Skip to Main content Skip to Navigation
Conference papers

Fault Tolerant Control Strategy Using Two-Layer Multiple Adaptive Models for Plant Fault

Menglin He 1, 2 Zetao Li 1 Boutaib Dahhou 2
2 LAAS-DISCO - Équipe DIagnostic, Supervision et COnduite
LAAS - Laboratoire d'analyse et d'architecture des systèmes
Abstract : Fault tolerant control (FTC) is always a popular research direction in the domain of automatic control. Inspired by the concept of adaptive model and corresponding approaches in [1], this paper proposed an FTC design strategy for plant fault by introducing these adaptive models into the two-layer multiple model structure. The two-layer multiple model structure describes a hyper-system which considers the nominal and faulty situations of a complex system. A group of local models are selected to present the system in its full range of operation and this is the first layer multiple model. At the second layer, we create a group of model bank to describe the system in nominal and each faulty situations. By checking the validity of the second layer model banks, information of corresponding local models are used to initialize the adaptive models to have a precise approaching to the real system. Besides, model predictive control (MPC) is designed for the reference model of the adaptive process to generate proper reference input for achieving control goals while dealing the FTC problem. Simulations are given to show the validity of the proposed method.
Complete list of metadata
Contributor : Boutaib Dahhou <>
Submitted on : Tuesday, February 9, 2021 - 10:16:26 AM
Last modification on : Thursday, June 10, 2021 - 3:05:42 AM
Long-term archiving on: : Monday, May 10, 2021 - 6:02:36 PM


Files produced by the author(s)



Menglin He, Zetao Li, Boutaib Dahhou. Fault Tolerant Control Strategy Using Two-Layer Multiple Adaptive Models for Plant Fault. 2020 International Conference on Control, Automation and Diagnosis (ICCAD), Oct 2020, Paris, France. pp.1-6, ⟨10.1109/ICCAD49821.2020.9260565⟩. ⟨hal-03105849⟩



Record views


Files downloads