Optimization of a self-twisting composite marine propeller using genetic algorithms

The design for a self-twisting composite marine propeller is optimized using a genetic algorithm (GA). The objective is to optimize the efficiency of the composite marine propeller for many inflow velocities, an objective unattainable for rigid propellers, and possible only due to the material properties of the self-twisting propeller. Parameters to be optimized are the orientation angles of the fibers in each layer. A generic GA is modified to include a binary-tree memory and local improvement procedures. The GA is also modified to interface with fluid mechanics solvers and ABAQUS. To test the efficacy of a simplified model, two optimizations are performed, one for a simple beam model, and another for the full-fledged propeller model, and these results are compared. Overall, the results for the optimization of the propeller show a significant increase in efficiency for a range of operating conditions.