Check out our science

Discover the stories of researchers at the
Faculty of Mechanical Engineering.

Multi-scale modelling of hydrogen embrittlement inspires state-of-the-art simulations for hydrogen storage

For the past two years, quantum sensing technology specialist Richard Norte (Department of Precision Microsystems Engineering) has been working intensively with machine learning expert Miguel Bessa (Department of Materials Science & Engineering) - a collaboration that has led to a real breakthrough in science.

Circular progression: turning palm oil trunks into biofuel

What if agricultural waste such as oil palm trunks can be turned into high values products such as biofuels? Experimental research by Delft University of Technology and Guatemalan partners using hydrothermal liquefaction is yielding promising results.

Underwater localisation using magnetic fields

How do you navigate equipment over the seabed when there’s no light and no GPS signal? The most energy efficient and least environmentally disruptive solution may be to create magnetic fields maps. Manon Kok of the Delft Centre for Systems and Control (DCSC) and Rudy Helmons at the department of Maritime and Transport Technology (MTT) are collaborating on a 4-year Cohesion Project entitled Enhanced Underwater Localisation using the Magnetic Field.

Metamaterials to reduce noise disruption during installation of wind turbines

Europe’s commitment to a rapid energy transition demands more sources of renewable energy including more and bigger wind turbines in the North Sea. A common way to install a wind turbine is to hammer a monopile foundation into the seabed but this is an extremely noisy process and harmful to marine life. As part of a Cohesion project, researchers from different departments of the Faculty of Mechanical, Maritime and Materials Engineering (3mE) have developed a metamaterial interface to filter out the sound energy that is harming marine biodiversity.

Even better operations with LED photoacoustic imaging

Technology is helping the medical world progress and it is an important motivating factor for Saskia van Heumen, who graduated cum laude at the end of 2021 from the Faculty of 3mE. Her research into using LED photoacoustics in medical imaging brought the medical world a step further. It earned her the title Best Graduate of her Faculty.

