Design Practice and Method Modeling
of Composite Material Chip Metadata Based on MGE Data General Rules
Yongchao
LU, Lanting ZHANG*, Hong WANG
Materials Genome Initiative Center, School of
Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240,
China
ABSTRACT: With the advancement of the
concept of material genetic engineering, the new paradigm of data-driven
materials research and development has become increasingly clear. The problem
of material data standardization has become more and more prominent, and the
disorder of data format content restricts the speed of material law exploration
under the new situation. The FAIR (findable, accessible, interoperable, and
reusable) principle is the benchmark specification for scientific data. Based
on this, the committee of China Standards of Testing and Materials (CSTM)
formulated and issued the group standards named General Rule for Materials
Genome Engineering Data (General Rule),
which stipulate the composition of material genetic engineering data and the
information that must be included in it. However, it did not provide actual
methods and practical cases on how to formulate specific material data
standards. Material metadata is the structured data of material information,
which determines the type range and framework of subsequent specific data.
Based on the metadata design practice of the composite material chip
preparation and characterization experiment, this work models the metadata
design method that meets the specific material data standards of the General Rule to provide a reference for the
standardization of material data.
This work uses ion beam deposition (IBD) composite material chip
preparation technology to sequentially combine and integrate different
elements/materials on the substrate to form a sample with continuous distribution
of components. Then XRD, XRF and other characterization methods are used to
characterize & analyze the composition and phase of the sample. According
to the classification of material data in the General Rule, sample
information metadata, characterization metadata, and analysis derived metadata
are respectively formulated. Combining the principles, processes and related
equipment parameters of each link to conduct metadata collection, screening and
correlation analysis to ensure the comprehensive effectiveness and
interoperability of information. Use unique and permanent digital resource
identification technology to identify metadata to ensure data findability.
Relational structure and Internet architecture are used in metadata design to
ensure the availability of data. In addition, focus on t the readability of
information to ensure the reusability of information.
Combined with the
standardized metadata design practice of the combined chip, establish a universal
standardized "material metadata information flow"
model, as shown in Figure 1: I.
Combining principle analysis, process drawing, equipment parameter collection
and other methods, establish the carrying entity of information flow, that is,
the main body "channel" of information flow. II. The circulation channel of the
information flow should be closed and continuous to avoid information
interruption, which means that metadata should be collected strictly according
to the information flow channel, and important information should be selected
to ensure the integrity of the information flow. III. The information circulating in the
information channel should be intuitive and visible. For example, for
non-standard documents, the software and version must be clearly read to ensure
the readability of the information.
IV. Each information stream
should be distinguishable, which means that a single piece of information
should be encoded with a unique identification code to ensure its uniqueness
and searchability within the information channel. V. Use the standard vocabulary of material
genetic engineering data to avoid ambiguity during interaction.
Figure 1. Material metadata information flow model
Keywords: Data-driven; FAIR; Combined material chip; Material metadata information flow model
Lanting Zhang,is currently a Professor of Materials Science & Engineering at Shanghai Jiao Tong University (SJTU) and Deputy Director of the Materials Genome Initiative Center (MaGIC) of SJTU. He is also the group leader of the High Performance Metallic Materials Laboratory in School of Materials Science and Engineering (SMSE). He received his BSc, MSc and Ph.D. in materials science from SJTU in 1991, 1994 and 1997 respectively. His current research interests include high-throughput characterization of materials, development of rare-earth permanent magnetic materials for traction motor and heat-resistant steels for turbine and ultra-super critical power plant etc. He is now leading a National Key Research and Development Program project on high-throughput characterization of materials combinatorial chips. He has published over 90 peer-reviewed journal articles in Acta Materialia, Scripta Materialia, PRB, JAP, Intermetallics and JALCOM etc.