An Improved Data Assimilation Scheme for High Dimensional Nonlinear Systems

Nonlinear/non-Gaussian filtering has broad applications in many areas of life sciences where either the dynamic is nonlinear and/or the probability density function of uncertain state is non-Gaussian. In such problems, the accuracy of the estimated quantities depends highly upon how accurately their...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Monajemi, Hatef, Kitanidis, Peter K
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Monajemi, Hatef
Kitanidis, Peter K
description Nonlinear/non-Gaussian filtering has broad applications in many areas of life sciences where either the dynamic is nonlinear and/or the probability density function of uncertain state is non-Gaussian. In such problems, the accuracy of the estimated quantities depends highly upon how accurately their posterior pdf can be approximated. In low dimensional state spaces, methods based on Sequential Importance Sampling (SIS) can suitably approximate the posterior pdf. For higher dimensional problems, however, these techniques are usually inappropriate since the required number of particles to achieve satisfactory estimates grows exponentially with the dimension of state space. On the other hand, ensemble Kalman filter (EnKF) and its variants are more suitable for large-scale problems due to transformation of particles in the Bayesian update step. It has been shown that the latter class of methods may lead to suboptimal solutions for strongly nonlinear problems due to the Gaussian assumption in the update step. In this paper, we introduce a new technique based on the Gaussian sum expansion which captures the non-Gaussian features more accurately while the required computational effort remains within reason for high dimensional problems. We demonstrate the performance of the method for non-Gaussian processes through several examples including the strongly nonlinear Lorenz models. Results show a remarkable improvement in the mean square error compared to EnKF, and a desirable convergence behavior as the number of particles increases.
doi_str_mv 10.48550/arxiv.1208.0065
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1208_0065</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1208_0065</sourcerecordid><originalsourceid>FETCH-LOGICAL-a655-aa7ef2a213d35420b256afc44df0949163d250ce396705bd2edabd467fdba6e43</originalsourceid><addsrcrecordid>eNotzz1PwzAUhWEvDKiwM6H7BxIcf6UZoxZoRQVDu0c38TW1FDuVHVX030OB6QyvdKSHsYeKl2qpNX_C9OXPZSX4suTc6Fv21kbYhlOazmRhjTNCm7MPfsTZTxH2w5ECgZsSbPznEdY-UMw_BUd4n-LoI2GC_SXPFPIdu3E4Zrr_3wU7vDwfVpti9_G6XbW7Ao3WBWJNTqCopJVaCd4LbdANSlnHG9VURlqh-UCyMTXXvRVksbfK1M72aEjJBXv8u_3FdKfkA6ZLd0V1V5T8Bu95R4I</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>An Improved Data Assimilation Scheme for High Dimensional Nonlinear Systems</title><source>arXiv.org</source><creator>Monajemi, Hatef ; Kitanidis, Peter K</creator><creatorcontrib>Monajemi, Hatef ; Kitanidis, Peter K</creatorcontrib><description>Nonlinear/non-Gaussian filtering has broad applications in many areas of life sciences where either the dynamic is nonlinear and/or the probability density function of uncertain state is non-Gaussian. In such problems, the accuracy of the estimated quantities depends highly upon how accurately their posterior pdf can be approximated. In low dimensional state spaces, methods based on Sequential Importance Sampling (SIS) can suitably approximate the posterior pdf. For higher dimensional problems, however, these techniques are usually inappropriate since the required number of particles to achieve satisfactory estimates grows exponentially with the dimension of state space. On the other hand, ensemble Kalman filter (EnKF) and its variants are more suitable for large-scale problems due to transformation of particles in the Bayesian update step. It has been shown that the latter class of methods may lead to suboptimal solutions for strongly nonlinear problems due to the Gaussian assumption in the update step. In this paper, we introduce a new technique based on the Gaussian sum expansion which captures the non-Gaussian features more accurately while the required computational effort remains within reason for high dimensional problems. We demonstrate the performance of the method for non-Gaussian processes through several examples including the strongly nonlinear Lorenz models. Results show a remarkable improvement in the mean square error compared to EnKF, and a desirable convergence behavior as the number of particles increases.</description><identifier>DOI: 10.48550/arxiv.1208.0065</identifier><language>eng</language><subject>Mathematics - Optimization and Control ; Physics - Chaotic Dynamics ; Physics - Data Analysis, Statistics and Probability ; Statistics - Computation</subject><creationdate>2012-07</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1208.0065$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1208.0065$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Monajemi, Hatef</creatorcontrib><creatorcontrib>Kitanidis, Peter K</creatorcontrib><title>An Improved Data Assimilation Scheme for High Dimensional Nonlinear Systems</title><description>Nonlinear/non-Gaussian filtering has broad applications in many areas of life sciences where either the dynamic is nonlinear and/or the probability density function of uncertain state is non-Gaussian. In such problems, the