Diagnosis of air-conditioner by using its dynamic property

Abstract

To avoid failures of an air-conditioner, we need suitable condition monitoring technique. Unfortunately, an air-conditioner has many parameters in it, i.e., inlet temperature, outlet temperature of a condenser or an evaporator, pressure of each component and input power for a compressor and so on. Furthermore, it is very difficult to identify failures in an air-conditioner because of small difference of parameters between normal and abnormal condition. In this paper, we proposed diagnosis technique by using support vector machine (SVM) with identified system parameters by ARX model. Because dynamic property of an air-conditioner identified by ARX model did not depend on operation condition, i.e., room temperature, outside temperature, our proposed technique did not need a lot of data. Finally, our proposed technique well identified the clogging in heat exchanger very well.