Modified robust adaptive process control with improved transient performance and its application to atmospheric plasma spray process

Abstract

This chapter presents the σ-modification and the low-frequency learning to the model reference adaptive control (MRAC) to improve the transient performance of a closed-loop control system under uncertainties and external disturbances. The σ-modification improved the robustness to external bounded disturbances, and the low-frequency learning enabled fast adaption without creating high-frequency oscillations. The feasibility of the resulting robust adaptive control architecture is tested for the atmospheric plasma spray process (APSP) to achieve consistency in ceramic coatings through numerical simulations. The robustness and adaptability of the modified and the standard MRAC architecture are investigated under artificially induced disturbances. The proposed architecture shows better performance than the standard MRAC under external disturbances. The results depict the improved transient performance with faster convergence and smooth variations in closed-loop control signals without creating high-frequency oscillations or steady-state errors.