Buildings with shared spaces such as corporate office buildings, university dorms, etc. involve multiple occupants who are likely to have different temperature preferences. The building temperature control system needs to balance the desire of all users, as well as take the building energy cost into account. Centralized temperature management is challenging as the use comfort function is held privately and not centrally known. This paper proposes a distributed temperature control algorithm which ensures that a consensus is attained among all occupants, irrespective of their temperature preferences. Occupants are only assumed to be rational, in that they choose their own temperature set-points to minimize a combination of their individual energy cost and discomfort. We establish the convergence of the proposed algorithm to the optimal temperature set-point that minimizes the sum of the energy cost and the aggregate discomfort of all occupants. Simulations with realistic parameter settings illustrate the performance of the algorithm and provide insights on the dynamics of the system with a mobile user population.
American Control Conference (ACC), Chicago, 2015.