Global optimization of large scale HVAC system

A large-scale Variable Air Volume (VAV) air-conditioning system composed of several subsystems is established. It´s a multi-variable, strongly coupled, time variant, nonlinear and large time delay system. The system is wholly analyzed with large-scale system theory(LSST) in this paper. Iterative learning control (ILC) is adopted to solve the problems of long steady-state time and large overshoot when set-points change. The system is divided based on “decomposition and coordination” strategy. Decentralized steady-state and energy consumption models are built for global optimal control. For the evaluation of the large scale hierarchical control system theory, ILC and steady-state optimization strategy (LSHC-ILC-SO), a case is studied on the VAV experimental platform. The results show that all devices work under coordinated conditions with coordination strategy under variable loads. The global steady state optimization can reduce the total energy consumption.