Nonlinear System Identification for Damage Detection

Nonlinear System Identification for Damage Detection
Charles R. Farrar, Keith Worden, Michael D. Todd, Gyuhae Park, Jonathon Nichols, Douglas E. Adams, Matthew T. Bement, Kevin Farinholt
CONTENTS
LIST OF FIGURES . viii
LIST OF TABLES .xv
ABSTRACT .1
1. INTRODUCTION 2
1.1 The Structural Health Monitoring Process 2
1.2 Annual Workshops 4
2. COMMONLY USED DAMAGE FEATURES AND THEIR LIMITATIONS 4
3. TYPES OF DAMAGE THAT CAN PRODUCE NONLINEAR
SYSTEM RESPONSE 5
4. MOTIVATION FOR EXPLORING NONLINEAR SYSTEM IDENTIFICATION
METHODS FOR SHM AND DEFINITIONS OF TERMS .9
5. A DISCUSSION OF NONLINEAR SYSTEM RESPONSE VERSUS
NONSTATIONARY SYSTEM RESPONSE .12
6. NONLINEAR INDICATOR FUNCTIONS .17
6.1 Basic Signal Statistics .18
6.2 Coherence Function 24
6.3 Linearity and Reciprocity Checks .26
6.4 Harmonic Distortion .37
6.5 Frequency Response Function Distortions .42
6.6 Probability Density Function 47
6.7 Correlation Tests .49
6.8 The Holder Exponent 50
6.9 Linear Time Series Prediction Errors 56
6.10 Nonlinear Auto-Regressive Moving Average with Exogenous
Inputs Models 61
6.11 Hilbert Transform .65
6.12 Frequency Domain ARX Models 68
6.13 Higher Order Spectra 75
6.14 Nonlinear Acoustics Methods .81
vLA-14353-MS Nonlinear System Identification for Damage Detection
7. APPLICATIONS OF NONLINEAR DYNAMICS SYSTEMS THEORY .83
7.1 Modeling a Cracked Beam as a Bilinear System 84
7.2 Chaotic Interrogation of a Damaged Beam .88
7.2.1 Local Attractor Variance .88
7.2.2 Detection of Damage in a Cantilever Beam Using the
Local Attractor Variance .90
8. NONLINEAR SYSTEM IDENTIFICATION APPROACHES .93
8.1 Restoring Force Surface Model 93
8.2 Gradient Descent Methods for Nonlinear System Identification 99
8.3 Using Nonlinear System Models to Design Inputs for
Active Sensing 102
8.4 Extended Kalman Filter 106
9. CHALLENGES AND FUTURE RESEARCH NEEDS .110
9.1 Nonlinear Behavior Does Not Generalize .110
9.2 Nonlinear Approaches are Computationally Cumbersome,
Expensive, and Require too Many Parameters to be Defined .111
9.3 Many Existing Engineers, Technicians, and System Certifiers
are not Comfortable with and/or even Trained with Nonlinear
Processes or Analysis Techniques 112
9.4 Other Technical Challenges 112
10. RECOMMENDATIONS 113
10.1 Fundamental Benchmark Platforms 113
10.2 Increased Industry and Government Investment 114
10.3 Education 114
11. SUMMARY 115
12. REFERENCES 117
13. APPENDIX A. UC-IRVINE CONCRETE COLUMN TESTS .125
13.1 Test Structure Description .125
13.2 Static Loading .127
13.3 Dynamic Excitation .129
13.4 Data Acquisition .130
13.5 Data Files 132
14. APPENDIX B: FOUR-DEGREE-OF-FREEDOM TEST STRUCTURE 133
14.1 Test Structure Description .133
14.2 Data Acquisition .133
14.3 Data Files 134
viNonlinear System Identification for Damage Detection LA-14353-MS
15. OTHER ENGINEERING INSTITUTE STRUCTURAL HEALTH
MONITORING REPORTS AND THESES .136
16. WORKSHOP PARTICIPANTS .138
17. DISTRIBUTION .
كلمة سر فك الضغط : books-world.net
The Unzip Password : books-world.net