Automated fault diagnosis method for a variable air volume air handling unit

Abstract

Automated control methods in building HVAC systems have been widely applied. It is known that poorly maintained, degraded, and improperly controlled equipment wastes 15% to 30% of energy used in commercial buildings. So, we need to detect and cope with abnormal event of target system using intelligent control system. The expert rules were developed for application to single-duct variable air volume or constant air volume air handling unit. Expert rules are composed of 28 rules using only a small number of temperatures and control signals. If any one of rules is satisfied, a fault is indicated. In this study, we tested the performance of FDD method by expert rule using simulation data under single fault and multiple faults cases. Test of the expert rule set was performed using no-fault and faulty data produced by a simulation of a three-zone building conditioned with a VAV AHU. The simulations were carried out using summer season weather data in Seoul during the occupied period under no-fault and faulty conditions. FDD method using expert rules was found to indicate each fault accurately under single fault and multiple faults. ？ 2009 SICE.