Delayed generalized predictive control

This paper presents an alternative approach to the traditional generalized predictive control (GPC) presented by Clarke et al. (1987). The specific aim is to provide a new controller with control increment u(t) depending on the past values of the output y(t-k), y(t-k-l), ... at the present time t in order to represent the computational delays and/or compensate for system structural characteristics. A comparison is made between the GPC and the modified algorithm. The modified algorithm shows better performance than the traditional GPC