Beschrijving

The research (and teaching) activities of KU Leuven/ESAT/ELECTA cover a wide spectrum from power systems over power electronics to control and socio-economic issues. The main research topics include power system operation and control; distributed control and optimisation; power electronics; energy markets; and multi-scale, multi-physics system modelling. Since more than fifteen years, smart grids are a focal point, where we look into the power grid infrastructure, ICT infrastructure, smart grid applications, algorithms, business processes and stakeholders. As smart grids are about managing flexibility, this makes ELECTA's experience and knowledge an indispensable input for this project.
The research group has been active since the 1960s and currently has 6 full-time professors, 6 part-time professors, 8 post-doctoral and senior researchers, 13 administrative and technical persons and 45+ PhD students. ELECTA is a founding member of the Energy Institute (www.kuleuven.be/ei) (1997), which is recognised as a KU Leuven centre of excellence. ELECTA is also one of the driving forces behind EnergyVille (www.energyville.be), which unites KU Leuven, VITO, imec and UHasselt for research on sustainable energy and intelligent energy systems. Since 2016 EnergyVille has its research facilities Genk, where it hosts advanced smart grid laboratories. Within this project, KU Leuven has full access to these EnergyVille facilities. Finally, ELECTA is one of the key partners in InnoEnergy (EIT – Knowledge and Innovation Community) with focus on Intelligent Energy Efficient Buildings and Cities.

This PhD research is part of the project “Control algorithms for flexibility in power-to-X and industrial processes (InduFlexControl-2)”, the general objective of which is to develop data-driven modelling techniques and control methodologies, harness flexibility in the energy-intensive industry and enable its active participation in the energy transition while remaining cost-efficient and low in CO2 emissions.  

The PhD focus will be consider/develop and compare white-box models (purely based on physical principles, governing equations), black-box models (purely based on data) and grey-box models (combining physical principles and data) to help redesigning and improving energy-intensive processes (reduced cost and more environmentally friendly), with particular attention to energy conversion (possibly between unconventional energy carriers, so-called power-to-X) and energy storage technologies. 

Fully-fledge physical models (e.g., finite elements) are parametrizable and provide the most accurate solution at the expense of a high computational cost and are thus unfeasible for real-time control, or optimisation loops. Applying reduced-order model techniques has proved an efficient workaround, but there is still room for improvement, particularly in case of nonlinearities and multiple timescales. 

Data-driven models allow identifying both linear and nonlinear systems from data. Most machine learning techniques are data-driven, blind the governing physics but still able to predict accurate results if enough data are available, they are good interpolators. However, when highly nonlinear dynamics are present in the system, the research problem remains wide open. We need techniques that allow for reliable extrapolation.

The current tendency is to combine both approaches, i.e., to provide the physical governing equations as a constraint to the data-driven models and take advantage of all the available information. 

The modelling effort at the component/system-level must result in robust, reliable, modular, scalable, and computationally cheap algorithms suitable to be embedded in e.g., an optimisation framework or characterise a complex energy system. 

Profiel

You are a highly motivated and enthusiastic researcher with a strong interest in the development of numerical simulation methods. You have:

  •  A master’s degree in electromechanical/energy engineering OR electrical engineering or mechanical engineering OR mathematical engineering OR computer science OR data science.
  • Critical and creative thinking skills. 
  • Independence and initiative.
  • Reliability in teamwork and good communication skills.
  • Excellent proficiency in English (speaking, listening and writing). 

Please indicate clearly in the motivation letter your research interests and your previous experience and knowledge. 

 

  • A four-year doctoral scholarship and a PhD in Engineering Science, if successful.
  • The opportunity to participate in exciting scientific projects carried out within ELECTA and EnergyVille research centre. EnergyVille actively collaborates with industry, helping you to expand your professional network.
  • Access to state-of-the-art technologies, core facilities and technical support
  • The opportunity to be part of in an exciting and international research environment, engage in research collaborations and participate at international conferences
  • A flexible working culture with opportunity to up to 40% remote working.
  • Benefits such as health insurance, access to university sports facilities, etc.

Locatie

Leuven

Regime

Voltijds