Advances in development of dedicated evolutionary algorithms for large non-linear constrained optimization problems

Efficient optimization algorithms are of great importance in many scientific and engineering applications. This paper considers development of dedicated Evolutionary Algorithms (EA) based approach for solving large, non-linear, constrained optimization problems. The EA are precisely understood here as decimal-coded Genetic Algorithms consisting of three basic operators: selection, crossover and mutation, followed by several newly developed calculation speed-up techniques. Efficiency increase of the EA computations may be obtained in several ways, including simple concepts proposed here like: solution smoothing and balancing, a posteriori solution error analysis, non-standard use of distributed and parallel calculations, and step-by-step mesh refinement. Efficiency of the proposed techniques has been evaluated using several benchmark tests. These preliminary tests indicate significant speed-up of the large optimization processes involved. Considered are applications of the EA to the sample problem of residual stresses analysis in elastic-plastic bodies being under cyclic loadings, and to a wide class of problems resulting from the Physically Based Approximation (PBA) of experimental data.