20 Postdoctoral Fellowship at Delft University of Technology (TU Delft), Netherlands

Are you interested in new problems at the interface between fluid dynamics and biology? Are you excited about hands-on experimental work in a multidisciplinary environment? We are looking for two enthusiastic Post Doc candidates to develop new experiments to track microbial life in flows.

Photosynthetic microalgae hold promise for the sustainable production of high-value products, bioplastics, biofuels, and future engineering living materials. Flowing suspensions of living microalgae cells represent a new kind of living fluid that physiologically responds to the environment and flow conditions. New insight into the flow dynamics of these living microalgae suspensions is crucial to model algae blooms or develop new flow technologies for bioreactors.

The postdoc position is available within the project “Flow4Algae”. This experimental project aims to understand the multiscale fluid dynamics of living microalgae suspensions. We are particularly interested in the interplay between flow conditions, cell physiology, and cell growth. Each different project will focus on flow conditions ranging from linear shear flows in microfluidics to weak turbulence. In this project, you will design new multiscale experiments and analytical tools for the different flow regimes. These studies will be conducted in the BioFluids Laboratory in TU Delft and will use the infrastructure in the laboratory, including advanced flow diagnostics (Particle Image Velocimetry, Particle Tracking Velocimetry, and Laser Induced Fluorescence), microscopy (3D tracking multi-view microscopy and Fluorescence), rheology tools, microfluidics, and existing flow dynamics setups.

As a postdoc, you will be part of a vibrant team of researchers working on the project Flow4Algae, with diverse backgrounds and expertise. The group is part of the Process & Energy Department, which thrives to conduct world-class research & education focusing on process & energy technologies for sustainable development.

Hydrogen is expected to play a key role in reaching climate neutrality for hard-to-abate sectors. Today, producing hydrogen via water electrolysis is prohibetively expensive, due to high investment costs, low efficiencies and a high operational cost related to the electricity input. At the same time, the regulatory framework requires future electrolyzer projects to operate in tandem with nearby renewable electricity production, requiring dynamic operation of the electrolyzer plant.

In this research project, we investige the interplay between (i) electrolyser technology design, flexibility and costs, (ii) different configurations and downstream integration concepts, (iii) the ability of the asset to capture different revenue streams in energy and secondary markets and (iv) flexibility requirements imposed by the required matching of renewable electricity production and green hydrogen.

To address this question, you will develop state-of-the-art equilibrium models describing linked electricity and hydrogen markets, and the role of different electrolyzer technologies in these markets. You will build on existing implementations and ongoing work within our research group. You will focus on how data-driven approaches, such as inverse optimization, may be used to enhance the realism of resulting market outcomes and to quantify “model” uncertainty.

Note that this postdoctoral position is one of the two positions funded by the WinHy project at TU Delft (financed by NWO). This position focusses on the interaction of the electrolyzer assets with the overall energy system, whereas the other position focusses on the design of the electrolyzer plant itself.

In WinHy, the consortium partners Repsol (a technology integrator and developer of electrolyzer plants), Sunfire (a leading OEM and electrolyzer manufacturer) and research insititutes DLR and TU Delft are seeking to address shortcomings of the current alkaline water electrolysis technology by developing advanced stack designs and plant integration concepts. This unique consortium spans technology development and lab-scale experiments to energy system and economic integration, which is the focus of this postdoctoral position lies.

Your daily supervisor will be Kenneth Bruninx. You will work closely with the research group led by Mar Pérez-Fortes on the assessment and implementation of emerging technologies and the consortium partners. You will join a lively community of internationally renowned interdisciplinary energy researchers in the Department of Engineering Systems and Services at TU Delft. Our faculty hosts a diverse group of people from different countries and disciplines, and we welcome candidates that contribute to and enjoy this diversity.

Hydrogen is expected to play a key role in reaching climate neutrality for hard-to-abate sectors. Today, producing hydrogen via water electrolysis is prohibitively expensive, due to high investment costs, low efficiencies and a high operational cost related to the electricity input. At the same time, the regulatory framework requires future electrolyzer projects to operate in tandem with nearby renewable electricity production, requiring dynamic operation of the electrolyzer plant.

In this research project, we aim to bring down the cost of green hydrogen produced via water electrolysis by developing improved operational and design decision support methods, geared towards (prospective) electrolyzer plant owners. These novel approaches allow for better operational decisions (e.g., which position to take in the electricity market,, how fast to ramp up/down the asset, or better thermal management of the asset) as well as design choices (e.g., how many electrolyzer stacks should make up the plant, and how are they connected).

