+353-1-416-8900REST OF WORLD
+44-20-3973-8888REST OF WORLD
1-917-300-0470EAST COAST U.S
1-800-526-8630U.S. (TOLL FREE)

Vibration-based Condition Monitoring. Industrial, Automotive and Aerospace Applications. Edition No. 2

  • Book
  • 448 Pages
  • June 2021
  • John Wiley and Sons Ltd
  • ID: 5212100

Vibration-based Condition Monitoring

Stay up to date on the newest developments in machine condition monitoring with this brand-new resource from an industry leader

The newly revised Second Edition of Vibration-based Condition Monitoring: Industrial, Automotive and Aerospace Applications delivers a thorough update to the most complete discussion of the field of machine condition monitoring. The distinguished author offers readers new sections on diagnostics of variable speed machines, including wind turbines, as well as new material on the application of cepstrum analysis to the separation of forcing functions, structural model properties, and the simulation of machines and faults.

The book provides improved methods of order tracking based on phase demodulation of reference signals and new methods of determining instantaneous machine speed from the vibration response signal. Readers will also benefit from an insightful discussion of new methods of calculating the Teager Kaiser Energy Operator (TKEO) using Hilbert transform methods in the frequency domain.

With a renewed emphasis on the newly realized possibility of making virtual instruments, readers of Vibration-based Condition Monitoring will benefit from the wide variety of new and updated topics, like:

  • A comprehensive introduction to machine condition monitoring, including maintenance strategies, condition monitoring methods, and an explanation of the basic problem of condition monitoring
  • An exploration of vibration signals from rotating and reciprocating machines, including signal classification and torsional vibrations
  • An examination of basic and newly developed signal processing techniques, including statistical measures, Fourier analysis, Hilbert transform and demodulation, and digital filtering, pointing out the considerable advantages of non-causal processing, since causal processing gives no benefit for condition monitoring
  • A discussion of fault detection, diagnosis and prognosis in rotating and reciprocating machines, in particular new methods using fault simulation, since “big data” cannot provide sufficient data for late-stage fault development

Perfect for machine manufacturers who want to include a machine monitoring service with their product, Vibration-based Condition Monitoring: Industrial, Automotive and Aerospace Applications will also earn a place in university and research institute libraries where there is an interest in machine condition monitoring and diagnostics.

