Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls

In this work, we present the ensemble-marginalized Kalman filter (EnMKF), a sequential algorithm analo- gous to our previously proposed approach [1, 2], for estimating the state and parameters of linear parabolic partial differential equations in initial-boundary value problems when the boundary dat...

Full description

Bibliographic Details
Main Authors: Iglesias, Marco, Sawlan, Zaid, Scavino, Marco, Tempone, Raul, Woodard, Christopher
Format: Article
Published: IOP Publishing 2018
Online Access:https://eprints.nottingham.ac.uk/51562/
_version_ 1848798524386312192
author Iglesias, Marco
Sawlan, Zaid
Scavino, Marco
Tempone, Raul
Woodard, Christopher
author_facet Iglesias, Marco
Sawlan, Zaid
Scavino, Marco
Tempone, Raul
Woodard, Christopher
author_sort Iglesias, Marco
building Nottingham Research Data Repository
collection Online Access
description In this work, we present the ensemble-marginalized Kalman filter (EnMKF), a sequential algorithm analo- gous to our previously proposed approach [1, 2], for estimating the state and parameters of linear parabolic partial differential equations in initial-boundary value problems when the boundary data are noisy. We apply EnMKF to infer the thermal properties of building walls and to estimate the corresponding heat flux from real and synthetic data. Compared with a modified Ensemble Kalman Filter (EnKF) that is not marginalized, EnMKF reduces the bias error, avoids the collapse of the ensemble without needing to add in- flation, and converges to the mean field posterior using 50% or less of the ensemble size required by EnKF. According to our results, the marginalization technique in EnMKF is key to performance improvement with smaller ensembles at any fixed time.
first_indexed 2025-11-14T20:21:08Z
format Article
id nottingham-51562
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:21:08Z
publishDate 2018
publisher IOP Publishing
recordtype eprints
repository_type Digital Repository
spelling nottingham-515622020-05-04T19:37:05Z https://eprints.nottingham.ac.uk/51562/ Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls Iglesias, Marco Sawlan, Zaid Scavino, Marco Tempone, Raul Woodard, Christopher In this work, we present the ensemble-marginalized Kalman filter (EnMKF), a sequential algorithm analo- gous to our previously proposed approach [1, 2], for estimating the state and parameters of linear parabolic partial differential equations in initial-boundary value problems when the boundary data are noisy. We apply EnMKF to infer the thermal properties of building walls and to estimate the corresponding heat flux from real and synthetic data. Compared with a modified Ensemble Kalman Filter (EnKF) that is not marginalized, EnMKF reduces the bias error, avoids the collapse of the ensemble without needing to add in- flation, and converges to the mean field posterior using 50% or less of the ensemble size required by EnKF. According to our results, the marginalization technique in EnMKF is key to performance improvement with smaller ensembles at any fixed time. IOP Publishing 2018-05-22 Article PeerReviewed Iglesias, Marco, Sawlan, Zaid, Scavino, Marco, Tempone, Raul and Woodard, Christopher (2018) Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls. Inverse Problems, 34 (7). 075008. ISSN 0266-5611 http://iopscience.iop.org/article/10.1088/1361-6420/aac224 doi:10.1088/1361-6420/aac224 doi:10.1088/1361-6420/aac224
spellingShingle Iglesias, Marco
Sawlan, Zaid
Scavino, Marco
Tempone, Raul
Woodard, Christopher
Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls
title Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls
title_full Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls
title_fullStr Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls
title_full_unstemmed Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls
title_short Ensemble-marginalized Kalman filter for linear time-dependent PDEs with noisy boundary conditions: application to heat transfer in building walls
title_sort ensemble-marginalized kalman filter for linear time-dependent pdes with noisy boundary conditions: application to heat transfer in building walls
url https://eprints.nottingham.ac.uk/51562/
https://eprints.nottingham.ac.uk/51562/
https://eprints.nottingham.ac.uk/51562/