Kar Hoou Hui

Dempster-shafer evidence theory for multi-bearing faults diagnosis

Support vector machines (SVMs) are frequently used in automated machinery faults diagnosis to classify multiple machinery faults by handling a high number of input features with low sampling data sets. SVMs are well known for fault detection that involves binary fault classifications only (i.e., healthy vs. faulty). However, when SVMs are used for multi-faults diagnostics and classification, they result in a drop in classification accuracy; this is because the adaptation of SVMs for multi-faults classifications requires the reduction of the multiple classification problem into multiple subsets of binary classification problems that result in many contradictory results from each individual SVM model. To overcome this problem, a novel SVM-DS (Dempster-Shafer evidence theory) model is proposed to resolve conflicting results generated from each SVM model and thus increase the classification accuracy. The analysis of results shows that the proposed SVM-DS model increased the accuracy of the fault diagnosis model from 76% to 94%, as SVM-DS continuously refines and eliminates all conflicting results from the original SVM model. The proposed SVM-DS model is found to be more accurate and effective in handling multi-faults diagnostic and classification problems commonly faced in the industries, as compared to the original SVM method.

Vibration analysis of multi stages rotor for blade faults diagnosis

Blade fault is one of the most common faults in turbomachinery. In this article, a rotor system consists of multiple rows of blade was developed. The effectiveness of conventional FFT spectrum and wavelet analysis in the diagnosis of multi stage blade rubbing faults is examined at different stages, variety of blade fault conditions, and different blades rubbing severity. Blade fault caused impacts and the use of wavelets as analysis tool to detect the blade faults was studied. Results showed that, vibration spectrum can clearly depict the location and the stage of blade rubbing, while it is difficult to be identified in wavelet analysis. The limitations of wavelet analysis for multi stage blade fault diagnosis were identified. Some probable solutions to improve wavelet time-frequency representation in blade fault diagnosis were also presented.

Time-Frequency Signal Analysis in Machinery Fault Diagnosis: Review

Growing demand of machines such as gas turbine, pump, and compressor in power generation, aircraft, and other fields have yielded the transformation of machine maintenance strategy from corrective and preventive to condition-based maintenance. Real-time fault diagnosis has grabbed attention of researchers in looking for a better approach to overcome current limitation. The parameters of health condition in machinery could be monitored thus faults could be detected and diagnosed by using signal analysis approach. Since some fault signals are non-stationary or time dependent in nature, therefore time-frequency signal analysis is crucial for machinery fault diagnosis. Common time-frequency signal analysis methods are such as short time Fourier transform (STFT), wavelets analysis, empirical mode decomposition (EMD), Hilbert-Huang transform (HHT), etc. This review provides a summary of the basic principle of signal analysis, the most recent researches, and some advantages and limitations associated to each types of time-frequency signal analysis method.

An improved wrapper-based feature selection method for machinery fault diagnosis

A major issue of machinery fault diagnosis using vibration signals is that it is over-reliant on personnel knowledge and experience in interpreting the signal. Thus, machine learning has been adapted for machinery fault diagnosis. The quantity and quality of the input features, however, influence the fault classification performance. Feature selection plays a vital role in selecting the most representative feature subset for the machine learning algorithm. In contrast, the trade-off relationship between capability when selecting the best feature subset and computational effort is inevitable in the wrapper-based feature selection (WFS) method. This paper proposes an improved WFS technique before integration with a support vector machine (SVM) model classifier as a complete fault diagnosis system for a rolling element bearing case study. The bearing vibration dataset made available by the Case Western Reserve University Bearing Data Centre was executed using the proposed WFS and its performance has been analysed and discussed. The results reveal that the proposed WFS secures the best feature subset with a lower computational effort by eliminating the redundancy of re-evaluation. The proposed WFS has therefore been found to be capable and efficient to carry out feature selection tasks.

Analysis of Residual Wavelet Scalogram for Machinery Fault Diagnosis

Wavelet analysis is a very useful tool for machinery faults diagnosis. However, actual application of wavelet analysis for machinery fault diagnosis in the field is still relatively rare. This is partly due to the fact that visual interpretation of wavelet results is often difficult and very challenging. This paper investigates an effective method to present wavelet analysis results in order to simplify the interpretation of wavelet analysis result for machinery faults diagnosis. Analysis of residual wavelet scalogram was proposed in this study as a mean to display and extract key faults signatures from raw sensor signals. Simulated signals were generated to test the feasibility of the proposed method. Test results showed that the proposed wavelet method provides a simple and more effective way to diagnose machinery faults.

Vibration Condition Monitoring: Latest Trend and Review

Vibration analysis has proven to be the most effective method for machine condition monitoring to date. Various effective signal analysis methods to analyze and extract fault signature that embedded in the raw vibration signals have been introduced in the past few decades such as fast Fourier transform (FFT), short time Fourier transform (STFT), wavelets analysis, empirical mode decomposition (EMD), Hilbert-Huang transform (HHT), etc. however, these is still a need for human to interpret vibration signature of faults and it is regarded as one of the major challenge in vibration condition monitoring. Thus, most recent researches in vibration condition monitoring revolved around using Artificial Intelligence (AI) techniques to automate machinery faults detection and diagnosis. The most recent literatures in this area show that researches are mainly focus on using machine learning techniques for data fusion, features fusion, and also decisions fusion in order to achieve a higher accuracy of decision making in vibration condition monitoring. This paper provides a review on the most recent development in vibration signal analysis methods as well as the AI techniques used for automated decision making in vibration condition monitoring in the past two years.

Time Frequency Analysis for Blade Rub Detection in Multi Stage Rotor System

Blade fault is one of the most causes of failure in turbo machinery. This paper discussed the time frequency analysis for blade rubbing detection from casing vibration signal. Feasibility of Short Time Fourier Transform (STFT), Wigner-Ville distribution (WVD) and Choi-Williams distribution (CWD) were examined for blade rub detection in a multi stage blade system through an experimental data. Analysis results of the experimental data showed that these time frequency analysis methods have some inevitable deficiencies in segregating the blade passing frequency (BPF) components of the three rotor stage signals. However, CWD demonstrated a better time-frequency resolution in analyzing the multi stage rotor system signal.

Introduction of Fingerprint Recognition Method for Machinery Faults Diagnosis

This paper presents a novel method to diagnose turbine blade faults by analysing wavelet result based on fingerprint recognition method. The study focuses on applying the fingerprint features extraction method to extract faults information from wavelet displays. Fingerprint recognition method studied in this paper is known as fingerprint bifurcations points. Experimental results show that this method could potentially be used as a features recognition method to identify the different types of machinery faults such as rubbing and rotor unbalance.

Machine Learning Tools in Machinery Faults Diagnosis: A Review

Machinery faults can be detected by various signal processing tools; however, they require human expertise to achieve maximum success. Machine learning tools can help to achieve automatic machinery-faults diagnosis. This paper provides a brief review of the most common machine learning tools.

Equipment Aging, Aging Detection, and Aging Management: A Review

It is becoming increasingly difficult to ignore the importance of equipment aging in critical industries such as power generation plants, oil and gas plants, etc. since many system components are approaching the end of their designed operational lifetime. New installations of system components in these critical industries will be extremely costly whilst any increase in demand will be too small to warrant totally new facilities. Aging management can be an alternative for this aging equipment to ensure its safety and reliability. This review aims to provide a brief understanding of equipment aging, aging detection and recent research into aging management.