A modified functional observer-based EID estimator for unknown sampled-data singular systems. Issue 10 (27th July 2019)

• Record Type: Journal Article
• Authors: Tsai, Jason Sheng-Hong; Chang, Chia-Yuan; Chen, Yang-Fang; Guo, Shu-Mei; Shieh, Leang-San; Canelon, Jose I.
• Is Part Of: International journal of systems science. Volume 50:Issue 10(2019)
• Page Start: 1976
• Page End: 2001
• Publication Date: 2019-07-27
• Subjects: Optimal linear quadratic digital tracker -- disturbance estimator -- functional observer -- observer/Kalman filter identification -- singular system -- equivalent input disturbance estimator; System analysis -- Periodicals; 003.3
• DOI: 10.1080/00207721.2019.1645919
• Languages: English
• ISSNs: 0020-7721

Abstract

Abstract : This paper presents both an observer/Kalman filter identification (OKID) method-based linear quadratic digital tracker (LQDT) and a new functional observer-based discrete equivalent input disturbance (EID) estimator, for unknown square/non-square singular sampled-data systems with unknown input and output disturbances. Initially, it is shown that even if the solution of the impulsive mode-free singular linear quadratic tracker problem exists, under some quite general conditions the tracker-based generalised algebraic Riccati equation (GARE) might have no solution. To overcome this issue, an innovative methodology that involves an introduced reduced-order equivalent proper regular model (EPRM) of the singular system is presented, to make the tracker-based GARE solvable for a wide class of servo control problems for a singular system. Furthermore, a new functional observer-based design methodology for the discrete EID estimator of the discrete EPRM is proposed in this paper, for a proper sampled-data system with unknown matched/mismatched input and output disturbances. Finally, a discrete-time EPRM of the unknown singular sampled-data system is constructed using the off-line OKID method and the robust prediction-based state-estimate optimal LQDT, associated with the (plug-in) discrete EID estimator, is developed for the singular sampled-data system.