A Distributed Multi Party Consensus Algorithm for Temperature Control in Smart Buildings

Abstract

Smart building technology with multi occupants with shared spaces such shopping mall, airport lunges, corporate house, hotels and school colleges etc. are often occupied by multiple occupants that typically have different temperature requirements. Attaining a preferable temperature based on the occupants in a multi zone area is a challenging problem. Also in a multi zone space if temperature is controlled at different set point then it also increases the energy savings in the overall system. Using the concept of consensus we propose a distributed multiparty control strategy to maintain the different temperature set point in different zone. The space under concern are partitioned into different zones (or parties), among which each zone is describes by a temperature model, is consider an agent. The proposed method within the same party are synchronise with each other, but agents or zones from different parties have different target trajectories. One typical treatment on this is to use complex numbers to describe relationship among all target trajectories in different parties, which nevertheless significantly reduces its applicability to problems with complex target formations. In this paper we propose a matrix-based technique to capture information interactions among agents of different parties, making it suitable for tackling consensus problems for a larger variety of formations. Simulation studies demonstrate the effectiveness of the proposed approach.