Multi-objective optimization of building climate control systems using fuzzy-logic

Frequently the user of multi-variable control systems is interested in operating them in a demand- or event-responsive manner according to various, sometimes opposing performance criteria. E.g. within well isolated low-energy houses there is an increasing requirement to coordinate the control of heating, ventilation and air conditioning systems (HVAC) in such a way that both economy and comfort criteria can be considered with a user-specific tradeoff. In order to find an on-line solution of this multi-objective process optimization problem, a new supervisory control concept has been developed at IITB. By means of a simple slide button the user is enable to choose his individual weighting factors for the economy and comfort criteria which are taken to optimize the reference commands of heating and ventilation controllers. The disturbing influence of external climate changes is considered as well as variations of the room occupancy. The performance of the fuzzy-based multi-objective optimization concept which has been implemented in a test environment at IITB is analyzed and discussed by means of experimental results.