To address this question, you will develop state-of-the-art optimization models to capture (i) the technical limitations of the electrolyzer plant, the stacks and the downstream process, as well as the associated costs, (ii) the interaction with electricity and ancillary services markets and (iii) the impact of the prevailing regulatory requiremements, such as the Delegated Act on RFNBO’s.

Note that this postdoctoral position is one of the two positions funded by the WinHy project at TU Delft (financed by NWO). This position focusses on the design of the electrolyzer plant, whereas the other position focusses on interaction of the electrolyzer assets with the overall energy system and markets.

In WinHy, the consortium partners Repsol (a technology integrator and developer of electrolyzer plants), Sunfire (a leading OEM and electrolyzer manufacturer) and research insititutes DLR and TU Delft are seeking to address shortcomings of the current alkaline water electrolysis technology by developing advanced stack designs and plant integration concepts. This unique consortium spans technology development and lab-scale experiments to energy system and economic integration.

Your daily supervisors will be Kenneth Bruninx and Mar Perez-Fortes. You will join a lively community of internationally renowned interdisciplinary energy researchers in the Department of Engineering Systems and Services at TU Delft. Our faculty hosts a diverse group of people from different countries and disciplines, and we welcome candidates that contribute to and enjoy this diversity.

The ’Near wellbore characterisation and architectural element modelling of heterolithic tidal reservoir zones’ project (REVFlow) aims to develop a faster method to determine the key differences of the flow and transport behaviors occurring in heterolithic tide-dominated sedimentary succession. The specific focus is to classify groups of small- and medium-scale 3D lithofacies heterogeneities using rock property characterization and link them to arrival time distributions in the near well-bore area. The role holder will create a library of the spatial distribution of multiscale architectural elements of these lithofacies and develop a classification scheme to group the elements. Petrophysical properties for each class will be varied using a design-of-experiment approach. Flow and transport behaviour will be computed for single- and multiphase flow processes for each class (incl. petrophysical variability) using upscaling and flow diagnostic techniques to compute average flow properties and arrival time distribitions. The flow processes are generic and representative for a broad range of subsurface applications. Finally, machine learning techniques will be used to cluster and classify the flow and transport behaviours and identify they key heterogeneities that control the different classes of dynamic behaviours. The research will be conducted in close cooperation with an industry partner, so coordination and discussions with other scientists will be essential; discussions. Involvement in teaching and (co-)supervision of BSc and MSc students is greatly welcomed.     

Semiconductor chips store and process information in everything from pacemakers and cars to the electronic device you are reading this on right now. As these chips evolve, they’ve shrunk to the point that makes them hard to see and measure. Controlled manufacturing of increasingly small chips is one of the biggest challenges faced by the semiconductor industry today. To solve this problem, TU Delft joined forces with ARCNL and ASML to push current boundaries in applying electron beam imaging instead of light imaging to inspect and measure chips in the production process.

We are inviting applications for a 1,5-year position (0.8 FTE) as a postdoc researcher in the interdisciplinary project ReBioClim – Restoring urban streams to promote Biodiversity, Climate adaptation and to improve quality of life in cities. ReBioClim is an Interreg Central Europe project that brings together 12 partners around four urban stream restoration cases in Dresden (Germany), Poznan (Poland), Bratislava (Slovakia) and Jablonec nad Nisou (Czechia), with external expertise on spatial analysis and design from TU Delft.

Building on the tradition of the Department of Urbanism of designing and planning cities with water, as well as the involved research group’s experience in urban riverspace analysis and design, the TU Delft team will conduct urban geospatial analyses considering the morphological, social, ecological and environmental particularities of urban stream corridors. Carried out as a spatial multi-criteria analysis in all case study locations, this activity will reveal how local spatial patterns influence stream restoration potentials and to what extent they are generic and thus transferrable. Conflicts (e.g., between traffic and ecological corridors) and synergies (e.g., low-impact recreation in a riparian zone) across the different layers of analysis will be highlighted.

TU Delft will also lead the joint development of a co-design workshop methodology. Following a multi- and mixed-method approach to data collection in the workshops, the research team will triangulate nuanced information on different facets of stream restoration and its perception by citizens. The co-design methodology will be set up in such a way that comparison and integration of information across the cases will be possible in subsequent activities. The workshop will be designed to engage citizens effectively and make their co-design experience enjoyable, insightful and empowering.

Moreover, TU Delft will contribute to the development of transferrable design principles for social-ecological integration that rely on the nature-based solutions collected or developed during the project. The integrative character of the design principles (e.g., how they meet social and ecological targets jointly) will be made explicit in the description of the principles. All principles will have an illustration that explains the essence of the proposed intervention and thus facilitates their uptake in implementation. After being evaluated and validated by experts, the principles will be incorporated into a best-practice guide and a handbook for integrated urban stream restoration.

