Keynotes and Invited Presentations

Ivan Krajinović
Rimac Technology

Yves Nadot and Carole Nadot-Martin
Institut Prime, ISAE-ENSMA, Université de Poitiers

Irina Munteanu
Dassault Systemes

Cloud-based Simulations in Rimac Technology

Cloud computing is one of the ongoing trends in numerical simulations. Across multiple industries there exist trend to decrease costs of new product development, mostly by decreasing the investment physical prototyping and testing. Additionally, physical testing has its limitation, from number of sensors per device under test (DUT) to frequency of testing, which is not limited only by speed and costs of production, but also with the availability of testing facilities. Virtual prototyping and virtual testing using sophisticated numerical codes is becoming more and more dominant way of doing development. To successfully perform such task powerful HPC cluster are needed. They are either built on-premise within the companies or rented from different government organizations or universities. Another way to access HPC computing power is HPC cloud computing. There are barriers to uptake, either having on-premise HPC cluster or a cloud solution. Two leading barriers for on premise HPC cluster are high software and hardware costs, while for a cloud-based solution those are total cost of ownership and data security. Rimac Technology HPC journey started with utilization of one university HPC cluster, then moved to on-premise HPC cluster and finally having a hybrid solution combining on-premise HPC cluster with a cloud solution. Initially, during first half of 2021, detailed Proof-of Concept (PoC) was done with several cloud providers. The results, speed and Total Costs of Ownership (TCO) has been compared to Rimac Technology on-premise solution. Microsoft Azure with UberCloud platform has been selected as preferred vendors. Now, almost two years using cloud solution there are multiple lessons learned from POC, creating custom solution, and putting it in production. Multiple benefits of using cloud computations are also going to be discussed.

²Computing Platforms for Engineering Simulations 2015 vs 2020, Lee Margetts, University of Manchester, Benchmark , July 2021

Dr. Ivan Krajinović is Head of Simulations at Rimac Technology, Croatia. He organized and leads the Simulations department, which is responsible for structural, electrical and thermal simulations of different components of electric powertrain. Simulations department is working on all major Rimac Technology’s projects for major automotive OEMs. Showcase of what has been done are components that power Rimac’s electric hypercar Nevera, fastest accelerating production car in the world.

Fatigue Design of Additive Layer Manufacturing Materials

Additive layer manufacturing (AM) has been rapidly evolving over the last decade, and more and more industrial application examples are to be seen in the aerospace sector. This can be attributed to the compatibility of this technology of the characteristic objectives from this sector; namely small production values with flexible delivery schedule and lightweighting. The critical technical challenge being the certification of structural performance and reliability, fatigue research is at the forefront of engineering development. In the keynote lecture a summary of the AM specific fatigue process is presented, where the role of surface and subsurface defects is highlighted, since they have a dominant impact by controlling the fatigue properties. A modelling strategy is proposed to evaluate the influence of defect morphology on the fatigue limit of additively manufactured Al alloys by;

• (a, b, c) obtaining a 3D image of defect population from X-ray micro-computed tomography (µ-CT) or process simulation, (d) computing the Equivalent Inertia Ellipse of each individual pore,
• (e) modelling the influence of the defect on the fatigue limit through the DSG approach and,
• (f) 3D mapping the criticality of each individual defect.

The Methodology is validated through fatigue tests conducted on AlSi10Mg alloy manufactured by selective laser melting. Modelling strategy

Figure 1: Modelling strategy: (a) fatigue sample: fatigue limit and critical defect identified, (b) 3D µ-CT image of the gauge section, (c) each porosity is imaged and labeled, (d) the EIE is computed, (e) the DSG criterion is computed for each individual defect, (f) 3D rendering of the criticality of all defects

Yves Nadot is a Full Professor at the National School of Mechanics and Aerotechnics, located at Poitiers Futuroscope. He leads the research teams “Damage” and “Physics and Mechanics of Materials” and is an active contributor to the research on the defect induced fatigue process. He worked on industrial collaboration projects with the companies SAFRAN, Zodiac Aerospace, Renault, Airbus, Knorr-Bremse, SNCF, and RR.

5G Applications in the Context of Human Exposure

Long awaited and discussed, the 5th generation of wireless networks – 5G – is becoming a reality. While the typical perception is that 5G is a consumer-focused upgrade of the previous cellular network communication protocols, its impact on industry will be tremendous, 5G being a key part of the industry digitalization process for several industries. It will connect not only people, but also equipment and vehicles, powering the Industrial Internet of Things and the development of smart factories. All this requires innovative antenna systems, designed to function in extremely complex environments and in various frequency ranges.

One essential prerequisite of the 5G success is to make these connections reliable – taking into account electromagnetic, mechanical and thermal performance, but also safe for the people using them to ensure public acceptance.

