Neural-network-based identifier/controller for modern HVAC control

This paper reports the application of an artificial neural network (ANN) to serve both as a system identifier and as an intelligent controller for an air-handling system. A comprehensive software model has been established based on the specifications of a standard air-handling unit (AHU) on the market. The model is appropriate for testing various control algorithms including our new ANN identifier/controller. The ANN behaves as an identifier by continuously keeping track of all the real-time parameters associated with the whole air-handling system. Five actuating signals are produced based on the nonlinear error optimization of the outputs of the ANN, now served as a controller. The control target involves the minimization of two weighted factors - the errors between setpoints and control variables and the total energy consumption. The excellent performance of the ANN identifier/controller is illustrated by comparing it with that of a conventional proportional-integral-derivative (PID) controller.