accuracy of the estimated quantities depends highly upon how accurately their posterior pdf can be approximated. In low dimensional state spaces, methods based on Sequential Importance Sampling (SIS) can suitably approximate the posterior pdf. For higher dimensional problems, however, these techniques are usually inappropriate since the required number of particles to achieve satisfactory estimates grows exponentially with the dimension of state space. On the other hand, ensemble Kalman filter (EnKF) and its variants are more suitable for large-scale problems due to transformation of particles in the Bayesian update step. It has been shown that the latter class of methods may lead to suboptimal solutions for strongly nonlinear problems due to the Gaussian assumption in the update step. In this paper, we introduce a new technique based on the Gaussian sum expansion which captures the non-Gaussian features more accurately while the required computational effort remains within reason for high dimensional problems. We demonstrate the performance of the method for non-Gaussian processes through several examples including the strongly nonlinear Lorenz models. Results show a remarkable improvement in the mean square error compared to EnKF, and a desirable convergence behavior as the number of particles increases.</description><subject>Mathematics - Optimization and Control</subject><subject>Physics - Chaotic Dynamics</subject><subject>Physics - Data Analysis, Statistics and Probability</subject><subject>Statistics - Computation</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzz1PwzAUhWEvDKiwM6H7BxIcf6UZoxZoRQVDu0c38TW1FDuVHVX030OB6QyvdKSHsYeKl2qpNX_C9OXPZSX4suTc6Fv21kbYhlOazmRhjTNCm7MPfsTZTxH2w5ECgZsSbPznEdY-UMw_BUd4n-LoI2GC_SXPFPIdu3E4Zrr_3wU7vDwfVpti9_G6XbW7Ao3WBWJNTqCopJVaCd4LbdANSlnHG9VURlqh-UCyMTXXvRVksbfK1M72aEjJBXv8u_3FdKfkA6ZLd0V1V5T8Bu95R4I</recordid><startdate>20120731</startdate><enddate>20120731</enddate><creator>Monajemi, Hatef</creator><creator>Kitanidis, Peter K</creator><scope>AKZ</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20120731</creationdate><title>An Improved Data Assimilation Scheme for High Dimensional Nonlinear Systems</title><author>Monajemi, Hatef ; Kitanidis, Peter K</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a655-aa7ef2a213d35420b256afc44df0949163d250ce396705bd2edabd467fdba6e43</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Mathematics - Optimization and Control</topic><topic>Physics - Chaotic Dynamics</topic><topic>Physics - Data Analysis, Statistics and Probability</topic><topic>Statistics - Computation</topic><toplevel>online_resources</toplevel><creatorcontrib>Monajemi, Hatef</creatorcontrib><creatorcontrib>Kitanidis, Peter K</creatorcontrib><collection>arXiv Mathematics</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Monajemi, Hatef</au><au>Kitanidis, Peter K</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>An Improved Data Assimilation Scheme for High Dimensional Nonlinear Systems</atitle><date>2012-07-31</date><risdate>2012</risdate><abstract>Nonlinear/non-Gaussian filtering has broad applications in many areas of life sciences where either the dynamic is nonlinear and/or the probability density function of uncertain state is non-Gaussian. In such problems, the accuracy of the estimated quantities depends highly upon how accurately their posterior pdf can be approximated. In low dimensional state spaces, methods based on Sequential Importance Sampling (SIS) can suitably approximate the posterior pdf. For higher dimensional problems, however, these techniques are usually inappropriate since the required number of particles to achieve satisfactory estimates grows exponentially with the dimension of state space. On the other hand, ensemble Kalman filter (EnKF) and its variants are more suitable for large-scale problems due to transformation of particles in the Bayesian update step. It has been shown that the latter class of methods may lead to suboptimal solutions for strongly nonlinear problems due to the Gaussian assumption in the update step. In this paper, we introduce a new technique based on the Gaussian sum expansion which captures the non-Gaussian features more accurately while the required computational effort remains within reason for high dimensional problems. We demonstrate the performance of the method for non-Gaussian processes through several examples including the strongly nonlinear Lorenz models. Results show a remarkable improvement in the mean square error compared to EnKF, and a desirable convergence behavior as the number of particles increases.</abstract><doi>10.48550/arxiv.1208.0065</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1208.0065
ispartof
issn
language eng
recordid cdi_arxiv_primary_1208_0065
source arXiv.org
subjects Mathematics - Optimization and Control
Physics - Chaotic Dynamics
Physics - Data Analysis, Statistics and Probability
Statistics - Computation
title An Improved Data Assimilation Scheme for High Dimensional Nonlinear Systems
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-14T20%3A32%3A08IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=An%20Improved%20Data%20Assimilation%20Scheme%20for%20High%20Dimensional%20Nonlinear%20Systems&rft.au=Monajemi,%20Hatef&rft.date=2012-07-31&rft_id=info:doi/10.48550/arxiv.1208.0065&rft_dat=%3Carxiv_GOX%3E1208_0065%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true