ABSTRACT | 第1-9页 |
ACKNOWLEDGEMENTS | 第9-10页 |
TABLE OF CONTENTS | 第10-14页 |
LIST OF FIGURES | 第14-19页 |
LIST OF TABLES | 第19-20页 |
LIST OF ABBREVIATIONS | 第20-22页 |
CHAPTER 1 INTRODUCTION | 第22-30页 |
·Background and motivation | 第22-25页 |
·Problem statement | 第25-26页 |
·Research objectives | 第26-27页 |
·Contributions | 第27-29页 |
·Organization | 第29-30页 |
CHAPTER 2 LITERATURE REVIEW | 第30-43页 |
·Introduction | 第30页 |
·Data processing | 第30-37页 |
·Time domain | 第31-33页 |
·Frequency domain | 第33-34页 |
·Time-frequency domain | 第34-37页 |
·Intelligent fault diagnosis and prognosis | 第37-41页 |
·Feature extraction | 第38页 |
·Feature selection | 第38-39页 |
·Fault diagnosis and prognosis | 第39-41页 |
·Summary | 第41-43页 |
CHAPTER 3 A FAST AND ADAPTIVE VARYING-SCALE MORPHOLOGICAL FILTER FOR BEARING FAULT DIAGNOSIS | 第43-59页 |
·Introduction | 第43-44页 |
·Mathematical theory of morphological analysis | 第44-46页 |
·Theoretical fundamental of morphological filter | 第44-45页 |
·The structure elements(SEs) | 第45-46页 |
·Principle of the varying-scale morphological analysis | 第46-49页 |
·Simulation study | 第49-52页 |
·Experimental validation | 第52-58页 |
·Summary | 第58-59页 |
CHAPTER 4 A DOPPLER TRANSIENT MODEL FOR LOCOMOTIVE BEARING FAULT DIAGNOSIS | 第59-82页 |
·Introductions | 第59-60页 |
·Theoretical Background | 第60-63页 |
·Doppler effect | 第60-61页 |
·Transient Model Based on the Laplace Wavelet | 第61-62页 |
·Correlation Analysis | 第62-63页 |
·Proposed Doppler Transient Model Based on Laplace Wavelet and Spectrum correlation Assessment | 第63-68页 |
·Doppler Distortion of the Transient Model Based on the Laplace Wavelet | 第64-66页 |
·Envelope Spectrum Correlation Assessment | 第66-67页 |
·Parameter Identification and Locomotive Bearing Fault Detection | 第67-68页 |
·Simulation Validation of the Proposed Method | 第68-71页 |
·Application of the Proposed Method to Real Locomotive Bearing Fault Diagnosis | 第71-80页 |
·Summary | 第80-82页 |
CHAPTER 5 A GENERIC SUPPORT VECTOR REGRESSIVE CLASSIFIER FOR IDENTIFYING THE ROTATING MACHINERY FAULT PATTERNS | 第82-106页 |
·Introduction | 第82页 |
·Theoretical background | 第82-88页 |
·Wavelet packet transform | 第82-83页 |
·Support Vector Machine | 第83-85页 |
·Support Vector Regression | 第85-88页 |
·Proposed health status identification scheme | 第88-92页 |
·Fault feature extraction | 第88-90页 |
·Fault feature selection | 第90-91页 |
·Fault pattern recognition using a generic multi-class solver | 第91-92页 |
·Experimental validation of the proposed intelligent machine fault diagnosis scheme | 第92-104页 |
·Case 1:The proposed scheme validated by bearing fault data | 第92-99页 |
·Case 2:The proposed scheme validated by gear fault data | 第99-103页 |
·Discussion on the effect of wavelet basis function on the performance of the proposed scheme | 第103-104页 |
·Summary | 第104-106页 |
CHAPTER 6 TW0-LAYER SUPPORT VECTOR REGRESSIVE MACHINES FOR RECOGNIZINGBEARING FAULT PATTERNS AND SIZES | 第106-123页 |
·Introduction | 第106-107页 |
·Two-layer SVRMs for bearing health evaluation | 第107-111页 |
·Fault feature extraction | 第107-108页 |
·Using two-layer SVRMs for accurate bearing fault diagnosis | 第108-110页 |
·Two-layer SVRMs parameter selection and optimization | 第110-111页 |
·Validation of the proposed method | 第111-118页 |
·Comparisons with other SVM and ANN methods | 第118-121页 |
·Summary | 第121-123页 |
CHAPTER 7 REMAINING USEFUL LIFE ESTIMATIONFOR THE IMPELLERS OF SLURRY PUMPS USING THEHEALTH STATUS PROBABILITY ESTIMATION | 第123-138页 |
·Introduction | 第123-125页 |
·The tested slurry pump | 第125页 |
·The proposed method for RUL prognostics | 第125-133页 |
·Sensitive features extraction | 第127-132页 |
·Health status probability estimation provided by support vector machine | 第132-133页 |
·Remaining useful life estimation by combining the health status probability and the historical data | 第133页 |
·Case studies | 第133-137页 |
·Summary | 第137-138页 |
CHAPTER 8 CONCLUSIONS AND FUTURE WORK | 第138-143页 |
·Conclusions | 第138-139页 |
·Future work | 第139-140页 |
·Awards and publications | 第140-143页 |
BIBLIOGRAPHY | 第143-156页 |
中文简介 | 第156-161页 |