4th World Congress on Integrated Computational Materials Engineering (ICME 2017): Integration Framework and Usage - IIB
Program Organizers: Paul Mason, Thermo-Calc Software Inc.; Michele Manuel, University of Florida; Alejandro Strachan, Purdue University; Ryan Glamm, Boeing Research and Technology; Georg J. Schmitz, Micress/Aachen; Amarendra Singh, IIT Kanpur; Charles Fisher, Naval Surface Warfare Center
Wednesday 2:00 PM
May 24, 2017
Room: Salon II, III
Location: Ann Arbor Marriott Ypsilanti at Eagle Crest
2:00 PM Invited
An ICME Framework for Incorporating Bulk Residual Stresses in Rotor Component Design: Vasisht Venkatesh1; 1Pratt & Whitney
Abstract Coming Soon!
Materials Integration System being Developed for Relating Among Process, Structure, Property, and Performance: Masahiko Demura1; Junya Inoue1; Makoto Watanabe2; Manabu Enoki1; Toshihiko Koseki1; 1The University of Tokyo; 2National Institute for Materials Science
Still now, process, structure, property, and performance are the key elements in materials science and engineering, and any types of modellings can be regarded as links relating among them. Building a chain of links among them and predicting performance should be helpful and would be crucial for accelerating the R&D of new innovative structural materials. Based on this guiding principle, we have proposed a new concept named Materials Integration (MI) and have been developing the MI system in the SIP* Structural Materials for Innovation program led by the Cabinet Office, Japan. It is a modular-type system, where one designs and executes a workflow connecting several computational modules of simulations, empirical or theoretical formulas, and experimental or computational databases. Each workflow can be regarded as a chain of modellings linking among the four elements, and yields an integrated pathway to predict the performance of a target structural part. In this talk, we will introduce our MI system with a couple of workflow examples for predicting creep lifetime of a heat-resistant engineering part. *SIP: Cross-ministerial Strategic Innovation Promotion Program.
An Ontological Framework for Integrated Computational Materials Engineering: Sreedhar Reddy1; BP Gautham1; Prasenjit Das1; Raghavendra Yeddula1; Sushant Vale1; Chetan Malhotra1; 1TCS Research, Tata Consultancy Services
ICME is expected to significantly reduce the dependence on trial and error based experimentation cycles for materials development and deployment in products. However, modelling and simulation is a knowledge intensive activity. In an integrated design, choosing right models for different phenomena, at right scales, with right parameters, and ensuring integration across these models is a non-trivial task. The gaps in modelling and simulation need to be filled with tacit knowledge and co-engineered with product knowledge. Therefore, an IT platform having capabilities such as, a) a repository of building-block models, templates and workflows with an intelligent means to choose and compose right workflows for a given problem, b) a knowledge engineering framework for knowledge management, c) a simulation services framework for simulation tool integration and simulation execution, d) tools for decision support, optimization, robust design etc., is essential for scaling up ICME for industrial applicability. This requires a unifying semantic foundation. Ontologies can provide the common substrate for integration of different models, the common language for information exchange, and the means for capturing and organizing knowledge. However, ontology engineering is a challenge when we consider the diversity of the material systems, products, processes and mechanisms involved in ICME. This calls for a flexible ontological framework that provides a means for modelling the generic structure of a subject area (e.g. materials) and a means for instantiating subject specific ontologies from this generic structure. We describe a model driven framework and how it has been used for developing an enabling platform for ICME.
Establishment of a Vocabulary-inventory System for Data Utilization in Materials Science and Engineering: Takuya Kadohira1; Toshihiro Ashino2; Hiroaki Ishiki3; Satoshi Minamoto1; Kaita Ito4; Makoto Watanabe1; Junya Inoue4; Manabu Enoki4; Toshihiko Koseki4; 1National Institute for Materials Science; 2Toyo University; 3ITOCHU Techno-Solutions Corporation; 4University of Tokyo
We show an idea to construct systems of vocabulary-inventory for research and development of materials with data utilization. Based on the idea, the list of vocabularies with persistent identifiers will be created using collective intelligence, i.e. more than one user can register descriptors used in his/her research and organize them under his/her own data structure such as data schema, notes and so on. After evaluation of the idea using wiki, we have created an experimental system under the idea. In this talk, we will make a brief introduction about the experimental system, and also discuss a potential of the idea to establish an unified ontology for data integration from scattered data-sources, which will lead acceleration of data-utilized research in the field of Materials Science and Engineering.
