Application of hybrid predictive control for intelligent buildings

Recent technological innovations have caused increased interest in examining processes that have both continuous and discrete nature. Mathematical model of the process is traditionally associated with differential and difference equations typically derived from related physical laws. The paper deals with the advanced control design predictive strategy for hybrid systems in intelligent buildings consists of continuous - and discrete-time sub processes. A typical feature of hybrid systems is the interaction between continuous and discrete parts. Using hybrid control structure is one effective way to control such systems. A typical control structure is a two level computation structure. The upper level coordinates individual subsystems on the process level and derives the time and energy optimal switching control actions. Process level consists of individual subsystems which are controlled by discrete commands from the upper level such that the overall performance is achieved. Hybrid complex systems for intelligent buildings are working in multiple operating modes where each mode is controlled by its own dynamic characteristics. Transitions between modes are triggered when some state variable exceeds the threshold (state events), after expiration of specified time period (time events) or by external inputs (entry events).