Optimal Control Theory Applied to Dwelling Heating System

Optimal Control Theory has been applied to the design of controller to be implemented in a dwelling heating system. The system under consideration includes a floor basic heating system completed by additional electric convectors. A simplified model has been derived, including solar inputs and user significant variables. The criterion to be minimized is the integral over a finite receding horizon of energetic costs plus a penalty quadratic function taking user’s comfort into account. Minimum principle application leads to the solution of a two point boundary problem which integrates atmospheric disturbances over the horizon. The derived control law yields a series of control actions which optimality depends on local weather forecasts. From statistical analysis of a set of weather data, predictors have been derived and coupled with the controller.