3:30 PM Break
Digital Infrastructure for Driving Rapid Insertion of New Materials: Deborah Mies1; Will Marsden1; 1Granta Design, Ltd.
ICME was founded to drive the rapid insertion of new materials into parts manufacturing. The methodologies to achieve this have evolved dramatically in recent years with the evolution of additive manufacturing. New polymeric and metallic feed materials, for instance, are being developed at a rapid pace to meet the demands of this manufacturing method as well as new materials developed as a result of the manufacturing process itself. This digitally-driven manufacturing process can easily be integrated into a data management infrastructure that supports the capture and management of all process inputs and outputs, supports data discovery while securing IP, and enables certifiable knowledge extraction from integrated computational analysis tools. When execution is consistent with developing standards for certification, these new materials can be quickly approved for manufactured parts. Granta Design presents its experience in advancing the digitization of advanced materials development activities, such as ICME or Additive Manufacturing.
Context Aware Information Retrieval from Materials Publications: Sapan Shah1; Dhwani Vora1; B P Gautham1; Sreedhar Reddy1; 1TCS Research
Knowledge of material properties as a function of material composition and manufacturing process parameters is of significant interest to materials scientists and engineers. A large amount of information of this nature is available in publications especially in the form of experimental measurements, simulation outcomes, etc. In a typical problem solving context, when information required is not available in standard databases, an engineer has to first go through a large collection of publications, either internal reports or open literature, to filter the right set of documents, containing information relevant to the context. Having filtered the publications, an engineer has to then go through each of them to extract the relevant pieces of information. Our goal is to help automate some of these steps. In this paper, we present our ongoing work on a system that provides information search and extraction based on material entities such as elemental composition, manufacturing processes (along with processing parameters) and properties. To specify search criteria, the system provides a domain specific query language. The accuracy of our system critically depends on the accuracy of the information extraction component. To address this, we are developing a sophisticated extraction component that combines rule based and machine learning based approaches. We have conducted an experiment on a small library of publications on steel on which searches such as “get the list of publications about steel which has carbon composition between 0.2 and 0.3 and on which tempering is performed for about 30 to 40 min” are performed.
Symbiotic Resources for Atomistic Simulations: nanoHUB + OpenKIM: Sam Reeve1; Christopher Chow1; Alejandro Strachan1; 1Purdue University
Molecular dynamics (MD) simulations are increasingly being utilized within ICME frameworks and used by a wider range of the materials community. However, there are numerous potential issues in running MD simulations due to i) lack of computational resources or limited computational modeling and coding experience or ii) low level of access to interatomic models or difficulties in determining the ideal interatomic models. We describe the implementation and uses of an online simulation tool combining the capabilities of the nanoHUB and OpenKIM projects (both National Science Foundation supported). The Nanomaterials Mechanics Explorer (nanoHUB.org/resources/nanomatmech) leverages the cloud computing infrastructure and graphical user interfaces (GUI) of nanoHUB.org, alleviating the first set of difficulties. The tool internally queries the interatomic model database of openkim.org, shows the user all available potentials, downloads and builds the model chosen on the fly, and enables running MD simulations and directly comparing multiple models, addressing the second set of issues. The tool allows a wide range of user input (through the GUI), atomistic structure and thermodynamic outputs, and use the uncertainty quantification framework within nanoHUB, all within internet browsers, extending the usefulness further. In future work we hope to close the loop between nanoHUB and openKIM, using the nanomatmech simulations to provide properties (model predictions) and improve MD tests (standard calculation methods) available through OpenKIM.
Ensuring Reliability, Reproducibility and Transferability in Atomistic Simulations: The Knowledgebase of Interatomic Models: Ryan Elliott1; Ellad Tadmor1; James Sethna2; 1University of Minnesota; 2Cornell University
Atomistic simulations using empirical interatomic potentials play a key role in realistic scientific and industrial applications. This talk describes an NSF-funded effort to develop an open source online tool for promoting the use and reliability of interatomic models. The Knowledgebase of Interatomic Models (https://openkim.org) allows users to compare model predictions with reference data, to generate new predictions by uploading simulation test codes, and to download models conforming to an application programming interface (API) standard that has been developed in collaboration with the atomistic simulation community. An overview will be given of the KIM project and its main components which include the KIM API, the KIM data structure for representing arbitrary material properties, the KIM processing pipeline, and the KIM visualization framework.