CONDITIONAL MOMENTS ANALYSIS OF TRANSIENTS WITH APPLICATION TO HELICOPTER FAULT DATA

Conditional time-frequency moments are discussed and applied to the Westland helicopter test data. The conditional moments we use have a simple physical interpretation, namely, they are the mean, median and mode frequencies at a given time, and the spread about the mean frequency at a given time (the `instantaneous spectral bandwidthʹ). We show that they characterise the faults well and can differentiate between different fault classes. Using spectrograms for their calculation enables real-time computation. Accordingly, these conditional moments are a sparse feature set that can potentially serve as an on-line diagnostic tool for detecting and monitoring developing faults.