Type: Bachelor assignment
Educational program: Technical Computer Science
Contact: Gerwin Hoogsteen and Nataly Bañol
Energy management systems (EMSs) play an important role in the context of smart grids and the energy transition, featured by the deployment and ICT based control of several distributed energy resources (DERs) such as photovoltaic systems, wind turbines, energy storage systems, electric vehicles, controllable loads, among others. Both generation and consumption units hold high fluctuating and unpredictable behavior, making it difficult to secure a reliable operation of the electrical grid.
An EMS allows monitoring and controlling DERs interconnected through electrical and communication networks in large or small-scale systems, e.g., distribution grid, buildings, or factories. It employs control algorithms to optimize the grid performance in the near real-time, aiming for a cost-effective and sustainable operation. The future energy grid cannot function without an EMS, as it would lead to both energy system instability and local grid overloading, potentially resulting in service interruptions (local and global blackouts).
However, the addition of EMSs brings up a new vulnerability, where the reliability and operation of the electrical grid may be compromised by technical failures or even cyberattacks targeting communication networks used by EMSs and the EMSs themselves. Therefore, we need innovative control algorithms and backup systems to secure the grid operation when the EMS is unable to respond. In this way, we can discover alternative solutions to cope with system failures/cyberattacks in EMSs and gain insights into the requirements to provide more robust control solutions.
The assignment consists of researching, developing, and validating backup control algorithms and methods able to detect problems (e.g., by measuring the grid state) and react to secure the operation of the electrical grid in case of EMS failures or a cyberattack. These systems are limited to simple communication (single direction) using low bandwidth communication. Furthermore, these events rarely occur (in the best case never), for which machine learning cannot be applied. Hence, the challenge is to develop a system that can respond adequately and rapidly, using limited communication and knowledge, and without direct feedback.
Workload:
- 30% theory
- 30% implementation
- 20% simulating
- 20% writing
For more information, contact Gerwin Hoogsteen and Nataly Bañol:
