tesi R. Valiante.pdf - EleA@UniSA

More documents

Recommendations

Info

$3Eterov e)ce(te/rou sofo\v to/ te pa/lai to/ te nu=n. Ou ... - EleA@UniSA$

22 the damaged region and that damage could be quantified by establishing a relationship between changes in curvature and damage extent. The evaluation of changes in the curvature of dynamic deflection shapes as a tool for damage detection and localization were also investigated in [10] and subsequently in [11]. In these studies, the dynamic behavior of beams with notch defects and delaminations was studied analytically and experimentally. Analytical models described the dynamic behavior and the curvature modes of cracked beams through perturbation of the modal parameters of the undamaged beams, so that approximated analytical expressions for the damaged modes were obtained. The analytical studies were supported by experimental investigations performed on simple beam structures. Their results show the potentials of the technique when applied to the first mode of the beam. The limitation to a single mode was mainly dictated by the limited spatial resolution available in the accelerometers-based experiments. Ho and Ewins [24] formulated a damage index defined as the quotient squared of the corresponding modal curvatures of the undamaged and damaged structure. The damage index was found to be highly susceptible to noise, as measurement errors were amplified due to second derivative computations based on numerical techniques. They also demonstrated that spatially sparse measurements adversely affect the performance of the damage index. As an alternative Ho and Ewins [25] investigated the changes in the square of the slope of mode shapes of beams. Oscillations in the slope computations were reduced through polynomial fit of the measured mode shape as compared to the use of finite difference approximation. They also reported that higher derivatives can be more sensitive to damage, but are more subject to numerical errors. The same technique has then been extended to plate structures, where accelerometers are used to measure the detections to be interpolated for successive differentiation. The results presented in [30, 27], show the effectiveness of the technique, and are used as a basis for the developments presented in the present study.
INTRODUCTION 23 1.6. GUIDED WAVES-BASED DAMAGE MEASURES Guided waves are one of the most encouraging tools for SHM applications. In plates these waves can propagate for large distances, they are sensible even to small defects and they involve the whole material thickness, meaning that a defect can be identified also if it is in the middle of the plate. One distinction that can be done lies in the selection of the mode used for the investigation. The chosen mode should be able to offer a very low dispersion, low attenuation with the distance, high sensitivity to damages, easy excitability and good detectability. Some authors [30] consider the S 0 mode as the proper one, for its low dispersion, low attenuation in amplitude and its velocity of propagation. Other authors [31, 32] prefer the A 0 mode, because, even if it is more dispersive for low frequencies, it has lower wavelength than the S 0 mode, thus it can detect smaller defects. Guided waves result from the constructive interference of bulk longitudinal and shear waves propagating within a confined geometry such as a railhead, wing skin, or pipeline. An illustration of the guided wave phenomenon, thought of as the superposition of bulk longitudinal and shear waves, is shown in figure 1.2. To satisfy the boundary conditions at each interface, mode conversion into to both longitudinal and shear waves occurs due to an incoming bulk wave. With enough propagation distance, the large number of mode converted bulk waves result in bulk wave resonances, otherwise known as guided waves. Fig. 1.2 - Guided waves propagating within a confined geometry.
Page 1: Unione Europea Università degli St
Page 5 and 6: RINGRAZIAMENTI E’ per me doveroso
Page 7: Vorrei, infine, disporre del vocabo
Page 11 and 12: CONTENTS Abstract of dissertation .
Page 13 and 14: LIST OF FIGURES AND TABLES Fig. 1.1
Page 15 and 16: Fig. 3.32 - : RMS of the field of r
Page 17 and 18: ABSTRACT OF DISSERTATION 7 ABSTRACT
Page 19 and 20: ABSTRACT OF DISSERTATION 9 are: The
Page 21 and 22: INTRODUCTION 11 Ideally, routinely
Page 23 and 24: INTRODUCTION 13 Feature extraction
Page 25 and 26: INTRODUCTION 15 1.3. GLOBAL VERSUS
Page 27 and 28: INTRODUCTION 17 damage. These inves
Page 29 and 30: INTRODUCTION 19 between shifts at v
Page 31: INTRODUCTION 21 Space Shuttle Orbit
Page 35 and 36: INTRODUCTION 25 Fig. 1.3 - Differen
Page 37 and 38: INTRODUCTION 27 Considering the arg
Page 39 and 40: INTRODUCTION 29 Fig. 1.6 - Differen
Page 41 and 42: INTRODUCTION 31 2 ω c 2 k = 2 ω
Page 43 and 44: INTRODUCTION 33 Fig. 1.8 - Frequenc
Page 45 and 46: INTRODUCTION 35 where 1 1 p1 c k =
Page 47 and 48: INTRODUCTION 37 Fig. 1.12 - Group v
Page 49 and 50: INTRODUCTION 39 Fig. 1.13 - Directi
Page 51 and 52: INTRODUCTION 41 for very simple cas
Page 53 and 54: INNOVATIVE SHM TECHNIQUES BASED ON
Page 63 and 64: EXPERIMENTAL ANALYSIS 53 3. EXPERIM
Page 65 and 66: EXPERIMENTAL ANALYSIS 55 Fig. 3.3 -
Page 67 and 68: EXPERIMENTAL ANALYSIS 57 same point
Page 73 and 74: EXPERIMENTAL ANALYSIS 63 of the ene
Page 75 and 76: EXPERIMENTAL ANALYSIS 65 Fig. 3.13
Page 77 and 78: EXPERIMENTAL ANALYSIS 67 Fig. 3.16
Page 79 and 80: EXPERIMENTAL ANALYSIS 69 Analysis g
Page 81 and 82: EXPERIMENTAL ANALYSIS 71 Analysis g
Page 83 and 84:
EXPERIMENTAL ANALYSIS 73 and in spa
Page 85 and 86:
EXPERIMENTAL ANALYSIS 75 and are tr
Page 87 and 88:
EXPERIMENTAL ANALYSIS 77 obtained w
Page 89 and 90:
EXPERIMENTAL ANALYSIS 79 Fig. 3.26
Page 91 and 92:
EXPERIMENTAL ANALYSIS 81 (a) t = 68
Page 93 and 94:
Page 95 and 96:
Page 97 and 98:
Page 99 and 100:
Page 101 and 102:
Page 103 and 104:
Page 105 and 106:
Page 107 and 108:
EXPERIMENTAL ANALYSIS 97 The data p
Page 109 and 110:
Page 111 and 112:
EXPERIMENTAL ANALYSIS 101 To increa
Page 113 and 114:
EXPERIMENTAL ANALYSIS 103 Both of t
Page 115 and 116:
Page 117 and 118:
EXPERIMENTAL ANALYSIS 107 Complete
Page 119 and 120:
FINITE ELEMENT MODELING OF WAVE PRO
Page 121 and 122:
Page 123 and 124:
Page 125 and 126:
Page 127 and 128:
Page 129 and 130:
Page 131 and 132:
Page 133 and 134:
Page 135 and 136:
Page 137 and 138:
Page 139 and 140:
Page 141 and 142:
CONCLUSIONS AND FUTURE WORK 131 5.
Page 143 and 144:
CONCLUSIONS AND FUTURE WORK 133 cas
Page 145 and 146:
REFERENCES 135 [6] Rose, J. L. “A
Page 147 and 148:
REFERENCES 137 [20] Ko, J. M., Wong
Page 149:
REFERENCES 139 [35] J. D. Achenbach
show all

tesi R. Valiante.pdf - EleA@UniSA

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?