In this study, we focused on the layout optimization of a wind farm with the goal of minimizing the Levelized Cost of Energy (LCOE). LCOE depends on the farm power output and its costs. Optimization was done by single-objectuve and multi-objective Genetic Algorithm and Neural Network Algorithm.
Many of the previous researches focused on maximizing the total power output of the farm and overlooked the effects of the optimization of the costs of the wind farm. However, a decrease in the costs and increase in the power output both have a direct effect on decreasing the LCOE. For this purpose, in this study, along with maximizing the total power of the wind farm, we broke down farm costs to different parts(turbines, cables and transformers, foundations, O&M service vessels) and minimized each of them separately.
We considered three different wind scenarios in this project:
Wind Scenario I: Fixed direction wind at a constant speed,
Wind Scenario II: Multi-directional wind at a constant speed,
Wind Scenario III: Multi-directional wind at a variable speed.
We also studied the effect of wind speed on the performance and thrust coefficients. As a result, in the case of considering changes of the coefficients, power output falls because thrust and performance coefficients show a decrease by increasing the wind speed.
We also practiced using different sizes of wind turbines. As a result, using wind turbines with different sizes although increase the power output of the farm slightly, leads to a huge increase in the costs of the farm which makes it unpractical to use this method. Put differently, for a wind farm with specified output power, it is better to use a fewer number of turbines with a larger size rather than using a mix of different sizes of small and large turbines, for it has less LCOE.