Thermal Control of the New-borns Using a Cascade Approach

Abstract

: This paper proposes a novel control method for the skin temperature of a preterm new-born and the air temperature and relative humidity of a preterm infant incubator. In the proposed control structure two generalized predictive controllers (GPC) are connected in series with a third GPC for the incubator humidity control. In this paper, a decoupling method is presented, which is meant to reduce the coupling between the infant`s skin temperature and the incubator air space humidity. In order to implement the new proposed controller, a prediction model for an infant incubator with an ultrasonic humidification system was developed. For this purpose, this system was divided into three compartments, namely the preterm infant, the incubator air space, and the active humidification system. Each compartment was modeled by a transfer function using chaotic particle swarm optimization. The effectiveness of the new control structure was compared with that of two other feedback loop-based controllers, namely a simple skin servo-control method (S-GPC) and a skin servo-control cascade method without decoupling (C-GPC). The obtained results demonstrate that the suggested controller performs better than S-GPC and C-GPC in terms of integral absolute error, settling time, response time and disturbance rejection.