Data constrained modelling

Data constrained modelling (DCM) refers to a research and development methodology where the sample-specific data and information, and a theoretical model are tightly integrated for quantitative modeling of a physical system. The DCM methodology has been developed by Dr Y. Sam Yang Et Al at CSIRO since 2007 in relation to 3D characterization of material compositional microstructures and microstructure-based properties ,^[1] and has been implemented as a DCM software ^[2] .^[3] In contrast to the conventional image segmentation technique to derive microstructures from X-ray CT (computed tomography) images, DCM does not impose a length-scale cut-off at the image pixel size scale. Using DCM, consistent 3D compositional microstructures are obtained with different X-ray CT imaging resolution .^[4] The fine length-scale information below image pixel size is preserved as compositional partial volumes in voxels ^{[5] [6] [7]} .^[8]

DCM has been successfully used in 3D microstructure characterization of aerospace paint primers ^{[9] [10] [11] [12] [13] [14]} ,^[15] conventional and unconventional hydrocarbon reservoir rocks ^{[16] [17] [18] [19] [20] [21] [22]} ,^[23] coal ^{[24] [25] [26] [27]} ,^[28] and other materials ^[29] .^[30]

A number of DCM characterized material compositional microstructure data sets are available online ^{[31] [32] [33]} .^[34] Such data sets make it possible for quantitative modeling of microstructure-based material properties, such as fluid transport ^{[35] [36] [37] [38] [39]} .^[40]

For accurate characterization of microstructures, DCM would require multi-energy and quantitative X-ray CT. Care should be taken to minimize the factors which could influence the quantitative nature of the X-ray imaging .^[41]

Additional functionalities can be added to the DCM software using various plug-ins ^[42] .^[43]

References