Texture Componets Scattering Parameter Selection for Polycrystalline Materials when Clustering Rodrigues Space

Abstract

The article considers the problem of maximum scattering cluster value selection at texture components formation within Rodrigues space for texture analysis representation. Texture analysis is carried out in terms of direct pole figure. Orientation cloud is being formed within Rodrigues space by calculating Rodrigues vectors for each found orientation. Texture components describing the texture are being formed by maximin algorithm clustering. The algorithm is considered finished when a predetermined value of maximum scattering in a cluster is reached. In this case, maximum scattering in clusters will be determined automatically. The algorithm enables texture presentation with required detail level corresponding to the posed problem. Implementation of Rodrigues space to form texture components enables texture analysis results visualization. The article gives principal expressions for Rodrigues-vector operations. The calculation of mean Rodrigues vector in a cluster is shown. The algorithm that allows comparison of texture analysis results obtained by two pole figures, taken for one sample. Comparison is carried out by calculation of distances between mean Rodrigues vectors of two pole figure clusters for different scattering parameter values. Texture representation is chosen as the best correspondence of texture components obtained by two pole figures. An example of aluminum tape texture analysis is presented. It is shown that formation of eight texture components with maximum cluster scattering 11.3° by pole figure of family {111} and of six texture components with maximum cluster scattering 10.15° by pole figure {200} their correspondence is 85 %.