The successful candidate will contribute to the XCROSS project, which is part of an EU-funded international innovation program. This project involves collaboration with partners across Europe, focusing on the work package “Development of Advanced Digital Twin.” This package aims to develop and validate numerical models for optimizing crossing geometry profiles, enhancing the efficiency and reliability of wheel-rail interactions, and ultimately contributing to better asset management in the railway sector.

Autonomous vehicles have come a long way from initial public testing with safety drivers to fully driverless deployments in multiple cities across the world. However, to further improve these vehicles, they need to be trained from ever growing data sources collected by their fleets. Due to the vast cost of storing and processing this data, car manufacturers face the problem that they can only store a fraction of it (storing 1h of data costs about $5000 per year). Therefore new paradigms are needed to reduce the storage and processing cost.

Federated learning is such a novel paradigm that revolutionizes how Machine Learning systems are trained. Rather than storing all data on a central server and training there, federated learning allows us to learn in a distributed way. In the case of autonomous vehicles, this means that every vehicle can perform training steps while the system is idle, thereby saving energy and bandwidth. Furthermore, federated learning is much less privacy-sensitive, since sensor data of faces does not need to be transferred to the cloud.

The Intelligent Vehicles section at TU Delft, the Netherlands, invites applications for a fully funded two-year (up to 30 months) Post-Doctoral research position in the area of federated learning for autonomous vehicles. This position is part of the EU Horizon project Cynergie4MIE and in collaboration with the chip company NXP. You will be one of the first to train modern perception and/or prediction (a.k.a. motion forecasting) algorithms in a federated learning setup and provide an analysis of the pros and cons of this approach. While the exact task (detection, segmentation, prediction) is flexible, it should make use of radar sensors, potentially fused with other sensors. Your results will be published in top tier conferences like CVPR, ICCV, ECCV, ICRA and NeurIPS. For your work you will have access to the compute resources of TU Delft, ranging from personal machines, to shared GPU servers, the Delft AI Cluster that is shared across departments, as well as DelftBlue, which is one of the top 250 supercomputers in the world. Your main supervisor will be Dr. Holger Caesar, creator of the nuScenes dataset and co-author of the PointPillars method for lidar-based object detection. You will receive hands-on mentoring for your career development and help with applying for future research grants, to kick-start your continued academic career.

The most advanced X-ray based techniques are availble at synchrotron radiation sources, such as the ESRF in Grenoble or DESY in Hamburg. Access to these extremely bright sources is limited, and Dutch researchers are in a disadvantageous position as the Netherlands are not directly involved in any of these centres of research. The Smart*Light project (https://interregvlaned.eu/en/smart-light-2-0/) aims to make advanced X-ray analysis techniques available in the Netherlands. Several institutes in The Netherlands (TU Delft, TU Eindhoven, DIFFER) and Flanders (UAntwerpen, UGent) collaborate to develop a compact instrument that comprises an intense and energy-tunable X-ray source, X-ray optics and detectors. The source is based on the Inverse Compton Effect, where visible laser light, scattered by a relativistic electron beam, is promoted into the X-ray range, yielding a brilliant, coherent X-ray beam with a narrow bandwidth.

Your main task will be to build and commission the experimental endstation of Smart*light. This endstation is meant for a characterisation of the beam and proof of principle experiments with a focus on cultural heritage objects (art objects or archaeological objects). This involves testing different components in the laboratory with X-ray tubes, verify their integration into the control software, determine spectroscopic figures of merit, optimize the instrument and demonstrate its capabilities in the first experiments. Based on the early experiments you will be involved in the planning of future Smart*Light upgrades, especially regarding optics and beam shaping.

This work will be done within the team of the Alfeld Lab at the Department of Materials Science and Engineering of the Delft University of Technology. The Department of Materials Science is a diverse institution, working in small laboratories, headed by independent PIs on a range of topics, including optimising steel microstructure, corrosion, recycling and the investigation of cultural heritage. The Alfeld Lab is working on the multi-modal investigation of cultural heritage, with a special focus on X-ray investigations and the development of new instruments and data evaluation approaches.

You will work with the PI of the group, M. Alfeld, on the Smart*Light endstation in the lab in Delft and with the team of the TU/e on the installation of the endstation in Eindhoven. It is possible to supervise Master students during the project and to pursue independent research endavours related to spectroscopy and/or cultural heritage. 

