Generalized modal analysis techniques for the characterization and modeling of dissipationless planar waveguide structures and discontinuities encountered in microwave and optical integrated circuits, as well as of quantum waveguide structures and devices, are presented. The frequency-dependent transmission properties of the curved microstrip bend are derived by utilizing a second-order perturbation analysis of the equivalent modified curved waveguide model and a mode-matching method which includes the higher order modes. An extension of the mode-matching method for characterization of microstrip right-angle bends and T junctions having a rectangular notch is formulated. The modal solutions for an arbitrary graded-index dielectric slab waveguide are derived by applying the generalized telegraphist's equations to the equivalent inhomogeneous parallel-plate waveguide model with electric or magnetic walls. These modal solutions are employed in a mode-matching procedure to calculate the transmission properties of a step discontinuity in typical diffused optical dielectric slab waveguides. Power loss calculations for an abrupt offset in a diffused optical waveguide show a smooth increase in radiation loss whereas a sharp transition from almost zero to nearly total radiation loss is found for an abrupt change in diffusion depth. In the analysis of quantum waveguide structures, the modal expansions of the wave function together with a mode-matching technique are utilized. The computed generalized scattering matrices (GSMs) of junctions and uniform waveguide sections are combined via an extended GSM technique to obtain the scattering parameters of composite quantum waveguide structures. Results for cascaded right-angle bends and periodic multi-waveguide structures in a split-gate configuration are presented assuming hard wall confinement. For the multisection structures, strong resonant behavior similar to that in resonant tunneling diodes is found. Calculated current-voltage characteristics for a double constriction in GaAs are shown, exhibiting a region of negative differential resistance for temperatures up to approximately 60K with a maximum peak-to-valley ratio of over 80:1. Finally, a uniform narrow constriction with an assumed parabolic - like lateral potential confinement is analyzed by utilizing the modal expansion techniques developed for dielectric waveguides. / Graduation date: 1991

Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 396-400). Also available via the Internet.

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 122-123). Also available via the Internet.

Thesis (Ph. D.)--Mechanical Engineering, Georgia Institute of Technology, 2008. / Dewey H. Hodges, Committee Member ; Olivier Bauchau, Committee Member ; Aldo A. Ferri, Committee Member ; Itzhak Green, Committee Member ; Jerry H. Ginsberg, Committee Chair.

The sensing technique described in this dissertation produces modal coordinates for monitoring and active control of structural vibration. The sensor array is constructed from strain-sensing segments. The segment outputs are transformed into modal coordinates by a sensor gain matrix. An adaptive algorithm for computing the sensor gain matrix with minimal knowledge of the structure's modal properties is proposed. It is shown that the sensor gain matrix is the modal matrix of the segment output correlation matrix. This modal matrix is computed using new algorithms based on Jacobi rotations. The procedure is relatively simple and can be performed gradually to keep computation requirements low. The sensor system can also identify the mode shapes of the structure in real time using Lagrange polynomial interpolation formula. An experiment is done with an array of piezoelectric polyvinylidene fluoride (PVDF) film segments on a beam to obtain the segment outputs. The results from the experiment are used to verify a computer simulation routine. Then a series of simulations are done to test the adaptive modal sensing algorithms. Simulation results verify that the sensor gain matrix obtained by the adaptive algorithm transforms the segment outputs into modal coordinates. / Ph. D.

No description available.

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1992. / Vita. Abstract. Includes bibliographical references (leaves 133-135). Also available via the Internet.

No description available.

Vibration in internal turning is a problem in the manufacturing industry. Vibrations appear under the excitation applied by the material deformation process during the machining of a workpiece. In order for a lathe to perform an internal turning or boring operation, for example, in a pre-drilled hole in a workpiece, it is generally required that the boring bar should be long and slender; therefore extra sensitive to vibrations. These vibrations will affect the result of machining, in particular the surface finish, also the tool life may be reduced. As a result of tool vibration, severe acoustic noise frequently occurs in the working environment. This thesis comprises three parts and the first part presents a method for active control of boring bar vibration. This method consists of an active boring bar controlled by, for example, an analog controller. The focus lies on the analog controller and the advantages that may be obtained from working in the analog domain. The controller is a lead-lag compensator with digitally controlled parameters, such as gain and phase. However, signals remain in the analog domain. In addition, the analog controller is compared with a digital adaptive controller and it is found that both controllers yield an attenuation of the vibration by up to 50 dB. The second part of this thesis concerns the dynamic properties of a clamped boring bar used by the industry. In order to design a robust controller for a certain system, knowledge about the system's dynamic properties is required. On the workshop floor, a boring bar is dismounted and remounted, and reconfiguration of boring bars will alter the dynamic properties of the clamped boring bar. The dynamic properties of a standard boring bar and an active boring bar for a number of possible clamping conditions, as well as for a linearized clamping have been investigated based on an experimental approach. Also simple Euler-Bernoulli modeling of clamped boring bars incorporating simple non-rigid models of the boring bar clamping are investigated. Initial simulations of nonlinear SDOF systems have been carried out: one with a signed squared stiffness and one with a cubic stiffness. The purpose of these simulations was to identify a nonlinearity that introduces a similar behavior in the SDOF system dynamics as the nonlinear behavior observed in the dynamic properties of a clamped boring bar. The third and final part of this thesis focuses on vibration analysis methods in engineering education. A signal analyzer (which is a commonly used instrument in signal processing and vibration analysis) was made accessible via the Internet. Assignments were developed for students to learn and practice vibration analysis on real signals from a real setup of a relevant structure; a clamped boring bar. Whilst the experimental setup was fixed, the instrument and sensor configuration nonetheless enable a variety of experiment, for example: excitation signal analysis, spectrum analysis and experimental modal analysis.

In this thesis, the dynamics of two plates overlapping and connected by three bolts are studied. The data collected in the test are used in modal analysis. The vibrational test and the modal analysis were made using an LMS system. Hammer excitation is used for the tests. The main purpose of this thesis is to study how the suspensions affect the extracted eigenfrequencies and modal dampings. In this thesis, more than 10 suspensions were examined. Another objective in this thesis work is to build an FE-model. This model is made using the software Abaqus. To improve the reliability of the FE-model, a set of reliable experimental data is used to calibrate the model. The calibrated FE-model, using the measurement data, has a dynamic behavior close to the measurement data.