A Preliminary Empirical Analysis of Termination Criteria in the Genetic Algorithms for Wind Farm Micrositing

Wind energy has evolved as a leading source of renewable energy. The harnessing of maximum wind energy from a wind farm is governed by many factors. One key factor is the optimal layout design of the wind farm. This layout defines the optimal placement of wind turbines within the farm. The sheer complexity of identifying this optimal layout in presence of various technical constraints makes the wind farm micrositing (WFM) an NP-hard optimization problem. The problem is generally solved with nature-inspired algorithms (NIAs) whose performance depends on several parameters. Among these, the algorithm termination condition plays a crucial role since it determines the right amount of time required by the algorithm to converge. In the context of WFM problem, this study proposes several termination criteria while using the genetic algorithm as the test bench. Performance of six termination criteria is empirically evaluated with respect to the quality of solutions produced and the execution time involved. Results indicate that among the various criteria tested, the Running mean and Phi were the best in terms of quality of solution and runtime, respectively.