Design and Simulation of a Fuzzy Supervisory Control System Integrated in a Small Public Building

Public administrations are encouraged to adopt actions for energy requalification of public buildings, which are responsible for considerable amounts of energy consumption. Low-cost technology improvements for energy renovation with high return on investment is a good opportunity when resource shortage hampers the adoption of expensive refurbishment. To this purpose, the integration of advanced control policies may optimize energy use of buildings with minimum hardware enhancement. In this contribution, we evaluated the performances of a control system based on fuzzy logic, supervising local Proportional-Integrative-Derivative (PID) controllers, through numerical simulation. First, the design process will be described, with particular concern to the proper selection of the rules in the knowledge base and the design of the inference engine. Then, its performances will be compared with the ones provided by the use of just ordinary PID control units. The fuzzy based control system was simulated as an addiction to the heating system of an existing small public building, that is the ?San Elpidio a Mare? community clinic located in Italy. The performances were assessed by means of simulations of an energy dynamic model developed in the DymolaTM environment, that had been previously validated with real consumption measurements. In this paper we will show not only what energy saving can be obtained by means of the fuzzy supervisor, but also its satisfactory performances in terms of stability of the control outputs.