This paper presents a fault detection method for a Variable Air Volume (VAV) system by means of stochastic qualitative reasoning. The VAV system, which is a building air conditioning system, controls the temperature and the volume of supplied air in the refrigerator and in the VAV valves for air conditioning. In order to ensure that the system is operating properly, it is necessary to detect the malfunctions in it from the perspective of the indoor environment. However, the various and changeable structure of each air conditioning system makes it difficult to formulate the physical relationships between its factors. In addition, because of cost concerns, the numerical information that is measured by sensors of limited numbers is not enough to detect problems in a complex system of that type. The goal of this study is to detect faults in this kind of system by means of the stochastic qualitative reasoning method. In this method, the complex structure of the VAV system can be expressed by a qualitative model. On the basis of the qualitative models that correspond to normal and abnormal operation, the behaviors of the system are analyzed in order to pinpoint the cause of the faults.
Fault detection, Qualitative reasoning, Qualitative model, Building air conditioning system, Simulation
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).