After an introduction to the 5G characteristics and challenges, the talk will present some complex scenarios and will show how simulation can help answer questions such as: Can a stable connection between a multitude of devices and sensors in a particular industrial setting be achieved? How does the movement of equipment and people influence the connection quality? How to evaluate the electromagnetic field level around and inside the human body and assess safety levels?

Prof. dr. Irina Munteanu is Strategy and Business Development Director at Dassault Systemes, leading strategic projects in various industries and areas of technology, in particular electronics. She also teaches electromagnetics and simulation-related courses at the Technical University of Darmstadt, Germany and authored/co-authored more than 120 papers and 6 books. Her research interests include: numerical methods for electromagnetic field simulation, electromagnetic compatibility, signal integrity, lightning strikes simulation, optimization and reduced order modeling.

Lubos Pirkl
CFDSupport

Gabor Zipszer
eCon Engineering Kft

Tomasz Łukasik
Tenneco

Artur Król
Tenneco

Various Aspects of Using Open Source in Computer-Aided Engineering - CAE

Open-source software is a large topic in technology and it affects all kinds of human activity. This also applies to engineering. Open-source has strong pros and cons and no wonder its reputation is very contradictory among engineers and even more among managers. While open-source is typically well accessible, flexible, and ready for further development, it is often missing a user interface, professional services like technical support, best practices, benchmarks, documentation, training, and a clear roadmap of its development for the future. The goal of this presentation is to make a summary of various aspects of using open-source for engineering and share 15 years of experience with the open-source CAE and virtual prototyping.

Luboš Pirkl is the managing director of CFD SUPPORT. Luboš’s specialisation is strategy, product management, and business development. He studied at the Czech Technical University in Prague, Faculty of Mechanical Engineering (FME). He got his master’s degree in Mathematical modelling in engineering in 2007, later he successfully finished his Ph.D. studies in 2011, where he majored in Mathematical modelling in Hemodynamics.

He worked as a lecturer of programming at the Department of Technical Mathematics of CTU. In 2009, still, during his Ph.D. studies, Lubos co-founded CFD SUPPORT LTD. Lubos likes riding a motorbike, playing cards, chess, football, and fishing. He has a wife and four children, who are now besides business his main source of entertainment.

Aircraft air-intake optimisation using AI supported Virtual Blade Model

The water-cooler intake of an ultra-light aircraft was optimised accounting for propeller induced flow field by the utilisation of a 3D corrected Virtual Blade Model. The 3D aerodynamical corrections were derived by an Artificial Intelligence module linked to the in-house developed VBM. The CFD model was built-up using Ansys Fluent, while the optimisation was governed by OptiSLang. Three of the optimised intake designs were built and tested during in-flight measurements.

Gabor Zipszer is a senior CFD simulation engineer at eCon Engineering Kft. He was the leader of a 3-year long tender project focusing on the development of CFD and FEM simulation and optimisation methodologies for small- and medium-aircraft verified by an extensive in-flight test campaign. As part of the project an AI driven 3D corrected Virtual Blade Model was developed and applied during the optimisation tasks. Gabor previously worked as an aero-performance engineer at Dowty Propellers (UK) and were involved in new product development and wind tunnel test campaign activities.

Increasing simulation efficiency and accuracy of prediction by means of Digital Twins concept.

Due to demanding current business situation, recently OEM expectation is to reduce the number of prototype builds and tests in design and validation phase of the design and development project. The background of this trend is related to decreased overall new vehicle production process and release on market. The trend to avoid costly and time consuming prototyping and testing phase, made a space for different types of simulations, taking the role of design selection of design failure detection. That’s raise the questions how accurate those simulations are. This question is especially important due to ongoing automotive market transformation. Electrification of the vehicles leads to change of requirements, often leading to increasing loads due to higher vehicle weight. How can we provide the proof of correlation by structural way. The digital twin concept can be used to provide solution for that requirement. The smart connection the simulation results with test data significantly improve the confidence of engineering judgement. Digital Twins might take a role of smart design selection tools at the time demanding quotation phase of product development project. Additionally, Digital Twins will become source of the engineering knowledge, potentially replacing current forms of knowledge capturing methods, e.g. design guidelines written as text document.

PhD. Eng Tomasz Łukasik, CAE manager in Tenneco. Specialization: Strength and durability FEA analysis, SPDM, Digital Twin Education: Silesian University of Technology PhD degree: Fatigue of welded joints with application of local approach. The responsibility is functional management of simulations of suspension components dampers, modules. Additional activities is engineering process digitization implementation.

Artur Król Engineering Manager in Global Engineering organization of Tenneco company – Global supplier of automotive components. The responsibility area is connected to driving quality standards in engineering processes, e.g. DFMEA process, KPI reporting, business process improvement. Additional role is Six Sigma Master Black Belt, ensuring Six Sigma methodology and tools are correctly used in Engineering organization. The next level of personal development is interest in Digitalization of Engineering Processes, especially by Digital Twins implementation.