Uncertainty analysis and field implementation of a fault detection method for residential HVAC systems

Abstract

The vast majority of fault detection and diagnosis (FDD) methods for air conditioning systems are developed for packaged commercial systems. This paper presents a method for obtaining key operating parameters for air conditioning systems (airflow rate, cooling capacity, system efficiency, and refrigerant mass flow) in a way that is well-suited for the residential sector. The method relies on an air-side capacity estimate and will be compared and contrasted with a more traditional method that relies on a refrigerant-side capacity estimate. These methods are compared in terms of their sensitivity to input parameters, their uncertainty in the outputs, and their sensor requirements. The proposed air-side sensing method requires fewer sensors and has a significant advantage for residential split systems because it requires no outdoor unit sensors. The air-side sensing method is then successfully implemented on a field operating residential system. The experimental results show that the proposed method is sensitive to manually introduced airflow faults.