Optimal placement of wind turbines on non-flat terrain using cluster identification and multi-objective genetic algorithm
To date, wind power has become very popular to produce electricity due to climate change, greenhouse gases and the fossil fuels crisis. Although using wind turbine technology to produce electricity is very mature, industries are looking to achieve the best utilization of the wind energy in order to fulfill the electrical needs for cities at a very affordable cost. In this thesis, a method entitled Cluster Identification Algorithm (CIA) and an optimization approach called a Multi-Objective Genetic Algorithm (MOGA) are integrated and implemented to maximize wind power efficiency and wind power in wind farms and minimize cost caused by the size and quantity of wind turbines installed in wind farms located on non-flat terrain (i.e., terrain with different heights). An analysis evaluating the fitness function for different populations and generations in order to select the best option was performed. Furthermore, assumption of one wind direction and different factors like different turbine capacities and different quantity of turbines available are considered in this thesis. Also, it considered how the downstream decay model from wind energy theory caused for a wind turbine positioning ahead on the wind farm layout affected the remaining. Finally, a model related to layouts of the wind farm with optimal combination of efficiency, power and cost is developed. A case study that solved three dimensional terrain optimization problems using the combination of CIA and MOGA is also discussed. This thesis forms the basis for solving many other similar problems that occur in renewable energy industries. ^
Alternative Energy|Engineering, Mechanical
Garcia Rosales, Carlos Alejandro, "Optimal placement of wind turbines on non-flat terrain using cluster identification and multi-objective genetic algorithm" (2012). ETD Collection for University of Texas, El Paso. AAI1533226.