Evolutionary Search Algorithms for Boundary Layer Control

Abstract

In order to achieve an overall reduction in energy consumption when using suction to delay the transition from laminar to turbulent flow, it is necessary to apply the suction in an efficient if not optimal manner. Here an investigation is conducted into the use of distributed surface suction with multiple suction panels. Cost functions based on minimising the suction effort while maintaining transition in a fixed position and minimising the total energy consumption of the system are employed. Evolutionary methods, Genetic Algorithms and Simulated Annealing, are used to perform the optimisation. Realistic suction distributions are produced. In general better behaviour was found with the Genetic Algorithm than with Simulated Annealing, which was prone to premature convergence.