Open AccessAnalysis and experimental application of a dead‐time compensator for input saturated processes with output time‐varying delays
Author(s) -
Alves Lima Thiago,
Tarbouriech Sophie,
Gouaisbaut Frédéric,
Prudêncio de Almeida Filho Magno,
García Pedro,
Claure Torrico Bismark,
Gonzalez Nogueira Fabrício
Publication year - 2021
Publication title -
iet control theory and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.059
H-Index - 108
eISSN - 1751-8652
pISSN - 1751-8644
DOI - 10.1049/cth2.12063
Subject(s) - control theory (sociology) , compensation (psychology) , robustness (evolution) , computer science , smith predictor , dead time , actuator , stability (learning theory) , mathematics , control (management) , control engineering , temperature control , engineering , pid controller , psychology , biochemistry , chemistry , artificial intelligence , machine learning , psychoanalysis , gene , statistics
Dead‐time compensators (DTCs) are a family of classical controllers derived from the Smith Predictor. Their main characteristic is that they explicitly employ the model of the open‐loop process to feedback a predicted value of the non‐delayed system, thus obtaining compensation of the delay. Such a perfect compensation is not achievable in the case of time‐varying delays. This paper addresses stability analysis of a DTC structure in this situation, in addition to considering saturating actuators and disturbances of limited energy. Specific challenges related to the DTC closed loop are taken into account in the developed theoretical conditions, which are expressed in terms of linear matrix inequalities by using an adequate Lyapunov–Krasovskii functional and generalised sector conditions. Furthermore, a new approach for the definition of the set of initial conditions in an augmented space in conjunction with the Lyapunov–Krasovskii functional is presented. Besides theoretical innovations, practical discussion about the relation between the tuning of DTC controllers and robustness for this class of systems is presented through numerical examples. An experimental application on a neonatal incubator prototype is carried out to emphasise the effectiveness of the results.