Hans Stenglein

M.Sc. Hans Stenglein

Resilient control design for electrical power systems

+49 (0) 6151 16-21719
fax +49 (0) 6151 16-21712

S3|10 310
Landgraf-Georg-Str. 4
64283 Darmstadt

Research Interest

Research Interest:

  • cyber-physical systems, electrical power systems
  • distributed control
  • sensor placement, structured controller design
  • complex systems, network science
  • resilience

Research Project:

Electrical power systems are a critical infrastructure and their vulnerability is mainly driven by the strong physical coupling of all components. Electrical power system failures lead to serious societal consequences.

Crisis events, such as line outages due to overloading, winter storms or cyber attacks, change the underlying physical network structure. Resilience against physical changes demands for either largely over-sized system components or an adaption of the automated control system.

The research project investigates the complementary couplings in the physical and the cyber (communication + control algorithms) domain of electrical power systems. Considered events are mainly line outages and system splits. The project aims to develop analysis and synthesis methods that enable designers to build medium scale power systems which function on their own, even after a crisis event.

Open theses

Supervisor: Hans Stenglein
Earliest start: immediately
Type: Master Thesis


We ask: How do different MILP formulations describing graph partitions perform when included into network design problems with planning and operational constraints?
The aim of the proposed work is to compare different MILP formulations calculating system splits with constraints and objectives uncommon in the mathematical programming literature, but relevant to power systems (graph partitionings with power systems related constraints). The work should identify the fastest to solve formulation for small example networks.
Further, their usefulness as component of a multilevel optimization program should be evaluated, where resolving with changed right-hand side (upper level decisions) is common.
The work is expected to perform the following tasks:
  1. Basic cut / connectivity model and solution quality
    • Understand and implement multi-component connectivity / max k-way cut MILP formulations from the literature (virtual flows, representatives)
  2. Compare the fit of the basic models to power systems related constraints
    • Analyze resulting model sizes and possible synergies with submodels for power-system-specific objectives and constraints, especially
      • Objectives other than cut size (number of edges removed), using the cut size as a constraint
      • Unknown, but limited number of islands / partitions
  3. Test the solve times and robustness of the formulations against parameter changes on the same graph, i.e. upper-level decisions

The work can be done as a Master Thesis or Project Seminar.

Required Background Knowledge and Skills
  • Graph theory basics
  • Power system basics (operation, dispatch)
  • Good (scientific) programming skills (preferred languages: Julia, Python; GAMS, Matlab is possible)
  • Mixed Integer Linear Programming
  • Attendance in lecture "Energy management & Optimization" could be helpful

Literature
  • Hojny, C., Joormann, I., Lüthen, H., & Schmidt, M. (2021). Mixed-integer programming techniques for the connected max-k-cut problem. Mathematical Programming Computation, 13(1), 75–132.
  • Healy, P., Jozefowiez, N., Laroche, P., Marchetti, F., Martin, S., & Róka, Z. (2024). A branch-and-cut algorithm for the connected max-k-cut problem. European Journal of Operational Research, 312(1), 117–124.
  • Dena-Studie Systemsicherheit 2050. (2020). Deutsche Energie-Agentur.

Short Bio

Hans started his academic journey after graduating from a high school with engineering courses.

He started his studies in information systems engineering with focus on optimization and control. During his Master's degree programme, he specialized in power engineering and embedded systems design. Parallel to his M.Sc. courses, he studied philosophy and history of technology.

  • since 2024: Research Associate at EINS
  • 2023: M.Sc. Informationssystemtechnik (information systems engineering), TU Darmstadt

 

Publications

Minimizing Worst-Case Cyber Graph Reconfigurations in Resilient Cyber-Physical Systems

[Journal]
Andrew Eliseev, Hans Stenglein, Florian Steinke:
Minimizing Worst-Case Cyber Graph Reconfigurations in Resilient Cyber-Physical Systems.
In: IEEE Control Systems Letters 9 , P. 733-738, 2025

Robust and Chance-Constrained Dispatch Policies for Linear Power Systems

[Conference]
Hans Stenglein, Timm Faulwasser, Florian Steinke:
Robust and Chance-Constrained Dispatch Policies for Linear Power Systems.
In: 12thIFAC Symposium on Control of Power and Energy Systems (CPES 2024), Rabat, Morocco, 2024