Open subjects for BSc and MSc theses

In general, any application with a thesis proposal dealing with our research lines (see below) will be considered for support, upon an evaluation of its technical and strategic soundness, and taking into consideration the resources availability (tutors time and infrastructure) within our Department. Applications can be sent directly to the contact person of the working groups for consideration.

Moreover, the following subjects are currently open within our research groups and the research projects of our Department:

Research Group Renewable Energies

M.Sc. Topic I: “The Use of Artificial Intelligence in Energy System Modeling"

Supervisor: Dr. Danial Esmaeili Aliabadi & Dr. Ali Forootani | E-MAIL

The student will investigate the role of AI in energy system modeling. There are two applications of AI models that we are interested in:
  1. Convolution Neural Networks for image processing, and
  2. Large Language Models for natural language processing.
Tasks:
  • Organizing the inputs for the AI model
  • Implementing and fine-tuning the AI model
  • Preparing the dataset/scenario for the optimization model to use

M.Sc. Topic II: “Spatially-explicit modeling of the bioenergy in Germany”

Supervisor: Dr. Danial Esmaeili Aliabadi | E-MAIL

  • Increasing the spatial resolution of the Bioenergy Optimization (BenOpt) model using GIS information.
  • Developing new tools to facilitate the interaction between backend (i.e., the GAMS solver) and frontend (i.e., user-interface).
  • Analyzing consumers’ behavior using an agent-based model in connection with the produced optimization model.

M.Sc. Topic III: “Optimizing Electric Vehicle Charging Station Operations Using Reinforcement Learning with Stochastic Arrival and Departure Dynamics”

Supervisor: Dr. Ali Forootani | E-MAIL

Reinforcement Learning (RL) can be defined as a machine learning paradigm in which an agent learns to make decisions by interacting with an environment in a sequential manner. The goal is for the agent to learn an optimal policy, which dictates the best actions to take in any given state to maximize cumulative rewards over time. In the case of an electric vehicle (EV) charging station, RL can be applied to manage and optimize the allocation of charging resources to incoming vehicles.

In this scenario, the charging station serves as the environment, while each arriving vehicle is treated as a request for service. Vehicles arrive at the charging station stochastically, which can be modeled using a Poisson process to represent random inter-arrival times. The Poisson process is appropriate here, as it is commonly used to model random events occurring independently over time, such as customer arrivals in service systems.

Once a vehicle arrives at the charging station, it stays for a random duration to charge, which can also be modeled as a stochastic process. The vehicle's departure depends on factors such as the remaining battery level and the charging speed. Both the arrival and departure processes of vehicles are influenced by real-world uncertainties, such as traffic patterns, weather, or the driver's schedule, making the problem dynamic and complex.

Reinforcement Learning can help the charging station operator make real-time decisions on various aspects, such as:
1. Allocating charging slots efficiently when multiple vehicles are present.

2. Dynamic pricing to balance demand and incentivize vehicle owners to charge at off-peak times.

3. Managing energy distribution to prevent overloading the grid while minimizing waiting times.

Through trial-and-error interactions, the RL agent (charging station management system) learns from experience how to optimally handle fluctuating demand and supply while adapting to stochastic elements like vehicle arrivals and charging durations. The agent receives feedback in the form of rewards or penalties (e.g., maximizing customer satisfaction by minimizing waiting times and energy costs), thus improving its performance over time.

Python libraries:

Gymnasium: https://github.com/Farama-Foundation/Gymnasium

Stable Baselines3: https://github.com/DLR-RM/stable-baselines3

Pandapower: https://github.com/e2nIEE/pandapower

M.Sc. Topic V: “Wind energy in forests: ownership, structural parameters of forests, regional comparison”

Supervisors: David Manske & Nora Mittelstädt | E-MAIL

  • Analyse forest ownership data, forest properties and wind energy site data
  • Compare regional tendencies

Please note: As we are currently supervising a large number of theses, we will unfortunately not be able to accept new supervision requests until 2025.

B.Sc. Topic VI: “Closest to the energy transition: expansion of renewable energies in (former) lignite mining areas.”

Supervisor: Nora Mittelstädt | E-MAIL

  • Registration of German lignite areas in GIS software and overlap with renewable energy additions and land cover classes
  • Assessment of administrative and political conditions (for example: re-cultivation obligations) for (former) lignite mining areas
  • Qualitative analysis, for example by method of causal-process tracing

Please note: As we are currently supervising a large number of theses, we will unfortunately not be able to accept new supervision requests until 2025.

B.Sc./M.Sc. Topic VII: “Image recognition of wind turbines on aerial photographs with machine learning and deep learning methods”

Supervisor: David Manske | E-MAIL

  • Existing geodata on wind turbines could be incorrect in terms of their localization. To validate these datasets in terms of their geographical location, image recognition can help identify datasets with incorrect geographical positions.
  • Validation of existing geodata with wind turbines using open source aerial imagery
  • Image recognition with Machine Learning/deep learning methods with Python or R

B.Sc./M.Sc. Topic VIII: “How fair is the energy transition?”

Supervisors: David Manske & Nora Mittelstädt | E-MAIL

  • With the energy transition, there are region-specific winners but also losers. But who are they? Which stakeholders/regions benefit from the energy transition and which are left out in the cold?
  • Assessment of local hotspots of the energy transition in conjunction with socio-economic indicators
  • Stakeholder analysis

Please note: As we are currently supervising a large number of theses, we will unfortunately not be able to accept new supervision requests until 2025.

B.Sc. Topic IX: “Regional analyses: Wind energy in forests: ownership, structural parameters of forests”

Supervisors: David Manske & Nora Mittelstädt | E-MAIL

  • Analyse forest ownership data, forest properties and wind energy site data for one selected region

Please note: As we are currently supervising a large number of theses, we will unfortunately not be able to accept new supervision requests until 2025.

Research Group System Analysis of the BioEconomy

M.Sc. Topic I: Development of socio-economic end-point indicators for the bioeconomy in Germany.

Supervisor: Dr. Alberto Bezama | E-MAIL

  • Defining social issues in German waste management of biobased materials / wood
  • Applying the (modified) UNEP/SETAC guidelines to assess social impacts Of recycling
  • Application in a case study: recycling of bio-based materials in Germany
  • Identifying possible aggregation options for the socio-economic indicators to develop end-point indicators

M.Sc. Topic II: Evaluation of selected decision alternatives for internalization or outsourcing of heat utility services in the context of eco-industrial cooperations in bio-based-industries

Supervisor: Dr. Alberto Bezama | E-MAIL

  • Efficiency analysis for selected wood and fiber drying processes
  • Technical-economic pre-feasibility studies of efficiency enhancement alternatives for wood and fiber drying processes
  • Analysis of innovative financing schemes for efficiency enhancement alternatives in complex industrial networks

MSc. /BSc. Topic III: Assessment of woody biomass flows in a regional context – Life cycle inventory and scoping LCA

Supervisor: Dr. Alberto Bezama | E-MAIL