Multidimensional prognostics for rotating machinery: A review

Abstract

Determining prognosis for rotating machinery could potentially reduce maintenance costs and improve safety and availability. Complex rotating machines are usually equipped with multiple sensors, which enable the development of multidimensional prognostic models. By considering the possible synergy among different sensor signals, multivariate models may provide more accurate prognosis than those using single-source information. Consequently, numerous research papers focusing on the theoretical considerations and practical implementations of multivariate prognostic models have been published in the last decade. However, only a limited number of review papers have been written on the subject. This article focuses on multidimensional prognostic models that have been applied to predict the failures of rotating machinery with multiple sensors. The theory and basic functioning of these techniques, their relative merits and drawbacks and how these models have been used to predict the remnant life of a machine are discussed in detail. Furthermore, this article summarizes the rotating machines to which these models have been applied and discusses future research challenges. The authors also provide seven evaluation criteria that can be used to compare the reviewed techniques. By reviewing the models reported in the literature, this article provides a guide for researchers considering prognosis options for multi-sensor rotating equipment.

Predictive repair scheduling of wind turbine drive‐train components based on machine learning

“…Besides the above listed examples applied to real failures, the prediction of warnings is regarded as an alternative monitoring approach (Leahy et al). Since mechanical failures are generally associated with multivariate processes, which can be treated via a sensor data‐fusion, the idea of the present study is to capture the information contained in the full set of SCADA data available close to an observed fault‐induced event, in order to predict future occurrences of the same type of event. A fault‐induced event is defined as an occurrence that leads to a complete turbine shutdown.…”

An incipient fault diagnosis method for rotating machinery based on bilateral spectrum and precession energy difference density spectrum

“…Time-frequency transform analysis of vibration signal is one of the most widespread and effective method among the existing methods to incipient fault diagnosis [1,2]. Most of the vibration signals of rotating machinery are non-stationary and nonlinear signals [3,4]. The main methods of dealing with such signals are including Short-time Fourier Transform [5], Wavelet transform [6], quadratic time-frequency distribution, Empirical Mode Decomposition, etc.…”

Canonical Variate Residuals-Based Fault Diagnosis for Slowly Evolving Faults

“…Rotating machines, such as centrifugal compressors, are widely used due to their high performance and robustness [1]. These machines typically operate under adverse conditions such as high pressures and speeds.…”