This paper presents a methodology for the feature estimation of a new fault indicator focused on detecting gear mechanical degradation under different operating conditions. Preprocessing of acoustic emission signal is performed by applying chromatic transformation to highlight characteristic patterns of the mechanical degradation. In this study, chromaticity based on the computation of the hue, light, and saturation transformation of the main acoustic emission intrinsic mode functions is performed. Then, a topology preservation approach is carried out to describe the chromatic signature of the healthy gear condition. Thus, the detection index can be estimated. It must be noted that the applied chromatic monitoring process only requires the characterization of the healthy gear condition, being applicable to a wide range of operating conditions of the gear. Performance of the proposed system is validated experimentally. According to the obtained results, the proposed methodology is reliable and feasible for monitoring gear mechanical degradation in industrial applications.

Peer Reviewed