A model-based air balancing method of a ventilation system

In this paper we propose a novel and easily implementable model-based air balancing method for a ventilation system. The proposed method provides a well-defined form of balancing for a ventilation system, which can be effectively solved. The method follows three procedures: (i) mathematical modeling for duct system which is based on steady-state pressure balance; (ii) model parameters identification by supervised machine learning; (iii) determination of damper positions based on the model. The performance of this method is validated through testing in a duct testbed with five terminals. The results show that the final flow rates in all terminals match with the desired values with no more than 6.8% relative error.