To take the advantages of both generalized predictive control (GPC) and proportion-integral-differential control (PID), it is feasible to combine them together to form up a new predictive control algorithm (PID-GPC). However, there is large online computational burden in GPC algorithm itself. And PlD-GPC algorithm has also increased the computation quantity a little. The overall computational burden is so heavy that it restrains the on-line application of PID-GPC. Hence, it is imperative to improve the control algorithm to broaden its application. This paper is concerned with lightening computational burden by exploring its internal mechanism and puts forward an improved algorithm, which just needs to pick the reciprocal of a number, instead of calculating the inverse matrix. The simulation results show that this improved algorithm is feasible and effective.
Published in:
Machine Learning and Cybernetics, 2008 International Conference on
(Volume:4
)
Date of Conference: 12-15 July 2008