The following is an excerpt.
Marcos Fernandez’s research brings us closer to that day when physical tests become history. His efforts also pave the way for safer cars, planes, ships, and other structures.
On 21 June, after almost four years of intensive research, he will defend his thesis. He has given it the title «On the use of a virtual laboratory for aluminium alloys: application to large-scale analyses of extruded profiles». The overall aim is to improve the cutting-edge technology for large scale modelling and simulation of aluminium components. In particular, he focuses on extruded profiles designed to absorb energy. You will find them in the wheel suspensions, bumper systems, engine cradles, crash boxes, and other car parts.
A VIRTUAL LABORATORY WITH A ROW OF BENEFITS
Marcos Fernandez’s work extends and refines the Virtual Laboratory (VL) to design aluminium alloys. The VL has been developed in close collaboration between SFI CASA and the parallel project FractAl (2016-2022), led by Professor Odd Sture Hopperstad. The idea is to replace costly and time-consuming physical tests with virtual testing on a computer.
It brings a row of benefits for those who put it to use, and it is one of the main achievements of CASA. A carmaker that switches to an entirely virtual design process can reduce the time from when a new idea is born until thoroughly tested, from 2 months to 24 hours.
PART OF THE ALLDESIGN PROJECT
Marcos Fernandez is also part of the NTNU Digital Transformation project Rational Alloy Design (ALLDESIGN). The project aims to create a digital materials design platform for intermetallic alloy design. The primary focus is aluminium-based alloys, which also are of great importance for the Norwegian industry. His main supervisor is Associate professor David Morin, who plays a vital role in establishing and developing the Virtual Laboratory.
READ MORE: David Morin Shares Virtual News
MORE ROBUST AND FASTER SIMULATIONS
Fernandez feels confident that his work will make a difference in the industry. «How big or how small, I do not know. Nevertheless, I am sure it does», he says. Crash-testing cars and components are expensive, takes much time and harms the environment. The VL enables more robust and faster simulations of greater variety. Also, as the producers run virtual tests before making the prototypes, products and components become safer.
A WORK THAT COVERS ALL THE SCALES
You can think of the VL as a modelling chain with four links, starting with nanostructure modelling. Then it goes up in scale to crystal plasticity and further to unit cell modelling. Then localisation analysis is used to establish a failure criterion for the alloys.
Most scientists in this field prefer to delve into one of the links in the chain. According to Marcos Fernandez, his modelling work is unique. He covers all the scales from the nano-level to full-scale components.
A UNIQUE MIXTURE OF ATTRACTIVE PROPERTIES
The backdrop of Fernandez´s doctoral thesis is the rapidly growing use of aluminium, especially in the automotive industry. The global effort to reduce CO2 and greenhouse gas emissions has fueled the demand. Carmakers embrace the many attractive features of extrusion-based products. They can extrude as cross-sections, single- and multi-chambers, thin-walled profiles and other complex designs. Also, the superb blend of lightweight, stiffness, strength, and the ability to deform before fracture, makes them optimal in energy-dissipating structures.
ENSURING RELIABLE AND SAFE CRASHES
The component’s ability to absorb energy is obtained by different physical mechanisms deep down in the metal’s micro-structures.
Even the most minor variations in the geometry, material properties, and loading conditions can affect the behaviour and thus the component’s ability to protect fragile human bodies. «Engineers and industrial designers must consider these variations», says Marcos Fernandez. Some extruded aluminium profiles are designed to fold up into an accordion shape during impact. This shape points to effective absorption of shock and force in an accident and thus ensures a reliable and safe crash. The illustration shows a deformed extruded aluminum profile during a front-end-impact
SMALL VARIATIONS – HUGE EFFECTS
«If the energy dissipates inappropriately because of a non-robust design, this can produce unexpected accelerations and intrusions inside the vehicle», says Fernandez.
He has studied how various features, such as chemical composition, plasticity and anisotropy affect the materials and the behaviour of specific components. While plasticity is the material’s ability to undergo permanent deformation, anisotropic materials show different properties in different directions.
Even if he focuses on the automotive industry, the results are helpful in other sectors where aluminium is crucial. Examples are the aerospace industry, shipbuilding, and transportation. Regardless of the application, the aim is more accurate simulations and, thus, safer products.
MORE ROBUST AND ACCURATE FINITE ELEMENT MODELS
The industry uses large-scale finite element models to understand the structural capacity and prediction of the response from components to final products.
The robustness and accuracy of these models are imperative, according to Fernandez. Now, he hopes his work can help engineers and designers make proper, faster, and cost-effective decisions. He also hopes that it improves the behaviour of these metallic components in terms of weight and crash performance.
PART I PERFORMING ROBUST MODELLING TECHNIQUES
The PhD candidate’s work is entirely numerical and based upon experimental data provided by his fellow researchers at SIMLab and SFI CASA. «This is my working tool», he says, smiling, lifting his laptop from the desk.
His thesis consists of three parts, each reflecting three specific goals. The first was to perform robust modelling techniques.
In this part, he investigated single and double-chambered extruded aluminium profiles subjected to axial crushing and three-point bending. He pays special attention to deformation patterns and force levels and how different shell drilling stiffness values affect the profiles’ behaviour.
INVESTIGATING EFFECTS OF CHEMICAL COMPOSITION
In the second part, he evaluates how variations in the chemical composition affect the mechanical response of the double-chambered profiles. This part includes large-scale analyses of full-sized extruded aluminium profiles and the use of a nanostructure model. These profiles were later subjected to a variety of loading conditions. After that, he compared the results from the simulations with experimental results.
PLASTICITY AND ANISOTROPY
The third part also includes large-scale analyses. Fernandez studies how plastic and failure anisotropy variations influence the structural behaviour of the double-chambered profiles. Anisotropy is the property of a material which allows it to have different strengths in different directions.
«We created ten different material models with as many anisotropic features. Then we ran the large-scale simulations and saw the variations in behaviour. Of course, this is very relevant for the industry».
Fernandez says that including this information in the Virtual Lab makes the simulations more accurate.
Of course, there have been tough periods. But after all, doing a PhD is demanding. You must put all your efforts into it.
ADDED VALUE TO THE VIRTUAL LABORATORY
When asked what he thinks makes his work stand out, the PhD candidate says:
«The Virtual Laboratory approach used in this thesis covers all the scales, from the nano-level to full-scale analyses, which is unique. The most rewarding part has been seeing how variations in the chemical composition, plastic and failure anisotropy in large-scale analyses can certainly influence the structural response for the considered components. I believe this thesis will help improve the state-of-the-art in large-scale modelling and simulation of energy-absorbing structures for industrial applications. Therefore, this Virtual Laboratory can make a difference. We have proved that it works and that can work well».
FROM GALICIA TO TRONDHEIM
Marcos Fernandez earned his MSc degree at the University of A Coruña (UDC) in Spain in 2016. After that, he worked as a mechanical engineer at the Calculation and Simulation Department at the Automotive Technology Centre of Galicia (CTAG). In his own words, he «did much mathematical stuff which led to the doctoral position in Trondheim».
Despite Covid-19, lockdown, social restrictions, and the like, he declares that he will remember the PhD period as happy years of his life. Also, he praises his supervisors, David Morin and Professor Odd Sture Hopperstad, and fellow researchers at SIMLab for their excellent collaboration.