The significance of the maritime industry is undeniable in the sectors of global trading, renewable energy, and oil and gas industry. As a growing industry, there is continuous research on the development of better ships, efficient wind farms and reliable floating structures for which wavetanks are extensively used to perform scaled-model testing. However, such experimental campaigns are not only expensive but also time-consuming, therefore, limiting the number of tests required for data acquisition. To address this problem, researchers have proposed numerous numerical models to replace extensive experimental testing but these models are limited in their applications. Hence, we present a solution which is a novel approach for developing numerical wavetank models that can simulate a broad spectrum of water dynamics, i.e. shallow- and deep-water dynamics, and can be extended to analyse water-wave interactions with flexible offshore structures, thereby offering applicability to a wide array of maritime industrial challenges.
The establishment of such mathematical and numerical wavetank models is a challenge which can only be accomplished systematically.
The dynamics of linear and nonlinear water waves are governed by a variational principle (VP) that emphasizes the conservative structure of nonlinear water-wave dynamics. This is because energy and mass conservation, as well as the conservation of phase-space volume, are intimately connected with the conservative structure. A novel feature of our model is that it implements the time-discretized variational principle directly in the finite-element-based environment, Firedrake, which automates the derivation of time-discretized weak formulations and subsequently reduces the time and effort for the model implementation. At first, we developed a depth-averaged numerical wavetank model based on linear and nonlinear shallow water dynamics and established our novel approach through extensive testing. After that, we increased the model's complexity by developing a piston-driven numerical wavetank model based on linear and nonlinear potential flow theory. Our time-discretized VP-based model is capable of generating a numerical representation of actual wavetanks by including a piston wavemaker for wave generation and beaches for wave absorption. The results from the novel approach are promising and we are confident that this approach will simplify the development process of waveflap-driven numerical wavetank model, a widely utilized tool in the maritime industry. Furthermore, we have also developed and shared a hyperelastic structure model that can be coupled with the numerical wavetank to solve fluid-structure interaction (FSI) problems. However, the coupling is still in the development stages and is not presented in this thesis.
The numerical wavetank models must undergo experimental validation before they are deemed suitable for industrial use. Recognizing this necessity, we designed an experimental setup which is capable of measuring incoming waves, the structure's accelerations in response to the wave interactions, and waves reflected from the structure, simultaneously. After designing the setup, we conducted a series of experiments under a wide range of sea conditions ranging from regular-to-irregular and moderate-to-extreme wave height and steepness. The study of such a wide range of conditions makes the experiments suitable for providing reliable data in the validation of a suite of mathematical and numerical FSI solvers, i.e., linear, nonlinear and high-fidelity. The data from the experiments has been made publicly available through open-source data-sharing platforms. Lastly, we have used the experimental data to validate MARIN's in-house linear and high-fidelity FSI solvers, which confirms that the data is suitable for validation purposes.
