Design of Large Scale Hydrodynamic Models of Offshore Structures

This paper discusses the design of hydrodynamically scaled models of offshore structures subjected to wave loading. A considerable amount of engineering design and analysis is required to ensure the high performance of a model/dynamometer system, especially when dealing with large scale models (for example, 1:40). The system is designed to be as rigid as possible so that the dynamometer measures wave loading without dynamic amplification caused by structural vibrations. (Ideally, the system would consist of a low mass, rigid model and a dynamometer having a stiffness which is as high as possible while having the required load sensitivity and capacity). The primary constraints are geometry (related to the hydrodynamic scaling of the model), load measurement accuracy over a specified range, and specified wave parameters. In order to satisfy these constraints, the dynamometer components must be appropriately sized and the model/dynamometer system must be designed to have a sufficiently high natural frequency. Primarily, this paper deals with the problem of designing a high natural frequency model/dynamometer system. In order to avoid dynamic amplification effects, it is necessary to ensure that the lowest natural frequency of the system is several times larger than the highest wave frequency. The stiffness of the system is dictated by the structural configuration and material of the model and by the flexibility of the dynamometer. The added mass associated with the water entrained by the model can be significant and must be taken into account, in addition to the structural mass, when estimating natural frequencies. Examples dealing with previously tested hydrodynamically scaled models of offshore structures such as gravity base structures and jacket platforms are presented. Other design considerations are briefly discussed. INTRODUCTION Model testing of offshore structures subjected to wave loading is an important phase of the prototype design process and is used to verify numerical or analytical predictions of global and local hydrodynamic loads on the prototype and its components. Examples of global loads are shear and overturning moment at the base of the structure. Local loads include wave impact, run-up, etc. Recent projects [1,2,3,4] at the National Research Council of Canada Institute for Marine Dynamics (NRC/IMD) have involved the testing of various large scale models models of structures in a wave basin. Examples of such projects include the testing of models of various structures in the Ekofisk oil field in the Norwegian sector of the North Sea (client: Phillips Petroleum, Norway) and the Hibernia gravity base platform (client: Doris Development Canada). In these tests, the scales of the models were relatively large, of the order of 1:40. For example, the Hibernia gravity base structure (GBS) is approximately 100 metres in diameter in a water depth of approximately 80 metres. This corresponds to a model of approximately 2.5 metres in diameter in a water depth of approximately 2 metres. These types of models may have a mass of several thousand kilograms, not including the dynamometer. As indicated in a paper by Hansen et al [6], the problem of testing large scale models of offshore structures subjected to wave loading often requires novel engineering solutions. A National Research of Council of Canada report by Mogridge et al [5] also describes the design and testing of a large scale model of an offshore structure. The objective of these types of tests is to accurately measure the global and local wave loading on the model. It should be emphasized that the model is hydrodynamically scaled (as opposed to hydroelastically scaled) because the objective is to determine the wave loading based on the assumption that the structure is rigid. Because the model is mounted on a dynamometer, it is necessary that the model/dynamometer system should be as stiff as possible. However, as discussed in this paper, the model/dynamometer system is not rigid, because of the inherent flexibility of the dynamometer and the flexibility of the model itself. Therefore, part of the model design involves assessing the stiffness of the model/dynamometer system and, in particular, determining its lowest natural frequency so that the measured loads are not amplified by vibrations caused by the dynamic loading on the model. This paper discusses, in general terms, the design of a typical large scale model of an offshore structure, with particular emphasis on the design of the dynamometer and its components and on the calculations and tests required to ensure that the system is sufficiently stiff. MODEL/DYNAMOMETER SYSTEM DESIGN Based on the specifications provided, the design of the model/dynamometer system involves a number of criteria to be satisfied. The following sections of this paper discuss some of the important design criteria relating to the model, the dynamometer and the model/dynamometer system. Test Specifications Specifications provided by the client include the following quantities. • Water depth • Ranges of wave heights and periods (for regular waves), wave spectra (for irregular waves) • Range of wave headings • Dimensions of prototype • Hydrodynamic scale (to be applied to structure, water depth and wave parameters) • Types of loads to be measured (for example, global wave loading, local loads caused by wave impact) • Quantities to be measured (for example, resultant wave force in various directions and overturning moments about various axes) • Accuracy or resolution of load measurements General Layout of Model/Dynamometer System An example of a GBS model/dynamometer system is shown in Figure 1. (Note that the upper part of the model/dynamometer system (shafts, deck and associated dynamometers) are not shown in this figure). The dynamometer is fixed to the floor of the wave basin and the model is attached to the dynamometer at various locations where loads are measured by individual force transducers. In order that water does not enter the interior of the model, a flexible seal is attached around the periphery of the model base and the wave basin floor. In this case, there is a small gap between the base of the model and the basin floor. The seal [1,10] must be sufficiently strong to withstand hydrostatic pressure and hydrodynamic forces, yet not too stiff so that it influences the load measurements. Model The external dimensions of the model are determined from the prototype dimensions and specified hydrodynamic scale. In the case of a cylindrical caisson, the primary dimensions are height and external diameter. For a jacket platform, the individual lengths, orientations and outside diameters of the tubular members are required, for hydrodynamic scaling purposes. For the caisson type structure shown in Figure 1, in order to produce a rigid model, the cylindrical shell is internally braced with ring stiffeners and vertical stiffeners, as well as with stiff brackets for connection to the dynamometer. Depending on the particular design, the TOOTH JACKET -