Array design of a wave farm composed of multiple WECs is an essential issue during the design phase of an offshore wave energy farm. Although various research has looked into the optimal array design, the optimization process usually takes a lot of time.

The objective of this thesis is to reduce the time required during the WEC design phase. The study focuses on performing a hydrodynamic analysis of a wave farm consisting of four single-point absorbers, examining the relationship between excitation force and array spacing, and determining the average total power production of the wave farm.

This thesis employs the Boundary Element Method (BEM) to analyze the hydrodynamic interactions of waves and oscillatory motions in WECs. Using the open-source software Capytaine, it estimates key hydrodynamic coefficients and solves the system of equations via Green's theorem to evaluate added mass and excitation forces. By selecting excitation force as a design parameter, a linear regression model is established. Furthermore, a machine learning model based on the Random Forest algorithm is developed to capture the complex relationship between total power output and the design parameter. After the machine learning model is created, the optimal array spacing for a specific site can be estimated quickly by combining results with a wave scatter diagram.

These findings aim to contribute to wave energy systems design by providing a reliable link between hydrodynamic results and design parameters and accelerating development in the wave energy sector.