| Acknowledgements | 第6-7页 |
| Abstract | 第7-8页 |
| 摘要 | 第9-11页 |
| List of abbreviations and symbols | 第11-20页 |
| 1. Introduction | 第20-29页 |
| 1.1. Machine vibration causes | 第20-21页 |
| 1.2. Bearing fault diagnosis | 第21-24页 |
| a. Simulated outer race fault | 第22-24页 |
| b. Simulated inner race fault | 第24页 |
| 1.3. Signal acquirement | 第24-25页 |
| 1.4. Fault diagnosis literature review | 第25-28页 |
| 1.5. Stochastic Resonance | 第28-29页 |
| 2. Stochastic Resonance with Duffing Oscillator-Gauss potential for weak signal extraction | 第29-55页 |
| 2.1. Principle of Stochastic Resonance | 第29-35页 |
| 2.1.1 The periodic response analysis | 第30-32页 |
| 2.1.2 Frequency spectrum analysis | 第32-33页 |
| 2.1.3 Re-scaling frequency stochastic resonance | 第33-35页 |
| 2.2. Optimal Bandwidth detection based on Bayes Segmentation | 第35-36页 |
| 2.3. Application of the B-S algorithm into the frequency spectrum | 第36-37页 |
| 2.4. Peak energy index | 第37-39页 |
| a. Frequency Spectrum | 第37页 |
| b. Envelope Spectrum | 第37-38页 |
| c. Butterworth filter | 第38-39页 |
| 2.5. BS-PE analysis | 第39-40页 |
| 2.6. Duffing Oscillator | 第40-43页 |
| 2.7. Procedure of the proposed model 1 | 第43-44页 |
| 2.8. Experimental verification | 第44-54页 |
| 2.8.1. Case Western Reserve University bearing data set | 第44-48页 |
| 2.8.2. Bearing data from self-experiment developed at Zhejiang University | 第48-54页 |
| 2.9. Chapter conclusion | 第54-55页 |
| 3. Stochastic resonance polynomial index and its application in bearing fault diagnosis | 第55-70页 |
| 3.1. Underdamped second-order stochastic resonance model | 第56-57页 |
| 3.2. Stochastic Resonance Polynomial Index | 第57-63页 |
| 3.2.1 Indexes used in the proposed method | 第58-59页 |
| 3.2.2 Index variation with different noise intensities | 第59页 |
| 3.2.3 Stochastic Resonance Polynomial Index | 第59-63页 |
| 3.3. Procedure of the proposed model 2 | 第63-64页 |
| 3.4. Experimental verification | 第64-68页 |
| 3.4.1 CWRU bearing data set | 第64-67页 |
| 3.4.2 ZJU bearing data set | 第67-68页 |
| 3.5. Chapter conclusion | 第68-70页 |
| 4. FFT-sliding window in combination with adaptive stochastic resonance for bearing fault diagnosis | 第70-82页 |
| 4.1. Weighted power spectrum kurtosis index | 第71-72页 |
| 4.2. Improved adaptive stochastic resonance method | 第72-73页 |
| 4.2.1 Sliding window construction | 第72-73页 |
| 4.3. Procedure of the proposed model 3 | 第73-76页 |
| 4.4. Experimental verification | 第76-78页 |
| 4.4.1. CWRU bearing data set | 第76-77页 |
| 4.4.2. ZJU bearing data set | 第77-78页 |
| 4.5. Enhancement signal extraction by using sliding window mechanics | 第78-79页 |
| 4.6. Benchmark comparison | 第79-81页 |
| 4.7. Chapter conclusion | 第81-82页 |
| 5. Proposed Models Comparison | 第82-88页 |
| 5.1. Considerations for the model selection | 第82页 |
| 5.2. The quality house | 第82页 |
| 5.3. User's voice | 第82-83页 |
| 5.4. Researcher's voice | 第83页 |
| 5.5. Models proposed | 第83-84页 |
| 5.5.1. Stochastic Resonance(Model 1) | 第83页 |
| 5.5.2. Improved Stochastic Resonance (Model 2) | 第83-84页 |
| 5.5.3. Improved Adaptive Stochastic Resonance (Model 3) | 第84页 |
| 5.6. Solutions evaluation | 第84-87页 |
| a) Evaluation of the specific weight of each criterion | 第84-85页 |
| b) Fault identification | 第85页 |
| c) Performance in noise | 第85页 |
| d) Computational cost | 第85-86页 |
| e) Algorithm complexity | 第86页 |
| f) Adaptive method | 第86页 |
| g) Future research | 第86-87页 |
| h) Conclusions table | 第87页 |
| 5.7. Chapter conclusion | 第87-88页 |
| 6. Conclusions and Future work | 第88-90页 |
| 6.1. Conclusions | 第88-89页 |
| 6.2. Future work | 第89-90页 |
| Bibliography | 第90-93页 |
| Achievements | 第93页 |
