طراحی طرح مزرعه بادی با استفاده از الگوریتم کلونی مورچه ها

The wind is a clean, abundant and entirely renewable source of energy. Large wind farms are being built around the world as a cleaner way to generate electricity, but operators are still searching for more efficient wind farm layouts to maximize wind energy capture. This paper presents an ant colony algorithm for maximizing the expected energy output. The algorithm considers wake loss, which can be calculated based on wind turbine locations, and wind direction. The proposed model is illustrated with three different scenarios of the wind speed and its direction distribution of the windy site and, comparing to evolutionary strategy algorithm available in literature. The results show that the ant colony algorithm performs better than existing strategy, in terms of maximum values of expected energy output and wake loss.

Wind Farm Layout Optimization Problem (WFLOP) is a problem, which has an objective function that tries to minimize wake effects of turbines by each other. Therefore, the expected power production of the farm is maximized [1]. There are many researches show that using heuristics to solve WFLOP is possible [2], [3], [4], [5], [6], [7], [8], [9] and [10]. Mosetti et al. [3], Grady et al. [4], Huang [5], and Emami and Pirooz [6] used a genetic algorithm to solve WFLOP. The wind farm is divided into a square grid, to facilitate the encoding of a binary solution. The general framework of optimizing the offshore wind turbine layout was presented by Elkinton et al. [11]. Mora et al. [7] also used a genetic algorithm to maximize an economic function. Mora et al. [7] and Gonzales et al. [8] used an evolutive algorithm. Bilbao and Alba [9] worked on the same problem and conditions within ref. [7] and also developed a simulated annealing algorithm. Öztürk and Norman [10] compared discrete with continuous WFLOP and decided that using continuous location model was better than using discrete one. They constructed six test problems with different siting area sizes and used a Greedy Search Algorithm [10]. Kusiak and Song [2] developed an evolutionary strategy algorithm for WFLOP with continuous variable for turbine locations. The annual energy production was improved by optimizing the wind farm layout design, specifically minimizing the wake loss. From the papers and reviews available in literature [3], [4], [5], [6], [7], [8], [9], [10] and [11], it has been noted that the use of heuristics to address this problem in practice is intensive. This paper presents a new co-operative agents approach – the Ant Colony Search Algorithm – for solving WFLOP. The main goal of this paper is to investigate the applicability of an alternative intelligent search method in the design of wind farm layout optimization. The effectiveness of the proposed scheme has been demonstrated on the Kusiak and Song’s energy output model which calculates the wake loss based on turbine locations [2] and the results were compared with evolutionary strategy algorithm developed by Kusiak and Song [2]

A heuristic ant colony optimization algorithm based on a novel pheromone updating scheme has been presented in this paper for in-land wind farm layout continuous optimization problem. Novel pheromone updating is used to compute pheromone quantity at the end of the each iteration and allows ants to generate new solutions by concentrating to better ants. The constraints of the problem were integrated to algorithm. Thus, the optimization procedure worked on only as a non-linear maximization problem. The optimal solutions via from more accurate layout designs with less energy losses were illustrated with farm layouts and the performance of the proposed algorithm was evaluated on three benchmark problems (first, second, and third scenarios). The first and second scenarios were solved with bi-objective evolutionary strategy algorithm available in literature. It is concluded that the use of ACO algorithm can help to find better wind farm layouts than prior study without being trapped in local maximum in selected problem within a reasonable solution time. The performance of the proposed algorithm was generally good than that of existing algorithm proposed for continuous problems, so it is obvious that the algorithm by using ACO is useful for finding global maximum such as our continuous function.