Bearing Severity Fault Evaluation Using Contour Maps—Case Study

Abstract

Increasing industrial competitiveness has led to an increased global interest in condition monitoring. In this sector, rotating machinery plays an important role, where the bearing is one of the most critical components. Many vibration-based signal treatments are already being used to identify features associated with bearing faults. The information embedded in such features are employed in the construction of health indicators, which allow for evaluation of the current operating status of the machine. In this work, the use of contour maps to represent the diagnosis map of a bearing, used as a health map, is presented for the first time. The results show that the proposed method is promising, allowing for the satisfactory detection and evaluation of the severity of bearing damage. In this initial stage of the research, our results suggest that this method can improve the classification of bearing faults and, therefore, optimise maintenance processes.

Construction of Fault Evolution Maps for Damage Level Assessment of Railway Bearings

“…There is a lack of criteria to evaluate the state of the components of a bearing and to relate them to a fault level. However, the use of contour maps [14] allows to characterise the behavioural pattern of a bearing defect, by relating the failure frequency of a component to the amplitude changes it experiences as the failure level increases.…”

“…In recent years, studies developed using artificial intelligence have increased significantly, where surface learning and deep learning algorithms are protagonists in the detection, diagnosis, and prognosis of bearings [2], [3], [8]- [11]. On the other hand, an increase in studies associated with traditional vibration signal processing techniques is also evident in the literature [7], [12]- [14]. In these studies, envelope analysis takes centre stage in the demodulation of resonant zones of signals associated with frequencies of bearing failures.…”

“…In these studies, envelope analysis takes centre stage in the demodulation of resonant zones of signals associated with frequencies of bearing failures. Moreover, a novel methodology based on contour maps has been presented for the characterisation and construction of fault evolution maps, which serve as a reference in the diagnosis of bearings [14].…”

“…The main objective of this work focuses on the evaluation of the performance of the bearing diagnosis methodology based on contour maps, presented in Soto-Ocampo et al [14]. Contour maps are graphical illustrations used to represent spatial functions by incorporating isolines.…”

The Use of Machine Learning Methods for Binary Classification of the Working Condition of Bearings Using the Signals of Vibration Acceleration

“…Цей потужний підхід дає спеціалістам з діагностики можливість використовувати перелік емпіричних індикаторів для детекції несправностей підшипникових вузлів, однак передбачає ручний вибір ширини фільтру та аналіз спектрів сигналу, що робить цей підхід затратним до людських ресурсів. Інший підхід до аналізу огинаючої пропонується у роботі [7]. Автори статті використовують синтезований набір даних сигналів прискорень підшипників для розрахунку ізоліній та побудови діагностичних карт.…”