A study on the quantitative single and dual fault diagnosis of residential split type air conditioners in static operation using support vector machine method

Owing to the recent uprise in summer temperatures, the use of air conditioners has been increasing accordingly. Air conditioners consume a significant amount of energy, and defects in air conditioners usually could lead to even more consumption of energy. Hence, early detection of defects could not only enhance user satisfaction, but also conserve energy. In the present work, quantitative fault detection models for single-and dual-failure modes have been developed using a support vector machine technique based on refrigeration cycle simulation data including normal and defective conditions. The defect modes investigated in the present work include refrigerant shortage and degraded air flow rates for the evaporator and condenser of an air conditioner. The results indicate that the proposed method can predict the values of more than 95% of the defective parameters within +/- 5% for the single-failure mode, and more than 90% of the data within +/- 10% for the dual-failure mode.