Real Time Monitoring Of Fault Occurrences And Fault Prediction For An Industrial Air Handling Unit

Abstract

Electronically Commutated Brushless DC (EC-BLDC) motor driven fan is an integral part of the Air Handling Unit (AHU). Maintenance of AHU is difficult as it is placed in remote location and high to reach areas. Evaluating the health of AHU helps in determining the possible failures associated with the system at an earlier stage. Efficient maintenance, reliable operation and accurate detection of incipient faults with sufficient lead time are the motivation factors for condition monitoring. In this work, the real time data of EC-BLDC fans such as actual speed, DC link voltage, current and temperatures of motor and the entire AHU module are acquired, and their performances are analyzed. To acquire real time data, a communication processor with MODBUS protocol is used.  With the help of a Virtual Instrumentation software, the acquired data are continuously stored and monitored.  The analysis of pre-fault data helps in identification of root cause of the fault occurrences in AHU. This paper also describes the methodology for collecting annotated time series data that can be used for prediction of anomalies as recommended by the Industrial Internet Consortium (IIC) Analytics framework.