Offshore Wind Farm Layout Optimization

Abstract

Nowadays an increasing number of offshore wind farms (OWF) is getting built. Several disciplines are considered into this subject. The manual and sequential design approach used so far is not sufficient to guarantee optimized systems because interactions between the different components are disregarded. Therefore, engineering companies are looking for a multidisciplinary, modular, user-friendly and easy tool to be applied during the preliminary design assessment. This should take into consideration almost every crucial aspect of OWF design. Moreover, it should be able to perform some overall layout optimization, i.e. providing the best turbine positioning w.r.t. the energy yield, the costs, the cables, the installation and the O&M, ensuring some good constraint-handling techniques and guaranteeing a sufficient degree of robustness, precision, flexibility and speed. Therefore, the concept of MDAO (Multisciplinary Design Analysis and Optimization) workflow has been introduced. With respect to the common sequential design procedure, which optimizes the wind farm components separately, the MDAO analysis helps the user deal with the interactions between different design areas whilst automating the design process. The result is a design where all the parts are jointly optimized.Within the MDAO domain, two components are identified. The former is called analysis block and it comprehends all the modular tools which refer to a specific physical/economic discipline; the latter is the driver, i.e. an optimizing procedure which calls the analysis block in each iteration.This thesis focuses mostly on the analysis of the driver. In particular, the first goal is to identify the best optimizing strategy to be adopted, thanks to several quantitative assessment criteria. The second research goal is studying to what extent the different design areas - aerodynamics, support structure, cable topology etc. - affect the result of the optimization. The third research objective is to see whether the chosen optimizing procedures are able to deal with more complex design situations, such as more constraints or longer design vectors. In the end, an additional analysis on the effect of neighbouring wind farms on the layout optimization has been done, in order to see whether the optimization should consider the effect of those wind farms (disturbed wind conditions) or whether the results from the undisturbed case can be still valuable if re-computed in disturbed conditions.