To gain insights on how hydrogen interacts with the material and leads to the observed changes in mechanical properties, atomistic simulations based on the Density Functional Theory (DFT) and Molecular Dynamics (MD) will be performed within this project. These simulation techniques provide the necessary insight into the hydrogen interaction with the underlying microstructural features of the material at the nanoscale. To this end, the adsorption energies of hydrogen on Fe-based metal surfaces will be computed using DFT (incl. effect of oxygen and moisture). MD simulations will be performed to model the diffusion of hydrogen atoms and their interaction with the microstructure.  

This position is primarily scientific in nature, but the Post-Doc is expected to contribute to the creation of a large-scale follow-on project. The project will be carried out in the research groups of Dr. Othon Moultos and Dr. Poulumi Dey. The two departments involved in this project are Materials Science and Engineering (MSE) and Process & Energy (P&E) of TU Delft.

We are inviting applications for a 1,5-year position (0.8 FTE) as a postdoc researcher in the interdisciplinary project ReBioClim – Restoring urban streams to promote Biodiversity, Climate adaptation and to improve quality of life in cities. ReBioClim is an Interreg Central Europe project that brings together 12 partners around four urban stream restoration cases in Dresden (Germany), Poznan (Poland), Bratislava (Slovakia) and Jablonec nad Nisou (Czechia), with external expertise on spatial analysis and design from TU Delft.

Building on the tradition of the Department of Urbanism of designing and planning cities with water, as well as the involved research group’s experience in urban riverspace analysis and design, the TU Delft team will conduct urban geospatial analyses considering the morphological, social, ecological and environmental particularities of urban stream corridors. Carried out as a spatial multi-criteria analysis in all case study locations, this activity will reveal how local spatial patterns influence stream restoration potentials and to what extent they are generic and thus transferrable. Conflicts (e.g., between traffic and ecological corridors) and synergies (e.g., low-impact recreation in a riparian zone) across the different layers of analysis will be highlighted.

TU Delft will also lead the joint development of a co-design workshop methodology. Following a multi- and mixed-method approach to data collection in the workshops, the research team will triangulate nuanced information on different facets of stream restoration and its perception by citizens. The co-design methodology will be set up in such a way that comparison and integration of information across the cases will be possible in subsequent activities. The workshop will be designed to engage citizens effectively and make their co-design experience enjoyable, insightful and empowering.

Moreover, TU Delft will contribute to the development of transferrable design principles for social-ecological integration that rely on the nature-based solutions collected or developed during the project. The integrative character of the design principles (e.g., how they meet social and ecological targets jointly) will be made explicit in the description of the principles. All principles will have an illustration that explains the essence of the proposed intervention and thus facilitates their uptake in implementation. After being evaluated and validated by experts, the principles will be incorporated into a best-practice guide and a handbook for integrated urban stream restoration.

The Faculty of Civil Engineering and Geosciences, Department of Engineering Structures, is seeking a postdoctoral researcher to advance the application of AI/machine learning in the field of concrete engineering. The successful candidate will focus on predicting concrete strength properties, optimising mix designs, characterising materials, discovering new materials and ensuring production quality control. This research will utilise both laboratory and field data, facilitated by the implementation of advanced sensor systems.

The primary objective of the project is to deepen the understanding of the intricate relationships between raw material properties, mix proportions, processing conditions and the resulting engineering properties of concrete. This understanding will lead to significant improvements in process optimisation, reducing the need for large safety margins in mix design. Ultimately, the results of this research will help to reduce the energy consumption, costs and carbon emissions associated with concrete production. This role provides a unique opportunity to make a significant contribution to the field of sustainable construction through advanced data-driven methodologies.

Your main responsibility will be to carry out the postdoctoral project and conduct research in data-driven concrete science. You will have the opportunity to influence the details of the research according to the needs of the project, with the possibility of combining experimental work with modelling and working in close collaboration with industry. As a postdoc, you will be involved in teaching and supervising students in the department.

The project will be carried out at TU Delft in the Faculty of Civil Engineering and Geosciences (CEG) in collaboration with the Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS). The position is located in the Concrete Structures Group within the Engineering Structures Department of the CEG.

The Department of Engineering Structures focuses on the development of resilient, smart and sustainable structures and infrastructures. Our aim is to meet societal demands in transportation, the energy transition and sustainable reuse. Research themes include dynamics of structures, mechanics of materials related to e.g. climate change, modelling and design of railway systems, multi-scale modelling of pavement materials and structures, reuse of materials, structures and parts of structures, assessment methods for structures, smart monitoring techniques, design methods, replacement and renovation of civil infrastructure and development of new materials and maintenance techniques. Our unique Macro Mechanics Laboratory facilities support full-scale testing, monitoring and modelling of structures to facilitate implementation of innovations. The department delivers both groundbreaking research and world-class education for undergraduate and graduate students. As a team, we represent different backgrounds, skills and views. We foster an inclusive culture, as our combined identities, attitudes and ambitions widen our perspective and make up our strengths.

