Adaptive fixed-time terminal sliding mode control of a Peltier cell fused with a fuzzy fixed-time perturbation estimator

Abstract

This study focuses on constructing an enhanced dynamic model for the Peltier cell to develop a model-based temperature controller. To simplify the control system design, a reduced-order model of the system is developed. Perturbations in the reduced-order model with nominal parameters are compensated using a novel fixed-time observer, aligning it with the actual system. The proposed scheme utilizes measurements from thermometers on both the cold and hot sides of the cell to estimate the system states and perturbations. The parameters of the proposed observer are tuned automatically using a fuzzy inference system to provide an accurate state estimation and attenuate the effects of noise in measurements. Mathematical analyses demonstrate the fixed-time convergence of the estimation method. Accordingly, a novel adaptive fixed-time terminal sliding mode controller is designed based on the enhanced model to maintain the desired temperature on the cold side. The proposed controller adapts to the actual system and is reliable and cost-effective due to the use of the reduced-order model. Additionally, mathematical analyses show the fixed-time convergence of the tracking error to the sliding surface and guarantee convergence to the origin within a fixed time. Experimental tests conducted on a constructed Peltier platform demonstrate the improved efficiency of the proposed control method. Comparative results with prevalent controllers highlight the superior accuracy of the suggested controller in tracking the desired temperature despite the presence of perturbations.