This paper proposes a hybrid predictive controlstrategy composed of local predictive controllers which interactwith each other. This control strategy is applied to a solarpowered hot water storage (SHWS) system that suppliesenergy for a building cooling system. Two local controllers areproposed for the SHWS system and the building cooling systemrespectively. The main objective of the control approach is tooptimize the use of the solar energy in order to ensure thecooling requirement of the building. The proposed interactionsbetween the controllers offer a degree of freedom to theSHWS controller in order to satisfy the energy demand by thebuilding cooling system. The performances of this interactingcontrol architecture are illustrated using a TRNSYS model ofa solar cooling system.