Application of a Hierarchical Chromosome Based Genetic Algorithm to the Problem of Finding Optimal Initial Meshes for the Self-Adaptive hp-FEM

Abstract

The paper presents an algorithm for finding the optimal initial mesh for the self-adaptive hp Finite Element Method (hp-FEM) calculations. We propose the application of the hierarchical chromosome based genetic algorithm for optimal selection of the initial mesh. The selection of the optimal initial mesh will optimize the convergence rate of the numerical error of the solution over the sequence of meshes generated by the self-adaptive hp-FEM. This is especially true in the case when material data are selected as a result of some stochastic algorithm and it is not possible to design optimal initial mesh by hand. The algorithm has been tested on the non-stationary mass transport problem modeling phase transition phenomenon.