Tuning fuzzy logic controllers for energy efficiency consumption in buildings

In EU countries, primary energy consumption in buildings represents about 40% of total energy consumption and more than a half of this energy is used for indoor climate conditions. The use of appropriate automatic control strategies for passive systems control could result in energy savings when compared to manual control. Fuzzy Logic Controllers (FLCs) have been successfully applied to a wide range of real-world problems from different areas. However, due to the complexity of the problem, the development of smart tuning techniques for these controllers will enable a rational operation and improved performance of FLCs and it is a necessary condition to solve it. The problem has some specific restrictions that get it to be very particular and complex because of there are large time restrictions due to the long computation of the model and to the necessity of a multiobjective fitness function, that enlarges the solution search space. In order to solve the first restriction, a two-stage tuning methodology is proposed, in which the first stage is a rough global tuning (using a Genetic Algorithm) and the second one is a refined local tuning (using a Simulated Annealing). The second restriction is solved by multiobjective techniques that allow to work with fitness functions with competitive objectives. This combination of tuning with different search space granularity and multiobjective approaches is a novel way of facing up to these kinds of problems. This work is being carried out within the framework of the GENESYS European Project.