interTwin co-designs and implements the prototype of an interdisciplinary Digital Twin Engine (DTE), an open-source platform that provides generic and tailored software components for modelling and simulation to integrate application-specific Digital Twins (DTs). Its specifications and implementation are based on a co-designed conceptual model – the DTE blueprint architecture – guided by the principles of open standards and interoperability. The ambition is to develop a common approach to the implementation of DTs that is applicable across the whole spectrum of scientific disciplines and beyond to facilitate developments and collaboration.
Co-design involves DT use cases for High energy physics, Radio astronomy, Astroparticle physics, Climate research, and Environmental monitoring, whose complex requirements are expected to significantly advance the state of the art of modelling and simulation using heterogeneous distributed digital infrastructures, advanced workflow composition, real-time data management and processing, quality and uncertainty tracing of models, data fusion and analytics. As a result, a consolidation of software technologies supporting research will emerge.
The validation of the technology with multiple infrastructure facilities will boost the accessibility of users to technological capacity and the support of AI uptake in research. interTwin builds on the capacities of experts from pan-European research infrastructures and the long tail of science, an Open-Source Community of technology providers that will deliver TRL 6/7 capabilities to implement the interdisciplinary DTE, experts of the European Centre of Excellence in Exascale Computing, and infrastructure providers from the EGI Federation, PRACE and EuroHPC supporting data and compute intensive science. interTwin key exploitable results will be continually co-developed and aligned with the contribution of external initiatives such as Destination Earth, EOSC, EuroGEO and EU data spaces.
General aims
General objectives
The overall objective of interTwin is to co-design and implement the prototype of an interdisciplinary Digital Twin Engine – an open-source platform based on open standards that offers the capability of integrating with application-specific Digital Twins. Its functional specifications and implementation are based on a co-designed interoperability framework and conceptual model of a DT for research – the DTE blueprint architecture.
CMCC role
CMCC participates in WP2 – Innovation Management and Communications, specifically Task 2.2 – Communications, Dissemination and Engagement, and WP6 – Digital Twin Engine Core Modules, specifically Task 6.1 Advanced workflow composition, real-time acquisition, and processing, Task 6.2 Quality and uncertainty tracing (validation framework of model quality) and Task 6.4 Big Data Analytics. CMCC is also involved in WP4 – Technical co-design and validation with research communities, where it leads Task 4.5 – Climate Change Future Projections of Extreme Events (storms & fire), and WP7 – Digital Twin Engine Thematic Modules, where it leads Task 7.4 Climate analytics and data processing.
Activities
The overall objective of the project is broken down into specific objectives that provide detailed descriptions, indicators and means of verifying achievability for the different aspects of the high-level goal:
1) Co-designing, developing, and providing a Digital Twin Engine that simplifies & accelerates the development of complex application-specific DTs that benefit researchers, business and civil society: The project will co-develop a Digital Twin Engine, a software platform that provides generic and tailored functional modules for modelling and simulation and promotes sharing and reuse of solutions across communities. The aim is to facilitate the ubiquitous use of DTs for research, digital transformation of economy, and the protection of our environment and society by providing near-real time information for example to manage emergencies related to landslides, storms, floods, fire, droughts.
2) Co-designing a Digital Twin Engine blueprint architecture that provides a conceptual framework for the development of DTs supporting interoperability, performance, portability, & accuracy: interTwin will deliver a common conceptual model and interoperability framework of a DT Engine to enable interoperable DTs. This approach can be improved by relying on a DT blueprint architecture, scientific programming libraries, standards, and interoperability best practices to be adopted by users in their own development environment to separate scientific code from hardware-dependent software layers. The interTwin DTE Blueprint Architecture describes a reference architecture that can be used to implement DT solutions for any research collaboration.
3) Extending the technical capabilities of the European Open Science Cloud with modelling & simulation tools integrated with its compute platform: The EGI Federation will offer the DTE as a managed service through the EOSC Portal for testing and validation by research communities external to the project. The DTs developed in the project will be also published in the EOSC Portal. The DTE will build on the compute and storage federated platform of EOSC based on the EGI Federation by supporting the EOSC interoperability guidelines for access and orchestration. It will be integrated with the EOSC core services such as AAI and accounting and will extend the EOSC accounting capabilities with features to include HPC resources. As a result, heterogeneous compute capabilities will be accessible through orchestration to run complex coupled applications across a range of compute architectures. interTwin federated data will integrate the thematic APIs developed by research communities where available in order to discover and access research data available in EOSC.
4) Ensuring trust and reproducibility in science through quality, reliability, & verifiability of the outputs of Digital Twins: Building credibility on computer modelling and numerical simulations requires collecting evidence on the accuracy of the simulated data with respect to the observation of the reality under consideration. Generally, this entails providing support to processes such as traceability, reproducibility, or data provenance. The flow of the information regarding uncertainty quantification along the whole chain of model validation, or the investigation of the trade-off between simulation performance and accuracy are key topics to address. interTwin extends well- known software engineering techniques for Quality Assurance, to the whole lifecycle of the model(s) included in a given DT, i.e., introducing quality indicators in the software, data with associated error budgets, and models. In interTwin a new concept of “Model Quality Verification and Validation as a Service” is established, based on open-source software and customizable by the research domains, that will constitute a major step forward towards the definition of quality-enabled DTs.
