE network is used to compress the state variables and pre-processed observation data. The DA updating is performed in the reduced-order latent space subsequently.…”

Section: Introductionmentioning

confidence: 99%

Generalised Latent Assimilation in Heterogeneous Reduced Spaces with Machine Learning Surrogate Models

Cheng¹,

Chen²,

Anastasiou³

et al. 2022

Preprint

View full text Add to dashboard Cite

Reduced-order modelling and low-dimensional surrogate models generated using machine learning algorithms have been widely applied in high-dimensional dynamical systems to improve the algorithmic efficiency. In this paper, we develop a system which combines reduced-order surrogate models with a novel data assimilation (DA) technique used to incorporate real-time observations from different physical spaces. We make use of local smooth surrogate functions which link the space of encoded system variables and the one of current observations to perform variational DA with a low computational cost. The new system, named Generalised Latent Assimilation can benefit both the efficiency provided by the reduced-order modelling and the accuracy of data assimilation. A theoretical analysis of the difference between surrogate and original assimilation cost function is also provided in this paper where an upper bound, depending on the size of the local training set, is given. The new approach is tested on a high-dimensional computational fluid dynamics (CFD) application of a two-phase liquid flow with non-linear observation operators that current Latent Assimilation methods can not handle. Numerical results demonstrate that the proposed assimilation approach can significantly improve the reconstruction and prediction accuracy of the deep learning surrogate model which is nearly 1000 times faster than the CFD simulation.

show abstract

“…Significant work has been carried out recently on predicting solution instances outside the training domain for a variety of fluid problems with discontinuities, wave propagation, and advection-dominated flows. Liu et al [24] presented a predictive data assimilation framework based on the Ensemble Kalman Filter (EnKF) and the DDROM model, which uses an autoencoder network for the compression of high-dimensional dynamics to lower dimensional space and then the LSTM method to model the fluid dynamics in latent space. The model capabilities were estimated using 2D Burgers' equation and flow past a cylinder test case.…”

Section: Introductionmentioning

confidence: 99%

Deep Convolutional Architectures for Extrapolative Forecasts in Time-Dependent Flow Problems

Bhatt

Kumar

Soulaïmani³

2022

SSRN Journal

View full text Add to dashboard Cite

EnKF data-driven reduced order assimilation system

Cited by 28 publications

References 53 publications

Ensemble latent assimilation with deep learning surrogate model: application to drop interaction in a microfluidics device

Ensemble latent assimilation with deep learning surrogate model: application to drop interaction in a microfluidics device

Generalised Latent Assimilation in Heterogeneous Reduced Spaces with Machine Learning Surrogate Models

Deep Convolutional Architectures for Extrapolative Forecasts in Time-Dependent Flow Problems

Contact Info

Product

Resources

About