Abstract
Renewable energy sources are progressively assuming a pivotal role in shaping our future, and wave energy stands out as a promising avenue due to its substantial potential and minimal ecological impact. Consequently, extensive research has been directed towards optimizing the layout of wave energy converters (WECs). However, as the number of optimized buoys increases, so does the complexity of calculating hydrodynamic interactions, placing great demands on computing power. Simultaneously, these dynamic interactions can yield either constructive or detrimental outcomes, amplifying the intricacy of layout evaluation. Effectively and promptly determining the optimal buoy arrangement to achieve high energy output efficiency emerges as a pivotal research challenge. We propose a chaos-based differential evolutionary algorithm with a three-layer information structure, including excavation, balancing, and recycling layers, which reasonably adjusts the population structure and makes full use of individual information to optimize the next exploration by combining with chaotic maps. We compared our method with other state-of-the-art intelligent algorithms applied to the problem of wave energy generators and tested it in four real wave scenarios (Perth, Adelaide, Tasmania, and Sydney) using numerical modeling to calculate its energy output. The experimental results show that our improved strategy improves the energy output of the oscillating buoy-type wave energy generator by an average of 101.5%, 93.1%, 23.1%, and 0.7% compared to the mainstream state-of-the-art algorithm for the four scenarios, respectively.