5) Demonstrating data fusion with complex modelling & prediction technologies: The sheer amount of data handled by many next generation experiments such as in physics, health and environmental science requires a rethinking of the way modelling, simulation and the running of data-intensive computing algorithms is approached. Modelling and simulation have increasingly complex AI and data analytics needs that require the ability to access third-party generated data. interTwin will deliver functional modules for easier integration of different observational data sources and model generated data; data driven AI/ML models or empirical or theoretical physical based models. interTwin addresses these issues in scale differences, data completeness and reference systems. Thematic modules for specific domains such as earth observation, climate, weather and physics will provide tools for creating harmonized data flows that, integrated with the core engine, will allow for mathematical and scientifically meaningful combinations of those data streams with the model data.
6) Simplifying DT application development with tools to manage AI workflows and the model lifecycle while reinforcing open science practices: The DTE will deliver tools for experienced and novice DT developers to advance their modelling and simulation applications with AI. New models created in the DTE can be exported, reused, and deployed, reinforcing open science practices.
Expected results
The major impact of the interTwin project will be to boost excellence of the European Research Area with a Digital Twin Engine prototype and a shared DT interoperability framework that targets the data- and compute-intensive challenges of some of the most demanding DT applications in science. Besides addressing the outcomes of the call, the project supports policies setting the EU long term strategy in: Open-Source Software, Data, AI, the new European Research Area and Open Science.
Key Exploitable Results:
1) Interdisciplinary Digital Twin Engine: A software platform that provides generic and tailored functional modules for modelling and simulation to facilitate the development and deployment of Digital Twins that address scientific problems in different domains.
2) Interoperability Framework: Guidelines, Specifications, and Blueprint Architecture: The interTwin interoperability framework aligns technical approaches and foster collaboration in modelling and simulation application development across scientific domains.
3) Toolkit for AI workflow and method lifecycle management: AI-based methodologies to extract application sector specific information from research data at the exabyte-scale level in a real-time manner and increase the efficiency and accuracy of simulation and modelling outputs.
4) Quality Framework: Tools for automated quality measures and trust, development of standard quality mapping and indicators for appropriately communicating differences in qualities of inputs and outputs from digital twins, addressing issues such as data and model pedigree, accuracy, and lack of knowledge.
5) DTE federated infrastructure integrated with EOSC and EU Data Spaces: Federated distributed compute platform providing access to distributed data and integrating HTC, HPC, Cloud and Quantum Computing capabilities for processing.
6) interTwin Open-Source Community: The community of DT application developers, users and operators that is responsible for the design, development and maintenance of the DTE code base.
Partners
- CYFRONET Cyfronet Akademia Gorniczo-Hutnicza Im. Stanislawa Staszica W Krakowie
- CERFACS Centre Europeen De Recherche Et De Formation Avancee En Calcul Scientifique
- CERN European Organization for Nuclear Research
- CESNET Cesnet Zajmove Sdruzeni Pravnickych Osob
- CMCC Fondazione Centro Euro-Mediterraneo sui Cambiamenti Climatici
- CNRS Centre National De La Recherche Scientifique
- CSIC Agencia Estatal Consejo Superior De Investigaciones Cientificas
- DELTARES Stichting Deltares
- DESY Deutsches Elektronen-Synchrotron
- ECMWF European Centre for Medium-Range Weather Forecasts
- EODC Earth Observation Data Centre for Water Resources Monitoring Gmbh
- ETHZ Eidgenoessische Technische Hochschule Zuerich
- EURAC Accademia Europea Di Bolzano
- GRNET National Infrastructures For Research And Technology
- INFN Istituto Nazionale Di Fisica Nucleare
- FJZ Forschungszentrum Julich Gmbh
- KIT Karlsruher Institut Fuer Technologie
- KBFI Keemilise Ja Bioloogilise Fuusika Instituut
- KNMI Koninklijk Nederlands Meteorologisch Instituut-Knmi
- LIP Laboratorio De Instrumentacao E Fisica Experimental De Particulas
- MPG Max-Planck-Institut für Radioastronomie
- WWU Westfaelische Wilhelms-Universitaet Muenster
- PSNC Instytut Chemii Bioorganicznej Polskiej Akademii Nauk
- TU Wien Technische Universitaet Wien
- UKRI (STFC) United Kingdom Research And Innovation
- UNITN Università Degli Studi Di Trento
- UHEI Ruprecht-Karls-Universitaet Heidelberg
- UPV Universitat Politecnica De Valencia
- JSI Institut Jozef Stefan
- IZUM Institut Informacijskih Znanosti
Social Media Networks of the project
https://twitter.com/intertwin_eu
https://www.linkedin.com/company/intertwin/