Research & Projects

In addition to students there is a vast network of TU Delft researchers active in various energy efficiency projects. Many of these researchers are in the steering board of this platform on smart energy efficiency or have ties to members of the platform. Many of these projects cross the borders of the TU Delft campus and even the borders of the Netherlands. Many projects are namely being executed as European initiatives in resourceful consortia in different countries. Below is an overview of some of these projects in which the TU Delft and platform members are involved.

MSc topic: Big data to improve our understanding of energy use of buildings

Buildings are responsible for about 40% of global energy use. It is surprising how little we know about what all this energy is used for. This is especially the case for buildings in the service sector (offices, schools, hospitals, etc.). We only have anecdotal information of how much of all the energy is used for various categories, like lighting, computers, heating, ventilation and air conditioning. We also do not know how efficient the energy is used.

At the same time, lots of data are gathered, e.g. from smart meters, ICT systems, wifi connectivity, and building management systems. The aim of the master thesis project is to investigate whether we can get a better understanding of energy use in buildings if we combine all these data in a smart way.

The idea is to use data for buildings at Delft University of Technology. Currently, this data is gathered. The research work will include energy analysis of the buildings combined with statistical analysis of the data.

Suitable master programmes are: Computer Science, Sustainable Energy Technology, Industrial Ecology, Complex Systems and Management.

For more information, please contact Kornelis Blok, k.blok@tudelft.nl.

MSc topic: Agent-based modeling of consumer behavior (Afstudeer Atelier)

​Please be invited to the master thesis project atelier on agent-based modeling of consumer behavior. This is set up in coordination to a H2020 project Cheetah, Changing Energy Efficiency Technology Adoption in Households. In this European project, a better understanding of household behavior is sought in order to find ways to increase the energy efficiency of energy appliances in households, it studies which policies work well in what contexts, and adopts a number of modelling approaches, amongst others agent-based modelling.

Assignments are framed in correspondence to the Cheetah project, but they are adapted according to the wishes and strengths of thesis candidates. The atelier runs continuously and there are up to four positions available per year. Please contact Emile Chappin if you're interested. Affinity with and a background in agent-based modeling is strongly recommended. All projects will be state-of-the art and in interactions with project partners. More on Cheetah can be found here.

Possible projects include:

  1. The analysis and comparison of models of the Theory of Planned Behavior on consumer behavior,
  2. New agent-based model analysis techniques applied to consumer technology adoption behavior,
  3. Understanding of intermediary parties and cooperatives and intermediary-specific policies,
  4. The importance of rebound effects through a second hand market for energy-intensive appliances,
  5. Realism of ABM studies on energy efficiency.

Specialisation - Electives package: Advanced Modelling Gaming and Design
SectionE&I
Program: EPA; SEPAM/CoSEM
Methods: Agent-based modelling)

For more information, please contact Emile Chappin

MSc topic: HVAC diagnostics benchmark

Researchers in the area of hybrid systems are often looking for benchmarks where to test their methodologies. Heating, ventilating and air conditioning (HVAC) systems offer this opportunity, because thermostatic behaviour and faults give rise to interesting hybrid dynamics. This assignment starts from the experience of DCSC in modelling HVAC system. You are required to develop a simulator of a multiple-room HVAC system, that can be made freely available (e.g. on GitHub) to the scientific community.

Material provided: report and scientific papers on the topic, initial Matlab software
Expected output: Matlab model, user's guide for the software
Requirements: neat programmer, advanced Matlab programming skills
Department/Faculty: Delft Center for Systems & Control at ME  

For more information, please contact Simone Baldi, S.Baldi@tudelft.nl