The successful candidate will contribute to the XCROSS project, which is part of an EU-funded international innovation program. This project involves collaboration with partners across Europe, focusing on the work package “Development of Advanced Digital Twin.” This package aims to develop and validate numerical models for optimizing crossing geometry profiles, enhancing the efficiency and reliability of wheel-rail interactions, and ultimately contributing to better asset management in the railway sector.

Autonomous vehicles have come a long way from initial public testing with safety drivers to fully driverless deployments in multiple cities across the world. However, to further improve these vehicles, they need to be trained from ever growing data sources collected by their fleets. Due to the vast cost of storing and processing this data, car manufacturers face the problem that they can only store a fraction of it (storing 1h of data costs about $5000 per year). Therefore new paradigms are needed to reduce the storage and processing cost.

Federated learning is such a novel paradigm that revolutionizes how Machine Learning systems are trained. Rather than storing all data on a central server and training there, federated learning allows us to learn in a distributed way. In the case of autonomous vehicles, this means that every vehicle can perform training steps while the system is idle, thereby saving energy and bandwidth. Furthermore, federated learning is much less privacy-sensitive, since sensor data of faces does not need to be transferred to the cloud.

The Intelligent Vehicles section at TU Delft, the Netherlands, invites applications for a fully funded two-year (up to 30 months) Post-Doctoral research position in the area of federated learning for autonomous vehicles. This position is part of the EU Horizon project Cynergie4MIE and in collaboration with the chip company NXP. You will be one of the first to train modern perception and/or prediction (a.k.a. motion forecasting) algorithms in a federated learning setup and provide an analysis of the pros and cons of this approach. While the exact task (detection, segmentation, prediction) is flexible, it should make use of radar sensors, potentially fused with other sensors. Your results will be published in top tier conferences like CVPR, ICCV, ECCV, ICRA and NeurIPS. For your work you will have access to the compute resources of TU Delft, ranging from personal machines, to shared GPU servers, the Delft AI Cluster that is shared across departments, as well as DelftBlue, which is one of the top 250 supercomputers in the world. Your main supervisor will be Dr. Holger Caesar, creator of the nuScenes dataset and co-author of the PointPillars method for lidar-based object detection. You will receive hands-on mentoring for your career development and help with applying for future research grants, to kick-start your continued academic career.

 

(18) Postdoctoral Fellowship Position

Postdoc Fellowship Position summary/title: Postdoc Foundation Models for Radar-based Autonomous Vehicle Perception

Deadline : 15 September 2024

View details & Apply

 

To gain insights on how hydrogen interacts with the material and leads to the observed changes in mechanical properties, atomistic simulations based on the Density Functional Theory (DFT) and Molecular Dynamics (MD) will be performed within this project. These simulation techniques provide the necessary insight into the hydrogen interaction with the underlying microstructural features of the material at the nanoscale. To this end, the adsorption energies of hydrogen on Fe-based metal surfaces will be computed using DFT (incl. effect of oxygen and moisture). MD simulations will be performed to model the diffusion of hydrogen atoms and their interaction with the microstructure.  

This position is primarily scientific in nature, but the Post-Doc is expected to contribute to the creation of a large-scale follow-on project. The project will be carried out in the research groups of Dr. Othon Moultos and Dr. Poulumi Dey. The two departments involved in this project are Materials Science and Engineering (MSE) and Process & Energy (P&E) of TU Delft.

Reaching energy transition goals of 2030 and 2050 require multiple increases in production capacity and cost reduction of support structures for wind turbines offshore. With the increasing size of turbines and the deeper location of wind farms, the monopile foundations of offshore wind turbines increase in size, weight, and cost. Jackets can be attractive alternative because multi-membered circular hollow section (CHS) support structures are lighter than monopiles. However, the complex welds result in reduced fatigue resistance of structural joints of steel circular hollow sections thus higher costs and longer production route for jackets than for monopiles.

In the project WrapNode-II, TU Delft leads research with industrial partners to build crutial knowledge that will enable implementation of the innovative, bonded, wrapped composite joint in jackets for offshore wind turbines. We aim to characterise and further understand full-scale and multi-axial behaviour and influence of offshore environment on durability of the joints. The project is done in collaboration with: Tree Composites,  Enersea, Siemens Gamesa Renevable Energy, AOC Nederland, Shell Global Solution International, Parkwind and Vattenfall.