Coordination of distributed model predictive controllers using price-driven coordination and sensitivity analysis

In this paper, a coordination control algorithm based on hierarchical scheme is presented to coordinate several non-linear model predictive controllers (NMPC) working in parallel, with an upper layer, where a price-driven coordination technique is used to drive the controllers in such a way that some global constraints are satisfied. To coordinate the lower layers, it is used a price-adjustment algorithm based on Newton's method, in which a reformulation of Fiacco's work is used in order to obtain the sensitivity analysis for a nonlinear system no matter the set of active constraints. The efficiency of the scheme is evaluated using a simulation of a four-tank benchmark.