A decision tree based data-driven diagnostic strategy for air handling units
Data-driven methods for fault detection and diagnosis of air handling units (AHUs) have attracted wide attention as they do not require high-level expert knowledge of the system of concern. This paper presents a decision tree based data-driven diagnostic strategy for AHUs, in which classification and regression tree (CART) algorithm is used for decision tree induction. A great advantage of the decision tree is that it can be understood and interpreted and therefore its reliability in fault diagnosis can be validated by both testing data and expert knowledge. A steady-state detector and a regression model are incorporated into the strategy to increase the interpretability of the diagnostic strategy developed. The proposed strategy is validated using the data from ASHRAE 1312-RP. It is shown that this strategy can achieve a good diagnostic performance with an average F-measure of 0.97. The interpretation of the diagnostic decision tree using expert knowledge showed that some diagnostic rules generated in the decision tree comply with expert knowledge. Nevertheless, the interpretation also indicated that some diagnostic rules generated are not reliable and some of them are only valid under certain operating conditions, which indirectly demonstrated the importance of the interpretability of fault diagnostic models developed using data-driven methods.
Please refer to publisher version or contact your library.