IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 28, NO. 3, AUGUST 2013 3075 A New and Ef ficient Method for Optimal Design of Large Offshore Wind Power Plants Javier Serrano González, Member, IEEE, Manuel Burgos Payán, and Jesus Riquelme Santos Abstract—This work addresses the problem of the optimal micro-siting of the wind turbines in large offshore wind power plants with the aim of maximizing the economic profitability of the project. To achieve this goal it is first necessary to estimate the required investment and, secondly, the yearly operation and maintenance costs as well as the yearly income resulting from the operation of the wind power plant over its life span. With this purpose, a complete and realistic model of economic behavior for offshore wind farms has been developed. The optimal turbines layout of a wind farm is a challenge both from a mathematical and technological point of view. The size of the solution space (computational complexity) of the problem ad- dressed in this work dramatically increases with an increase in size of the wind farm. In order to address this difficulty, a new and com- putationally efficient algorithm is proposed. The method is based in the division of available marine plot in smaller areas of suitable size, sequentially optimized by an improved genetic algorithm. Index Terms—Genetic algorithm, micro-siting, offshore wind farm, wake effect. I. INTRODUCTION G ROWING interest in renewable energy along with the maturity of the relevant technologies has caused offshore wind energy to experience significant development in recent years. The offshore installed capacity in Europe has risen from 4.95 MW in 1991 to the present (mid-2012) 3813 MW [1]. Offshore wind farms (OWFs) are not yet as developed as on- shore installations, but they have several advantages over on- shore facilities. The best wind conditions are at sea: the wind speeds are greater than on land (due to the reduced friction pro- vided by the water) and the wind is less turbulent (in the absence of obstacles on the sea). These factors lead to greater production and a major reduction in fatigue on the blades and the structural components of the turbines located offshore. Further advantages include the availability of large areas and the reduced visual and noise impact from marine facilities. At present, the largest operational OWF in operation is Walney Offshore Wind Farm in the UK, with 102 wind tur- bines (WTs) and a rated power of 367 MW. The progressive reduction in costs makes it more attractive to increase the size of OWFs. It is therefore expected that in the coming years, the rating of individual OWFs may reach 20 GW, as is expected to Manuscript received July 05, 2012; revised December 27, 2012 and February 18, 2013; accepted February 23, 2013. Date of publication April 15, 2013; date of current version July 18, 2013. This work was supported in part by the Spanish MCI under grant ENE2011-27984, the Andalusia Government under project Ref. P09-TEP-5170, and by IV Plan Propio de Investigación Universidad de Sevilla. Paper no. TPWRS-00771-2012. The authors are with the Department of Electrical Engineering, University of Seville, Seville 41092, Spain (e-mail: javierserrano@us.es; mburgos@us.es; jsantos@us.es). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPWRS.2013.2251014 be the case of Tianjin Hangu OWF (still in the early stages of project development) [2]. Currently the maximum power rating of WTs is around 5 MW and in the coming years it is expected that could reach values of approximately 10 MW. This limitation in the power rating of individual WTs along with the outstanding increase in the capacity of the offshore facilities involves the installation of a large number of turbines in a relatively small area: a 1-GW OWF would need 200 WTs of 5 MW. The optimum positioning of each WT is not a trivial problem since it depends on several fac- tors, many of which are interdependent. For example, although it would be desirable to separate the WTs as much as possible in order to maximize the energy capture (by minimizing the wake effect), this would increase the total length of the cables, with ensuing increasing in infrastructure costs and electrical losses. It is therefore necessary to reach a balance between these two opposing effects in order to maximize profitability. From a mathematical point of view, this optimization problem exhibits manifold optimal solutions (convexity) and cannot be completely described in an analytical form since some variables have a character of non-allowed values, (the solution space is not simply connected) while other variables are discrete. This renders the problem non-derivable, and prevents the use of any classic analytical optimization techniques. To date, optimal planning of wind farms has been addressed in several works [3]–[7]. The problem of optimal wind turbines positioning in a wind farm (WF) was introduced by Mosetti et al. [8] which aimed to determine the optimal placement of WTs for maximum energy capture at the lowest possible investment costs. Their optimization algorithm is guided by a simplified economic model of the WF (based on economies of scale and overlapping wakes) to search for an optimal layout based on genetic algorithms. Ozturk et al. [9] used a greedy algorithm as an optimization method in order to optimize a slightly different objective function. Grady et al. [10] included certain improve- ments in the economic model and wake effect model. Serrano et al. [11]–[14] developed a cost model more complex and realistic than in previous research by including aspects such as electrical installation design and modelling of the main features of civil work. Wan et al. [15] proposed a PSO algorithm as a method of optimization. The major innovation introduced in that paper is the use of a continuous domain instead of the discrete domain used in previous studies. Kusiak et al. [16] proposed the opti- mization of a multi-objective function using an SPEA algorithm also considering a continuous domain on a circular terrain. All previous work highlights the complexity of solving the problem of micro-positioning turbines in a WF. However, the extraordi- nary increase in the size of the solution space due to the poten- tial size of a WF has not been addressed. Although previously 0885-8950/$31.00 © 2013 IEEE