Multiplexed optimization for complex air conditioning systems

Optimization has been considered as an efficient tool to realize energy efficiency in the operation of air conditioning (AC) systems. With the increase of complexity of AC systems, the computational complexity of real-time optimization appears to be a challenge for practical applications. In order to overcome the challenge, this paper proposes a multiplexed optimization scheme. Unlike conventional optimization that optimizes and updates all decision variables simultaneously, the proposed scheme optimizes and updates the decision variables sequentially and one decision variable at a time. The proposed scheme is compared with a conventional optimization method (in which the genetic algorithm is adopted) as regards computational load, energy performance and system stability. Case studies show that compared with the conventional optimization method, the computational burden of the proposed scheme is greatly reduced, up to 98.3%; the energy saving achieved by the proposed scheme is 6.8%, which is comparable to that achieved by the conventional method (6.7%); and the system operation stability is significantly enhanced since the average tracking errors for several monitored variables were reduced around 50%.