Quote:
Originally Posted by freebeard
If one had 3D models of, oh say, the EV-1, the Volkhart-Sagitta, and the VW XL-1 that all had the same polygon counts, with equivalent vertices along the hard edges (wheel wells and rocker panels to start); then it would be possible to morph among the three to define an envelope of possible design solutions. Then turn evolutionary algorithms loose on the data set and see what falls out.
|
I don't think that day is too far off.
CFD is coming on like gang busters,and wind tunnel results show increasing accuracy with respect to the numerical models.
With an 'archive' of body vectors of known aerodynamic performance,as with EV-1,Volkhart,etc.,seems like the algorithms could solve for variations which remain within some geometric boundary,analyzing for bounded pressure and velocity distributions known to protect a turbulent boundary layer.
Any feature could be tweaked,then the computer would solve for appropriate changes to the rest of the body in order to satisfy the aero requirements.
