Diesel Engine Fault Diagnosis Using Wavelet Transforms Method Based on Labview Software

摘要

Experiment presented in this research, used vibration data obtained from a four-stroke, a 295 diesel engine.Fault of the internal-combustion engine was detected by using the vibration signals of the cylinder head.The fault diagnosis system was designed and constructed for inspecting the status and fault diagnosis of a diesel engine based on wavelet analysis and LabVIEW software.
　　The cylinder-head vibration signals were captured through a piezoelectric acceleration sensor that was attached to a surface of the cylinder head of the engine, while the engine was running at three speeds (620, 1000 and 1300 rpm) and four loads (0, 15, 30 and 45 N·m).Data was gathered from five different conditions associated with the cylinder head, such as single cylinder shortage, double cylinders shortage, intake manifold obstruction, exhaust manifold obstruction, and normal condition.Discrete wavelet transforms signal processing method on the engine cylinder head vibration signal with db5 and decomposition level 5 was used to decompose the signal into some of the details and approximations coefficients.Therefore, the energy was extracted from each frequency sub-band of normal and abnormal conditions as a feature of engine fault diagnosis.Thereby, the fault was distinguished by comparing the accumulations of energy in each sub-band of healthy and faulty conditions.
　　The results showed that detection of fault by discrete wavelet analysis is practicable.Statistical parameters as time-domain analysis such as standard deviation

目录

DEDICATION

声明

LIST OF CONTENT

LIST OF FIGURES

LIST OF TABLES

ABSTRACT

CHAPTER ⅠINTRODUCTION

1．1 objectives and problems statement

1．1．1 Problems statement

1．1．2 Objectives of the study

CHAPTER Ⅱ LITERATURE REVIEW

2．1 Diesel engine defect detection and monitoring methods

2．1．1 Vibration signal method

2．2 The vibration excitation sources of the diesel engine

2．2．1 Vibration response

2．2．2 Main sources of diesel engine noise

2．3 Time domain analysis

2．3．1 Feature extraction and selection from vibration signal

2．3．2 Time or statistical analysis

2．3．3 Standard deviation(STD)

2．3．4 Root mean square(RMS)

2．3．5 Peak level

2．3．6 Crest factor

2．3．7 Shape factor(SF)

2．3．8 Kurtosis

3．3．9 Skewness

CHAPTER Ⅲ MATERIALS AND METHODS

3．1 Materials and hardware design of fault diagnosis system

3．1．1 Location of experiment

3．1．2 The test diesel engine of experimental study

3．1．3 The CW40 electric dynamometer

3．1．4 Charge amplifier YE5853A

3．1．5 NI-Data acquisition card PCI6040 E

3．1．6 Shielded connection box(SCB-68)

3．1．7 Piezoelectric acceleration sensor-type CA-YD-106

3．1．8 Personal computer

3．1．9 LabVIEW software and engine accelerated vibration signal acquisition system

3．1．10 The virtual instrument construction and operation

3．2 Selection method for signal processing

3．2．1 Introduetion

3．2．2 Wavelet transform method

3．2．3 Continuous wavelet transforms

3．2．4 Multi-resolution analysis

3．3 Signal denoising

3．3．1 The threshold denoising method

3．3．2 Types of thresholding

3．4 Experimental settings and parameters selection

3．4．1 Setup of the experiment

3．4．2 Sensors installation on the diesel engine head

3．4．3 The selected sampling frequency and sampling points

3．4．4 Selection method for signal denoising

3．4．5 Selection of the optimum threshold level and mother wavelet decomposition for the denoising process

3．4．6 Selection of mother wavelet and wavelet decomposition level for signal analysis

3．5 Results and discussion

3．5．1 Wavelet analysis on cylinder head vibration signal

3．5．2 Characteristics of the signal energy and fault detection

3．5．3 Results of the analysis of time domain features extracted

CHAPTER Ⅳ BACK PROPAGATION NEURAL NETWORK AND SUPPORT VECTOR MACHINE

4．1 Back propagation neural network and support vector machine

4．1．1 Back propagation neural network(BPNN)

4．1．2 Architecture of backward propagation neural network

4．2 Support vector machine and signal pattern recognition

4．2．1 Construction of SVM algorithm

4．3 Results and discussions

4．3．1 Design of the back-propagation(BP)network

4．3．2 Design of the support vector machine training

4．3．3 Features extracted using SVM and BPNN

CHAPTER Ⅴ CONCLUSIONS AND RECOMMENDATIONS

5．1 Conclusions

5．2 Recommendations and future studies

ACKNOWLEDGMENTS

BIBLIOGRAPHY

APPENDIX A VI，FRONT PANEL AND BLOCK DIAGRAM