Automatic fault detection and diagnosis (FDD) in HVAC systems has the potential to ensure the comfort of building occupants, decrease energy consumption, and lower maintenance costs. Disadvantages of current FDD methods include the relatively high costs of engineering specific FDD systems for individual equipment and expensive sensor and microprocessor requirements. This paper presents two easy-to-implement FDD methods for rooftop air conditioners with thermal expansion valves. Both methods utilize simple rules and have low hardware and software requirements. One method requires a steady-state model, while the other is model independent. The two methods were tested using laboratory experiments for different fault types and fault levels under a variety of operating conditions and found to have good performance.

Units: I-P