Abstract
This thesis is about experimental investigations into the response of ships to waves. Experiments may be conducted using models or full-scale ships in either real seas, or in model simulations of the sea in a laboratory. The model simulation methods which are available extend from long-crested regular waves (which are assumed to contain a single frequency and direction) to short-crested random waves (which are characterised by a directional wave spectrum).Existing experimental practices are reviewed and methods of predicting and measuring natural directional wave spectra are described in some detail.
The use of a long-crested regular wave model implies that the ship response must be a linear function of the waves, and requires very many experiments to determine response functions over an adequate range of frequencies, directions, and speeds.
Greater economy is achieved using long-crested random waves in the laboratory. The simulation of long-crested random waves is reviewed. Methods and equipment for quantifying and maximising the accuracy of the simulation are developed. The instrumentation of models is examined in detail, again with the objective of quantifying and minimising errors. Analysis of the measurements is described in detail, particularly the use and advantages of frequency domain processing techniques. These methods are illustrated by describing the design and conduct of a series of experiments on a model of a surfaced submarine in waves.
The results of experiments can be used to aid the making of decisions about the design of new ships, or associated facilities, and also the making of operational decisions affecting the use of existing ships. To illustrate this, the applications of the results of the experiments on the surfaced submarine are reviewed in detail.
Laboratory experiments are not practical if the ship is small and fast, because of limitations on the economic size of laboratory test tanks. The need is demonstrated to conduct experiments in short-crested random waves, particularly if responses are suspected of being non-linear. The design of, and analysis of results from, such experiments is shown to be a logical extension of the methods developed for long-crested seas.
The conduct of experiments in natural seas is described to illustrate the selection of a suitable site, the experiment design, the methods of estimating the directional wave spectrum and measuring the model response during the tests, and the analysis and interpretation of the results. Extension of the methods to the testing of ships at full scale is discussed, and the possible application. of full scale test results are identified.
| Date of Award | 1991 |
|---|---|
| Original language | English |
| Awarding Institution |
|