The adaptive modification of the mechanical properties of structures has been described as a key to a number of new or
enhanced technologies, ranging from prosthetics to aerospace applications.
Previous work reported the electrostatic tuning of the bending stiffness of simple sandwich structures by modifying the
shear stress transfer parameters at the interface between faces and the compliant core of the sandwich. For this purpose,
the choice of a sandwich structure presented considerable experimental advantages, such as the ability to obtain a large
increase in stiffness by activating just two interfaces between the faces and the core of the beam.
The hypothesis the development of structures with tunable bending stiffness is based on, is that by applying a normal
stress at the interface between two layers of a multi-layer structure it is possible to transfer shear stresses from one layer
to the other by means of adhesion or friction forces. The normal stresses needed to generate adhesion or friction can be
generated by an electrostatic field across a dielectric layer interposed between the layers of a structure. The shear stress
in the cross section of the structure (e.g. a beam) subjected to bending forces is transferred in full, if sufficiently large
normal stresses and an adequate friction coefficient at the interface are given. Considering beams with a homogeneous
cross-section, in which all layers are made of the same material and have the same width, eliminates the need to consider
parameters such as the shear modulus of the material and the shear stiffness of the core, thus making the modelling work
easier and the results more readily understood.
The goal of the present work is to describe a numerical model of a homogeneous multi-layer beam. The model is
validated against analytical solutions for the extreme cases of interaction at the interface (no friction and a high level of
friction allowing for full shear stress transfer). The obtained model is used to better understand the processes taking place
at the interfaces between layers, demonstrate the existence of discrete stiffness states and to find guidance for the selection
of suitable dielectric layers for the generation of the electrostatic normal stresses needed for the shear stress transfer at the
The main cables of suspension bridges are often wrapped with a steel wire, in order to compact the cable and hold it in
shape. If a non-destructive evaluation by means of magnetic methods is performed on such a cable, disturbances due to
the wrapping can be expected in the measured signal. In the presented work, these disturbances shall be quantified and
compared to the flaw signals. Different approaches for the separation of the disturbance and the flaw signal are discussed.
Additionally, the possibility to detect wire breaks and corrosion within an unwrapped steel cross-section could be shown
in laboratory measurements. The influence of the wrapping was investigated using finite element (FE) simulations and
experimental laboratory measurements. A parameter study was performed in order to obtain data in which the components
from a flaw and the wrapping can be separated. The parameters varied in this study were chosen depending on the prospect
of success and the cost of the realization. Using these data sets different filtering methods, such as wavelet analysis, were
implemented. A final comparison of the different methods suggests the most efficient way to assess the condition of such
cable systems using magneto-inductive testing. Finally, it can be concluded that the use of FE simulation is a very useful
tool for the development of new data analysis methods, even if a real set-up and data from measurements exist.
The suppression of vibrations of a structure is commonly considered a necessary measure for the extension of its
lifetime, when high amplitude vibrations are observed. As an alternative to the introduction of discrete damping
devices, the modification of the stiffness of a beam is proposed as a means to suppress vibrations due to resonance,
thank to the ability to reject mechanical energy input at specific frequencies. Previous work has outlined the
principle and the potential advantages of such an approach based on the behavior of a small scale system. In
order to confirm the feasibility of the approach on macro-scale systems, such as a light weight pedestrian bridge,
experiments for the tuning of a 2.5 m long glass fiber reinforced polymer I-beam were performed. The results
of the experiments show that it is possible to modify the bending stiffness of structural elements that can be
used for real life engineering applications. Measurements show that it is possible to shift the resonance peak of
a beam while maintaining a reasonably good q-factor in the transfer function, thus indicating that the change
in behavior happens in connection with an increased stiffness rather than with the introduction of substantial
damping. Based on the presented feasibility study, the development of an adaptive bridge deck will be considered.
Vibration control and suppression in structures plays a central role in the extension of their service life and improvement of their reliability. While in many cases the solution of this problem implies the introduction of external damping devices, it is also conceivable to adaptively modify their vibratory properties, so that the occurrence of severe vibrations due to resonance phenomena can be curbed at its origin. The modification of the shear stress transfer at the interface between the core and the faces of a sandwich beam has been shown to have a remarkable effect on the bending stiffness of the structure. Such modification can be obtained by applying a normal stress between the core and the un-bonded, electrically insulated faces of the sandwich by means of a strong electrical field.
An intermediate behavior between fully bonded and un-bonded layers in terms of inter-laminar shear stress can be achieved by temporary electrostatic bonding of the components. The outlined approach to the reduction of transversal vibrations in thin multi-layer beams is promising and can in principle be applied to multi-layer plates.
Experimental work performed on several full-scale stay-cable models as well as on RAMA IX Bridge in Bangkok has confirmed that the application of magnetic flux leakage (MFL) methods is a viable approach to the non-destructive evaluation of large diameter steel cables. Such method allows for a high sensitivity and high-resolution detection of fractured wires in stay cable systems.
So far, the information obtained from the recorded data (intensity of the MFL on the surface of a cable) was limited to the accurate position of detected flaws along the axis of the cable and a qualitative indication of the position of the flaws within the cross-section. The ability to accurately determine the position of flawed wires within the cross-section of a cable is especially useful in the case of multi-strand systems, in which individual strands can be replaced if damaged.
Such information can be obtained by computation with finite element models or sophisticated dipole approximations. An alternative to such computing intensive approach, based on a simple mathematical model of the MFL function is proposed in this work. The function is used for a non-linear fit of the measured data. The method has been tested successfully on simulated and measured data.