DEVELOPMENT OF PARAMETER BASED COMPUTER MODELING TECHNIQUE FOR ONLINE FAULT DETECTION AND PREDICTION OF DAMPER AND SENSOR CONTROL FAULTS

Heating, Ventilation and Air Conditioning (HVAC) systems, often do not achieve the same level as itachieved at commissioning. An appropriate system analysis for Fault Detection and Diagnosis (FDD)can save material and energy significantly. FDD technology optimizes the use of the components ofthe system, so that they only be replaced or corrected when they are no longer useful. This paperpresents a complete methodology of introducing warning signal component in building managementsystem, which gets automatically activated as and when faults occur. The method is based upon AutoRegressive Exogenous (ARX) model and Recursive Parameter Estimation Algorithm. It is concludedthat the method is robust and can detect faults in dampers, sensors and PID control.