PNNL

Abstract

Constructing realizations of random media from limited statistical descriptors attracts strong research interest due to its importance in materials design. This study focuses on efficiency and accuracy of microstructure reconstruction using different statistical descriptors. Two point functions and lineal path functions are utilized in microstructure reconstruction. When more statistical descriptors are used, the reconstructed microstructures are closer to target microstructure. In the periodic microstructures studied, linear path function did not improve reconstruction quality dramatically. Minimizing error will improve reconstruction if the input statistical information is appropriate. Generally, efficiency represented by calculation time consumed decreased with more statistical descriptors introduced in microstructure representation.