Numerical Simulation Study on Adjoint-Based Optimization of a Propeller Using Reverse Engineering

In order to enhance the performance of existing propellers, a study has been carried out on the adjoint-based optimization of propellers based on reverse engineering. This method facilitates the rapid development of a more efficient propeller. In the initial phase of the study, the propeller was scanned with a 3D scanner to obtain point cloud data. The point cloud reconstruction technique was then employed to create a 3D model of the propeller. The propeller was then subjected to a thrust test on a propeller thrust test bench. The numerical simulation employed the SST k- $\omega $ turbulence model to solve the multiple reference coordinate system model of the Reynolds average equation. This was then compared with the original propeller thrust test data. The thrust average discrepancy between the simulation and experimental results was found to be 9.8%, thereby confirming the reliability of the reverse-engineered data. Consequently, the radial distributions of the propeller thrust and torque were analysed. Furthermore, optimization of the airfoil sections contributing in excess of 10% of the total thrust was undertaken to enhance the propeller’s thrust. The optimization target was set at a 10% improvement in the lift-to-drag ratio, and the adjoint-based optimization of the target airfoil sections at different angles of attack was performed using the gradient descent algorithm. Following the replacement of the original airfoil profiles with the optimized ones, an overall increase in hover thrust of 9.46% was observed, along with a significant increase in thrust at all speeds from 2000rpm to 5500rpm. The findings of this study illustrate the efficacy of the adjoint-based optimization approach and offer significant insights into the domain of propeller design and optimization in related fields.