On the numerical analysis of non-convex variational problems (Q1923306)

A numerical method for finding Young-measure-valued minimizers of non-convex energy functionals is discussed. A method for producing strongly-convergent minimizing sequences for an energy functional whose integrand is quasiconvex but not strictly quasiconvex is specified. This method is related to the relaxation of the original functional to different levels. Another goal of the paper is to show how uniqueness of the Young measure can be used for a numerical convergence theorem. In this case two model problems are considered, one involving scalar valued functions and a multiwell energy, the other from micromagnetics.