Damage detection in vibration signals using non parametric time-frequency representations

The vibration analysis is the most used tool for fault detection in rotating machines, thereby its low cost in comparison with others tool as ultrasound. This tool accomplished of digital signal processing is very powerful for make decisions and faults identification. Time-frequency (TF) representations are one of the most popular characterization methods for non-stationary vibration signals. Despite of their potential advantages, these representations suffer of large quantity of redundant and irrelevant data which makes them difficult to use for classification purposes. In this work, a methodology for reduction of irrelevant and redundant data is explored. This approach consists on removing irrelevant data, applying a relevance measure on the t-f plane that measures the dependence of each t-f point with the class labels. Results shows that the relevance measures increase the performance classification and the faults are totally differentiable.