A fault tolerant closed-loop anesthesia system based on internal model control and extended state observer

This paper proposes a fault tolerant strategy for a closed-loop anesthesia system. To the authors best knowledge, it is the first time that the concept of fault tolerance applied to intravenous anesthesia. Internal model control (IMC) is used to construct the basic closed-loop system, and bispectral index (BIS) is selected as a measurement of patients depth of anesthesia (DoA) as well as a feedback signal. In this study, a sensor fault, BIS artifact, is considered. An extended state observer (ESO) and a switching strategy are selected for fault detection and control reconfiguration. All of the above mentioned works are established on a closed-loop aesthesia simulation platform. Simulation results demonstrate that the proposed scheme can deal with the considered kind of fault effectively.