Temperature Control of Nonlinear Uncertain Systems via Model-Free Cascaded Controller: Application to an Infrared Emissometer

Abstract

This article introduces a cascade control system designed to precisely manage complex heating processes characterized by two actuator saturations, the absence of any cooling system, variable properties of the sample being heated, and a wide temperature span. The proposed control system combines an industrial PID for furnace heating element control with a model-free controller to control the temperature of the heated sample. We clearly describe the procedure to tune the controller and to accurately model the whole heating process. We have performed simulations to compare the performance of the control system we propose to the performance of conventional cascade controllers; our control system is superior in terms of accuracy and adaptability. We have successfully implemented the control system in an experimental instrument: an infrared emissometer. We have proved that the implementation is effective in controlling both tuned plants and unknown plants, without the need to modify the previous control configurations. The tests we have done also demonstrate that the achieved final model represents the whole experimental system with a high degree of accuracy, as the results from the simulations match with the experimental results.