Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1994. / Vita. Abstract. Includes bibliographical references (leaves 111-118). Also available via the Internet.

Thesis (M.S.)--Brigham Young University. Dept. of Civil and Environmental Engineering, 2005. / Includes bibliographical references (p. 87-89).

Thesis (MScIng)--University of Stellenbosch, 2008. / Bibliography. Also available via the Internet.

No description available.

The objective of this thesis is to study the vibration suppression capabilities of a double-octahedral adaptive truss with an appended flexible beam. Three cases are studied. The first case is an experimental and analytical investigation of flexible beam control using an adaptive truss positioned in an equilibrium configuration such that the beam is parallel to the gravity field. A linear model of the system is derived, and a linear LQR control law is designed. Control experiments are conducted and show that the adaptive truss performs good vibration control of the flexible beam with the effective modal damping increased from 0.012 for the uncontrolled vibrations to 0.19 for the controlled vibrations. Good agreement between the open-loop and closed-loop simulations and experiments is achieved such that a correlation coefficient of 0.92 is observed between the experiment and simulation open-loop data and a correlation coefficient of 0.85 is observed for the experiment and simulation closed-loop data. The second case is the analysis and vibration control of the flexible beam while the adaptive truss is positioned at some large-angle configuration such that the flexible beam is severely affected by gravity. In this case a nonlinear beam model is developed and a linear LQR control law is implemented. Simulations show vibration control of the flexible beam for large-angle configurations of the truss, and that the vibration control performance is highly dependent on the linearization point of the control law. The third case is the analysis and control of large-angle slewing of a flexible beam using the adaptive truss. In this case the nonlinear beam model is implemented with both linear LQR and nonlinear gain-scheduled control laws. Simulations indicate that the adaptive truss achieves vibration control of the beam during slewing manuvers for all control laws tested with the nonlinear control law achieving superior vibration control performance in comparison to all control laws tested. / Master of Science

Parallel manipulators are being thought of as a solution to many problems involving control and manipulation. They offer significant advantages over serial manipulators in terms of increased strength and rigidity. Variable geometry trusses (VGTs) are a special class of parallel trusses. Literature on VGTs has covered many interesting problems, yet there has been no conscious effort to formulate a definition for a VGT. The major emphasis of this thesis is developing a set of precise kinematic rules for defining and analyzing VGTs. Traditional mobility criteria, when applied to parallel geometries, predict results which are often confusing and sometimes inaccurate. Based on the set of rules developed, mobility equations are derived for planar and spatial VGTs. The equations are tested on a sufficiently large number of VGT configurations to convince the author about the validity of the set of rules and the equations derived. Using the mobility equations, different candidate geometries can be analyzed and compared. In addition, the equations can be used in the type and number synthesis of VGTs. / Master of Science

The analysis of trussed bents can be very satisfactorily executed by using the Cross method of moment distribution in conjunction with a method of distributing thrust. The use of such a method as this may be realized when we think of the contrasting time element involved in the solution of the problem by this method and the older method of least work. In the case of the first method, the problem may be solved, once the distribution constants are known, in a matter of minutes. The least work process may take hours, and not infrequently, days. It was the purpose of this thesis to determine what effect the fixed end thrust in trusses has on the solution of the problem by distributing moments. The results of this investigation empathetically warn us that when we are dealing with fixed supports, we must consider these thrusts, whereas, when we are concerned with a trussed bent that has only hinged supports, the effect of the thrust may be neglected. It is the opinion of the writer that, whereas, the effect of the thrusts in parallel chord trusses may not be of such great influence on the final results, the truss sections which approach the shape of an arch (Fink roof truss with knee bracing, for instance) will be seriously affected, whether the supports or hinged or fixed. This method of analysis would be very greatly benefitted if some future investigator would strive to develop a tool which would make for more rapid convergences in the distribution process, and would develop tables which could be used for finding carry-over and stiffness factors and fixed-end moments and thrust for different truss sections. / Master of Science

From the foregoing investigation the following conclusions seem justifiable: That moment distribution can be successfully applied to trussed bents, with comparatively little effort once the constants are available; and this with an accompanying saving of time and effort, as compared to the analysis of trussed bents by least work or any other classical method. That the results obtained by moment distribution check closely with those obtained by the method of least work, and for all practical purposes are therefore acceptable. What the compilation of constants in chart form, covering all types of trusses, is a perfectly feasible undertaking and promises to be the basis on which a practical, simple, and straight-forward method of analysis will eventually be established. That the adoption of moment distribution for the analysis of trussed bents together with the charting of the constants will aid in bringing about a standardization of truss construction heretofore unrealized. / Master of Science

The stability bracing behavior of trusses was investigated using experimental testing and computational modeling. The laboratory experiments were conducted on twin trusses fabricated with W4x13 sections for the chord and web members. Spans of 48 and 72 feet were used in the tests that included both lateral load tests and buckling tests. Most of the tests were done on the regular (Howe) truss, except the lateral stiffness tests which were also done on the inverted (Pratt) truss. Computational models were developed using the three-dimensional finite element program, ANSYS, which were validated using the laboratory test data. A variety of models were used to simulate both as-built and idealized truss models. The experiments demonstrated that the buckling capacity of the truss with torsional bracing largely depended on the brace stiffness and the number of intermediate braces. Similar behavior was observed in the truss with lateral bracing. The tests results demonstrated that cross sectional distortion dramatically reduces the effectiveness of the torsional braces. The experiments provided valuable data for validating the finite element models that were used to conduct parametric studies on torsional bracing of truss systems. The results from the parametric studies were used to develop stiffness requirements for torsional bracing of trusses. / text

No description available.