Experimental Assessment of a Preliminary Rule-Based Data-Driven Method for Fault Detection and Diagnosis of Coils, Fans and Sensors in Air-Handling Units

Data-driven Automated Fault Detection and Diagnosis (AFDD) methods represent one of the most promising options for improving energy, environmental and economic performance of Air-Handling Units (AHUs). In this paper, a curated experimental faulted and unfaulted dataset associated to the field operation of a typical real AHU is firstly presented; a new rule-based data-driven AFDD method for fault detection and diagnosis of coils, fans and sensors is developed and its accuracy has been assessed in contrast with measured data.