The thesis proposes a methodology for a Surrogate Based Optimisation (SBO) study and its application in the Naval Architecture field using the open-source software DAKOTA. Surrogate Based Optimisation is a useful and powerful way of reducing time and costs in hull shape optimisation. It allows the employ of exploratory data analysis techniques and machine learning methods to get more insight, discover hidden patterns, and detect anomalies in the design space. However, an SBO consists of several interconnected steps, and each increases complexity and uncertainty to the overall process.

The main goal is the investigation of the effects of different surrogate models on ship geometry optimisation from a resistance point of view. Also, different sampling plans, infill techniques, and optimisation algorithms are analysed and compared, as elemental steps for an SBO. This is done by the use of test functions that emulate the problem of ship optimisation. Furthermore, the intent is to try and list general guidelines to follow when building up an SBO routine with CFD simulations involved, with a focus in the marine field.

This research led to the assembly of Halton sampling sequence, Kriging meta-model, Expected Improvement function, Genetic Algorithms, and Pattern Search tools to implement a working SBO routine. This routine is used to demonstrate the success of the SBO method for a Hull Vane and an aft ship optimisations. Moreover, it is used to validate the research outcomes of this thesis and to prove that this work can be safely used for every-day commercial work.

The two design applications provided a resistance reduction, with respect to a benchmark hull, of about 19% and 7%. Thus not only they were successful, but also they were excellent examples to show how a surrogate model and its correct visualisation can give the Naval Architect the right tools to critically analyse the results, gain more understanding on the problem, spot and correct anomalies, and provide creative solutions to a client.

The competitiveness of the maritime sector together with energetic and environmental challenges have promoted a demand for more efficient marine vehicles. In response to this, research campaigns have been conducted, with the aim of developing new technologies. One of these technologies is the application of hydrofoils (e.g.: a Hull Vane) to marine vehicles. Although these allow a significant increase in efficiency, challenges and concerns rise due to their influence on seakeeping and manoeuvring performance. As a starting point, for a better understanding of the impact of hydrofoils on the dynamic behaviour of vessels, this research was focused on the: “Development of knowledge, methodologies and tools to perform manoeuvring predictions of vessels equipped with a Hull Vane.” For this, Virtual Captive Tests (VCT) were combined with a 4 DOF manoeuvring model, to perform manoeuvring predictions of vessels equipped with a Hull Vane. The convergence of the CFD solutions, for VCT, proved to be a challenging task with a lot of room for improvement. This is caused by the presence of complex flow features, e.g.: fore and aft-body vortices. In the end, an Adaptive Grid Refinement (AGR) algorithm, allowed an equilibrium between computational cost and discretization uncertainties.
Two types of VCT were performed and compared: Oblique Towing Tests (OTT) and Planar Motion Mechanism (PMM) tests. Due to the strong non-linear nature of hydrodynamic forces and moments, a Discrete Spectral Method was developed for the analysis of the PMM tests. On one hand, PMM tests represent an advantage in the determination of unsteady forces/moments compared to semi-empirical models. On the other hand, virtual PMM tests are more complex and more vulnerable to sources of errors, e.g.: flow memory effects and higher numerical residuals. At last, this research showed that the gain in accuracy in the determination of unsteady effects, hardly compensates the loss in accuracy in the determination of steady effects. Furthermore, it was shown that the uncertainties of captive tests can be amplified by the manoeuvring model, which can significantly penalize the precision of manoeuvring predictions. For the validation of manoeuvring predictions, a 25 m patrol vessel equipped with a Hull Vane (RPA8) was considered, for which experimental data is available. By comparing the results of the manoeuvring prediction with the experimental ones, it was shown that the manoeuvring prediction process over-estimates the course stability. Resulting in 20% (≈20 m) over-estimation of the 35° turning circle diameter and 50% (≈0.4°) under-estimation of the 3°-3° zig-zag overshoot. Regarding the impact of the Hull Vane on the manoeuvring performance, it was concluded that the Hull Vane increases the course stability of RPA8, leading to 7% increase of the 35° turning circle diameter and 18% decrease of 10°-10° zig-zag yaw overshoot. This is the result of a combination between: interaction phenomena, an increase in yaw damping caused by the Hull Vane struts and an increase in Munk moment caused by the change in dynamic trim. Therefore, the extrapolation of these results for other vessels is not trivial, due to the interaction between these effects. With the aim of reducing the cost of manoeuvring assessments, a Linear Pressure Distribution Method (LPDM) was developed, to model the effects of the Hull Vane on manoeuvring. This model was successfully validated, using the results of an integrated approach, i.e.: hydrodynamic coefficients determined using captive tests with the Hull Vane. The results show about 1% (≈1 m) difference in the prediction of the dimensions of a 35° turning circle, and less than 11% (≈0.3°) difference in the prediction of the 10°-10° zig-zag yaw overshoot. Finally, this research provides knowledge, tools and methodologies for a better understanding of the dynamic behaviour of foil-assisted vessels, particularly vessels equipped with a Hull Vane. The results of this research provide a solid foundation for further research in this field.