Aerodynamic Shape Optimization of the Common Research Model Based on Improved SQP Algorithm

Abstract

Aircraft Aerodynamic Shape Optimization is a complex, large-scale and expensive optimization problem. Adjoint-based gradient optimization method plays a significant role in aerodynamic shape design. In this paper, we mainly focus on the optimization algorithm applied to aircraft aerodynamic design field. Sequential Quadratic Programming (SQP) is employed here, and some experience-based improvements are made in line search to accelerate the convergence of the algorithm. After improvements, the physical change of variables in each iteration can be evaluated, set and control. The optimization model with the drag coefficient minimization objective and wing thickness constraints for Wing-Body-Tail Common Research Model (CRM) is established. The optimization strategy and the improved SQP algorithm are detailed and verified afterwards. Aerodynamic shape design for CRM with Wing-Body-Tail Configuration is carried out. The optimization results are compared and discussed. Our optimization procedure reduced the drag from 167.9 counts to 149.4 counts (an 11% reduction) within 40 iterations, calling CFD solver 49 times. The optimization results demonstrate the effectiveness of the improved SQP method proposed in this paper.