Evolutionary Optimization Algorithm (Evol)

The Evolutionary Optimization Algorithm (Evol) is an evolution strategy based on the works of Rechenberg and Schwefel that mutates designs by adding a normally distributed random value to each design variable. The mutation strength (standard deviation of the normal distribution) is self-adaptive and changes during the optimization loop. The algorithm has been calibrated to efficiently solve design problems with low numbers of variables and with some noise in the design space.

See Also
Configuring the Evolutionary Optimization (EVOL) Algorithm
Optimization References

The Evolutionary Optimization algorithm has the following features:

  • Design space discretization. The algorithm considers only discrete design points, controlled via the Minimum Discrete Step technique option (default is 2% of the design variable domain).

  • Repeat calculation check. The algorithm makes sure that no two design points submitted for evaluation are the same.

  • Sigma expansion. If only repeat calculations are being found after randomization, the algorithm increases the standard deviation of the random normal distribution.

  • Consecutive variable search. The algorithm can vary either all design variables simultaneously or one variable at a time.

  • Parallel execution. The algorithm has been parallelized to produce n children when n parallel resources are available and to use the best of the n children to feed forward to the next iteration.