Design of a Low-Boom Supersonic Business Jet Using Evolutionary Algorithms and an Adaptive Unstructured Mesh Method

Abstract

The purpose of this study is to establish the feasibility of using constrained, multiobjective, evolutionary algorithms for the design of supersonic, low-boom aircraft in combination with high-delity modeling of the external aerodynamic ow. The objective functions that we attempt to minimize simultaneously are the aircraft coecient of drag at a xed lift coecient, and two dierent measures of the noise level generated by the sonic boom signature at the ground. As part of this work we have developed an analysis tool, BOOM-UA, which fully automates the process of CAD geometry and mesh generation, solution adaptation, and signature extraction and propagation. BOOM-UA is based on the AirplanePlus unstructured tetrahedral solver for the Euler and Navier-Stokes equations. In previous work we have conducted a careful study regarding the number of solution adaptive renement levels and mesh density that are required to accurately predict the sonic boom signature at the ground. Since the cost of driving BOOM-UA directly for reasonable design problems (> 20 design variables) with even the most ecient GAs is beyond the capabilities of modern day supercomputers, Kriging approximation models were constructed from a database of CFD analyses and used in conjunction with the GAs in order to minimize the cost of the computation. This procedure is able to generate a set of Pareto-optimal solutions for the sonic boom and drag objectives that are also forced to satisfy the specied constraints. Optimized shapes for both minimum initial pressure jump, p, and preceived noise level, PNdB, are discussed and their boom signatures are compared. The result of this eort shows that the current generation of computer clusters is able to treat signicant design problems (between 20 and 50 design variables) with reasonable accuracy, cost, and turnaround time.