#	Pagina
attuale pagina	/open-h2020/projects/198314/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - AdMoRe (Empowered decision-making in simulation-based engineering: Advanced Model Reduction for real-time, inverse and optimization in industrial problems)

Teaser

Summary

Computational Mechanics tools are well integrated in the technological industrial practice. However, the global effort (pre-process, solve and post-process) is a major overhead for real industrial problems. Thus, simulation-based engineering are not extensively used in real-time for decision-making. Real-time (fast-queries) is critical for control of manufacturing processes, non-destructive testing and fast decision-making at production phases. This is also the case for multiple-queries: optimization and parameter identification with uncertainty quantification. Addressing multiple-queries in an efficient and reasonably accurate manner is crucial in applications of major industrial interest (independently of the manufactured product: engines, cars, planes, helicopters, rockets, medical devices). The actual bottleneck lies in the computational effort to be furnished in solving each of the queries with the desired accuracy.

Developing frontier research and training a new generation of engineers to empower simulation-based engineering for fast and multiple-queries in real industrial problems is the mission of AdMoRe.

In order to define precise R&D objectives, the industrial partners of AdMoRe, have identified within the corrent multidisciplinary challenges four precise Industrial Problems (IPs): composites & forming technology; fast simulations for geometrical design; inverse problems & identification; integration of reduced order models into commercial codes. These IPs involve multi-disciplinary modeling (i.e. solids, fluids, structures, electromagnetics, heat, acoustics) because the computational complexity mostly arises from complex models. They also require managing inter-disciplinary connections (i.e. fluid-structure interaction, electro-magneto-mechanics, thermo-mechanics, aerodynamic noise) bridging the gaps between different formulations.

Classical approaches fail in addressing these challenges for fast and multiple-queries within a high-fidelity and accurate perspective. Therefore, techniques required to face them have been selected among new emerging scientific and technical fields and frontier research is being performed on two main streams. On the one hand, accelerating the numerical resolution of each of the queries to vanquish the computational bottlenecks mentioned. Reduced Order Models, - in particular the Proper Generalized Decomposition (PGD) -, is being used to account for the multidimensionality in fast and multiple-queries without suffering a dramatic increase of the computational complexity. On the other hand, a new â€œfull-packâ€ of cutting-edge computational mechanics approaches is required to solve these IPs accurately and reliably.

Todaysâ€™ global economy requires shorter transfer periods between the uncovering of a new method until it is incorporated in a daily industrial production environment. AdMoRe aims to drastically reduce this time and facilitate the diffusion mechanisms of Transfer of Technology with a Knowledge to Codes (K2C) concept which will bridge the gap between Academia and Industry.
Five specific objectives have been identified by the AdMoRe consortium:
1. To train and empower a new generation of engineers in developing innovative solutions for existing IPs in collaboration with both academic and industrial researchers. These solutions are going to be based in industry-driven computational simulation tools.
2. To position European industry on the cutting-edge of simulation-based engineering to reinforce and consolidate its worldwide leadership.
3. To truly contribute with original research and innovations in simulation-based engineering in order to solve multidisciplinary problems with interdisciplinary tools arising from the proposed IPs for fast decision making.
4. To translate knowledge to codes in a daily industrial production environment (i.e. software prototypes, friendly user interfaces).
5. To cultivate the entrepreneurial spirit of the trainees and integrate translation

Work performed

During the initial 12 months, agreements among the partners of the consortium have been established and a thorough recruitment has been performed in order to identify and retain the best candidates for the project. During the following 12 months, the recruited Early Stage Researchers got acquainted with the state-of-the-art of reduced order models. Their initial contributions were presented in many scientific congresses as both poster and oral presentiations and disseminated within the partner industries in form of seminars and internal talks. In particular: verification of the existing models and validation of the numerical results have been performed for thermo-mechanical and non-linear mechanical problems in the aerospace industry; high-fidelity real-time evaluation of shape aerodynamics in automotive industry; advances towards the accurate and fast simulation of eddy currents, deformations and vibrations in MRI scanners; integration of modern reduced order models techniques into commercial softwares.

During the mid-term review meeting (October 25-26, 2017), AdMoRe was selected to participate to the pilot program Innovation Radar of the European Commission to identify possible sources of innovation and code-to-market opportunities within the research projects cdeveloped in the consortium.

Final results

Many computational models are not currently used to their full potential by the European Industry and therefore a competitive edge is being lost with respect to other well-established (e.g. USA, Japan) and new-emerging economies (e.g. China, Brazil, India). AdMoRe is an industry-driven program which aims to maintain the worldwide leadership of the four European industrial partners involved. In order to do so, the existing gap between academic knowledge and technical competences currently exploited by the industry will be filled.

AdMoRe is expected to have a major impact in different fields of European industry:
1. The production of large thermoplastic composites parts is a challenging issue for today\'s aerospace and automobile industry. The thermo-mechanical process involved is significantly affected by many material and process parameters and techniques for monitoring and controlling in real-time such procedures are extremely valuable in the industrial practice.
2. Design and optimization of vehicles and aircrafts require capturing complex flow features with high-fidelity computational tools. During production cycles, this is known to be extremely expensive due to the large number of configurations tested and the high cost of every simulation of the shape optimization process. A real-time feedback of the simulation pipeline represents a major advancement in the overall design and optimization pipeline.
3. Magnetic Resonance Imaging (MRI) scanners use pulsed field gradients to form an imatge but they are often affected by imaging artifacts. There is a clear need of the industry for fast analysis tools to predict the magnetic field variations and for high-fidelity benchmarking tools able to validate the acquisition procedures at diferent stages of the manufacturing of MRI scanners.
4. Solving the aforementioned problems in a daily industrial production environament requires the incorporation of scientific and technical developments into commercial codes. This in itself is another challenge where compromise is necessary between development costs, performance and efficiency.