A practical approach for Generalized Predictive Control within an event-based framework

This work presents a combination of the Generalized Predictive Control (GPC) algorithm with event-based sampling techniques. The proposed control scheme preserves all well-known individual advantages of GPC and event-based sampling algorithms, respectively. The main benefits of this combination are an important reduction of actuation load meanwhile the control system performance is maintained within an acceptable level. Guidelines for a tuning procedure are given and tested for a wide set of industrial process models. Furthermore, the resulting algorithm is simple to be implemented and allows to establish a tradeoff between control performance and the number of actuations. The performance of the proposed control algorithm is first verified for a first-order plus delay process and afterwards it is evaluated by using a case study based on the greenhouse temperature control problem.

► We proposed a combination of GPC with event-based sampling techniques. ► The proposed control algorithm provides an important reduction of actuation load. ► A trade-off between actuation load (economic costs) and performance is reached. ► The resulting algorithm is straightforward to be implemented.