Fault Detection Using the Zero Crossing Rate
A method of fault detection based on the zero crossing rate of the signal, Z1, and the zero crossing rate of the first order difference signal. Z2, is presented. It is shown that the parameter pair (Z1, Z2) possesses adequate discriminating potential to classify a signature as good or defective. The parameter pair also carries considerable information about the underlying system structure and it is shown that the different clusters for the pod ad defective signatures in the (Z1, Z2) space are related to the excited system poles. The sensitivity of the (Z1, Z2) parameter pair to the dominant system poles makes the technique very effective in causes where the excited system resonances in the good and defective signatures are different.