1 of 19

Reference Dataset Ontology

(based on the exemplary reference dataset from creep experiments)

Angelika Gedsun

IMTEK University of Freiburg

NDFI MatWerk

​

February 2024

2 of 19

IUC02 : Framework for curation and distribution of reference datasets�Developing a framework for reference material data sets usingreference creep datasets of single- & polycrystalline Ni-based super alloy as example.

E Granda-Gutiérrez, et al., Congreso Internacional de Ingeniería Electrónica ELECTRO 2010 ,32 .

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

2

28. Februar 2024

specifies the methods for:

​

a) uninterrupted creep tests with continuous monitoring of extension,

b) interrupted creep tests with periodic measurement of elongation,

c) stress rupture tests where normally only the time to fracture is measured,

d) a test to verify that a predetermined time can be exceeded under a given force, with the elongation or extension not necessarily being reported.

​

NOTE A creep test can be continued until fracture has occurred or it can be stopped before fracture.

​

ISO: 204:2023

3 of 19

IUC02 : Framework for curation and distribution of reference datasets�Developing a framework for reference material data sets using reference creep datasets of single- & polycrystalline Ni-based superalloy as example.

User Journey/Research task: Alloy design, optimization of mechanical properties

User: Laboratory, researcher

Why we need a reference dataset?

​

  • Scarcity of data (creep experiment take several weeks, testing material is expensive)
  • Quality of data (History/condition of material should be exactly defined) 🡪 defined material characterization/ defined experimental settings/documentation
  • Material has too fulfil specific requirements/ high standards (safety, good mechanical properties under extreme conditions) 🡪 standard material with benchmark properties

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

3

28. Februar 2024

4 of 19

IUC02 : Framework for curation and distribution of reference datasets�Developing a framework for reference material data sets using reference creep datasets of single- & polycrystalline Ni-based superalloy as example.

User Journey/Research task :

  1. Alloy design, optimization of mechanical properties
  2. Calibration of experimental setup
  3. Integration into simulation workflow

​

  • Required Services and Developments:
  • semantic description (metadata schema, ontology) 🡪 application profile
  • Repository /Storage (Coscine)
  • FDOs

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

4

28. Februar 2024

  • documentation,
  • standards,
  • curation process,
  • common/shared concepts (semantic description)

5 of 19

Vocabulary (provided by PP01+PP18)

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

5

28. Februar 2024

  • ~ 220 terms
  • 30 terms defined
  • Hierarchically structured
  • Terms can be mandatory/optional

🡪 to qualify a dataset as a reference dataset

  • Instantiation possible

​

6 of 19

Creep experiment

Measures:

  • Local displacement �🡪 Extensometer system
  • Time
  • Temperature 🡪 Thermocouple
  • Stress/ force 🡪 Load cell

​

  • Detailed description on concepts
  • Application-Level ontology

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

6

28. Februar 2024

Deformation at high temperatures

(> 0.3 to 0.4 Tm, absolute melting temperature)

and constant stress,

low deformation rates up to < ×10−7 s−1

RDOC: Reference Dataset Ontology- Creep

PMD: Platform Material Digital Core Ontology

RO: Relation Ontology

OBI: Ontology for Biomedical Investigations

7 of 19

Reference Dataset from �a creep experiment

  • additional concepts:

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

7

28. Februar 2024

  • Testing Parameter

​

  • Tested material (Condition/history/type/sample)

​

  • Reuse:
    • ongoing work on PRIMA ontology extension (on sample)
    • PMD core ontology modules
      • For Experimental workflow: PMD Tensile Test Ontology (TTL)

PMD: Platform Material Digital Core Ontologyc

BFO: Basic Formal Ontology

RO: Relation Ontology

Co: Collection Ontology

IAO: Information Artefact Ontology

8 of 19

Reference Dataset from a creep experiment

  • Additional information on:

​

  • Testing event (Operator, ID, research project, date,…)
  • Testing Parameter
  • Testing Norm

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

8

28. Februar 2024

PMD: Platform Material Digital Core Ontologyc

BFO: Basic Formal Ontology

RO: Relation Ontology

IAO: Information Artefact Ontology

OBI: Ontology for Biomedical Investigations

dp: data property

ex: example

FOAF: Friend of a Friend

schema: Schema

m4i: Methadata4Ing

9 of 19

IUC02 : Framework for curation and distribution of reference datasets�Requirements

