Combining discrepancy analysis with sensorless signal resampling for condition monitoring of rotating machines under uctuating operations

Abstract

This paper proposes a novel framework for monitoring the condition of a rotating machine (for example a gearbox or a bearing) that may be subject to load and speed fluctuations. The methodology is especially relevant in situations where no (or only noisy) shaft angular position measurements are available. Shaft angular position reference measurements are often not available due to physical constraints that render it difficult to install tachometers or encoders on the shaft of interest. The proposed methodology aims to simplify the task of monitoring a time-varying vibration signal by using a neural network to filter out the normal vibration components that generally tend to dominate the signal. The neural network may be optimised without the need for extensive datasets that are representative of different machine fault conditions. The envelope of the filtered signal is referred to as a discrepancy transform, since the discrepancy signal indicates the presence of fault-induced signal distortions. The discrepancy signal tends to be significantly simpler (smoother) than the original vibration waveform and may thus be resampled using a less accurate reference signal than would be required to resample the original waveform. A numerical gear model is used to illustrate the diagnostic potential of the proposed methodology.

An open set recognition methodology utilising discrepancy analysis for gear diagnostics under varying operating conditions

“…Discrepancy analysis uses a model of the features, extracted from data that were acquired from a healthy machine, to assign discrepancy measures to the features of new data. This discrepancy measure is used to form a discrepancy signal which is analysed using techniques such as synchronous averaging to infer the condition of the gearbox [35][36][37][38]. A brief overview of the features and models used in this paper are given in this section.…”

“…Discrepancy analysis is a powerful novelty detection technique that allows gearbox diagnostics to be performed under varying operating conditions [35][36][37][38][39]. In discrepancy analysis, a localised novelty score or discrepancy measure is assigned to the extracted localised features with a model optimised on the features of a healthy gearbox.…”

“…Gaussian mixture models [36], Gaussian models [39], hidden Markov models [38] can be used in the process. The synchronous average of the discrepancy signal is a very efficient technique for gear diagnostics under varying operating conditions [35][36][37][38]. Despite the fact that discrepancy analysis can be used to infer the condition of the machine and to characterise and trend damage, historical fault data will become available during the operational lifetime of rotating machines and need to be utilised to improve the condition monitoring task.…”

A tacholess order tracking methodology based on a probabilistic approach to incorporate angular acceleration information into the maxima tracking process

“…Combet and Gelman [11] improved the procedure proposed by Bonnardot et al [8], by automating the harmonic selection process. Heyns et al [16] used a clustering procedure to track an instantaneous frequency component in the spectrogram of the vibration signal. The IF estimate is low-pass filtered, since it contains noise due to the limited resolution provided by the spectrogram.…”

“…16) is minimised to find the frequency f = l∆f Γ that minimises the cost function −|Γ[n, l]| 2 and adheres to the constraints. The Lagrangian multiplier and slack variable associated with the ith constraint are denoted by λ i and s i , respectively.…”