Wave energy has potential of contributing significantly to the world's electricity production, but the cost of electricity is still too high to be competitive with other energy technologies. In this paper, an economical model for large-scale wave energy systems is developed and used to optimize parks of point-absorbing wave energy converters. The optimization is performed with a genetic algorithm, and the aim is to find optimal configurations with minimal levelized cost of electricity for the full system. The economical model includes capital and operational costs both for the individual wave energy devices, and for the electrical system including substations and transmission to shore. The results show that the cost of energy of parks of different sizes in a restricted ocean area is expected to have a minimum point, and that the genetic algorithm is able to find layouts with cost of electricity comparable with benchmark cases.
The park design of multiple wave energy converter units is a multiobjective problem that depends on a large number of variables and constraints, such as the layout of the park, the number of devices and their separating distances, incoming wave direction, wave energy converter (WEC) dimensions, etc. Successful optimization of wave energy converter arrays is one of the steps that would help enhancing the commercialization of the technology. For this reason, until now some different systematic methods and optimization routines have been proposed to optimize primarily, but not only, the layout of such a park of devices. Among those we can mention genetic algorithms (Child and Venugopal, 2010; Giassi and Goteman, 2018; Sharp and DuPont, 2018), parabolic intersection (Child and Venugopal, 2010; Child, 2011), covariance matrix adaptation evolution strategies and glowworm swarm optimization (Ruiz et al., 2017; DTOcean).
The genetic algorithm used in this paper was introduced by Giassi and Goteman (2017) for optimization of a single wave energy converter (WEC) and further used for layout optimization of up to 14 identical WECs (Giassi and Goteman, 2018). The method was then extended to optimize simultaneously the power take-off characteristics and geometrical layout of the array (Giassi et al., 2017). In these aforementioned studies the optimal results were obtained by considering the objective function equal to the power output value of the park. The optimization process was therefore a systematic search of the layout with the highest power production. This choice is reasonable as a first approach to the problem; however, an ideal objective and non biased cost function would be an economical one, since the power output is not the only variable which will affect the final profit of the wave energy project. A better way to evaluate optimal solutions would be calculation of the LCOE (levelized cost of energy), taking into account capital (CapEx) and operational (OpEx) costs of the installations in relation to the power output. Among these costs, the cost of the electrical system are very important and result in 20–25% of the overall CapEx (Collin et al., 2017).