PERFORMANCE COMPARISON OF UNSCENTED AND EXTENDED KALMAN FILTER IMPLEMENTATION ON A CONTINIOUS STIRRED TANK REACTOR

Search for this publication on Google Scholar
Mostafavi, S. (2014). PERFORMANCE COMPARISON OF UNSCENTED AND EXTENDED KALMAN FILTER IMPLEMENTATION ON A CONTINIOUS STIRRED TANK REACTOR. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

200 views ◎
156 downloads ⇓


Abstract

In this thesis work, it is intended to investigate not only the Extended Kalman Filter (EKF) but to further study the more recent nonlinear Kalman Filters for their application to the nonlinear problem of a continuous stirred tank reactor. The various filters studied in this thesis are:* Extended Kalman Filter (EKF)* Monte Carlo Kalman Filter (MCKF)* Unscented Kalman Filter (UKF) and its various forms and alternate editionsThe study goes on to provide a comparison of the computational and accuracy costs involved in the cases of EKF and UKF.Until recently, the EKF has proven to be a better filter in terms of accuracy for the majority of non-linear cases such as the various cases of a continuous stirred tank reactor. The research work in this thesis also applies these filtering methods to the system dynamics of an exothermic continuous stirred tank reactor. Results obtained in this research corroborate the notion that an Unscented Kalman Filter tends to have a better accuracy than the previously trusted family of Kalman Filters, however after investigating the application of the UKF to the problem studied, it is still not possible to make a strong case for the UKF in general and the conclusion is that the application of this filter remains to be case sensitive.

Details
Author

Mostafavi, Saman

Title

PERFORMANCE COMPARISON OF UNSCENTED AND EXTENDED KALMAN FILTER IMPLEMENTATION ON A CONTINIOUS STIRRED TANK REACTOR

Physical Description

1 online resource (63 pages) : PDF

Date

2014

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

In this thesis work, it is intended to investigate not only the Extended Kalman Filter (EKF) but to further study the more recent nonlinear Kalman Filters for their application to the nonlinear problem of a continuous stirred tank reactor. The various filters studied in this thesis are:* Extended Kalman Filter (EKF)* Monte Carlo Kalman Filter (MCKF)* Unscented Kalman Filter (UKF) and its various forms and alternate editionsThe study goes on to provide a comparison of the computational and accuracy costs involved in the cases of EKF and UKF.Until recently, the EKF has proven to be a better filter in terms of accuracy for the majority of non-linear cases such as the various cases of a continuous stirred tank reactor. The research work in this thesis also applies these filtering methods to the system dynamics of an exothermic continuous stirred tank reactor. Results obtained in this research corroborate the notion that an Unscented Kalman Filter tends to have a better accuracy than the previously trusted family of Kalman Filters, however after investigating the application of the UKF to the problem studied, it is still not possible to make a strong case for the UKF in general and the conclusion is that the application of this filter remains to be case sensitive.

Genre

masters theses

Subjects--Topics

Electrical engineering

Degree

M.S.

Keywords

Extended Kalman Filter
Kalman Filer
Unscented Kalman Filter

Subject Area

Electrical Engineering

Advisor(s)

Kakad, Yogendra

Committee Members

Miri, Mehdi
Kazemi, Mohammad
Kakad, Yogendra

Degree Note

Thesis (M.S.)--University of North Carolina at Charlotte, 2014.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Mostafavi_uncc_0694N_10699

Permalink

http://hdl.handle.net/20.500.13093/etd:220