Design and performance analysis of a modified MRAC for second-order processes

Conventional controllers with fixed tuning fail to perform satisfactorily for processes with time varying nature. In such cases adaptive controllers are preferred as they are capable to provide suitable corrective measure according to the changes of process dynamics and undesired process upsets. In model reference adaptive control (MRAC) desired process output specifications are provided in the form of a reference model. Adaptation algorithm adjusts the controller output according to the difference between the outputs of the actual process and reference model along with a fixed adaptation gain. This paper deals with an improved MRAC technique designed for under-damped second-order processes with dead-time. In the proposed modified MRAC design, proportional-derivative (PD) feedback is incorporated with the conventional MRAC structure and the resulting controller is termed as MRAC-PD. Improved set-point tracking with fewer oscillations and faster recovery after undesired load changes justify the superiority of the proposed MRAC-PD controller compared to conventional MRAC and this fact is also substantiated with smaller values of performance indices - integral absolute error (IAE) and integral time absolute error (ITAE) under close-loop operation.