Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/119564
Type: | Theses |
Title: | Wavelet methods and system identification |
Author: | Abu Bakar, Mohd Aftar |
Issue Date: | 2016 |
School/Discipline: | School of Mathematical Sciences |
Abstract: | I begin with a brief introduction to dynamic systems, the identification of system parameters from records of input and output, and also wave energy converters which provide case studies to motivate the research. The dynamic systems discussed are categorized as linear or nonlinear dynamic systems. I present brief reviews of strategies for identification of dynamic systems which cover the history and also the areas of applications. The discretization of differential equations for dynamic systems is a recurrent theme and I consider forward, backward and central differences in detail for linear systems. The estimation techniques discussed are the principle of least squares, the Kalman filter and spectral analysis. Several system identification techniques for nonlinear dynamic systems in the time domain and in the frequency domain are presented and compared. The main focus of the thesis is estimation methods based on wavelets. I present some introduction to the wavelet transforms, which cover both continuous and discrete wavelet transforms. Wavelet methods for system identification of linear and nonlinear dynamic systems are discussed. Throughout this research, I have published four research articles guided by my supervisors. The first article discusses the wavelet based technique for linear system, and the technique was compared to the spectral analysis technique. The second article compare two types of wave energy converters, where the heaving buoy wave energy converter (HBWEC) is modelled as a linear system and the oscillating flap wave energy converter (OFWEC) as a nonlinear system. The frequency domain technique for system identification of nonlinear dynamic systems have been applied on the OFWEC model. Unscented Kalman filter have been discussed in the third article where the nonlinear OFWEC system have been used as the case study. A wavelet approach for nonlinear system identification has been discussed in the fourth article together with the probing technique. The probing technique was used to find the generalized frequency response functions of the nonlinear dynamic systems based on the nonlinear autoregressive with exogenous input (ARX) model. Both technique were compared for two weakly nonlinear oscillators, the Duffing and the Van der Pol. Once again, we selected the OFWEC system as a case study. |
Advisor: | Metcalfe, Andrew Green, David |
Dissertation Note: | Thesis (Ph.D.) (Research by Publication) -- University of Adelaide, School of Mathematical Sciences, 2016. |
Keywords: | wavelet system identification wave energy dynamic system oscillator |
Provenance: | This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at: http://www.adelaide.edu.au/legals |
Appears in Collections: | Research Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01front.pdf | 138.44 kB | Adobe PDF | View/Open | |
02whole.pdf | 5.21 MB | Adobe PDF | View/Open | |
Permissions Restricted Access | Library staff access only | 400.56 kB | Adobe PDF | View/Open |
Restricted Restricted Access | Library staff access only | 6.81 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.