An optimization methodology, algorithms, and computer program for cable/lumped-body deployment system are developed for the design and installation of sonar packages in deep ocean ranges. In addition, a methodology and algorithms for the control of the system are proposed. A penalty function is developed in terms of the system parameters and the dynamic constraints are set up for the optimization of cable/lumped-body deployment system. Deployment procedures and segmentations characterized by their initial and terminal condition are proposed. Alternating Direction (ADM) and Modified Alternating Direction (MADM) methods are employed for the optimum search. The optimization algorithms are then implemented on a desk-top computer. The objective of the search is to achieve the optimum vessel speed and cable pay-out rate that result in the minimum penalty value. Numerical examples are presented to test the robustness of the algorithm (ADM) and the compute program capabilities. The results are compared to those of the MADM method and the entire space search. During actual installation, it is anticipated that randomness in the excitation will cause sonar packages to miss the exact target locations. Thus a continuous correction in the controllable parameters is necessary. Three alternative types of control algorithms are presented for these reasons. / Graduation date 1993

Mooring systems for floating facilities that are used offshore to produce oil and gas, consisting of individual mooring lines and foundations, are currently designed on the basis of individual components and on a case-by-case basis. The most heavily loaded line and anchor are checked under extreme loading conditions (hurricane and loop current) with the system of lines intact and with one line removed. However, the performance of the entire mooring system depends more directly on the performance of the system of lines and foundations rather than on the performance of a single component. In this study, a floating production system design originally developed by the industry consortium, DeepStar, was chosen for study. The mooring system was designed for three different nominal water depths: 1000, 2000 and 3000 m. It is a classic spar with steel mooring lines in 1000 m of water and polyester mooring lines in deeper depths. Based on simulated results of loads on mooring lines and foundations using a numerical model, reliability analyses were conducted using representative probabilistic descriptions of the extreme met-ocean conditions, hurricanes and loop currents, in the Gulf of Mexico. The probability of failure of individual mooring line components during a 20-year design life is calculated first, followed by that of a complete mooring line which consists of top and bottom chains, a steel cable or polyester rope at the middle and a suction caisson foundation, and finally that of the mooring system. It is found that foundations have failure probabilities that are more than an order of magnitude smaller than those for lines under extreme loading. Mooring systems exhibit redundancy in that the failure of the most heavily loaded component during an extreme event does not necessarily lead to failure of the system. The system reliability and redundancy are greater for the taut versus semi-taut systems and is greater for designs governed by loop current versus hurricane events. Although this study concerns about the mooring systems of a classical spar, the methodology of the reliability analysis and the conclusions made in this study may have important implications to the other deepwater mooring systems / text

No description available.

Cables in hoisting installations, due to their flexibility, are susceptible to vibrations. A common arrangement in industrial hoisting systems comprises a driving winder drum, a steel wire cable, a sheave mounted in headgear, a vertical shaft and a conveyance. This system can be treated as an assemblage of two connected interactive, continuous substructures, namely of the catenary and of the vertical rope, with the sheave acting as a coupling member, and with the winder drum regarded as an ideal energy source. The length of the vertical rope is varying during the wind so that the mean catenary tension is continuously varying. Therefore, the natural frequencies of both subsystems are time-dependent and the entire structure represents a non-stationary dynamic system. The main dynamic response, namely lateral vibrations of the catenary and longitudinal vibrations of the vertical rope, are caused by various sources of excitation present in the system. The most significant sources are loads due to the winding cycle acceleration/deceleration profile and a mechanism applied on the winder drum surface in order to achieve a uniform coiling pattern. The classical moving frame approach is used to derive a mathematical model describing the non-stationary response of the system. First the longitudinal response and passage through primary resonance is examined. The response is analyzed using a combined perturbation and numerical technique. The method of multiple scales is used to formulate a uniformly valid perturbation expansion for the response near the resonance, and a system of first order ordinary differential equations for the slowly varying amplitude and phase of the response results. This system is integrated numerically on a slow time scale. A model example is discussed, and the behaviour of the essential dynamic properties of the system during the transition through resonance is examined. Interactions between various types of vibration within the system exist. The sheave inertial coupling between the catenary and the vertical rope subsystems facilitates extensive interactions between the catenary and the vertical rope motions. The nature of these interactions is strongly non-linear. The lateral vibration of the catenary induces the longitudinal oscillations in the vertical system and vice-versa. In order to analyze dynamic phenomena arising due these interactions the nonlinear partial-differential equations of motion are discretised by writing the deflections in terms of the linear, free-vibration modes of the system, which result in a non-linear set of coupled, second order ordinary differential equations with slowly varying coefficients. Using this formulation, the dynamic response of an existing hoisting installation, where problematic dynamic behaviour was observed, is simulated numerically. The simulation predicts strong modal interactions during passage through external, parametric and internal resonances, confirming the autoparametric and non-stationary nature of the system recorded during its operation. The results of this research demonstrate the non-stationary and non-linear behaviour of hoisting cables with slowly varying length. It is shown that during passage through resonance a large response may lead to high oscillations in the cables' tensions, which in turn contribute directly to fatigue damage effects. The results obtained show also that the non-linear coupling in the system promotes significant modal interactions during the passage through the instability regions. The analysis techniques presented in the study form a useful tool that can be employed in determining the design parameters of hoisting systems, as well as in developing a careful winding strategy, to ensure that the regions of excessive dynamic response are avoided during the normal operating regimes. / Thesis (Ph.D.)-University of Natal, Durban, 1999.

No description available.

No description available.

No description available.

A large number of building inventory in Canada and elsewhere in the world consists of non-ductile reinforced concrete frames, with or without masonry infill panels. These structures suffer damage when seismic force demands are higher than their force capacities. Therefore, seismic retrofitting of such frame buildings for drift control remains to be a viable option for improved building performance. A retrofit methodology has been developed in the current research project, which involves diagonal bracing of frames with prestressing strands. An experimental research project has been conducted to assess the effectiveness of diagonal prestressing in non-ductile reinforced concrete frame buildings. The experimental program consists of two large-scale single-bay single-storey reinforced concrete frames, with a height of 3.0m and a span length of 3.5 m. The frames were designed and built to reflect the 1960’s practice in Canada, without the seismic requirements of current building codes, and hence are seismically deficient. They were retrofitted with diagonally placed prestressing strands, having two different areas of steel, prestressed to 40% of the strand capacity. One of the frames was retested after the failure of the strands, with a new set of strands without any prestressing, forming the third test. The results indicate that lateral bracing reinforced concrete frames with high-strength prestressing strands is an effective strategy for controlling lateral drift and hence potential damage in buildings during strong earthquakes. Prestressing of the strands increases initial stiffness, as compared to non-prestressed cables, and provide superior performance. The area of diagonally placed steel (including the number of strands) and the level of initial prestressing depend on the required level of upgrade in the building in terms of seismic force requirements. The design procedure recommended in this thesis may be employed for implementing the technology. The thesis presents the details of the experimental program, and the test results. It also provides analytical verification of the approach, with a step-by-step design procedure.

Thesis (M.S.)--Ohio University, August, 1989. / Title from PDF t.p.

Thesis (Ph. D.)--Université de Liège. / Includes bibliographical references (p. 243-263).