Predictive feedback control: An alternative to proportional-integral-derivative control

Abstract

Even though employed widely in industrial practice, the popular proportional-integral-derivative (PID) controller has weaknesses that limit its achievable performance. In this paper, an alternative control scheme that combines the simplicity of the PID controller with the versatility of model predictive control is presented. The result is a controller that combines the time-delay compensation capability of predictive control algorithms, the effectiveness of inferential control schemes for disturbance rejection, and the adaptation capabilities of switching controllers. The robust stability and performance of the controller are analysed. These results are then used to generate two tuning procedures. The design, implementation, and performance of the controller are illustrated via simulations on linear and non-linear systems.