A methodology for identifying information rich frequency bands for diagnostics of mechanical components-of-interest under time-varying operating conditions

• Published in: Mechanical systems and signal processing Vol. 142; p. 106739
• Main Authors: Schmidt, Stephan, Mauricio, Alexandre, Heyns, P. Stephan, Gryllias, Konstantinos C.
• Format: Journal Article
• Language: English
• Published: Berlin Elsevier Ltd 01.08.2020; Elsevier BV
• Subjects: [formula omitted]gram; Cyclostationarity; Damage detection; Diagnostic systems; Frequencies; Frequency Band Identification; Gearbox diagnostics; Gearboxes; Machinery condition monitoring; Mechanical components; Noise levels; Robustness (mathematics); Rotating machinery; Rotating machines; Time-varying operating conditions; Wind turbines; Gearbox diagnostics; Time-varying operating conditions; IFBIαgram; Frequency Band Identification; Cyclostationarity
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2020.106739

•The IFBIαgram is proposed for identifying information rich frequency bands.•It determines the optimal frequency band to detect specific damaged components.•It can be used for gear and bearing fault detection, identification and trending.•It is well suited for rotating machines operating under varying operating conditions.•Its potential for monitoring is highlighted on numerical and experimental datasets. Performing condition monitoring on rotating machines such as wind turbines, which operate inherently under time-varying operating conditions, remains a challenge. The signal components generated by incipient damage are masked by other signal components that are not of interest and high noise levels. In this work, a new method, referred to as the IFBIαgram, is proposed that is capable of identifying frequency bands that are rich with diagnostic information related to specific cyclic components. This allows the optimal frequency band to be determined for diagnosing the component-of-interest. It is shown on numerical and experimental gearbox data that this method is not only capable of detecting incipient damage, but is also robust to time-varying operating conditions. Therefore, it can be used to independently determine the condition of different mechanical components and it is robust to spurious transients.