Condition monitoring techniques for machine bearings in non-stationary operation

摘要: Abstract Condition monitoring of machines in non-stationary operations can be considered as one the major challenges for research field rotating machinery diagnostics. The applications are ubiquitous: transport systems, energy vehicles, production plants. On these grounds, present work is devoted to measurement techniques and signal processing methods condition bearings undergoing operation conditions. Several types experimental set ups included study advantages drawbacks each discussed. For example, on side an ad-hoc test rig precision measurements developed utilized; other side, real scale campaigns at operating conversion systems (wind turbines) performed. A special focus important because often this kind devices fault detection becomes much more challenging due interplay with electromechanical couplings. To face drawback, most advanced post-processing need used. application constitute part diagnosis, especially its interpretation, respect controlled laboratory collected analysed through appropriate employed signals. In time domain, statistical features signals addressed novelty indexes principal component analysis. results support that Mahalanobis distance effective index order monitor level severity actual machine condition.