This paper presents a new approach for the optimization of the layout of offshore wind farms. Almost all previous work on optimal micro-siting for large offshore wind farms have been based on irregular arrangements of wind turbines. However, most offshore wind farms already built are configured in symmetrical/ regular layouts. From a mathematical point of view, the geometrical relationships of such symmetrical layouts enable the problem to be defined by just a few variables. This presents a considerable advantage compared with irregular arrangements where the number of variables is directly linked to both the number of wind turbines and the number of cells in which the computational domain is discretized. In contrast, symmetrical layouts are more demanding with regard to the optimization process, since the problem constraints, such as the shape of the available exploration area to deploy the project, the maximum surface allowed, and the maximum number of wind turbines, drastically increase the nonlinearity of the objective function, which affects the ability of the optimization algorithm to achieve the optimal solution. This work compares the behaviour of two meta-heuristic optimization algorithms (the Genetic Algorithm and Particle Swarm Optimization) in solving the addressed problem and, more importantly, it introduces a series of improvements on the objective function, which enhance the behaviour of the optimization algorithms when dealing with realistic constraints, such as the shape of the concession zone and maximum deployable area. Finally, the performance of the proposed methodologies has been tested under two situations. The first scenario is a small-sized hypothetical offshore wind farm. In the second scenario, the layout of a real project (Horns Rev 3 offshore wind farm) has been optimized and compared with the solutions proposed by the Danish transmission system operator. The results obtained show the ability of the proposed tools to successfully show the ability of the proposed tools to optimize offshore wind farms under realistic considerations. (C) 2017 Elsevier Ltd. All rights reserved.