University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

This thesis proposes a novel graphical model for inference called the Affinity Network,which displays the closeness between pairs of variables and is an alternative to Bayesian Networks and Dependency Networks. The Affinity Network shares some similarities with Bayesian Networks and Dependency Networks but avoids their heuristic and stochastic graph construction algorithms by using a message passing scheme. A comparison with the above two instances of graphical models is given for sparse discrete and continuous medical data and data taken from the UCI machine learning repository. The experimental study reveals that the Affinity Network graphs tend to be more accurate on the basis of an exhaustive search with the small datasets. Moreover, the graph construction algorithm is faster than the other two methods with huge datasets. The Affinity Network is also applied to data produced by a synchronised system. A detailed analysis and numerical investigation into this dynamical system is provided and it is shown that the Affinity Network can be used to characterise its emergent behaviour even in the presence of noise.

Thesis (M.Sc.Eng.) / Dielectric elastomers (DEs) have attracted significant attention in recent years and have been found to provide excellent overall performance in actuation-based application. This thesis will introduce the fundamentals of DEs, derive the field and finite element equations for simulating its deformation, and then focus on numerically studying electromechanical instabilities, in particular electrostatically driven creep and the effect of pre-stretch on surface (creasing) instabilities. First, a nonlinear, dynamic finite element model coupled with a finite deformation viscoelastic constitutive law is utilized to study the inhomogeneous deformation and instabilities resulting from the application of a constant voltage to dielectric elastomers. Theoretical studies are performed of two problems that have been experimentally observed, i.e. electromechanical snap-through instability and bursting drops in a dielectric elastomer. In general, increasing the viscoelastic relaxation time leads to an increase in time needed to nucleate the electromechanical instability. However, it is found that the time needed to nucleate the instability of these two cases scales with the relaxation time. Second, the effect of pre-stretch on the performance of dielectric elastomers is studied. Two cases are studied, the electromechanical snap-through instability under equibiaxial pre-stress, and a strip under uniaxial pre-stretch. It is found that prestress markedly increases the stability of the elastomers, while pre-stretch increases the critical field for electro-creasing instability. / 2031-01-01

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology / Department of Electromechanical Engineering