Operational Control and Air Mixing Fault Detection and Diagnostics for Rooftop Units

Soft faults in rooftop units serving commercial buildings often go unnoticed during regular operation, masking underlying control and scheduling issues that lead to operational inefficiencies, energy waste, and compromised occupant comfort. Unlike hard faults, soft faults, such as scheduling errors, setpoint conflicts, and damper control issues, do not cause system disruption, making early detection critical to prevent escalating equipment wear. Conventional fault detection and diagnostics applications often employ simple rule-based algorithms with limited detection and diagnostic capabilities, leaving soft faults affecting rooftop units undetected. This study presents hybrid fault detection and diagnostic algorithms that employ rule-based thresholds, statistical analysis, and inverse modelling to identify operational mode anomalies, zone-rooftop unit mismatch faults, and mixing-box faults using data from building automation systems. The hybrid approach incorporates diagnostic capabilities to isolate fault patterns and pinpoint their root causes, enabling targeted maintenance, such as recalibrating schedules and adjusting dampers. The hybrid algorithms are tested using operational data from a rooftop unit serving a commercial building in Quebec, Canada. The methodology is demonstrated through practical examples that illustrate its effectiveness in detecting soft faults and isolating their root causes. For example, an operational-mismatch fault was detected in two of five thermal zones, with simultaneous heating and cooling observed in 8.77% and 8.16% of valid data points, respectively, during the cooling season, suggesting control misconfigurations or setpoint conflicts in the flagged zones. The proposed hybrid approach, which leverages existing BAS data, offers a cost-effective solution that supports building operational decisions, reduces energy consumption and greenhouse gas emissions, and enhances occupant comfort.