Dual-mode robust fault estimation for switched linear systems with state jumps. (February 2018)

• Record Type: Journal Article
• Authors: Johnson, Scott C.; Chakrabarty, Ankush; Hu, Jianghai; Żak, Stanisław H.; DeCarlo, Raymond A.
• Is Part Of: Nonlinear analysis. Volume 27(2018)
• Page Start: 125
• Page End: 140
• Publication Date: 2018-02
• Subjects: Fault detection -- Sliding mode -- Robust observer -- Switched linear systems -- State jumps; Nonlinear functional analysis -- Periodicals; Analyse fonctionnelle non linéaire -- Périodiques; Nonlinear functional analysis; Periodicals; 515.7248
• DOI: 10.1016/j.nahs.2017.08.009
• Languages: English
• ISSNs: 1751-570X

Abstract

Abstract: Fault detection and estimation is a critical part of modern control system design and ensures safety and reliability of expensive machinery. In this paper, a dual-mode state and fault estimation scheme is proposed for switched linear systems with a class of state jumps in the presence of simultaneous actuator and sensor faults. In the absence of sensor faults, a switched sliding mode observer is developed to estimate the state and actuator fault signal. A residual signal is computed to detect the inception of a sensor fault. Upon detecting the sensor fault, a robust state observer is triggered which guarantees ultimate boundedness of the state estimation error. The performance of the proposed dual-mode switched observer is illustrated using a numerical example.