From NFDI-MatWerk:

  • Steered community process (definition of criteria for reference data)
  • Data governance (usage analytics for reference data)
  • MDE backend (data registration, PID, authenticity of reference data)
  • Metadata store (search and discover of data and interrelationships)
  • Incentivation mechanisms
  • Legal aspects (licenses)

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

9

28. Februar 2024

10 of 19

Reference Dataset (Definition IUC02 meeting Berlin October 2023)

Reference data of materials are:

- documented according to certain standards as well as

- curated (i.e., generated, maintained, preserved, validated) according to certain standards

- quantitative data sets (i.e., data + their corresponding metadata) and

- have been measured or simulated to a particularly high standard

- for a material that has either experienced a well-documented process or is reference material or its composition and structure has been completely characterized.

Reference data of materials can also be derived data from reference data.

These data are distributed with a license (e.g., GPL)

The here addressed standards for documentation, curation and measurement need to be defined for specific materials characteristics (e.g., creep tests) in terms of domain standards. These domain standards are the result of a steered community process.

Standards should integrate as much as possible existing standards (e.g., the individual MatWerk domain standards should be as much as possible consistent).

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

10

28. Februar 2024

11 of 19

Reference Dataset�Additional concepts needs to be defined :

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

11

28. Februar 2024

12 of 19

Approach�

  • Scope and requirements: determination of ontology level (middle-domain-application level) and purpose together with domain experts
  • Competency questions: Definitions of questions (search queries) to be answered by the ontology
  • Controlled Vocabulary (definitions+ hierarchy) 🡪 publish via MDI Service: Evoks (Skos)
  • Modelling: graphical representation of concepts, list of statements
  • Reusing and Mapping to other existing Ontologies (domain and top-level)
  • 🡪 pmdco/bfo/mwo/iao/ro/m4i
  • Instantiation: link to data
  • Publication
  • Definition of the curation and validation process: discussions on changes, new versions

​

​

​

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

12

28. Februar 2024

Definition of scope and requirements

Competency questions

Controlled Vocabulary

Classes, properties, constrains

Modelling concepts

Reusing existing Ontologies

Mapping

Instantiation

Validation and curation process

Ongoing work

13 of 19

Formal Requirements

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

13

28. Februar 2024

Domain: MSE (High temperature application/ creep testing)

​

Scope: Middle/Application Level

​

Requirements for the ontology: 

    • Representation of creep testing in a conceptual way  for a reference material and based on a testing standard/norm
    • user can transfer the concepts in their own work

​

Status: Metadata annotation /agreement on controlled vocabulary (with BAM/RUB)

​

Competency Questions

14 of 19

Controlled Vocabulary

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

14

28. Februar 2024

15 of 19

Modelling Calibration Process

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

15

28. Februar 2024

Join Miro Board

PMD: Platform Material Digital Core Ontologyc

BFO: Basic Formal Ontology

RO: Relation Ontology

IAO: Information Artefact Ontology

OBI: Ontology for Biomedical Investigations

m4i: Methadata4Ing

16 of 19

Modelling: Temperature Measurement

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

16

28. Februar 2024

PMDco: Platform Material Digital Core Ontologyc

BFO: Basic Formal Ontology

RO: Relation Ontology

IAO: Information Artefact Ontology

OBI: Ontology for Biomedical Investigations

m4i: Methadata4Ing

17 of 19

Top Level Alignment

Terminological Knowledge (T-Box)

​

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

17

28. Februar 2024

Assertional Knowledge (A-Box)

18 of 19

Important ontologies�

​

  • PMD: Platform Material Digital Core Ontology and modules for (tensile test, heat treatment)
  • BFO: Basic Formal Ontology (Top Level Alignment)
    • IAO: Information Artefact Ontology
    • RO: Relation Ontology
  • MWO: MatWerk Ontology
  • m4i: Methadata4Ing
  • : PMD Tensile Test Ontology
  • PRIMA ontology

​

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

18

28. Februar 2024

19 of 19

Thank you for your attention

Acknowledgements

NFDI: NFDI (nfdi-matwerk.de)

Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under the National Research Data Infrastructure – NFDI 38/1 – project number 460247524

​

Contact:

Angelika Gedsun

angelika.gedsun@imtek.uni-freiburg.de

Albert-Ludwigs-Universität Freiburg | LOD Workung Group

28. Februar 2024