Robotic support improves rehabilitation

We are living longer and longer these days. And the older we get, the higher the chance of becoming afflicted with an age-related disease, such as stroke. As many as three million people have a stroke every year in Europe. There is a high probability (40%) that many of those who survive will not be able to walk by themselves, unassisted, after three weeks. At which point they will need rehabilitation. But if an increasing number of elderly people want to rehabilitate, then that’s going to put more and more pressure on therapists. Or will it? Robotic support Probably not, according to Heike Vallery, professor of Human Motor Augmentation at the Department of BioMechanical Engineering in the Mechanical Engineering Faculty at the Delft University of Technology. She invented a robotic support system - the RYSEN™. - with a Swiss-Dutch consortium that is 3 metres wide and 10 metres long, which is fastened to the ceiling. According to her and Michiel Plooij, postdoc at TU Delft at the time and simultaneously robotics system engineer at partner Motek Medical BV, the 3D body weight support system can be used during rehabilitation with people who have trouble walking. ‘More robotisation during rehabilitation gives people the ability to train more independently from rehabilitation therapists. Until now, therapists often have to support patients to prevent them from falling. Our robotic system takes over this supporting role, so therapists don’t have to provide as much physical labour and can focus more on the patients and their rehabilitation process.’ This would not only improve the quality of the rehabilitation, but it could potentially allow therapists to assist two patients simultaneously. ‘Though that does depend on how serious someone’s walking problem is,’ Vallery adds. More robotisation during rehabilitation gives people the ability to train more independently from rehabilitation therapists. Uncomfortable fall Is there currently no device that would prevent someone from falling? There is, but it’s usually a cable fastened to a single point in the ceiling, often above a treadmill. If someone’s risk of falling, then the cable will catch them. The disadvantage of such a simple system is that it doesn’t provide support everywhere, and the force cannot be regulated, so that kind of a fall can be very uncomfortable. Moreover, people can only walk forward during rehabilitation. The rehabilitation system that’s being co-developed by TU Delft addresses all of these disadvantages. RYSEN™ enhances the options that are open to therapists for providing rehabilitation therapy. Reality-based rehabilitation ‘Our 3D body weight support system allows people to train freely instead being restricted by a treadmill. As a result, there are many more possibilities of practising in a normal walking environment, so the patient’s rehabilitation is based more on reality. For example, not only can patients move forward and backward, but they can also walk sideways, practise walking on stairs and other obstacles or slowly sitting onto a chair,’ Vallery says. ‘What’s more, the force can be adjusted, so the system is not only able to catch people when they fall but can also support or disrupt,’ Plooij adds. ‘The RYSEN™ thus enhances the options that are open to therapists for providing rehabilitation therapy.’ Practice safely This disrupting is less important for rehabilitation patients. What the researchers from the Delft University of Technology are particularly interested in discovering is how healthy people respond when they are hindered. ‘We are studying how people walk, how robots walk and how robotic tools can help people to walk again,’ Vallery says. ‘The more we know about how we walk, the easier it will be to discover how to diagnose gait abnormalities and optimise people’s training regimes.’ Indeed, researchers initially believed that they had to get people to exhibit the most perfect gait pattern possible in order to improve their rehabilitation. This notion has now been rejected. ‘The new hypothesis is: you have challenge people and have them practice and make mistakes on their own as much as possible, because that’s when they start to learn things. People learn from mistakes, not from perfection. That can only be done in an environment where patients can practice safely, and we need the right tools for that. I think our relief system is an example of such a tool.’ Clinical tests It’s too early for the researchers to indicate how many patients will be helped or how much quicker patients will be able to rehabilitate. That depends on the initial clinical tests with the robot-assistive rehabilitation method, which should reveal how well it works with patients suffering from spinal injuries. TU Delft is working on this together with the Dutch company Motek, the Swiss company GTX Medical, the rehabilitation hospital CRR SUVA in Sion and the Swiss Federal Institute of Technology in Lausanne (EPFL), within a collaborative Eurostars project. A year ago, researchers from these organisations already demonstrated that patients were able to improve their gait through better equilibrium, better limb coordination and foot placement. Smart mechanics with small motors In this European project, TU Delft’s main focus is designing the mechanical and controls concept. ‘Patients get into a harness that is connected to the 3D body weight support system with cables. The greatest research challenge was that they can move sideways, because mechanically speaking that’s the most interesting part of the robot,’ Plooij says. ‘We looked for a way of achieving this by means of smart mechanics with small motors. And we succeeded: this device can be plugged in with a standard plug, like the ones we use at home. That’s how little power is needed.’ The researchers believe that algorithms, which still need to be developed, will enable the robot to learn what a person needs and automatically adapt to his or her needs in the future. At least, if that’s what the therapist has programmed. ‘In the short term the system could detect that someone is making weird steps on their right side and train them to unlearn that. In the long term the system could learn exactly what point someone has reached in their rehabilitation process,’ Vallery says. More information WWW.MOTEKRYSEN.COM Contact Heike Vallery (TU Delft), h.vallery@tudelft.nl , +31152783517 Michiel Plooij (Motek Medical), michiel.plooij@motekforcelink.com , +31203013020 Jasper de Beus (Motek Medical), jasper.debeus@motekforcelink.com , +31203013020 Claire Hallewas (press officer TU Delft), c.r.hallewas@tudelft.nl ,+31640953085 Text: Desiree Hoving / Images: Marieke de Lorijn H. Vallery +31 15 27 83517 H.Vallery@tudelft.nl 34.E-1-320

How to produce steel in a greener and cleaner way

Steel is indispensable in our modern society. Unfortunately, its production releases a lot of CO2. Due to steel’s large-scale functionality, we have not managed to find a good alternative yet. But we can solve this problem, thanks to scientific research and technological developments.