Condition Monitoring Of Rolling Element Bearings Research Articles

Co-design Model for Neuromorphic Technology Development in Rolling Element Bearing Condition Monitoring

This paper presents an end-to-end condition monitoring co-design model, from vibration measurement to anomaly detection, where conventional signal processing principles are combined with neuromorphic sensing and computing concepts to enable investigations of the potential improvements offered by brain-like information processing technologies.
 The use of machine learning in condition monitoring became increasingly popular for intelligent fault diagnosis in the last decade, taking advantage of the rapid developments in deep learning.
 However, the high computational cost of training and using deep neural networks prevents the use of such solutions for analysing the bulk of data generated by the resource constrained edge devices, i.e., the condition monitoring sensor systems, as only a minor fraction of data can be transmitted to the cloud or edge servers for analysis.
 There is an untapped potential to process this data and thereby improve intelligent fault diagnosis models using event-triggered sensing, spiking neural networks, and neuromorphic processors that substantially can improve the energy efficiency and capacity of embedded machine learning condition monitoring solutions.
 The proposed co-design model is evaluated on two use-cases involving rolling element bearing failures, one based on a labelled laboratory environment dataset, and one based on a wind turbine drivetrain bearing failure representing a real-world scenario with stochastic changes of machine state and unknown labels of the bearing condition.
 By adjusting co-design parameters, the resulting hybrid conventional/neuromorphic model show a comparable accuracy in detection performance for the laboratory dataset compared to the state-of-the-art reported in the literature.
 Similarly, for the wind turbine drivetrain dataset a bearing fault detection time comparable to that in previous work is obtained.
 This shows the successful implementation of a hybrid conventional/neuromorphic co-design model for condition monitoring applications, offering novel opportunities to investigate performance trade-offs and efficiency improvements enabled by neuromorphic technologies.

Experimental Studies on Outer Race Fault Diagnosis of Rolling Element Bearings

Rolling element bearing condition monitoring and fault diagnosis can predict the condition of bearing and its components such as outer race, inner race, rolling elements and cage at any given instant. Online condition monitoring is more advanced and preferred over conventional methods. Advanced signal processing techniques and statistical analysis techniques aid in the component level fault prediction, its severity, interaction of different defect frequencies and remaining life of the component in a given system. However, the basic concept in most of the techniques is vibration data acquisition and identification of amplitude of vibration at defect frequencies. In this paper, the basic concepts of bearing defect frequencies are discussed and shown experimentally the presence of these frequencies. In experimentation 50mm diameter bearings with different faults are mounted on a simple test rig and vibration data is acquired under different loads at 2500 rpm. A simple frequency domain has been carried out to identify the defect.

A robust detector for rolling element bearing condition monitoring based on the modulation signal bispectrum and its performance evaluation against the Kurtogram

Envelope analysis is a widely used method for rolling element bearing fault detection. To obtain high detection accuracy, it is critical to determine an optimal frequency narrowband for the envelope demodulation. However, many of the schemes which are used for the narrowband selection, such as the Kurtogram, can produce poor detection results because they are sensitive to random noise and aperiodic impulses which normally occur in practical applications. To achieve the purposes of denoising and frequency band optimisation, this paper proposes a novel modulation signal bispectrum (MSB) based robust detector for bearing fault detection. Because of its inherent noise suppression capability, the MSB allows effective suppression of both stationary random noise and discrete aperiodic noise. The high magnitude features that result from the use of the MSB also enhance the modulation effects of a bearing fault and can be used to provide optimal frequency bands for fault detection. The Kurtogram is generally accepted as a powerful means of selecting the most appropriate frequency band for envelope analysis, and as such it has been used as the benchmark comparator for performance evaluation in this paper. Both simulated and experimental data analysis results show that the proposed method produces more accurate and robust detection results than Kurtogram based approaches for common bearing faults under a range of representative scenarios.

Condition monitoring of low-speed bearings — a review

Machines with operating speeds less than 600 rpm are classified as low-speed machinery. These machines are usually the most critical items in the production line, generally large and have high rotating inertias. Until recently, there has not been much interest in condition monitoring of these machines as they do not fail easily. However, if an expected failure does occur, the downtime and replacement costs can be enormous, which can lead to massive production loss. Unlike those medium- to high-speed machines, condition monitoring of low-speed machinery presents a challenge as traditional velocity and acceleration sensors are only sensitive to responses with high impact rates. The moving components of these machines that require condition monitoring are mainly bearings and gears. Due to the lack of interest until recently, there is limited literature on condition monitoring of low-speed machinery. In this review, the authors have attempted to gather together the innovative and advanced approaches on condition monitoring of low-speed machinery, with the main focus on rolling element bearing condition monitoring.

Application of Degree of Cyclostationarity in Rolling Element Bearing Diagnosis

Minor and random slip between rolling elements and races in rolling element bearings makes vibration signals have periodically time-varying ensemble statistics, which is known as cyclostationarity. Two second-order cyclostationary methods, the spectral correlation density (SCD) and the degree of cyclostationarity (DCS), are talked about in this paper based on a statistical model of rolling element bearings. The SCD provides redundant information in bi-frequency plane and cyclic frequency domain embodies the majority of it, which is a series of non-zero discrete cyclic frequencies completely reflecting the fault characters of rolling element bearings. The DCS has virtues of less computation and clearer representation, at the same time keeps the same characters with SCD in cyclic frequency domain. And the DCS is also proved to be resistant to the additive and multiplicative stationary noise. Simulation and experiential results from three rolling element bearing faults: outer race defect, inner race defect and rolling element defect, indicate practicability of the DCS analysis in rolling element bearing condition monitoring and fault diagnosis.

Applications of shock pulse technology to rolling element bearing condition monitoring

The Shock Pulse technique has been widely and successfully applied. Recent testing indicates its capability to determine bearing condition and lubrication condition in the presence of background machine noise. Examples of applications to helicopter gear box bearing monitoring will be presented as well as current data on other applications ranging from Railroad Car bearing to rocket engine turbopump bearing monitoring.