Smart, sustainable and energy-efficient buildings have recently become a trend for intelligent and green building industry. One of the challenges for such a building is to minimise power consumption without compromising the occupant comfort. It is possible to sustain high comfort level with minimal energy consumption through intelligent multi-agent control and optimisation. Our work focuses on a new algorithmic aspect that makes buildings more intelligent, resource efficient and comfortable for occupants. A multi-agent control system with intelligent optimisation is proposed to optimise heating, ventilation and airconditioning (HVAC) processes using computational intelligence approaches. The optimum solutions generated through the optimisation engine and task scheduling for each agent will be used to sustain a high level of occupant comfort with minimal power consumption, increase energy utilisation efficiency, and consequently reduce energy cost. Multi-agent control system with intelligent optimisation 61 Experimental results and comparisons with other evolutionary approaches demonstrate the overall performance and potential benefits of the new framework.

Abstract. Smart, sustainable and energy-efficient buildings have recently become a trend for intelligent and green building industry. One of the challenges for such a building is to minimize power consumption without compromising the occupant comfort. It is possible to sustain high comfort level with minimal energy consumption through intelligent multi-agent control and optimization. Our work focuses on a new algorithmic aspect of a multi-agent control system that makes buildings more intelligent, resource efficient and comfortable for occupants. A multi-agent control system with intelligent optimization is proposed to optimize HVAC (heating, ventilation and airconditioning) processes using computational intelligence approaches. The optimum solutions generated through the optimization engine and task scheduling for each agent will be used to sustain a high level of occupant comfort with minimal power consumption, increase energy utilization efficiency, and consequently reduce energy cost. Experimental results and comparisons with other evolutionary approaches demonstrate the overall performance and potential benefits of the new framework.