A new type of multi-objective genetic programming (MOGP) for design exploration is proposed. The feature of the new MOGP is the simultaneous symbolic regression to multiple variables using correlation coefficients. This methodology is applied to Pareto-optimal solutions of the multi-objective aerodynamic design optimization problem of a bi-conical shape reusable launch vehicle. The MOGP presents symbolic equations which have high correlations to zero-lift drag at supersonic condition, maximum lift-to-drag at supersonic condition and volume of shape through single MOGP run. These equations also have high correlation to another parameter of the body geometry. These results indicate that MOGP is capable of finding composite more efficient design parameters from original design parameters.

This content is only available via PDF.
You do not currently have access to this content.