Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Evaluation of different aspect of fault detection methodologies based on Time Synchronous Averaging
Submission Author:
Racquel Knust Domingues , SC , Brazil
Co-Authors:
Racquel Knust Domingues, Julio Cordioli, Danilo Braga
Presenter: Racquel Knust Domingues
doi://10.26678/ABCM.COBEM2023.COB2023-1820
Abstract
The importance of proper maintenance planning in industrial sectors cannot be overstated, as it ensures maximum availability of assets and economic return. Predictive maintenance, which is based on monitoring the condition of the machine, is one of the most commonly used strategies. Several parameters can be monitored to predict the maintenance requirement of a machine, and monitoring the vibration response of a structure is the main solution for detecting defects and failures in machines or equipment. In general, vibratory signals contain two components, a deterministic part, and a random part. The deterministic part is related to the response of the structure and the operating condition, and can be used to identify any anomaly in the machine. The random part can be caused by random excitations, such as a flow or a slip, or electrical noise from the instrumentation. Thus, it has become essential to remove unwanted noise from a signal to make it an effective tool for fault detection in vibration analysis. In this context, the Time-Synchronous Averaging (TSA) method can be applied to separate deterministic and random signals. The method involves resampling the signal as a function of a frequency of a synchronization signal that can come from a tachometer or through rotation detection techniques using vibration. The method takes the average of the samples of the signal after resampling, and the resulting signal represents the deterministic component. This work focuses on the analysis of different parameters related to the application of TSA to detect failures in rolling bearings. In this sense, the approach is based on the analysis of frequency components related to the characteristic frequencies of the defect in order to use this information to infer the presence or not of the defect. The study evaluates the effectiveness of the methodology in detecting defects and the effects of varying its internal parameters, which may lead to the identification of the optimal parameters for its application in different scenarios.
Keywords
Time Synchronous Averaging, Fault Detection, Rolling bearing
