logo Evolutions designs

Airfoil Design Software

Select language web site

Airfoil design and optimization for possibly more than one design condition is not an easy engineering task. . The aerodynamic characteristics at a given design point (i.e. at given Reynolds and Mach numbers) of a wing section are strongly dependent on its shape through a number of geometric parameters: the leading edge radius, the mean line curvature, the thickness distribution to mention a few. The sensitivity of this dependency is high and from there comes the arduousness of trying to combine those parameters in such a way to obtain an optimal shape. Moreover, a wing section must satisfy to very often conflicting requirements. That is to say that trying to optimize a wing section means usually to seek for the best compromise (or «Trade-offs») between different demands.

The classic approach

A consolidated way to face this challenge is to pick up a few airfoils from a database of known ones, whose aerodynamic characteristics could suite to a single design point and then to select the most promising one for further optimization, for instance by means of the gradient method. Such a way to proceed on top of being very lengthy and requiring to have the possibility to access a wing sections database, doesn't guarantee to lead at the end of the day to the selection of a global optimum solution within the search space, but more likely to a relative one.

The evolutionary strategy

At Evolution Designs an evolutionary approach to the search and optimization problem is used instead.
A multi-objective genetic algorithm, EDGAR (Evolution Design's Genetic Algorithm, real-coded) was written, tested and tailored explicitly for airfoil design.

Genetic algorithms (GAs) offer a very valuable alternative approach to the search of an optimum solution by mimicking the mechanism of natural of evolution. The advantages of such an approach on the more classical ones, for instance the gradient method, are several:

Robustness

GAs work well in non-smooth search spaces characterized by several local maxima and tend to be global.

Search effectiveness

in EDGAR a fitness sharing tecnique was implemented in order to enhance the natural GA's capbility to seek the global maximum in the search space. On top of that, an archiving technique pushes the population toward a higher fitness than the one of those generated during the past generations.

Multi-objective optimization

Very often the design objectives happen to be in conflict with each other. EDGAR's deals with multi-objective optimization problems. Across the generations the solutions population will evolve to the so called «Pareto Front», on which the trade-off among the different optimal designs can be easily assessed by the engineer.

Start from zero

EDGAR's true potential lies in it's ability to generate wing section with no need of a database of possible initial solutions. The population is randomly gnerated and simlpy "observed" as it evolves.

Turn around Time

Last but not least, the time to obtain a front of solutions is dramatically shorter than the time required for one airfoil design via the classical ways.

The careful airfoil design it’s a key factor for the succesful design of an aerodynamic surface. Evolution Designs offers its expertise to help in developing high quality solutions suited to different applications, such as high speed wing sections, optimized fowler and slotted flaps, wing sections for high performance gliders, profiles for very-light aircraft, hydrofoils for racing sailboats, blade sections for eolic generators.

Evolution Designs will be proud to be your partner in developing a new successful design.

Login

login


Are you not yet registered?
Register now.!

I forgot my password!