In this paper the application of a novel robust predictive controller for tracking periodic references to a section of Barcelona's drinking water network is presented. The system is modeled using a large scale uncertain differential-algebraic discrete time linear model in which it is assumed that a prediction of the water demand is available and that it is affected by unknown and bounded uncertainties. The control objective is to satisfy the water demand while trying to follow a given reference of the level of the tanks of the network. The controller considered has been modified to account for algebraic equations and large scale models and it joins a dynamic trajectory planner and a robust predictive controller in a single layer to guarantee that the closed-loop system converges asymptotically to a neighborhood of optimal reachable periodic trajectory satisfying the constraints for all possible uncertainties even in the presence of sudden changes in the reference. To demonstrate these properties three different simulation scenarios have been considered.