Detection and diagnosis of AHU sensor faults using principal component analysis method

Abstract

A strategy based on the principal component analysis (PCA) method is developed to detect and diagnose the sensor faults in air handling units (AHU). Sensor faults are detected using the Q statistic (squared prediction error, SPE). They are isolated using the Q statistic and Q contribution plot supplemented by simple expert rules. Two models are employed to deal with the heat balance and pressure flow balance separately to reduce the effects of the system nonlinearity and to ensure the PCA method's validity in different control modes. The fault isolation ability of the PCA method is also improved using the multiple models. Simulation tests and measurements from the BMS of a building are used to verify the PCA based strategy for automatic validation of AHU monitoring instrumentations and detecting/isolating AHU sensor faults under typical operating conditions. The robustness of the PCA based strategy in detecting/diagnosing sensor faults when typical component faults occur is examined.