Optimal indoor temperature control using a predictor

This article deals with optimal control of the indoor temperature in a building. The control system attempts to find an economic optimum to supply heat to the building with the use of a predictor for the indoor temperature, while maintaining a comfortable temperature in the building. A general control structure is described that uses a linear objective function, which is minimized by linear programming. This general control structure is applied to a specific test facility called passive climate system, whose main feature is that it uses natural ventilation by means of adjustable windows for cooling purposes. This article also describes some field tests with the optimal predictive control system applied to the passive climate system.>