Second order sliding mode approaches to fault detection and control of infinite dimensional systems

This presentation focuses on two approaches, both based on second order sliding modes, for detecting faults and controlling certain classes of infinite dimensional systems. Two concrete applications will be presented. The first approach considers a three-dimensional distributed-parameter model of an industrial furnace and addresses the problem of detecting faults in the heaters by using a certain number of suitably located temperature probes internal to the furnace. A second order sliding mode observer is designed, which produces suitable residual signals allowing to detect faults in the furnace electrical heating system. Simulation results will be illustrated. Secondly, a novel electro-active polymer, Ionic Polymer-Metal Composite (IPMC), is considered. This class of materials bends when solicited by an external electric field, and it is well known to exhibit fractional-order systems behaviour. Second order sliding mode algorithms, suitably tailored to the fractional order systems setting, are exploited to address the positioning problem of an IPMC membrane. Experimental results demonstrates the effectiveness of the latter approach.