Usage
  • 48 views
  • 109 downloads

Identification of Switched Linear Systems

  • Author / Creator
    Wang, Jiadong
  • This thesis is concerned with identification of switched linear systems (SLSs), which is an important part in model-based control. There are a large number of physical systems that can be represented or approximated by SLSs. Therefore, the study of SLSs has attracted much attention over the past decades. As input/output data points of SLSs are sampled from a couple of linear modes (or subsystems), conventional methods are not applicable. For this reason, many research results on identification of SLSs have emerged in recent years. For offline identification of SLSs, many of the existing methods are designed with the assumption that the number of modes is known. This information is, however, not always available in practice. In this thesis, a set membership identification approach is employed to remove this restriction. In its implementation, a major challenge is how to find a maximum feasible subsystem in an efficient way. To achieve this goal, a relaxed heuristic (RH) solution is proposed. Moreover, for SLSs with multiple unknown noise levels, an extended version of the RH solution is subsequently developed. For online identification, a good mode detection or online data classification procedure is critical to estimation performance. One simple and effective way is to directly run a mode detection function before parameter estimation. However, this creates a problem that there may involve a lot of mode mismatches in the mode detection, which has negative impacts on estimation results. In the thesis, two effective algorithms are developed to overcome this problem from different perspectives. In addition to the above aspects, identification of periodically switched linear systems has also been considered in the thesis.

  • Subjects / Keywords
  • Graduation date
    2013-11
  • Type of Item
    Thesis
  • Degree
    Doctor of Philosophy
  • DOI
    https://doi.org/10.7939/R33999
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.
  • Language
    English
  • Institution
    University of Alberta
  • Degree level
    Doctoral
  • Department
    • Department of Electrical and Computer Engineering
  • Specialization
    • Control
  • Supervisor / co-supervisor and their department(s)
    • Chen, Tongwen (Electrical and Computer Engineering)
  • Examining committee members and their departments
    • Chen, Tongwen (Electrical and Computer Engineering)
    • Zhao, Qing (Electrical and Computer Engineering)
    • Lynch, Alan (Electrical and Computer Engineering)
    • Prasad, Vinay (Chemical and Material Engineering)
    • Wu, Fangxiang (Mechanical Engineering)