Table of Contents

Chapter 1 Introduction and Background

1.1 Introduction

1.2 Maintenance strategies

1.3 Condition monitoring methods

1.3.1 Vibration analysis

1.3.2 Oil analysis

1.3.3 Performance analysis

1.3.4 Thermography

1.4 Types and benefits of vibration analysis

1.4.1 Benefits compared with other methods

1.4.2 Permanent vs intermittent monitoring

1.5 Vibration transducers

1.5.1 Absolute vs relative vibration measurement

1.5.2 Proximity probes

1.5.3 Velocity transducers

1.5.4 Accelerometers

1.5.5 Dual vibration probes

1.5.6 Laser vibrometers

1.6 Torsional vibration transducers

1.6.1 Shaft encoders

1.6.2 Torsional laser vibrometers

1.7 Condition monitoring - the basic problem


Chapter 2 Vibration Signals from Rotating and Reciprocating Machines

2.1 Signal classification

2.1.1 Stationary deterministic signals

2.1.2 Stationary random signals

2.1.3 Cyclostationary signals

2.1.4 Cyclo-non-stationary signals

2.2 Signals generated by rotating machines

2.2.1 Low shaft orders and subharmonics

2.2.2 Vibrations from gears

2.2.3 Rolling element bearings

2.2.4 Bladed machines

2.2.5 Electrical machines

2.3 Signals generated by reciprocating machines

2.3.1 Time-frequency diagrams

2.3.2 Torsional vibrations


Chapter 3 Basic signal processing techniques

3.1 Statistical measures

3.1.1 Probability and probability density

3.1.2 Moments and cumulants

3.2 Fourier analysis

3.2.1 Fourier series

3.2.2 Fourier integral transform

3.2.3 Sampled time signals

3.2.4 The discrete Fourier transform (DFT)

3.2.5 The fast Fourier transform (FFT)

3.2.6 Convolution and the convolution theorem

3.2.7 Zoom FFT

3.2.8 Practical FFT analysis and scaling

3.3 Hilbert transform and demodulation

3.3.1 Hilbert transform

3.3.2 Demodulation

3.4 Digital filtering

3.4.1 Realisation of digital filters

3.4.2 Comparison of digital filtering with FFT processing

3.5 Time/frequency analysis

3.5.1 The short time Fourier transform (STFT)

3.5.2 The Wigner-Ville distribution

3.5.3 Wavelet analysis

3.5.4 Empirical mode decomposition

3.6 Cyclostationary analysis and spectral correlation

3.6.1 Spectral correlation

3.6.2 Spectral correlation and envelope spectrum

3.6.3 Wigner-Ville spectrum

3.6.4 Cyclo-non-stationary analysis


Chapter 4 Fault Detection

4.1 Introduction

4.2 Rotating machines

4.2.1 Vibration criteria

4.2.2 Use of frequency spectra

4.2.3 CPB spectrum comparison

4.3 Reciprocating machines

4.3.1 Vibration criteria for reciprocating machines

4.3.2 Time/frequency diagrams

4.3.3 Torsional vibration


Chapter 5 Some special signal processing techniques

5.1 Order tracking

5.1.1 Comparison of methods

5.1.2 Computed order tracking(COT)

5.1.3 Phase demodulation based COT

5.1.4 COT over a wide speed range

5.2 Determination of instantaneous machine speed

5.2.1 Derivative of instantaneous phase

5.2.2 Teager Kaiser and other energy operators

5.2.3 Comparison of time and frequency domain approaches

5.2.4 Other methods

5.3 Deterministic/random signal separation

5.3.1 Time synchronous averaging

5.3.2 Linear prediction

5.3.3 Adaptive noise cancellation

5.3.4 Self adaptive noise cancellation

5.3.5 Discrete/random separation (DRS)

5.4 Minimum entropy deconvolution

5.5 Spectral kurtosis and the kurtogram

5.5.1 Spectral kurtosis - definition and calculation

5.5.2 Use of SK as a filter

5.5.3 The kurtogram


Chapter 6 Cepstrum analysis applied to machine diagnostics

6.1 Cepstrum terminology and definitions

6.1.1 Brief history of the cepstrum and terminology

6.1.2 Cepstrum types and definitions

6.2 Applications of the real cepstrum

6.2.1 Practical considerations with the cepstrum

6.2.2 Detecting and quantifying harmonic/sideband families

6.2.3 Separation of forcing and transfer functions

6.3 Modifying time signals using the real cepstrum

6.3.1 Removing harmonic/sideband families

6.3.2 Enhancing/removing modal properties

6.3.3 Cepstrum pre-whitening


Chapter 7 Diagnostic Techniques for particular applications

7.1 Harmonic and sideband cursors

7.1.1 Basic principles

7.1.2 Examples of cursor application

7.1.3 Combination with order tracking

7.2 Gear diagnostics

7.2.1 Techniques based on the TSA

7.2.2 Transmission error as a diagnostic tool

7.2.3 Cepstrum analysis for gear diagnostics

7.2.4 Separation of spalls and cracks

7.2.5 Diagnostics of gears with varying speed and load

7.3 Rolling element bearing diagnostics

7.3.1 Signal models for bearing faults

7.3.2 A semi-automated bearing diagnostic procedure

7.3.3 Alternative diagnostic methods for special conditions

7.3.4 Diagnostics of bearings with varying speed and load

7.4 Reciprocating machine and IC engine diagnostics

7.4.1 Time/frequency methods

7.4.2 Cylinder pressure identification

7.4.3 Mechanical fault identification


Chapter 8 Fault simulation

8.1 Background and justification

8.2 Simulation of faults in gears

8.2.1 Lumped parameter models of parallel gears

8.2.2 Separation of spalls and cracks

8.2.3 Lumped parameter models of planetary gears

8.2.4 Interaction of faults with ring and sun gears

8.3 Simulation of faults in bearings

8.3.1 Local faults in LPM gearbox model

8.3.2 Extended faults in LPM gearbox model

8.3.3 Reduced FE casing model combined with LPM gear model

8.4 Simulation of faults in engines

8.4.1 Misfire

8.4.2 Piston slap

8.4.3 Bearing knock


Chapter 9 Fault trending and prognostics

9.1 Introduction

9.2 Trend analysis

9.2.1 Trending of simple parameters

9.2.2 Trending of “impulsiveness”

9.2.3 Trending of spall size in bearings

9.3 Advanced prognostics

9.3.1 Physics-based models

9.3.2 Data-driven models

9.3.3 Hybrid models

9.3.4 Simulation-based prognostics

9.4 Future developments

9.4.1 Advanced modelling

9.4.2 Advances in data analytics



Robert Bond Randall University of New South Wales, Australia.