LAUSR

Abstract Zone-level sensor and actuator faults can substantially affect the energy and comfort performance of heating, ventilation, and air conditioning (HVAC) systems in commercial buildings. As these faults leave their fingerprints on a building automation system (BAS), algorithms which use features generated from BAS data streams as symptoms to detect and isolate faults can be developed. This paper presents an inverse greybox model-based fault detection and diagnostics method for zone-level sensor and actuator faults. The method maps the BAS data to a simplified physical representation of the zone temperature and airflow response. If the parameter estimates of the greybox model are abnormal, we treated them as potential symptoms of zone-level sensor and actuator faults. The method was demonstrated on the BAS data from 35 rooms of an academic office building. Inverse greybox models with six sensor and three actuator regressors and nine parameters for each room were trained by using a genetic algorithm. Based on the findings of a point-by-point condition survey, it was identified that these rooms contained three perimeter heater, two variable air volume (VAV) unit damper, one VAV pressure sensor, and two VAV reheat coil faults. All except a reheat coil fault were correctly identified based on the interpretation of the physical significance of the greybox model parameters. The BAS data and verified fault states from the condition survey are made publicly available to support the further development and assessment of fault detection and diagnosis algorithms.